U.S. FOOD AND DRUG ADMINISTRATION

+ + + + +

NONPRESCRIPTION DRUGS ADVISORY COMMITTEE (NDAC)

WITH THE

GASTROINTESTINAL DRUGS ADVISORY COMMITTEE (EDAC)

+ + + + +

JOINT PUBLIC MEETING

+ + + + +

FRIDAY,

JUNE 21, 2002

+ + + + +

BETHESDA, MARYLAND

+ + + + +

The Joint Meeting commenced at 8:00 a.m. in the Versailles Rooms I and II at the Holiday Inn, 8120 Wisconsin Avenue, Bethesda, Maryland, Louis R. Cantilena, Jr., M.D., Chairman, NDAC, presiding.

Michael Camilleri, M.D., Member

Susan Cohen, Consumer Representative

Byron Cryer, M.D., Member

Ronald P. Fogel, M.D., Member

Nancy L. Geller, Ph.D., Member

George S. Goldstein, M.D., Guest Industry Rep.

John T. LaMont, M.D., Member

Robert A. Levine, M.D., Member

Eric P. Brass, M.D., Ph.D.

Edwin E. Gilliam, Ph.D.

Richard A. Neill, M.D.

Michael C. Alfano, D.M.D., Ph.D.

Jonca Bull

Charles Ganley, M.D.

Florence Houn

Victor Raczkowski, M.D.

Sandra Titus, Ph.D.

John A. Gans, Pharm.D., American Pharmaceutical Association

Linda Golodner, National Consumers League

Robert M. Niecestro, Adrix Labs

Susan Winckler, American Pharmaceutical Association

A-G-E-N-D-A

Louis Cantilena, M.D., Ph.D., NDAC Chair 5

Sandra Titus, Ph.D., Executive Secretary, NDAC 8

Victor Raczkowski, M.D., Deputy Director, ODE II & Acting Director, Division of Gastrointestinal and Coagulation Drug Products 11

Open Public Hearing 14

Keith C. Triebwasser, Ph.D., Senior Director, Regulatory Affairs, Procter and Gamble 44

David Peura, M.D., Associate Chief of Gastroenterology, University of Virginia 50

Douglas Ws. Bierer, Ph.D., Director OTC Drug Development, Procter and Gamble 56

Douglas Levine, M.D., Chief Medical Officer, Gastrointestinal Therapeutic Area,

AstraZeneca LP 73

Nora Zorich, M.D., Ph.D., Vice President Pharmaceuticals, Procter & Gamble 78

Keith C. Triebwasser, Ph.D. 91

Gamble 94

Mark Avigan, M.D., Medical Officer, Division of Gastrointestinal and Coagulation Drug Products 131

Karen Lechter, J.D., Ph.D., Office of Drug

Safety 144

Daiva Shetty, M.D., Medical Officer, Division

of Over-the-Counter Drugs 154

Questions from the Committee to the FDA 165

Linda Katz, M.D., MPH, Deputy Director,

Division of Over-the-Counter Drugs 207

Adjourn

P-R-O-C-E-E-D-I-N-G-S

(8:06 a.m.)

DR. CANTILENA: Good morning everyone and welcome to the Nonprescription Drug Advisory and the GI Drug Advisory Committee. I'm Dr. Lou Cantilena, Chief of Clinical Pharmacology at the Uniform Inservices University here in Bethesda. I'll be chairing this meeting.

We will start off with the conflict of interest statement. Actually we'll start off by going around and introducing ourselves. I have already done that so if we can start at this end over here and introduce this way.

DR. GOLDSTEIN: I'm George Goldstein. I'm the industry representative for the Advisory Committee.

DR. ALFANO: I'm Michael Alfano, Dean of Dental School at NYU and the ILR at the OTC.

DR. JOHNSON: My name is Julie Johnson. I'm from the University of Florida and I'm a member of the Nonprescription Drug Committee.

DR. CRYER: Byron Cryer, member of the Gastrointestinal Drugs Advisory Committee. I'm a gastroenterologist from the University of Texas Southwestern Medical School, Dallas.

DR. BROWN: I'm Eric Brown from Harvard UCLA Medical Center, Department of Medicine. I'm a consultant to the committee.

DR. CAMILLERI: I'm Mike Camilleri, gastroenterologist from the Mayo Clinic, Rochester. I'm a member of the Gastrointestinal Drugs Advisory Committee.

DR. FOGEL: I'm Ron Fogel, Division Head of Gastroenterology, Henry Ford Health System, and I'm a member of the GI Drug Advisory Committee

DR. WILLIAMS: I'm Henry Williams from Howard University and a member of NDAC.

DR. UDEN: I'm Don Uden from the University of Minnesota College and member of NDAC.

DR. GELLER: I'm Nancy Geller. I'm the Director of the Office of Vital Statistics Research at the National Heart, Lung, and Blood Institute and I'm a member of the GI Advisory Committee.

DR. CLAPP: I'm Leslie Clapp, pediatrician, Buffalo, New York, Associate Professor of Pediatrics at SUNY UB. Also a member of NDAC.

DR. TITUS: I'm Sandy Titus. I'm the Executive Secretary to NDAC.

DR. NEILL: I'm Richard Neill. I'm a family physician and consultant to NDAC.

MS. COHEN: (Inaudible).

DR. GILLIAM: I'm Eddie Gilliam. I'm a family nurse practitioner from Tucson, Arizona and a member of NDAC.

DR. LEVINE: I'm Bob Levine from State University of New York, Upstate Medical University in Syracuse. I'm a gastroenterologist and member of the GI Advisory Board.

DR. LAM: I'm Francis Lam. I'm a member of the NDAC committee. I'm from the University of Texas in San Antonio.

DR. PATTEN: I'm Sonia Patten. I'm an anthropologist on the faculty at McAllister College. I'm from Minneapolis, Minnesota, and I'm a consumer representative on NDAC.

DR. LaMONT: I'm Tom LaMont. I'm a gastroenterologist, Chief of the Division of Gastroenterology at the Medical Center in Boston and a faculty member at Harvard Medical School.

DR. DAVIDOFF: I'm Frank Davidoff. I'm an internist and former editor of the Anals of Internal Medicine and I'm a member of NDAC.

DR. KATZ: I'm Linda Katz, Deputy Director of the Division of Over-the-Counter Drug Products of the FDA.

DR. GANLEY: I'm Charlie Ganley, Director of Division of Over-the-Counter Drugs at FDA.

MS. BULL: Good morning. Jonca Bull, Office Director, Office of Drug Evaluation and Center for Drug Evaluation Research.

DR. RACZKOWSKI: Good morning. I'm Victor Raczkowski. I'm the Acting Director of the Division of Gastrointestinal and Coagulation Drug Products.

DR. HOUN: I'm Florence Houn, Director of Office of Drug Evaluation Three and FDA.

DR. CANTILENA: Thank you. Now Dr. Titus will go through the conflict of interest statement for the June 21st Meeting of Nonprescription Drugs and Gastrointestinal Drugs Advisory Committees

DR. TITUS: The following announcement addresses conflict of interest issues associated with this meeting and is made a part of the record to preclude even the appearance of such at this meeting.

Based on the submitted agenda for the meting and all financial interests reported by the Committee participants, it has been determined that all interests in firms regulated by the Center for Drug Evaluation and Research which have been reported by the participants present no potential for an appearance of a conflict of interest at this meeting with the following exceptions.

Ft. Michael Wolfe is excluded from participating in today's discussion and vote concerning Prilosec 1.

Dr. Eric Brass has been granted a waiver under 18 U.S.C. 208(b)(3) for unrelated consulting with competitors. He receives less than $10,000 a year from two of the firms and between $10,001 and $50,000 per year from the third firm.

Dr. Michael Camilleri has been grated a waiver under 18 U.S.C. 208(b)(3) for his participation as a consultant on unrelated matters for five firms that have financial interests in competing products. He receives less than $10,001 a year from each firm.

Susan Cohen has been grated waivers under 18 U.S.C. 208(b)(3) and 21 U.S.C. 355(n)(4) amendment of Section 505 of the Food an Drug Administration Modernization Act, for ownership of stock in competitors. The first two stocks in competitors are valued between $5,001 and $25,000. The other two stock holdings are valued between $25,001 and $50,000.

Dr. Byron Cryer has been granted waivers under 18 U.S.C. 208(b)(3) and 21 U.S.C. 355(n)(4) amendment of Section 505 of the Food and Drug Administration Modernization Act, for shares of stock in the manufacturer of the product at issue and a competitor; and for consulting on unrelated matters for a competitor.

The stock in the manufacturer of the product at issue and a competitor is valued at less than $5,001. The unrelated consulting for a competitor is valued between $10,001 to $50,000.

Dr. Ronald Fogel has been granted a waiver under 18 U.S.C. 208(b)(3) and 21 U.S.C. 355(n)(4) amendment of Section 505 of the Food and Drug Administration Modernization Act, for shares in a sector mutual fund valued between $5,001 and $25,000.

Dr. Robert Levine has been granted waivers under 18 U.S.C. 208(b)(3) and 21 U.S.C. 355(n)(4) amendment of Section 505 of the Food and Drug Administration Modernization Act, for shares of stock in a competitor valued between $25,001 and $50,000.

A copy of these waiver statements may be obtained by submitting a written request to the Agency's Freedom of Information Office, Room 12A-30 of the Parklawn Building.

We would like to note for the record that Michael Alfano, Ph.D., and George Goldstein, M.D., are participating in this meeting as industry representatives, acting on behalf of regulated industry. As such, these participants have not been screened for any conflicts of interest.

In the event that the discussions involve any other products of firms not already on the agenda for which an FDA participant has a financial interest, the participants are aware of the need to exclude themselves from such involvement and their exclusion will be noted for the record.

With respect to all other participants, we ask in the interest of fairness that they address any current or previous financial involvement with any firm whose products they may wish to comment upon.

DR. CANTILENA: Thank you, Dr. Titus.

We will now hear from Dr. Raczkowski from the FDA to open up the issues for discussion.

DR. RACZKOWSKI: Good morning. My name is Dr. Victor Raczkowski and I'm the Acting Director of the Division of Gastrointestinal and Coagulation Drug Products. On behalf of FDA I would like to welcome members of the Nonprescription Drugs Advisory Committee, the Gastrointestinal Drugs Advisory Committee, as well as members of the public

I would like to briefly set the stage for today's deliberations for the Joint Advisory Committee Meeting. This is the second time that this joint committee is meeting to discuss whether the data are sufficient to recommend approval for Prilosec 1, or omeprozole magnesium, for use in the over-the-counter setting.

The first time this joint committee met on this issue was October 20, 2000. At this time different over-the-counter uses are being sought by the sponsor, Procter and Gamble and AstraZeneca.

Today the new use that will be considered by the Joint Advisory Committee is for the prevention of the symptoms of frequent heartburn for 24 hours. This proposed use indicates that Prilosec 1 is only for those who suffer heart burn two or more days a week.

The directions for use proposed by the sponsor is that consumers swallow one tablet which is equivalent to 20 milligrams of omeprozole with a glass of water every morning and to take one tablet every day for 14 days.

This new use and a new direction proposed for Prilosec 1 by the company reflect the underlying properties of Prilosec 1. Unlike the histamine H₂-receptor antagonists or antacids which can be used to treat heartburn, for example, the ability of Prilosec 1 to inhibit gastric acid secretion has a delay in onset and requires several days of continued treatment to build up to its maximum effect.

In other words, this new use proposed by Procter and Gamble and AstraZeneca means that the drug is not to be used to treat acute symptoms of heartburn or to prevent meal induced heartburn. Rather, it is to prevent frequent heartburn.

Moreover, these directions mean that consumers should not take Prilosec 1 episodically missing doses or skipping doses. Rather, consumers should take Prilosec 1 daily for 14 days.

In support of these new uses, the sponsor has conducted new labeling comprehension studies and has conducted new actual use studies and we'll be hearing more about these later this morning.

In contrast, the sponsor has not provided additional efficacy, safety, or pharmacokinetic or pharmacodynamic data. Rather, the studies on efficacy, for example, that were used in the previous submission are being used to support this newly proposed use.

Today we have many issues we are asking the Advisory Committee to consider. One issue is that one of the cardinal manifestations of gastro esophageal reflux disease, or GERD, is frequent heartburn and, as I've indicated, the sponsor is taking a new use for the prevention of frequent heartburn in the over-the-counter setting.

One may believe, for example, and this is open for discussion by the Advisory Committee members, that this disease, GERD, requires diagnosis and monitoring and management by a healthcare professional such as a physician. If so, then consumers with this disease of GERD have to make a choice about whether to see a physician or whether to self-administer Prilosec 1. We will be seeking the Advisory Committee's advice on whether this is an issue or not.

Additional issues that will be discussed by the Advisory Committee are whether consumers can appropriately self-select to use Prilosec 1. A third is whether or not consumers use the drug appropriately. The fourth issue is whether 14 days is an appropriate treatment duration. Finally, we will be asking the committee about its recommendations for approvability of Prilosec 1.

We have a full agenda today and we look forward to your deliberations. Thank you very much.

DR. CANTILENA: Thank you very much for those comments.

We will now move into the open public hearing. For the open public hearing we have four speakers. I would like to remind the speakers that prior to starting, if they have any conflicts of interest those should be stated for the committee.

Also, each speaker has five minutes for his or her talk. The first speaker, Linda Golodner from the National Consumer's League in Washington.

Linda.

MS. GOLODNER: Thank you. My name is Linda Golodner. I'm President of the National Consumers League. America's oldest consumer advocacy organization is pleased to testify today before the committees on the possible switch of Prilosec to nonprescription status.

NCL has a long history of providing information and educational materials to consumers so they can safely and effectively use medications, both prescription and nonprescription.

I would like to inform the committee that occasionally the League receives financial support from pharmaceutical companies for specific consumer education projects in which we maintain full editorial control.

In addition pharmaceutical companies have supported our annual dinner and conferences. These contributions amount to less than one half of 1 percent of our annual operating budget.

I personally do not have stock in or consult with any pharmaceutical companies. The National Consumer League did not receive any financial incentive to appear at this meeting.

Many of the 60 million Americans who suffer from heartburn are not satisfied with current OTC medications available. Making Prilosec, a proton pump inhibitor, available without a prescription would provide more options when self-treating heartburn symptoms. Heartburn is one of those conditions that when you have it, you know you have it. Therefore, we feel that consumers can adequately self-diagnose this condition.

If this medication were available OTC, consumers would be able to effectively prevent frequent heartburn with Prilosec without having the trouble of going to the doctor or having the expense of going to a physician.

Consumers today are taking a more active role in their healthcare including self-diagnosing and self-medicating. NCL is working to help consumers understand what medications they are taking, why they are taking them, and how to take them effectively.

Because of this trend in self-medication any medication slips from prescription to nonprescription status must first be found to be safe by the FDA by this committee and reported to the FDA and, if allowed on the market, there should be clear understandable information available at point of sale and through advertising for consumers to use that product appropriately and safely.

In order to better understand what consumers know about OTC medications and how they are using them, the National Consumers League commissioned a survey on consumer's use and attitudes regarding OTCs.

According to survey, consumers generally have a favorable impression of OTC drugs and use them regularly to treat minor health conditions. But one-third of consumers do not regularly read the labels of OTC products before purchasing or using them. That includes all OTC products that they are taking.

One-quarter of those surveyed had some trouble reading and understanding the label. Another one-third of the consumers reported taking more than the recommended dose some or most of the time, while more than one in five consumers take OTC medicines for longer than recommended.

These survey results underscore the need to use clear, good-size type on the labels and that specific directions for dose and how long to take the medicine be emphasized on the package, on the label, and in any promotional material.

A recent survey by NCPIE, a patient education advocacy group of which NCL is on the board, also found that consumers need to be better informed about using OTC medications appropriately. The survey found that 95 percent of the consumers read some portion of the label but they do so selectively.

When buying an OTC product the first time only a third look at the active ingredient and one in five seek out warning information. Over a third of the consumers combine nonprescription medications when they have multiple symptoms.

On a positive note the survey also found the majority of consumers get their information about OTC medications from their health professionals, and that the health professionals were very willing to discuss OTC medication use with their patients.

What is clear from these surveys is that consumers need to be better informed when using OTC products, but also that the involvement of health practitioners could increase consumer understanding of OTC use. Therefore, I would hope that there would be efforts by the manufacture to encourage healthcare professionals to educate consumers in the use of an OTC Prilosec.

If the FDA determines that Prilosec, in 20 milligram doses for 14 days, can be taken safely by consumers with a prescription, we recommend that there be appropriate labeling on the medication to ensure proper use by consumers, including clear information about who should not be taking the medication, especially those who take drugs that would interact with Prilosec.

There must be clear label directions on how to take it, specific listing of warning symptoms of when consumer should go to the doctor. NCL wants to ensure that consumers seek medical attention if the recommended Prilosec regimen does not relieve their heartburn or if they experience certain symptoms.

Clear label warnings and information should help prevent consumers from delaying seeking necessary medical attention. Of course, the label should also list possible side effects and encourage consumers to continue to have regular physician visits while taking Prilosec. Consumer should also be instructed to inform their physician that they are taking Prilosec, and to contact their physician or pharmacist if they have any questions about the medication.

NCL recognizes that pharmaceuticals are an important component in assuring good health, however, whether by prescription or over-the-counter they must be taken seriously. Just because a drug is available on the shelf of a grocery store or discount store does not make it more safe than if a doctor prescribed this drug.

Physicians, nurses, physicians assistants, pharmacists, and other healthcare professionals play a vital role in ensuring that consumers use their medications effectively. Thank you.

DR. CANTILENA: Thank you very much. Our next speaker is Dr. Niecestro from Adrix Labs.

Dr. Niecestro.

DR. NIECESTRO: Good morning. I am Dr. Robert Niecestro, Senior Executive Director of Clinical Research for Adrix Laboratory. As most of you are aware, we have an FDA approved generic version of omeprazole. Prior to that I played a pivotal role in the submission and approval of rebeprazole, another proton pump inhibitor currently marketed in the United States.

I came before the committee because there are two issues that need to be addressed by this committee. These issues are as follows. What are the effects of food on OTC omeprazole, and are potential interactions with other acid reducers understood well enough to allow for the safe and effective use of OTC omeprazole?

The proposed dozing label that was put up on the web yesterday is inadequate for instructing consumers on how or when to take omeprazole in relationship to food.

The lack of adequate information on food is nothing new to omeprazole. This was noted during the initial review of the omeprazole NDA in 1988. At that time the Food and Drug Administration requested that a definitive drug/food interaction study under fasted versus feed conditions be completed.

I have reviewed the published literature and I have concluded that this FDA requested definitive drug/food interaction study has either not been done and/or reported.

My first question to the committee is has this definitive drug/food interaction study requested in 1988 by the FDA been completed? And how do these results impact the label for OTC omeprazole and the instructions given to consumers? More importantly, even under the direction of the physician, there is confusion on how to take prescription omeprazole with food.

Recently, and I have provided this reference in the briefing book, Gunaratnamm et al., have concluded that over 50 percent of patients taking proton pump inhibitors in a community setting were taking them incorrectly due to insufficient data on proper administration of proton pump inhibitors in relationship to food.

It is their opinion that patients have developed inappropriate dosing habits which have led to ineffective symptom control and inappropriate dose escalation.

My next question for the committee is as follows. Since current labeling for prescription omeprazole does not adequately instruct physicians, how can we expect consumers to take an OTC preparation of omeprazole safely?

Now I would also like to address some of the other label. Although the proposed label states, "Do not use with other acid reducers," it is important for this committee to remember that OTC omeprazole has been proven no better than placebo for key treatment of heartburn.

In clinical trials with OTC omeprazole subjects did indeed use other acid reducers. There included antacids, H₂ blockers, and Rx proton pump inhibitors. Thus, I believe it is critical that the interaction of antacids, H₂ blockers, and other proton pump inhibitors be well defined and characterized with OTC omeprazole.

I'd like for the committee to know that the co-administration of antacids would alter both the disillusion rate and pharmacokinetics of omeprazole. More importantly, there is conflicting information available in the published literature on how to administer omeprazole with antacids.

The co-administration of antacids with omeprazole may not be relevant under the care of a physician but it is extremely relevant when consumers are self-administering OTC omeprazole.

More importantly, how would patients be instructed to take OTC omeprazole with antacids given the fact that there is conflicting information available from two independent studies sponsored by AstraZeneca and reported in the summary basis of approval for omeprazole.

I would like to remind the committee that in these two studies co-administration of antacids with omeprazole in one study increased by availability but in the second study decreased it.

They sent another review of the published literature. There are no definitive datas on the effect of H₂ blockers on the disillusion rate, pharmacokinetics and pharmacodynamics of omeprazole. We have seen in clinical trials for OTC omeprazole that these agents are indeed taken and there is a great probability that more consumers will take it when they are self-medicating.

Furthermore, patients have switched from one PPI to another under the care of physicians. However, patients under the care of physicians usually do not take two proton pump inhibitors together as seen in the clinical trials conducted by the sponsor.

When two proton pump inhibitors are taken together, can they have additive or synergistic effects on the retina, thyroid, and what would be the effects on the proton pumps found in the kidneys?

In conclusion, since we do not know and fully understand drug/food interaction, drug/antacid interactions, and interactions with H₂ blockers and other proton pump inhibitors, I would like to recommend to the committee and the FDA that they not approve OTC omeprazole until these public safety issues have been fully addressed and consumers can be properly instructed. Thank you very much.

DR. CANTILENA: Okay. Actually, we have a question for you, Dr. Niecestro. As you are heading back to the podium, I have a question for the FDA.

Is there a drug/food interaction study on file for the drug by anybody?

DR. RACZKOWSKI: I would like to introduce Dr. Suleiman Al-Fayoumi from the Food and Drug Administration. He's better pharmaceutics reviewer.

DR. CANTILENA: Thank you.

DR. AL-FAYOUMI: I would just like to note that the sponsor has submitted as part of their NDA application to the omeprazole magnesium OTC product a food effects study to evaluate the effect of food on the pharmacokinetics of omeprazole and magnesium tablets and there is significant food effect. We are probably going to recommend it be administered an hour before meals.

DR. CANTILENA: Okay. Thank you. So that is on file. Then we had a question from the committee regarding what your company has done to improve the use of Rx omeprazole.

DR. NIECESTRO: That information is confidential and I wish not to disclose it at a public meeting.

DR. CANTILENA: Okay. And some of the interactions that you talked about with sort of the other heartburn drugs would also obviously apply to the Rx omeprazole. Is that true?

DR. NIECESTRO: That is correct, sir.

DR. CANTILENA: Okay. Thank you very much.

Our next speaker is Dr. Gans. I'm sorry. There's been a change. Dr. Susan Winckler from the American Pharmaceutical Association.

DR. WINCKLER: Good morning. Thank you for the opportunity to present the views of the American Pharmaceutical Association, the National Professional Society of Pharmacists.

I am Susan Winckler, a pharmacist and an attorney, and Vice President for Policy and Communications with APhA. We are pleased with the opportunity to be here this morning.

My comments will focus on the role of the pharmacist in helping consumers navigate the use of omeprazole in the over-the-counter environment should the agency choose to approve such availability.

In the interest of full disclosure, APhA frequently partners with Federal agencies, consumer groups, the pharmaceutical industry, and others to develop educational programs for pharmacists and consumers. The Association did not receive funding to participate in today's meeting, and the views I am presenting are solely those of the Association and its membership.

APhA represents pharmacists in all practice settings and in each of those settings we help consumers manage and improve their medication use including the appropriate selection and monitoring of prescription and OTC products.

APhA supports the transition of suitable prescription drug products from nonprescription status when supported by studies assessing the safety, efficacy, and appropriateness of such drug products for OTC use.

In the questions before the committee today this proton pump inhibitor is being considered for OTC use for the prevention of frequent heartburn. Omeprazole magnesium would be the first proton pump inhibitor to be available without a prescription.

This switch may improve clinical outcomes by expanding consumer access to a drug therapy class that is considered more effective in preventing heartburn than alternative therapies such as histamine H₂- receptor antagonists.

To determine if this product should be switched from prescription to OTC status, we urge you to consider a review of all existing therapies in the self-care market. If existing options for self-care raise questions of safety or effectiveness, the relative safety of the switch candidate increases and the risk-benefit analysis shifts in favor of OTC availability.

Decisions to classify products as either prescription or nonprescription should be based on substantial evidence of safety and efficacy in actual OTC settings. The use of the drug product in the actual OTC setting is especially important in the real-world setting of self-care.

The number of products moving from prescription to OTC status has increased markedly over the past several years, and consumers are increasingly making decisions regarding the self-diagnosis and treatment of health conditions. This is a positive trend. A challenge of this trend, however, is equipping consumers with information to help them select and use those products appropriately. This is an area where pharmacists can help.

As pharmacists we are in the ideal position to help consumers select an OTC medication when appropriate and help monitor use of the product. Pharmacists can and do play and role in help consumers use OTC products for the prevention and treatment of frequent heartburn.

We educate patients about heartburn and more serious conditions such as GERD, ensure that patients are appropriate self-treatment candidates, assist patients with appropriate products selection, and refer patients with symptoms that may suggest a serious condition to a physician.

We also work with patients to ensure that they understand how to use the product, how often to take the medication, what dose and for what duration, and can suggest lifestyle modifications to help lessen the occurrence and severity of symptoms.

The dynamics of the same medication potentially being available for one indication in the OTC environment and other indications in a prescription environment will be challenging. The challenge, however, is not new.

Histamine H₂-receptor antagonists have been available for years in both prescription and nonprescription form. As I described earlier, pharmacists assist patients in deciding whether they should use a nonprescription product for short-term relief of heartburn or whether a consultation with a physician is warranted to determine of a more serious chronic condition exist.

Our ability to manage the use of a product like omeprazole in a dual prescription and nonprescription environment will be directly related to the amount of information available to pharmacists.

The product sponsor must provide product labeling that clearly delineates when OTC use of the product is appropriate and directs consumers to a healthcare professional when use of the product falls outside of labeled parameters.

Additionally, an educational campaign to equip pharmacists with the proper tools to identify and select OTC therapies for frequent heartburn will be needed.

While most OTC products are purchased at the pharmacy, a recent survey showed that mass marketers such as supermarkets and discount stores without pharmacies are gaining a larger share of the OTC market. In these environments consumers make OTC decisions without the assistance of a healthcare professional. The lack of access to a pharmacists or physician places even greater responsibility on the consumer for interpretation and understanding of drug labeling and appropriate use of medications.

Two types of studies are particularly valuable in determining whether there is sufficient evidence to reclassify prescription products to OTC status, OTC label comprehension and actual use studies. It is out understanding that the product sponsors have conducted several label comprehension studies since the agency first considered the switch in October 2000 and adjusted the labeling accordingly.

In conclusion, APhA recommends that the agency consider the real world use of omeprazole magnesium in the OTC environment, existing OTC products available for heartburn, the risks and benefits of increasing access to the product, and the ability of consumers to appropriately select and use the product without a learned intermediary. APhA supports the transition of this product to OTC status pending the outcome of this review by the FDA.

Thank you for the opportunity to present the views of the nation's pharmacists.

DR. CANTILENA: Okay. Thank you very much.

Our next speaker is Dr. Michael Wolfe, Boston Medical Center. Prior to his talk Dr. Titus has a conflict of interest statement that she will read.

DR. TITUS: During the opening statement that addressed conflict of interest it was announced that Dr. Michael Wolfe was recused from participating in the meeting as a federal employee. He has asked to participate in the open public hearing as a member of the public.

We made the decision that he could participate in the open public hearing because we feel that the advisory committee members and the public are entitled to hear all sides of an issue and we do not want to be seen as suppressing or censoring any perspective.

The Advisory Committee Members and the public should take into account the recusal by FDA in evaluating his comments. Further, as we do for all participants in the open public hearing, Dr. Wolfe has been asked to disclose any financial interests he may have in the matter before the committee.

We specifically advise all open public hearing participants that it is important to disclose financial relationships such as being an investigator, consulting, and stock ownership with the sponsor and with any of the competitors. Thank you.

DR. WOLFE: Thank you for the opportunity to speak. As Dr. Titus said, I am actually Chair of the Advisory Board for GI Drugs and recused because I have invented what is considered a competing agent for treating episodic heartburn. I'm the inventor. I'm not the owner. It's been licensed. I have no say in what happens with this but I do receive royalties for this.

I'm presenting Professor of Medicine at Boston University School of Medicine and Chief of the Section of Gastroenterology at Boston University and have worked there for the last 23 years.

I'm going to present data primarily on some work we have recently completed in my laboratory which has been submitted for publication. I received permission from the Journal to present this information without jeopardizing a chance of being accepted.

Before I do so, I just want to briefly mention and just follow up on some of the questions about appropriate use. I'm not sure if you understand why appropriate use. Just very briefly, PPIs are designer drugs which are designed specifically as pro-drugs and require activation. That activation occurs after eating a meal.

The food situation is quite significant and there are studies in H₂ blockers in several animal species showing if you use an H₂ blocker and PPI at the same time, PPI doesn't work at all.

Now, I just also want to show you real world use. This is a study that we published. Only one slide is published in the American Journal of Medicine October 15, 2001. This is after several years of frustration of having patients who failed PPI therapy and decided to do a study to see why they did.

We surveyed physician to see how they use the drugs. The results are actually kind of astounding. Despite all the package inserts, instructions and numerous lectures, the fact of the matter is these drugs are designed to be taken before breakfast, before the first meal of the day.

In fact, nongastroenterologists prescribe it before breakfast in less than 30 percent of the cases. It is unlikely, in my view, that nonphysicians, consumers, would do a better job than physicians with 13 years of experience with this drug.

Gastroenterologists did better but still 3 percent incorrectly prescribed these drugs before bedtime and said it didn't matter. It does matter and all studies have been done where the drug is used very specifically.

Now I'll move on to what I'm supposed to be discussing and that is safety issues. Safety issue specifically is that PPIs are extremely potent agents, especially with certain ethnic groups which have not been studied very extensively.

For example, in Asian populations these drugs in 30 percent of the population can inhibit acid secretion for days with one dose. That hasn't been studied really.

What happens when you eat a meal? This is an introduction to what we'll be talking about, when we eat a meal the different phases of acid secretion and the two important phases of the gastric phase when food actually enters the stomach. There are three primary aspects for acid secretion. Part of the stomach extends which causes the transrelease of acid.

The protein content also intragastric pH. The pH goes up when we eat a meal. That is very important because that causes the release of gastrin. pH is clearly the most important of the three when gastrin is released and that accounts for 92 percent of the response during the gastric phase.

This causes increased acid secretion which eventually after the buffering capacity is diminished, pH goes down. We have release of another hormone and gastric release is turned off. The classic physiological negative feedback loop. That's normally how all our bodies function.

If we turn off acid secretion, we have what is called a vicious cycle. Gastric release goes up and continues to go up in certain individuals. Is this a problem in the acute phase for 10 days? No. The fact of the matter is that people are used to using PPI for the last 13 years they read the label and it says 10 days.

In the real world what people end of doing is taking it and ignoring it and not reading the label. They will take it continuously. What will happen in this situation is 27 percent, one in four people, have elevated serum gastrin levels.

But why worry? You can't make acid. That's correct but gastrin does have other properties. This is a patient, a fairly typical patient. We'll see this fairly commonly. This is a patient who has these multiple little bumps, little polyps.

Initially when omeprazole was first presented to the FDA and actually presented for use around the world, there was a lot of concern because of these little bumps, these little polyps, irritants. As a result, actually omeprazole carried a warning until 1995 because of these. Now some years later they are saying forget the warning and we'll put it over the counter.

These bumps actually never concerned me because these are benign tumors both in humans and in rodents which go away with cessation of therapy. There are other issues. Gastrin is a trophic hormone. It is trophic embryologically and during adult life as well with abnormal states.

What people have done in the past is they have looked at serum gastric levels and tried to correlate serum gastric levels in 100 patients to see if there is a correlation. Those studies don't really cut it. You need a large population. This is the largest they ever published, over 100,000 individuals. I'm going to quote to you the conclusions.

"...a gastrin level above normal was associated with increased risk for colorectal malignancy (odds ratio, 3.9; 95% confidence 1.5-9.8). If this association is causal, 8.6% of colorectal cancers could be attributed to high serum gastrin level. Conclusion: Hypergastrinemia is associated with an increased risk of colorectal carcinoma."

Before I go any further, gastrin doesn't cause colon cancer. It causes preexisting conditions to get worse. This is actually a cascade of how colorectal cancer develops. You have benign abnomas which then turn to carcinoma.

This transition from here to here has to do with the rate of proliferation. Any proliferative agent will make that rate occur quicker. Gastrin is one such agent. In fact, these cancer cells make their own gastrin and make their own receptors because they are parasites which want to grow. They use whatever they can to grow. This is the case with gastrin and colon cancer.

As far as reflux, I'm going to do this just really quickly because we're talking about reflux disease here. Heartburn is a manifestation of reflux disease. The disease is reflux, the symptom is heartburn. If you have heartburn two or three times a week you have mild GERD but it is GERD nevertheless.

It appears monthly in 40 percent of individuals in this country. Weekly 15 percent of the people will say they have heartburn and 7 to 10 percent of people in this country have heartburn every single day. That is why we are here.

In most, GERD is a nuisance. Ten to 20 percent develop complications. Three to 7 percent have Barrett's Esophagus. One percent of people endoscope for different reasons. Cancer of esophagus is the fastest growing cancer in the U.S. for unknown reasons despite the best GERD therapy ever available.

There are phenomenal drugs as far as treating reflux disease. Complications may occur without severe symptoms because of a poor correlation between symptoms and what the esophagus looks like.

These are the data looking at esophageal cancer in this country published by the Mayo Clinic. Back when I was training the most common cancer of the esophagus in the western countries was squamous which has come down in the prominent oil producing countries and places where hot beverages are consumed now. The fastest growing cancer is endocarcinoma of the esophagus, Barrett's Esophagus.

It is perplexing that the incidence of this neoplasm has increased dramatically during the very period in which highly effective acid-reducing therapies have provided symptomatic relief and healing in those individuals with mucosal injury due to the erosive effects of acid and other gastric contents.

This is what the esophagus is supposed to look like. This is a normal esophagus where it is dark because -- excuse me because the gastroenterologist can explain this and I'm not a gastroenterologist.

This is the lining of the stomach. The stomach is an amazing organ. In my view the most important organ in the body. The reason it's amazing is because it will withstand the effects of such potent hydrochloride acid that would eat a hole in a piece of paper if you took it out. The stomach withstands that acid very, very nicely.

The lining of the esophagus is different. It does not tolerate acid very well at all. What happens sometimes for unknown reasons is this. This is not a 45-year-old white man which we generally consider Barrett's to occur. This is a 32-year-old woman who has Barrett's Esophagus.

You can see the dark lining is extending upwards. Here is a squamous lining. This is Barrett's. Barrett's is a conversion of the esophagus from the squamous lining to columnar lining. Why it happens no one knows. If I can be tautological for a second, what is happening is the esophagus is saying, "I cannot deal with this acid anymore. I'm going to change my lining to be able to withstand the effects of acid."

That's fine and dandy. It's becomes an intestinal type metaplasia. Metaplasia means "changed growth." This data was published in 1997 showing 1 percent of people with Barrett's will convert to cancer. More recent data suggest 1 in 200, 1 in 250. Nevertheless, one in 200 or 250 is quite significant.

Mostly importantly is we have no evidence at all that treatment with any agent, surgical treatment or medical treatment will alter this conversion. We did a study recently to determine whether functional gastrin receptors are present on esophageal adenocarcinoma cells. These are human cells. Where gastrin is present, could it actually be participating in this increase in esophageal cancer.

We used SEG-1 cells which were obtained from David Beer at the University of Michigan. They were derived from a human with adenocarcinoma of the esophagus. We actually have two other humans with very similar results. I'm going to show you the SEGs of the most dramatic results.

Again, they are derived from a human with esophageal adenocarcinoma in association with Barrett's. We used alpha and gastrin. I can't go into detail again but there are other forms of gastrin which we now have data emerging where the precursor gastroms causing the exact same effect that you will see here.

This is PCR and I'm not going to explain PCR to you. This looks at the presence of different genes being expressed or being present. These are actually rat adenocarcinoma cell lines which have the gastrin receptor.

These are the SEG-1 cells of humans. Here is actually Barrett's Esophagus which you see. We have not investigated this further yet. We actually found this in a patient with Barrett's, that the receptor is indeed present.

We actually confirmed this with analysis showing this in human colon cancer, human gastric cancer. These gastrin receptors are there because they want to grow. They like the gastrin to grow. We also did a class of binding studies. These are fairly visual studies. This is confocal microscopy.

DR. CANTILENA: Excuse me. You've hit your time so if you could wrap it up, please.

DR. WOLFE: I have two more slides. This basically shows you gastrin is indeed present. It is bound to the receptor and internalized. These are control cells not showing it. Most importantly the cells grow. They do proliferate. These are two different assays. These are counting assays and MPT assays. They grow in a dose dependent fashion and inhibit with an antagonist. This is showing the signals through the regular pathways.

My conclusion is the following. From the study the presence of functional gastrin receptors and

esophageal adenocarcinoma. This raises the possibility hypergastrinemia associated with proton pump inhibitor therapy may stimulate the proliferation of preexisting esophageal adenocarcinoma.

Thank you for the opportunity to speak.

DR. CANTILENA: Thank you very much, Dr. Wolfe. I have one question if you will stay at the podium. The question is is there any evidence that the use of PPIs increases the risk of developing like esophageal issues that you have shown?

DR. WOLFE: This is the very first study that demonstrates it. As far as actually causing it, there is no evidence that what we are seeing is association that the occurrences continue to increase despite the value of the therapy. There is absolutely no evidence to show that PPIs do cause endocarcinoma of the esophagus.

DR. TITUS: In addition to the open public hearing speakers that we just heard from, the agency received three statements from organizations that could not participate today. They are available in our book that is out on the table and the committee members have them in front of them. We received a statement from Wellpoint, from APhA, and from the American Gastroenterology Association.

DR. CANTILENA: Okay. Thank you, Sandy.

We will now move to the sponsor presentation from Procter and Gamble. If I could introduce Dr. Keith Triebwasser who will start off and then introduce other members of his team.

DR. TRIEBWASSER: Good morning, Mr. Chairman, ladies and gentleman of the Advisory Committees. I'm Keith Triebwasser with Procter and Gamble. We want to thank you for the opportunity to come here today and discuss the Rx to OTC switch of omeprazole for the prevention of frequent heartburn symptoms.

Some of you may recall that we came before the Joint NDAC/GDAC Advisory Committees in the fall of 2000. At that time it was noted that we needed to define a suitable target population and labeling that were congruent with our data and was a safe and effective use of omeprazole by this population.

Since that time we have worked with the FDA to identify this target population and to develop new OTC labeling. We have conducted label comprehension and actual use studies with this new label.

We are here today to show you how these data support our proposed OTC label and to show you that the known benefits of omeprazole will outweigh any potential risks associated with its use in the OTC setting. Our presentation will address the questions posed to you by the FDA.

Our OTC target population is those people with frequent heartburn defined as heartburn symptoms two or more days a week. This is roughly 40 million people in the United States. Their heartburn affects their daily lives.

It affects what they eat, it affects what they can do at work and at leisure, and often their sleep. It's not surprising that the goal of people with frequent heartburn is to prevent these symptoms rather than to try and treat each occurrence.

Right now most people with frequent heartburn 77 percent of them are using OTC medications. These medications are not indicated for the prevention of frequent heartburn. The OTC medicines they use are antacids and H₂-receptor antagonists, and they frequently use these together.

Most people with frequent heartburn aren't satisfied with the OTC medications they are taking. In fact, only 19 percent say they are completely satisfied with the heartburn relief they achieve with over-the-counter medications.

Just why is satisfaction with current OTC so low among people with frequent heartburn? One of the reasons is the current OTC therapies are not well suited to prevent frequent heartburn symptoms. The pharmacology of these OTC products limits their effectiveness against frequent heartburn. The duration of action is limited and acids only last one to two hours and H₂-receptor antagonists only last eight to 12 hours.

As a result of these limitations, these medications lack all-day efficacy. Often more than one dose is needed to control heartburn and people with frequent heartburn symptoms find themselves using these therapies repeatedly, often without complete satisfaction, or without adequate acid control. OTC omeprazole can be the solution these people are seeking.

Omeprazole is ideally suited for the prevention of frequent heartburn symptoms. Simply stated, omeprazole has the right pharmacology to meet the unmet needs of this target population. The mechanism of action of omeprazole provides for prolonged acid suppression which reaches a maximum at three to five days of dozing.

The prolonged acid suppression means that a single daily dose prevents heartburn symptoms for 24 hours. This onset of action profile and the long duration of action of omeprazole match very well with the needs of the target population and they are ideally suited for the prevention of symptoms.

In addition, omeprazole has an excellent safety profile. This drug has been marketed for 15 years in more than 125 countries with more than 450 million patient treatments. This extensive experience with this drug has revealed no safety concerns.

Since our previous advisory committee we've had several productive discussions with the FDA and have developed a label that we think is simple, direct and, as you will see, understood and adhered to by the consumer. This label provides clear instructions on how to select and use the product and what course of action to take if symptoms continue or return.

As I have mentioned, we have identified a specific target population and an indication that are appropriate for OTC omeprazole. We believe omeprazole should be available over the counter at 20 milligrams, the current Rx dose. This dose provides superior acid suppression and shows both clinical and statistical significance in our efficacy trials.

The directions for use call for one tablet a day taken in the morning. This is the dozing in our clinical efficacy trials and ensures 24-hour prevention of frequent heartburn.

OTC omeprazole should be labeled to be taken on consecutive days. This is consistent with both the pharmacology of omeprazole and with the needs of the people with frequent heartburn to prevent heartburn symptoms from occurring.

We believe the OTC label should specify that this regimen be 14 consecutive days. The 14-day regimen on the OTC label is a conservative application of the current clinical guidelines for omeprazole use. It is also an appropriate period after which people should be directed to see a doctor before continuing to use omeprazole.

Our actual use study demonstrates that people understand and comply with a 14-days label. This is the duration of our clinical trials which support our application. We have included the instructions to see a doctor if warning signs of a more serious conditions are present.

We have been very clear that if symptoms continue or return, to seek physician direction before continuing to use the product. Today we will show you how omeprazole properly labeled will be safely and effectively used in the OTC setting.

With this as background, let me just take you through today's presentation flow. First, Dr. David Peura from the University of Virginia will provide his views on how OTC omeprazole can fill a crucial gap in existing heartburn therapy and how this fits with current clinical practice.

Dr. Doug Bierer will represent the results of the efficacy program, the results of the study showing that consumers used the product appropriately and according to label instructions.

Dr. Douglas Levine and Dr. Nora Zorich will discuss product safety and safe use of omeprazole in the OTC setting. Their presentations will address the risk benefit analysis for individuals who may chose to use omeprazole chronically without physician involvement.

Finally, I will summarize how the data support the safe and effective use of OTC omeprazole for the prevention of frequent heartburn symptoms.

Just a minor housekeeping note before we start. Please note there is a number in the upper right-hand corner of each slide. If you will just keep note of that number, it will help us in any follow-up questions and answers that you have.

Let me introduce Dr. Peura.

DR. PEURA: Thank you and good morning. As a practicing gastroenterologist for almost 30 years, I have treated thousands of patients with all kinds of heartburn. My clinical and research experience actually extends back to the BC era, before cymetadine, and I don't think I'm a dinosaur.

I've been involved in the evolution of all classes of heartburn medicines from the 1970s when we used to dispense the antacids by the caseloads to the 1980s when I was involved in the early trials of the H₂-receptor antagonists.

Finally, the 1990s with the latest generation of heartburn medicines the proton pump inhibitors, or PPIs. Today PPIs are the class of drugs that my GI primary care colleagues and I most frequently prescribe.

Because of their unique pharmacology and duration of action they are by far and away the preferred medication for the prevention of frequent heartburn.

My purpose today is to give you a clinical perspective on the proposed omeprazole Rx to OTC switch. I'll talk to you about how this switch will fit into established medical practice and how it will benefit people in the OTC setting.

Now, in considering the proposed switch of omeprazole to OTC, first it's important to understand the condition of frequent heartburn. Just as the severity of heartburn symptoms ranges from relatively mild to very severe, frequency of heartburn varies as well. Some people get heartburn only once in a while, maybe when they eat a pizza.

Now, omeprazole is probably not the right drug for them because they can get immediate relief with current OTC products like the antacids or the H₂s. Larger numbers of people get heartburn two or more times a week with varying degrees of severity.

Some of them have this frequent heartburn every week and some of them only get it intermittently. It's a very common condition. There is currently no effective over-the-counter option to help these people prevent their symptoms.

The H₂s just don't last long enough and antacids just aren't strong enough. While these heartburn medications are adequate for treating symptoms of occasional heartburn making you feel better after that pizza, clinical experience has taught us that only PPIs can prevent the symptoms of frequent heartburn. We're talking about 40 to 60 million people out there.

I don't think all of these people with frequent heartburn need to see a doctor before beginning treatment. I do think with proper OTC medication consumers can safely and effectively self-manage their symptoms including on deciding when to see a doctor.

To explain why, let's talk first about how physicians use PPIs to prevent frequent heartburn. Conservatively I would say that more than half the patients I currently see take PPIs and that would probably be also true for my GI colleagues and many of these patients have failed the H₂s.

Primary care physicians are also very comfortable prescribing this class of medicine for most of their patients with frequent heartburn. In fact, in the United States most PPI prescriptions are written by primary care physicians.

When patients come to me with frequent heartburn symptoms and there are no warning signs of any other condition, I give them a short course of a PPI like omeprazole. Neither my colleagues nor I routinely do diagnostic tests unless a patient has severe or refractory symptoms.

In fact, few doctors would recommend initial endoscopy for a patient who complains only of frequent heartburn. This is actually what the professional societies currently recommend, a therapeutic trial before endoscopy. When I start a patient on a PPI they usually get better very quickly. I'm confident of the diagnosis and I'm simply follow their clinical progress.

This treatment approach to frequent heartburn that I've just described in the sponsor's proposal for OTC omeprazole are very consistent with current practice guidelines. In fact, the most recent published guidelines specify that therapy should be aimed at treating or preventing heartburn symptoms with acid reducing medicines. If acid reducing medicines prevent symptoms, nothing further needs to be done. Therefore, symptom management is really the first stage of patient management.

These patients are quite knowledgeable and they can recognize their frequent heartburn for what it is. They know when they need to see a doctor. Given these factors and since symptomatic management and prevention would be the doctor's initial approach, I believe that with a safe and effective OTC medication consumers will successfully self-manage their frequent heartburn.

Let me elaborate on this. First the sponsor's proposed label educates consumers to take omeprazole for 14 days and not to take it any longer without first contacting their doctor. I remember, and I'm sure many of you do, that there was a concern when the H₂s when over the counter that people would stop going to see their doctors.

Studies show that didn't happen. People still see their physicians about their heartburn. I expect the same will be true with OTC omeprazole. In fact, the sponsor's data would suggest that consumers will follow instructions and will ask their doctors about their frequent heartburn even if they haven't done so already.

Now, undoubtedly there will be some consumers who will take the drug for longer periods of time without talking to their physician. This doesn't really concern me because it's likely that's what their doctor would have told them to do anyway.

We know that when patients with frequent heartburn are treated with PPIs they do much better than those who are treated with any current over-the-counter medicine primarily because PPIs are more effective at reducing acid.

Some of you might have concerns that OTC omeprazole might mask or delay a diagnosis. I'm comfortable that's not going to be the case. It didn't happen when the H₂s went over the counter.

As far as masking a more serious condition, while it's conceivable, in almost 15 years of using these drugs and thousands of patients I've not seen it. In my opinion the risk benefit here is very favorable.

In conclusion, the best way to manage frequent heartburn is to prevent symptoms. Taking a PPI is the best way to do that. Doctors know that and that's why doctors use PPIs in their patients with frequent heartburn.

Certainly physicians have a role in the management of these patients but that involvement doesn't have to be intensive. This is a common condition. Consumers who have it and understand it know when they need to see their doctor.

They can safely and effectively self-manage their own frequent heartburn symptoms and I believe they should be empowered to do so. The proposed dose and duration of therapy is appropriate, effective, and consistent with current medical practice.

From my 30 years of clinical experience I know omeprazole is safe and it works. More importantly, so to my patients with frequent heartburn. Thank you very much.

DR. BIERER: Good morning. Thank you for the opportunity to come here to present the results of our clinical efficacy and our consumer behavior program that supports the use of omeprazole for the prevention of frequent heartburn symptoms.

Our program consist of efficacy and consumer understanding and behavior studies. First, pivotal studies that shows omeprazole prevents the symptoms of frequent heartburn for 24 hours and over a two-week period.

Second, consumer understanding and behavior studies show that consumers understand the product label and they use this product appropriately in a naturalistic OTC setting. Our entire program supports that the product is efficacious and the consumers will use omeprazole safely and according to the label directions.

We conducted two well-controlled efficacy trials in which we evaluated the efficacy of omeprazole magnesium for the prevention of the symptoms of frequent heartburn over a 14-day period. As you know, these studies were presented at a previous advisory committee meeting and they support our proposed dose and our dosing duration.

The study population included subjects who had heartburn symptoms two or more days a week and did not have prior physician diagnosis for GERD or erosive esophagitis. The subjects in the study were instructed to take one tablet every morning for 14 consecutive days. The end points that support our proposed label are the percentage of subjects who are heartburn free after the first dose of the product and a percentage of heartburn free over 14 days of consecutive dosing.

Let's look at the results of these studies. This slide shows the percentage of subjects who were heartburn free for the entire day over 14 days of consecutive dosing. In both studies we achieved our primary endpoint. That is, the prevention of heartburn symptoms for 24 hours after the first dose of product.

A higher percentage of people taking 20 milligrams of omeprazole were heartburn free as compared to placebo. As also can be seen from this graph, heartburn prevention increases over the first few days of dosing.

This is consistent with the pharmacology of the drug and this prevention effect remains consistently high over the 14 days. The effect is both clinically and statistically significant for both studies on day one for day 14 and across all 14 days.

In summary, our efficacy studies show that 20 milligrams of omeprazole provides clinical and statistically significance in the prevention of heartburn symptoms. The study supports our proposed OTC label indication which is the prevention of the symptoms of frequent heartburn for 24 hours. It supports our proposed dose of 20 milligrams and the label's direction to take one tablet in the morning for 14 consecutive days.

Now let's look at our consumer understanding and behavior program. For our OTC target audience of people with frequent heartburn, our program objectives were to demonstrate the consumers correctly self-selected, this was a product that they were willing to use, they understood how to use the product, and they adhered to the product warnings.

In our consumer understanding and behavior program we conducted two types of studies labeled comprehension and an actual use study to understand how the general population will use this product in a naturalistic OTC setting.

All of these studies met our objectives in that the vast majority of the people understood the product label, they used the product appropriately, and they adhered to the warnings.

In our first labeled comprehension study, we recruited 684 subject from 12 geographically and social economically diverse sites across the U.S. This study population included people with infrequent or no heartburn, people from our OTC target population of people with frequent heartburn, low literate frequent heartburn people who had less than an eight grade reading ability as measured by the REALM test which evaluates the understanding of medical terminology, people with potential drug-drug interactions, and finally those who are pregnant or nursing.

In general all five groups scored very well on label comprehension. Let's take a closer look. First, let's look at subjects with infrequent or no heartburn. Seventy-eight percent with infrequent or no heartburn correctly chose this was a product they should not use for episodic or acute heartburn of one day a week or less.

Ninety-nine percent of frequent heartburn people correctly chose this is a product that they could use. These data demonstrate a high understanding of who should and should not use the product.

We were also interested in low literate subjects' comprehension of the label. For this portion of the study we gave each person in the low literate group descriptions of situations involving either frequent or infrequent heartburn. Then we asked them whether they should use this product in this situations.

In situations involving frequent heartburn 79 percent answered correctly. In situations involving infrequent heartburn, 49 percent answered correctly. Initially these results concerned us. As you will see later, in our actual use study the low- literate group scored much higher for appropriate self-selection.

Now, let's look at how well all subjects understood the dosing directions. Here we found high comprehension of the label dosing directions. Subjects clearly understood how much omeprazole to take, how often to take it, and to contact a healthcare professional before taking it for more than 14 days.

We also then looked at other circumstances which would require healthcare provider involvement. In our proposed label we included warning for people who have symptoms that could be mistaken for or occur with any heartburn of any severity. These include chest pain, trouble swallowing, frequent wheezing, unexplained weight loss.

It is important to note that these symptoms are not related to the use of any kind of heartburn medication. Because these symptoms can be the sign of a more serious condition, we advise people in our label to talk to their doctor about these symptoms if they have not done so already.

After testing several versions of the label, we achieved labeling which was understood by 81 percent of the people with these general warning signs. As you will see shortly in our actual use study, people were compliant with this warning.

Regarding drug-drug interactions, we would that people were much more likely to understand the nature of the drug-drug interaction when they were shown brand names in addition to the generic drug names.

As you see, 82 percent responded correctly when shown both the brand name and the generic name as opposed to 50 percent when shown just the generic name. Because of regulatory and policy consideration, we believe it is most appropriate to discuss this labeling issue with the agency at a later date.

Also, more than 90 percent of pregnant and nursing women correctly selected that they shouldn't use this product without consulting their physician.

Now, let's look at our actual use study. As you know, label comprehension studies indicate whether people understand what is written on the label. What is even more important is how people will use the product in a real world setting.

Our actual use study had three major objectives, to evaluate whether consumers correctly self-selected, whether they used a product appropriately to prevent the symptoms of frequent heartburn, and whether they complied with the label.

In our actual use study we found that consumers did correctly self-select. They used a product appropriately for prevention and they used it according to label directions. We designed our actual use study to mimic real world purchase experience using both traditional methods and some new features.

This study was conducted at a mall kiosk which was freely accessible to subjects rather than at a clinical site. To ensure that subjects were able to correctly self-select based on the product label alone, there were no healthcare professionals on site.

There was no contact with the subjects during the use portion of the study. And, as with any OTC medication, the subjects were free to contact their doctor at anytime during the study.

We also incorporated a couple of new features designed specifically to mimic real world consumer purchase decisions. Subjects purchased the product at a realistic market price and they were free to return to the kiosk to buy additional product.

We were also very careful not to present barriers for repurchase. The kiosk was highly accessible, it was open during regular mall hours, and it was close to where the subjects lived. The subjects were told numerous times that they could return to buy additional product.

Let's take a closer look at this study. We recruited subjects through local advertising in spontaneous mall intercept at five malls across the U.S. At the mall kiosk the subjects made a self-selection decision. That is, they had to determine whether this was a product that they could use, and also whether they were willing to purchase the product.

Our product is intended to be used for a regimen of two weeks and not to be used for more than two weeks unless directed by a doctor. It was especially important to find out whether people would incorrectly use the product on an extended basis without physician involvement.

Therefore, we made the product available for a total of eight weeks and during that time people could have bought as much product as they wanted and theoretically they could have used four courses of treatment.

Four weeks after the use period we contacted the participants by telephone and asked them whether their frequent heartburn returned and, if so, what were they doing about it.

Before I describe the results of this study, let's first look at the disposition of subjects. We approached 5,060 subjects at the mall and we asked them, "Do you get heartburn?" Of these 3,809 said they either did not get heartburn, they were not interested in being interviewed, or the drug was not appropriate for them to use.

Of the 1,251 people who said they could use the product, 385 were not willing to buy the product mostly because they were not interested in participating in a clinical study or they wanted to check with their doctor before taking a new medication.

866 people said that this was a product that they could use and were willing to buy the product and participate in the study. This 866 constitutes are self-selection population since they not only identified omeprazole as a product that they could use, but they were also willing to buy it and use it.

When people are asked a hypothetical question, "Could you use this product," they may say yes. But in the real world the key differentiator is whether they will actually buy the product.

Now, let's look at the demographics of our self-selection population. Slightly more than half were women, 68 percent caucasian, 16 percent African-American, 11 percent Hispanic. The average age was 48 years with a range between 18 and 91 years old, and 8 percent had a low reading ability as measured by the realm test of medical literacy.

Importantly, 90 percent of the subjects who purchased the product had frequent heartburn two or more days a week. This indicates that our label is well understood by our intended OTC target population. More than 90 percent of the people said that they used OTC medications to control their symptoms and 40 percent reported using prescription medications.

Now, let's look at whether these people correctly self-selected. As I will show you, 81 percent met all six self-selection criteria.

These criteria, which are specified on our label, include heartburn two or more days a week, greater than 18 years of age, not allergic to omeprazole, not pregnant or nursing, no general warning signs, and no drug-drug interactions as listed on our label.

In order to correctly self-select for this trial, subjects had to meet all six of these self-selection criteria. That is, if they did not correctly select any one of these answers correctly for the criteria, they were counted as a self-selection failure.

Again, of the 866 people who said they could use this product and wanted to buy the product, 81 percent met all six self-selection criteria. When we look specifically at the low-literate people, we found that almost 70 percent met all six self-selection criteria.

Let's look at the people who did not correctly self-select. Four subjects who incorrectly selected weren't allowed to buy the product. These included three of them who were under 18 years of age and another who was pregnant. There were no subjects who were allergic to omeprazole.

However, we did allow subjects who were in these last three groups on this slide at the bottom to purchase and use the product even though we counted these people as a self-selection failure. This is consistent with FDA's guidance to include self-selection failures when the risk is minimal.

There were 82 subjects who had not talked to their doctor before the study about a general warning sign listed on the label. We found that during the trial none of these subjects had a serious adverse event.

There were eight subjects who were taking a drug listed in a drug-drug interaction section on the label. While these eight subjects did not initially contact a doctor or pharmacist, we did find that five of them did contact the doctor about the congest use with omeprazole during the study. Again, none had a serious adverse event.

Finally, there were 86 subjects who had infrequent heartburn. Half of these people took the product in compliance with the 14-day labeling and the other half took it sporadically as you would expect for someone with infrequent heartburn. None of these people with infrequent heartburn exceeded 14 doses.

Now, let's move to look at the use and repurchase phase of the study. Of the 866 people in our self-selection population, 96 did not return a diary despite several contacts.

Four subjects withdrew consent and eight were not allowed to purchase product. This includes the three subjects I mentioned earlier who were less than 18 years of age and the one pregnant woman.

There were also four others who had participated in a previous use study with this product. But importantly, 90 percent of our self-selection population, that is 758 people used the product and returned the diary.

Our use directions call for taking one tablet per day. Let's look at whether people complied with this use direction. Here we found excellent compliance with the dosing directions. Ninety-six percent of the subjects took no more than one tablet per dose and 91 percent of the subjects took only one tablet per day.  Again, people clearly understood these label directions.

One of our key questions is whether people with frequent heartburn would take this product as a regimen for 14 consecutive days. In our protocol, we defined compliance to the 14-day dosing regimen based upon two criteria.

First, they had to take between 80 and 100 percent of the product and take it over 14 days plus or minus three days. This is a range of about 20 percent for the days. That is, they had to take between 11 to 14 doses of the product and take it within 11 to 17 days.

This criteria is similar to the industry convention and epidemiological conventions of patients taking at least 80 percent of study medication in the clinical trial, and is also consistent with the long-lasting pharmacology of the drug.

The second criteria was if a subject took more than 14 doses of a product, that is, they took even one more dose than 14 doses, they had to consult with their doctor in order to be defined as compliant. With this definition in mind, let's look at how people were compliant with the 14-day regimen.

This slide shows the compliance of the 14-day regimen of all 758 people who used the product and returned the diary. Seventy-nine percent of the people were compliant. That is, they took 11 to 14 doses and 11 to 17 days, or they contacted a doctor if they exceeded 14 doses.

Also as mentioned in the FDA briefing document, 64 percent of these people took exactly 14 doses in 14 days. These data demonstrate high compliance with a dosing regimen of 14 days and also use of this product for the prevention of frequent heartburn.

Nine percent of the people took the right amount of drug. That is, 11 to 14 doses, but they took it over a longer period if time, greater than 17 days. An additional 9 percent of the people took fewer than 11 doses. As expected, many of these people had infrequent heartburn.

Note that none of these group of people, this 18 percent took more than 14 doses of the product. We also saw that fewer than 1 percent took multiple daily doses of product. There were only three of these people and none of them exceeded three doses per day, and also none had a serious adverse event.

We also found that only 3 percent of the subjects took more than 14 doses without healthcare professional contact. It is important to note that 75 percent of these people had either talked to their doctor about their heartburn before the study, or soon after the study ended.

In summary, the label was clearly understood. People understood that they were to use this product for the prevention of frequent heartburn symptoms, and they achieved high compliance with the dosing directions.

Now, as I explained earlier, four weeks after the study ended, we wanted to find out whether people's frequent heartburn returned and, if so, what they did about it. We were able to follow up by phone with about 85 percent of the people who used the product and returned the diary. Forty-three percent said that their frequent heartburn had not returned.

Of those who said their frequent heartburn had returned, 8 percent were not using any medication, 22 percent reported taking antacids, 9 percent reported taking an OTC H₂-RA, and 3 percent reported taking a combination of both antacids and H₂-RAs.

Finally, 15 percent returned to a previous Rx medication or started a new prescription. This level of physician involvement is consistent with the habits and practices we have found of people with frequent heartburn.

Our actual use study showed that consumers appropriately self-selected, they understood the label directions, and they took the product as on a regimen basis for the prevention of frequent heartburn. Finally, they used a product in accordance with label directions.

In summary, our efficacy and consumer behavior data supports our proposed label. The efficacy data supports our indication for the prevention of the symptoms of frequent heartburn for 24 hours. It supports our proposed dose of 20 milligrams and it supports our dosing directions to take one tablet in the morning for 14 consecutive days.

Our consumer understanding and behavior program shows compliance with these use directions and general adherence to the label warnings when this product is used in a naturalistic OTC setting. Thus, we have demonstrated that our proposed label, our efficacy data, and the consumer's ability to understand and use this product safely and appropriately are all congruent.

DR. LEVINE: Good morning everyone. I would like to present a perspective, perhaps another perspective, on safety issues related to over-the-counter use of omeprazole.

One type of safety consideration is product safety. This is defined as adverse events occurring in relation to product use during the short or the long term.

Omeprazole has a excellent safety profile. The product related adverse event profile is very well established based on data from clinical trials with the prescribed product, post-marketing surveillance with the prescription product, as well as the OTC clinical trials.

As you recall, the most common adverse events are reversible symptomatic side effects including things like headache, diarrhea, and abdominal pain. This profile of omeprazole makes it acceptable for over-the-counter use.

The use of omeprazole is intended to be short term as indicated on the proposed label instructions. However, if any unintended long-term over-the-counter use were to occur without medical supervision, the product adverse event profile for the situation is considered acceptable based on the extensive experience with omeprazole in the prescription setting.

Another type of safety consideration would include potential consequences of consumer behavior involving long-term use of the product without medical supervision but such consequences would not be directly linked to omeprazole.

This type of safety issue involves considerations of medical diseases other than acid reflux disease, as well as of the natural history of acid reflux induced esophageal damage which is not completely understood today from an epidemiologic perspective.

Again, the sequelae of these diseases are not directly linked to omeprazole. First, use of the over-the-counter product for alarm symptoms such as these would most likely be self-limiting because these symptoms would not be expected to subside with omeprazole use.

The proposed label instructed not to use the product and to seek medical attention if these symptoms are present. However, there are a variety of conditions with symptoms that could certainly respond to omeprazole and might lead to some behavior involving long-term use without medical supervision despite the label instructions.

The most common of these would be non-neoplastic upper GI conditions resulting from acid peptic injury. These include esophageal erosive disease or peptic ulcer disease involving the duodenum or the stomach.

Such abnormalities, though, would be well managed with chronic omeprazole therapy with little chance of adverse consequences. It is important to address medical diseases that include either overt malignancy or conditions that predispose to malignancy of the upper GI track.

Individuals with cancer of the esophagus or of the stomach might have frequent heartburn. However, these tumors more typically produce different symptoms such as dysphagia, nausea, vomiting, early satiety, and weight loss which do not respond to treatment with omeprazole and would likely led the consumer to seek medical attention.

Unfortunately, these types of tumors are commonly diagnosed as part of the very first medical presentation for medical care. This suggest that these diseases commonly evolve without producing significant heartburn or other symptoms during their precancer stages.

Additionally, in endoscopic survey studies of individuals with heartburn, cancer of the esophagus or the stomach is rarely identified.

Another area of concern is a condition associated with cancer risk and this is Barrett's Esophagus. Barrett's is a complication of chronic acid reflux disease with resulting esophageal damage. While Barrett's Esophagus is common, its progression to esophageal adenocarcinoma is unusual.

It is, in fact, the inconsistent correlation between frequent heartburn and the present of Barrett's plus the rarity of progression from Barrett's to esophageal adenocarcinoma that makes it difficult for the medical community to manage this risk. If desired by the advisory committee, we can provide additional data on this topic during the question and answer period or this afternoon's discussions.

In the context of over-the-counter treatment of heartburn, available data showed that omeprazole is, in fact, a neutral factor. It does not increase cancer risk and it does not reliably induce regression of Barrett's Esophagus.

To summarize the current situation on heartburn and esophageal adenocarcinoma, the increasing incidence of esophageal adenocarcinoma in the United States since the early 1970s is not related to acid reducers but research continues to evaluate many of the factors that may contribute to this rise in incidence.

Patients afflicted with esophageal cancer generally present without a prior history of heartburn of a prior diagnosis of Barrett's Esophagus. The status of esophageal adenocarcinoma in 2002 is as medically challenging as it is sobering.

Presently the incidence rate for esophageal adenocarcinoma is about the same as the mortality rate, again reflecting the fact that these cancers are discovered at late incurable stages without antecedent clinical signals that might lead to diagnosis of an earlier stage cancer or its precancerous precursor.

There is no evidence to suggest that acid reducers are masking any signals of these diseases. Fortunately, the development of this cancer is rare among individuals with heartburn or among those with documented Barrett's Esophagus and omeprazole specifically does not increase the risk of this cancer.

In conclusion, I've tried to frame today's discussions on safety related to over-the-counter omeprazole use. The product related safety profile of omeprazole is acceptable for over-the-counter use. The natural history of acid reflux damage to the esophagus can involve rare serious consequences. However, omeprazole does not directly increase the risk of esophageal adenocarcinoma.

Based on the overall safety considerations I have presented today, omeprazole is acceptable for over-the-counter use. Thank you for your attention to my remarks. I would like to introduce Dr. Nora Zorich, my medical colleague, who will continue the sponsor safety presentation.

DR. ZORICH: Good morning. Thank you, Dr. Levine.

I'm going to take a few minutes to settle down here and address in more depth the question of whether there is any concern what the long-term use of omeprazole without physician involvement.

In order to do this, I'm going to consider three key factors important in the benefit risk assessment. First, we'll look at data that provides insight into what people do now and what they might do in the future if omeprazole was available over the counter. I'm going to specifically address what proportion of consumers might use the product on a more regular basis.

Then I'll discuss data that examines physician involvement by people with frequent heartburn and specifically address the concern that once omeprazole is available, people with heartburn won't seek the care of their physician.

Finally, I'll recap the issue of potential risk and discuss the known benefits that might result for consumers who use the product even without physician involvement.

Now, before we can address how often consumers might use OTC omeprazole, I think it's worthwhile to talk a little bit about how they use it today. In order to do that, I'm going to take you all the way up to the 40,000 foot view of omeprazole use looking at a very large administrative claims database of prescription drug use.

Here is an analysis of the NDC database which collects and analyzes the prescription data from multiple managed care organizations that cover millions of lives in the U.S. We examine the drug records of almost 100,000 people who were prescribed omeprazole for the first time in January of 2001.

Then we followed whether they had any further omeprazole dispensed to them over the next year. These data represent people who had heartburn symptoms that were significant enough that they drove them to their physician and then their physician prescribed omeprazole.

Now, let's look at this data. As you can see, about 44 percent of them were dispensed omeprazole only once in a one-year period with about 70 percent of this group of people receiving three or fewer treatments dispensed to them over a year.

Then we get down to a group that I have collected together of people being dispensed six or more treatment courses and that's about 20 percent of this group. From that perspective we can look at chronic use of those people who would be taking the drug on more days than they would not be over a year.

Chances are what we know about chronic users these would be the same people who would go on to future use on a chronic basis. That's the really big picture of omeprazole use. Let's take a step further into understanding how this drug is used and look at a population specifically diagnosed as having GERD.

Here is a study published by Bardhan. It was a randomized clinical trial in which people with symptomatic frequent heartburn were evaluated and how often they required omeprazole to control their symptoms.

The reason I selected this particular study to discuss is that in contrast to the numerous studies that have been done looking at maintenance therapy, few studies have actually examined the strategy of intermittent therapy.

This particular study is helpful in that it captures the kind of use decision behavior that you are likely to see in the OTC setting. That is, the patients going back to the clinic for more treatment if and when their symptoms returned.

The most relevant group for our discussion are those people in this trial who took an initial course of therapy 20 milligrams of omeprazole for 14 days. Then they self-managed their frequent heartburn by requesting additional courses of therapy throughout the year.

Let's look at the results. There were 704 people on this trial and 526 were available for a final assessment. Bardhan found that 72 percent of these individuals were able to self-manage their frequent heartburn by taking additional 14-day regimens of omeprazole intermittently. Of the 72 percent 68 percent requested three or fewer treatments.

To put that in perspective, that is 42 days of omeprazole in a year. This pattern of use is actually quite similar to what was seen in that big NDC database where we saw about 70 percent of use being three or fewer treatment courses over a year.

Bardhan found that symptom control after two weeks of therapy was a powerful prognostic indicator of future need for therapy with almost 30 percent of the people requiring no further treatment if they had a good response to that first 14 days.

At the other end of the spectrum of use, only 28 percent of the study participants had ongoing symptoms which required maintenance therapy at some point during that one-year period.

Let's compare these data to our actual use trial, what you just heard from Dr. Bierer, where we found that well over 90 percent of the participants in the actual use trial purchased only one box of 14-day omeprazole during the two-month period that the drug was available.

When we contacted them three months after they entered the study, 43 percent of them said they did not have frequent heartburn and 42 percent while they said their frequent heartburn was back, over half of them had elected to manage their symptoms simply with antacids. And 15 percent of the participants were taking a prescription therapy.

Within this 15 percent half of them had just gone back on the prescription therapy they were on prior to their participation in the study. The other 15 percent had gone to see a physician and were prescribed a prescription therapy, generally a PPI.

After three months of observation, we found that very few people had used more than the one 14-day regimen of OTC omeprazole even when they had the opportunity to purchase it.

In summary in answering my first question, you could look across these three diverse types of data and see that even in the prescription setting it's clear there is a range of use of PPIs which undoubtedly reflects the range of symptoms with only on average about 25 percent of the use being chronic. We conclude that most people won't choose to use omeprazole chronically even in the OTC setting.

Now, returning to the larger consideration of chronic use without physician involvement, we need to ask will these people who may use omeprazole chronically see a physician. We have literature in our actual use trial to address that.

To begin with, it is helpful to review what we know about how often people with heartburn talk to their physician about their condition. Here is a publication from Oliveria who was at Cornell at the time. They surveyed more than 2,000 people with heartburn. The study was designed to capture people's understanding of heartburn, how they manage their symptoms, and importantly how often they consulted a physician.

More than 90 percent of these people were on some kind of therapy with 75 percent of them taking over-the-counter therapies. In regard to physician involvement, what the survey found was logical. The more frequent and severe people's symptoms were, the more likely they were to have seen a physician.

In fact, people with the most frequent heartburn were four times more likely to have seen a physician than those who had heartburn less than twice a week with 78 percent of the people with frequent heartburn having discussed their heartburn with their physician.

This is reassuring data and it's really very typical of what you find in the literature and in the surveys that have been conducted by the professional societies. Relative to our discussion today is whether omeprazole's availability in the OTC market would change this consumer behavior.

Now, as Dr. Peura mentioned, this question has been asked before. The recent move of H₂-RAs to OTC status in the 1990s provides a historical perspective on this question.

We are going to look at three studies that specifically examined what happened to physician visits when the H₂-RAs became available. I take the position that this is actually a relevant comparison because even though the H₂-RAs were switched at half the prescription dose, the introduction of H₂-RAs into a world that only had antacids was a meaningful therapeutic jump.

For perspective, a study by Simon and colleagues published in the American Journal of Therapeutics in 1995 stated that 70 percent of the people who obtained only some degree of relief with antacids claimed they experienced complete symptom relief for episodic heartburn using the H₂-RAs.

Yet, as you'll see, there was no negative consequence relative to physician visits for heartburn. Here is our first study. It's a publication from Andrade studying 2,000 people within the Fallon Community healthcare system who had a GERD related diagnosis and were receiving a prescription medication in 1994. What they found was that the yearly number of clinic visits in those patients with GERD stayed the same and averaged just under one visit per year per patient.

The second study was published by Shaw. It was a cross-sectional survey of adults in Minneapolis in 1993 which would be before the switches and again in 1997 after the switches. Within this general population the percent of people who went to the physician with complaints of dyspepsia or heartburn did not change when the H₂-RAs became available.

Finally, we took another look at a very large dataset, the MEDSTAT Marketscan Database of Administrative Claims, that covers again millions of lives. We looked at the people who had continuous coverage from the period of 1995 through '98. That's the period during the switch of the H₂-RAs.

We found that the percent of people undergoing endoscopy was not changed. And importantly, the mean number of visits for acid related diseases was constant with a mean of about two visits per year.

So we see from these three studies we found there was no evidence that potent acid-reducing drugs kept people from seeking physician care. Of course, the question for today is would omeprazole be different. For that we can look at our actual use trial.

What was the behavior of these participants who had omeprazole available to them nonprescription? When we looked at physician contact in the study during the period of our trial, we found that on average the monthly rate of physician visits actually rose. The rate of physician contact per month was twice what it was per month in the year prior. This was in part due to the behavior of two very important groups of people.

First, we were very pleased to see that 20 percent of those consumers who are potentially harder to reach, these were people who had never before discussed their heartburn with their physician. Twenty percent of them opted to talk to a physician about their heartburn when they participated in this study.

Importantly, over 50 percent of those individuals who took more than the 14-day regimen were in contact with their physician either during the study or when we contacted them at three months they told us that they did have a visit scheduled to discuss their heartburn.

We were very encouraged by this behavior and we believe this increased physician contact was driven in part by the label and the package insert. So from the published literature and the result of our actual use study, we see a consistent picture. The majority of people with frequent heartburn do talk about this with their physician.

People continue to see their physician even when the H₂-RAs became available. The label we propose for OTC omeprazole encourages people to go to the doctor and we believe that is important. As we saw in our actual use trial, the rate of physician visits actually increased.

Consequently, we believe that physician visits will not decrease and there is a chance that they could increase with OTC omeprazole available over the counter.

Now, while the data we have just reviewed is reassuring in that most people are not going to be using this product chronically, and those who do are going to be the same people that are most likely to be talking to their physician, what about the product use on a more chronic basis with people who simply will not seek physician involvement? What are the risks and benefits to these consumers?

Now, let me restate before I go on to talk any further about the benefits to these consumers that our position is clear. The label advocates physician involvement for those individuals who are using the product beyond 14 days.

In considering this population, probably most of these people are going to have nonerosive disease. For these individuals if their symptoms are well managed then there is a benefit and there is little chance of any adverse consequences we've heard today.

However, it's clear that some of these people could, in fact, have some degree of erosive disease. Consequently, it's important to examine the potential benefit for these consumers.

Studies published in the literature have clearly shown that PPIs provide better healing of mucosal injury compared to other therapies. The easiest way to summarize all of that is just to look at one single publication, a medianalysis by Chiba which I have shown one graph from that analysis.

This is a compilation of 43 studies looking at the healing of erosive esophagitis grades two to four, and it compares PPIs, H₂-RAs, and importantly placebo. I think you don't ever really see this result much but if you look at placebo after 12 weeks you get about 25 percent of healing in people not taking any therapy.

What I want to emphasize from this graph is that people taking PPIs you'll see that at two weeks there is more percent of people healed compared to even much longer treatment periods on H₂-RAs.

Importantly, considering the placebo response, when you look at the overall percent of healing due to PPIs, you can see that the vast majority of that healing actually occurs within the first two weeks.

Consequently, people who are not healed at two weeks experience only a modest benefit from any additional therapy. I think this data clearly showed that for those few people who have erosive esophagitis and choose not to be in contact with their physician, a 14-day regimen of omeprazole is going to be a much better option for them compared to any other medication that they would take over the counter.

In summary, we believe that the data in the literature, our analysis of large databases, and our actual use trials support that the majority of consumers will not use the product chronically.

Those consumer who may choose the product on a more frequent basis will do so with the involvement of their physician. Considering what we have heard from Dr. Peura and Dr. Levine, the potential risk for chronic dosing is minimal, while the known benefits are substantial.

DR. TRIEBWASSER: I would like to just briefly summarize what you have heard here today. You have heard that omeprazole will fill a critical gap in the OTC options for those people who suffer from frequent heartburn and who currently have no adequate OTC therapy to prevent their symptoms.

Omeprazole is ideally suited to meet the needs of these people for effective prevention of heartburn symptoms for 24 hours. From Dr. Peura you heard that OTC omeprazole fits into current medical practice and can be a safe and effective medication over the counter.

You heard that we have identified the right target population for OTC omeprazole. This target population is those people with frequent heartburn. This is the population that will benefit most from omeprazole and it is the population in our pivotal clinical studies.

We have shown you that the proposed label provides clear instructions for use. The actual use study demonstrated that this label is understood by consumers and that they appropriately self-selected and used the product.

The study also demonstrated that individuals who had a recurrence of heartburn symptoms responded appropriately. This label is congruent with out proposed use of omeprazole. It brings the right target population and the right indication for OTC omeprazole together with clear use instructions and a set of clear directions for physician involvement when appropriate.

You heard that OTC omeprazole should be labeled for 14-day regimen of therapy. Our efficacy trials and actual use studies showed that people received maximum symptomatic benefit within 14 days and that they understood and complied with the 14-day label.

Fourteen days is a conservative application of clinical guidelines. It also represents an appropriate period of OTC use after which a person should be instructed to contact a doctor if symptoms continue or return.

You also heard that 14 days of omeprazole provides healing to most of that fraction of individuals in the target population who may have erosive esophagitis. For all of these reasons, the OTC label should not specify a longer treatment period.

You have heard that some people may use OTC omeprazole on an ongoing basis without seeing their doctor. Some of these individuals may have GERD or erosive esophagitis. They will be receiving a more beneficial therapy than the medications available today over the counter. Any potential risks will be outweighed by these known benefits of omeprazole.

In conclusion, omeprazole can be safely and effectively used over the counter. Thank you for your attention. We will be glad to answer any questions.

DR. CANTILENA: Okay. Thank you very much for your presentation. What I would now like to do is actually entertain questions from the committee members to the sponsor. We will go sort of around the table. Well, actually, if there is anyone who has a question, let's just start with a show of hands and we'll identify you.

Okay. Dr. LaMont, Dr. Davidoff, Dr. Brass, and Dr. Camilleri for starters. Please don't put your hands down after.

DR. LaMONT: I wonder if we can get some clarification on the issues that were raised regarding the effect of food and antacids and other drugs on absorption. Your briefing book states on page 28 that omeprazole is completely absorbed after oral administration.

Food, antacids, and H₂-RAs have no clinically meaningful influence on the extent of omeprazole absorption. Yet, we've heard from others this morning that there are important effects of these agents on absorption. I wonder if we can have some clarification.

DR. TRIEBWASSER: Certainly. In the Rx clinical trials for ethical reasons antacids were always allowed. These were placebo-controlled trials. We have extensive data with a prescription product where it was utilized with antacids without deleterious consequences.

As was indicated before with regard to food interactions, I'm not sure if you want to take time now but we have submitted data showing that with the tablet formulation that is proposed for OTC use, there is really not a significant food effect. So we've done those studies, yes.

DR. LaMONT: Just a question of further clarification. You say on the extent of absorption. Do you really mean on the extent of clinical efficacy? Because we don't measure absorption clinically. The briefing book says absorption. I think you are referring to clinical effectiveness. Is that correct?

DR. TRIEBWASSER: Clinical effectiveness on the antacids with regard to food absorption. That's where we actually did those kinds of studies. I'm not sure if I'm pinpointing your question exactly.

DR. LaMONT: I'm trying to find out what clinically meaningful influence on absorption because we don't measure absorption clinically and we don't care about that. What we are really interested in is effectiveness. If I understand your response correctly, it had no effect. That is, antacids and other agents had no effect on clinical effectiveness.

DR. TRIEBWASSER: That is our belief. Yes.

DR. CANTILENA: Okay. Dr. Davidoff, please.

DR. DAVIDOFF: I have some questions about what I guess I've been thinking of as the dog that didn't bark. The barking dog is the potential relationship between omeprazole and cancer.

In the company's materials on page 83 there's the statement that there is no evidence that there is a causal relationship between the use of omeprazole and the development of gastrointestinal cancer in humans. Dr. Levine's slides 54 and 56 say the same thing.

I am somewhat puzzled, however, because in the materials that were given to us from the 2000 hearings on omeprazole, the summary statement from the medical review by the FDA says as follows:

"Although in the general and undifferentiated population of the U.S. there is no clear tumor association with omeprazole, the possibility that there are oncogenic effects in subceptable groups exposed to omeprazole for very long periods of time has not been ruled out.

Phase IV studies (and this is in italics) to investigate the incidents of GI adenocarcinoma and other malignancies using long-term prospective or nested control cohort study designs of large numbers of exposed individuals should be established."

They base this statement on the concern that the studies available were too short and not appropriately designed. In effect, they were saying there is absence of evidence and that is basically what the company statements have said. The statisticians are fond of reminding us that absence of evidence is not evidence of absence.

In 1930 if you asked anyone if there was a connection between smoking and lung cancer or lots of other diseases, they would say there is no evidence, but that was not because there was an absence of effect because no one had looked.

My question is has the company done appropriate prospective or nested cohort, or case control studies as suggested in the year 2000 on this aspect of safety? If they haven't actually themselves done those studies, have you at least done an exhaustive review of the literature and, if so, what did you find?

DR. TRIEBWASSER: Let me respond. I don't think the characterization of our review in any way resembles anything that has been done with the tobacco company. In fact, we have looked extensively. I would submit that the FDA position is conservative but I can review for you the nature of the data that the company is prospectively and proactively obtained over the 20-year history of the availability of this product.

First, as you may recall, the initial target tissue of concern was the ECL cell based on animal studies. In response to obvious concerns of the study that Astra at that time conducted actually several prospective controlled studies ranging in length anywhere from two to five years.

But then also open-label observational compassionate use studies in which individuals were treated in doses as high as 40 milligrams of omeprazole and endoscoped on a regular basis to actually look at not just the gastric musocal but actually the entire GI tract.

These are data that were prospectively submitted to FDA in response to agency requests over the past few years they have actually resubmitted and reanalyzed those kinds of studies.

In addition, there's been extensive post-marketing reports that we've collected and there are certain weaknesses with those kinds of data but they are available data.

We have always complied with a prospective reporting of those, but also submitted a number of analyses looking at not only upper GI tract cancers but pre-cancerous lesions as well as precancerous lesions and cancers of non-GI tract tissues.

Those data once analyzed did not reveal any safety signal whatsoever, although the entire marketing history of the product, and those data were submitted to FDA as well.

Response to this is that: "Yes, there is no evidence." The nature of the kinds of studies, one could argue, may not be the perfect scientific result, but this would be utterly impractical to perform.

DR. DAVIDOFF: Thank you very much. That is somewhat reassuring, although those studies are obviously very difficult to do and hard to interpret.

I have on other question. It has to do with the related issue of the potential masking of symptoms and the delay in diagnosis of GI cancer because in the data presented to us again in the 2000 material from the FDA, it was pointed out that there were in the patients taking the drug 49 cases of adenocarcinoma of the stomach occurred and, "In at least four of these cases, OMP therapy caused masking of symptoms and/or temporary healing of gastric mucosal with a one to 12 month delay in diagnosis of malignancy.

I note that this occurred in patients who, of course, were already seeing physicians because that's where they got the prescription for the PPI. I would ask the question, therefore, wouldn't the risk for the masking of serious symptoms like PPI be somewhat greater than this in the OTC environment, particularly since we know the post-marketing reporting of such events is notoriously incomplete.

DR. TRIEBWASSER: Again, the sponsor perspective is that the screening and surveillance detection of upper GI tract cancers are amazingly difficult to do. The medical community does not know who to screen and how identify as risk populations and that we have a chance to intercede.

I think there are certainly going to be documentation of these rare instances, but the bulk that we rely on are the instructions which are going to be for limited use and certainly strongly advising that individuals in the consumer population keep their physicians in the loop.

I think that this whole area is ripe for a new investigation because I think it's just a very challenging issue even when such individuals are in the hands of physicians.

DR. DAVIDOFF: Thank you.

DR. CANTILENA: Dr. Brass.

DR. BRASS: I would like to begin -- actually, do you know the soccer score?

DR. CANTILENA: No.

DR. BRASS: Okay. Thank you. I would like to begin by asking for a clarification from the sponsor. You cited efficacy data from studies 171 and 183 as the basis for the efficacy conclusion. Could you just clarify whether the endpoint cited were the prospective primary endpoints of that study or were they secondary endpoints of that study?

DR. PEURA: The data that I presented on the screen showed both the primary and the secondary. The primary variable was complete prevention of heartburn on day one and our secondary variables were those across the 14 days.

DR. BRASS: Thank you. Second, I think it's extremely likely that if this drug is made available OTC there will be pregnancy exposures despite any warnings. Therefore, could you update us on your experience with pregnancy exposures from your safety database?

DR. TRIEBWASSER: Certainly. We are relying on the data that we have submitted to the FDA in which several large epidemiologic studies have failed to identify a signal among women who were inadvertently exposed to the product during first trimester. This material is currently under FDA review and we've had discussions with FDA.

DR. BRASS: For the committee's benefit, could you give us an estimate of just the magnitude of that experience? How many exposures are you talking about to reach this safety conclusion?

DR. TRIEBWASSER: Sorry. We were able to accumulate data from three large epidemiologic studies where approximately 1,400 women were exposed to the drug and over 1,000 were exposed during the first trimester. It's on the basis of that data that we evaluated and failed to see any signal with regard to fetal risk.

DR. BRASS: Thank you. Then under your proposed labeling under warnings, you indicate that one should notify your doctor if you had heartburn for three months or longer without talking to your doctor. That would seem to capture 100 percent of the intended target population. I was wondering what your experience in the actual use study was compliance with that particular warning.

DR. PEURA: We found that approximately 65 percent of the people that had previous heartburn for greater than three months had seen their physician previously.

DR. BRASS: About their heartburn?

DR. PEURA: About their heartburn specifically.

DR. BRASS: Thank you.

DR. CANTILENA: Dr. Camilleri, Dr. Fogel, Dr. Uden, and others.

DR. CAMILLERI: Thank you very much. I would like to ask for some further clarification regarding the effectiveness of this therapy for the proposed target population. I would like to refer you to figures 62 and 63 in your dossier and also slide 25 which was the day-by-day 14-day efficacy which you demonstrated.

On the one-day response you actually show in your dossier that less than 50 percent actually achieve the desired no heartburn over 24 hours. That is really quite acceptable because we know from the pharmacological action of this drug it is going to take three to five days to really kick in.

I think this slide in particular shows an important point which we should remember, and that that 30 percent or more of these patients do not achieve relief. I keep that in mind as I also note from the 171 and 183 studies that 57 percent of the patients had frequent heartburn return at four weeks. In the Bardhan study 68 percent of patients required three or more courses a year of the omeprazole at the same dose of 20 milligrams.

The clarification I would like, if I may ask this, is is this truly a benefit for this particular population of patients who have what I would regard as a quite significant level of heartburn, more than twice a week occurring over a period of 30 days or more which I think was the conclusion criteria for your particular study.

What is the overall benefit for a patient to receive treatment for two weeks if the likelihood of needing yet another treatment two or three times in that next year is going to be at least 60 percent. Also bearing in mind that only 65 percent or so actually respond to the treatment. Thank you.

DR. BIERER: Well, first I would just like to say that for the people who are using OTC products to prevent frequent heartburn, this is a very considerable benefit to those people because they currently do not have products available to them in the OTC environment that can achieve this kind of symptom prevention. I think for those individuals it is clearly a benefit.

I think the question then of are there additional benefits for those people if they use the product over and over, certainly for those periods of time that they use them they will derive some dramatic benefit from those products.

DR. ZORICH: And I would just add that we do not consider people who are unresponsive to therapy after 14 days to actually be in the target audience and that is why I think a 14-day direction clearly advising them to seek additional physician interaction distinguishes them as what we consider to be the appropriate target audience.

DR. CANTILENA: Go on, Dr. Fogel.

DR. FOGEL: I have a question about potential unintended consequences on utilization of physicians if this drug is approved. We all know that over-the-counter H₂-receptor antagonists have limited efficacy in the treatment of reflux.

The cost of the over-the-counter H₂-receptor antagonists is somewhere between 20 and 30 cents a pill depending on what you buy and the quantity that you buy at any one time. These costs are not covered by insurers or managed care organizations.

The cost of omeprazole in the study that you did I believe was $12 for 14 pills. If individuals have to pay for the medication out of pocket, the odds are after using it once or twice they will seek a doctor to try and reduce their expenses.

Do you have any insights or knowledge about whether insurers and managed care organizations will pay for an over-the-counter omeprazole? If they do, that would actually be a disincentive to see the doctor because the cost of care would be covered by your insurance and you would not have to have a co-pay. You would not have to go through the discomfort of seeing a physician. The question is is there any sense as to whether insurers will pay for an over-the-counter medication?

DR. ZORICH: I would say this is an area in terms of cost effectiveness in the general area of distribution of our healthcare dollars is constantly debated but at this time we don't have any -- we have had no indication if people will be picking this up, particularly when generics will be on board.

DR. CANTILENA: Okay. Dr. Uden. And then, just in general, usually cost of out of bounds for the Advisory Committee.

DR. UDEN: I have questions about the actual use studies and label comprehension studies. It relates to industry established standards and how to interpret those. I saw information up there that the literacy people 49 percent understood the label or were able to self-select.

I think there was 70 -- understood the label and 70 percent were able to self-select. The general warning signs on the label were understood 81 percent of the time. I see numbers like 50 percent, 70 percent, 81 percent in terms of actual use and label understanding.

Has the industry established any standards which would give us some guidelines or use some guidelines as to what is reasonable for us to expect for understanding labels, being able to follow labels, or shall we just leave it to our imaginations?

DR. PEURA: I think as far as industry, there are no set standards for what is desired on label comprehension. I think it has to be determined on a case-by-case basis depending upon what the indication, the warning, and the statement is.

Certainly for some indications you would want a high level of comprehension. In other ones it may not make that much difference and I think we have to look at the risk for each one of those.

DR. UDEN: And where do you fall on omeprazole then?

DR. PEURA: On omeprazole I would say that we have determined it is appropriate, that people do use it appropriately, even the low literate group. I would point out in the low literate group the number they showed of 49 percent, these people were low literate with frequent heartburn.

We presented this scenario to them or a hypothetical question that said, "Now, imagine yourself having infrequent heartburn and you woke up at 3:00 in the morning and you had eaten pizza but you hadn't taken the product for three days before that. What would you do in this situation?"

It's very hypothetical but we found out probably a truer reflection is when we look at the actual use study of people with low literacy that these people actually scored better and they had to meet all six self-selection criteria, not just the one I spoke earlier about infrequent heartburn. That's probably a more real world realistic situation.

DR. CANTILENA: Okay. We have Dr. Geller and other hands, Cohen, Levine. Dr. Geller first, please.

DR. GELLER: I just would like to point out how optimistic your reporting of the actual use data is. I'm beginning with slide 41. 886 patients agreed to purchase the product. I imagine that means they did purchase the product. But then you report on only 758 of them because that is the number that returned the diary. To begin with, here you decrease your population to 87 percent of those who purchased the product.

Then if you take 87 percent of the next group of numbers, then the compliance goes down by about 10 or 12 percent. You have also here reported on the individual compliance, the three conditions. You have reported on only two of them and individually not together.

Then your definition of compliance on slide 43 is quite broad. Then you get an over estimate of the compliance because you define compliance broadly. That reduces the 79 percent to follow the labeling directions by about 10 percent.

Then if you look at slide 45 and talk about the return of frequent heartburn four weeks after the trial, well, now you are reporting on 83 percent of the 87 percent.

I guess it should be said that if you go back to slide 41 the 758 patients who used the product and returned the diary were more likely to be compliant with dosing conditions than all of those others. Those 13 percent of patients that you lost are likely to be noncompliant with the dozing directions.

I think then you are reporting on 83 percent of the 87 percent and, therefore, the percents that have no frequent heartburn are just a misrepresentation of what actually happened from those who actually bought the product. I believe you are reporting very optimistically.

Last, you didn't ask if the patients who were in the actual use study used the product for relief instead of prevention. I don't know if you asked if they used antacids or other drugs to get relief.

I also have a question about the labeling. I would like to know the difference between prevention of the symptoms of frequent heartburn and prevention of frequent heartburn.

DR. PEURA: I think I heard six questions. Let me start with the first one from this graph. From the 866 we did not get diaries back from 96 people. We were actually looking at how did people actually use the product in this situation so that's where we came to the number of the 758. We assume that those people since we didn't have a diary from them, you could assume that they --

DR. GELLER: I think you can assume we are noncompliant.

DR. PEURA: Well, we can't really assume that because I think we would -- I mean, you could make a case that they were noncompliant but I think I could also make a case that they could have been compliant with this. But we do know -- the best number that we do know is from the 758 where we actually do have diaries from those people.

If you can flip to the next slide. Next slide, please, 42 -- 43, the one with the description. This description gave us a range of 80 to 100 percent which I mentioned before is an epidemiologic standard that people use for compliance with an Rx dosing regimen.

The choosing between 11 to 14 doses for 11 to 17 days, well, that seems like a wide range. There is actually less than 1 percent of the people within that range. Less than 1 percent of the people who took 11 doses in 17 days.

Most of the people if they missed a dose took it on the 15th day, not on the -- they may have taken 14 consecutive. Some would have missed a dose. If they missed one dose, they would have taken it on the 15 day so the range is very tight within that.

Onto the next slide, please, 44. In this you had mentioned that the range of -- if I can remember the question -- the range was 79 percent. You thought -- the position was there were three dosing instructions in that making sure they had the right number of tablets per dose, the right number of tablets per day.

As the FDA did, multiply that times the correction factor in this one. That would come out to pretty close to probably about 75 percent because those were in the 90s that were there.

Also, as I did report earlier, if you look at the people who took exactly 14 doses in 14 days, they did exactly what was on the protocol, the number is 63 percent. Not of the 79 but of the total pie.

There was one other question after that before we get to the labeling question.

DR. GELLER: There were three, I think.

DR. PEURA: Pardon? Okay.

DR. GELLER: It was on the percent with no frequent heartburn that now involved 83 percent of the 87 percent of those who purchased the product. That was the first one.

DR. PEURA: Okay. It's the same general theme. We are looking at people that we actually have data upon to make a judgement. I think the labeling question was in terms of between symptoms of frequent heartburn and why do we put the word symptoms of frequent heartburn on the label. That was really --

DR. GELLER: In fact, I did include the word prevention because I think that ends up -- when you say prevention of symptoms, it seems to me it kind of introduces the possibility of taking it for relief.

DR. PEURA: Okay. Let me come back to you asked about how do we know that the people took it for prevention. I believe that was one of the questions in there. The people that took it for the 14-day period, the 79 percent of the people were taking it over a regimen of therapy. They were maybe missing one day within that over the time.

These people were probably not taking it in response to a symptom because they would have to have it every day. They were most likely taking it for prevention since they were taking it on a regimen basis.

The reason that we included the words "prevention of the symptoms of frequent heartburn" is that we did not want to imply that we are preventing frequent heartburn, preventing it was ever reoccurring whatsoever. It's not promising the cure. We're trying to define to the consumers the prevention of the symptoms which is more a consumer term.

DR. LEVINE: Symptoms is more OTC. Symptomatic relief or symptomatic prevention is what this is about.

DR. GELLER: I guess I just don't know what heartburn is if it's not a symptom.

DR. PEURA: It is. Perhaps we are being a bit redundant with it but we wanted to get the message across to the consumer that we are not preventing heartburn from ever reoccurring again. It is symptomatic treatment of heartburn. Prevention of symptomatic heartburn.

DR. GELLER: The two questions you missed were I asked if you asked in the actual use study if anybody used the -- if people used the product for relief. The other question was did you ask about concurrent use of other drugs for treating these symptoms.

DR. PEURA: Okay. Thank you for reminding me. If I could go back to the pie chart again. The pie chart before this, please. 44, please.

DR. GELLER: 45, I think, you want.

DR. PEURA: The full pie. Asked if anybody had used it for symptomatic relief. I think the best answer we can give you is that probably the people in this pie chart here between the 9 percent and the 9 percent looked as if they were using it not on a regimen basis.

Of that 3 percent of these people took only four doses of the product and then stopped. If we take those out, perhaps as much as 15 percent would be the max that probably we are using as a sporadically or for a PRN type basis. And accordingly --

DR. GELLER: Once again, that's of the 83 percent of the 87 percent.

DR. PEURA: Importantly they didn't use it beyond the 14 days. The last question was?

DR. GELLER: I see the data on -- oh, no. Did you ask if subjects also used other products at the same time?

DR. PEURA: We did collect it in the diary whether they used other antacids or H₂-RAs. Unfortunately the way the information was collected in the diary, if the person said they were on an

H₂-RA, it was counted throughout the whole range. We didn't record on a day-by-day basis so we really cannot answer that question.

DR. CANTILENA: Thank you. Ms. Cohen.

MS. COHEN: I have a lot of questions. We have 44 million Americans --

DR. CANTILENA: How about if I suggest that you ask them sort of one at a time.

MS. COHEN: All right. We have -- I will try the best I can because I suffer from GERDs and esophageal rings so I'm a good person to know about diet. The word diet has not been mentioned at all. Could we look at the label? The last time I asked for a copy of the label and the packaging. Did anyone think to bring one today? This is large. I'm sure this is much larger than is going to go on the box. Right?

DR. PEURA: Correct.

MS. COHEN: We really don't know the size of the print or if people can read it.

DR. PEURA: We do know that people can read the print since we did label comprehension studies and actual use studies and we used the actual box that we would market.

MS. COHEN: Thank you. When you look at the label it says "uses." It seems to me you could add, "Not for infrequent cases," like they say on page 21 in this report.

Dr. Bill mentioned three months along that some of these suffered from angina or something else. Three months is a long time to tell people not to see their doctor. "Do not use if heartburn continues after 14 days. See a physician," for those people that can afford to see a physician. Then we have 44 million Americans who can't.

I think it's very important that some foods can cause heartburn and you should check with a nutritionist or a physician or a library because a lot of foods can be eliminated that will stop people from getting heartburn which is a lot less expensive than having to buy drugs.

I don't see anywhere here the importance of diet like tomato sauce or red wine. A lot of foods cause heartburn and we can prevent that with people. I think the label, I would really like to see the size of the label how it is.

In reading the report that was put out, my question is it apparently says that a substantial portion of subjects experience no heartburn on day one or day 14 in the placebo group. Now, what was given to the placebo group?

DR. PEURA: Okay. Let me try and answer your questions in order. The first one you asked about diet and diet restrictions. We do have a package insert with this product where we talk about tips for managing heartburn which includes diet, certain foods not to eat, elevating the bed, not eating dinner late at night and before lying down.

This is the label. Take the label off, please. Actually, we do contain that on the package insert. There is too much information that we cannot put on the back label of the package. We have discussed this with the agency. In the back it's called, "Tips for managing heartburn," down here which is what one would want to do. It's a primary one.

Second, you had said that some people may have chest pain or angina. We do clearly say in the label under "do not use" to ask a doctor before use if you have chest pain with shortness of breath, sweating, pain spreading to the arm, etc., things that could be confused with potential heart attack.

That we have included and believe it is important to include that on the label. In fact, it probably ought to be included on the label for all heartburn medications OTC.

The last one was -- I have forgotten your last question.

MS. COHEN: I was interested in the placebos. If people were give just over-the-counter antacids.

DR. PEURA: They were given the placebo called the matched placebo, everything that was in the active pill with the exception of the omeprazole, the active ingredient.

MS. COHEN: Well, they said that there was a 40 percent treatment failure rate after 14 days in subjects with high-frequency heartburn.

DR. LEVINE: Forty percent.

DR. PEURA: Right. For some people placebos do work well for heartburn. As Dr. Weintraub once told us, a glass of water works fairly well for some people. For this the failure rate in omeprazole does not work in every patient. There is a therapeutic range in which it works. The failure rate is the 30 percent of people at the top of that graph.

MS. COHEN: Your statistics all involve people in the study itself. That is already a special class of people and it's not the typical and average consumer who would use the product. These people are already more conscious. They are in a study and they should be doing things that are expected of them. How does that represent the average and typical consumer?

DR. PEURA: In our actual use study we actually ask people how they evaluated the product. We asked them in terms of their global understanding was this product good, very good, excellent, poor. We found that 90 percent of the people rated the product good, very good, or excellent.

MS. COHEN: And these are the people in the study?

DR. PEURA: They are people in the actual use study.

MS. COHEN: So this is not the person who would go in and buy it?

DR. PEURA: It is the person who would go buy it.

MS. COHEN: But these are the people that knew about it because of the study you were doing.

DR. PEURA: They were recruited from a low intercept and asked, "Do you get heartburn?" They were close to a purchase decision that people would want to make in a drug store or outlet store.

MS. COHEN: Thank you very much.

DR. CANTILENA: Dr. Levine.

DR. LEVINE: I'll ask one question. In reference to slide 77, you mentioned that the majority of consumers in slide 78 won't be using omeprazole chronically. There is a subgroup, of course, who have chronic heartburn, mainly patients with gastrosophic geo-reflux disease.

In slide 77 you show Chiba's work. I believe that was with 20 milligrams but most of us recognize that it's at least 30 days or so where there is complete healing, much better healing, then what is shown in this particular slide at four weeks versus two weeks.

While we will discuss that later about the duration, as well as possibly dose of undertreatment populations, do you have any prospect of data in your studies that go longer than the 14 days to give us an idea, or other literature because the literature that I am familiar with clearly shows a better response rate between two and four weeks.

DR. ZORICH: Yes. I think that -- I don't mean in anyway to imply that this data should be taken very literally that there is a flattening here. I think it's more than fair to consider that if you smooth this line out that there is, indeed, a benefit.

But if you take into account the fact that there is also benefit that accrues with placebo, you can see that while there is a benefit, it is not as much of a benefit as you might anticipate.

But I think more importantly than simply t his data which is, as I said, a medianalysis of 43 studies, we could look at each one of the individual studies and each one would support that there is an incremental benefit as you go longer.

More importantly, we are not seeking to treat people with erosive esophagitis with this. We believe that those people should remain within the medical care system getting their medication from their physician. That is even more reason why people who are not responding to 14 days should be directed to their healthcare professional for evaluation.

I think 14 days in the OTC environment is a very logical place to say if you are not responding by 14 days, you may very well indeed have higher grades of erosive esophagitis best managed with the advice of your physician.

DR. LEVINE: Thank you. We'll discuss that later, I think.

DR. CANTILENA: Yes, we will.

Dr. Alfano and then Dr. Cryer.

DR. ALFANO: Yes. The reference is at slide 37. On slide 37 you list 385 people who elected not to participate. As I recall, you said some of them elected not to participate because they wanted to check with their physician first. How many of the 385 wanted to check with their physician first?

DR. PEURA: About a third of these people.

DR. ALFANO: And yet that was excluded from your analysis. In other words, these are people who said the product was appropriate for them but selected out before they ever hit your database.

DR. PEURA: Correct. We excluded them from our analyses.

DR. ALFANO: The second question is slide 57. Dr. Levine states that the increased incidence of adenocarcinoma beginning in the '70s is not related to acid reducers. This is related to a question I had on something Dr. Wolfe showed us where he showed the epidemiological trend, which coincidentally ended, at least in his slide, before omeprazole was launched in this country.

My question is what is the basis, Dr. Levine, that you say that this change is not related to acid reducers?

DR. TRIEBWASSER: Actually, several lines of evidence. First, the superficial look at the time relationship to this cancer and use of products isn't sufficient to really draw any correlation. In fact, the rising incidents of this cancer predated the introduction of the H₂ blockers. The H₂ blockers, as I'm sure you can appreciate, were introduced initially for treatment of peptic ulcer disease and GERD.

In addition, the initial introduction of omeprazole is the first PPI that was conservatively introduced for individuals with hypersecretory conditions like Zollander Allison syndrome and fully responsive in peptic ulcer disease. There was probably several years, in fact, of this rising incidence that bears no relationship from an epidemiologic perspective to the use of these drugs.

In addition, there have been careful epidemiologic studies that have, in fact, looked at the relationship of this type of cancer and acid reducers and the acid reducers basically seem to go along with the underlying condition which does increase the risk which is chronic persistent heartburn.

DR. ALFANO: Would it then also be true that these drugs bear no relationship to the decline in squama cell CA.

DR. TRIEBWASSER: I have no evidence to even suggest that, no.

DR. CRYER: I'd like to get back to a comment that Dr. Camilleri made a little earlier which was that in his opinion that heartburn of frequency of greater than two times per week is considerable heartburn and I would agree. I guess one of the things that has really been ostensibly absent from this discussion is a description of frequent heartburn as being GERD, gastro esophageal reflux disease.

As a gastroenterologist I'm having a difficult time understanding the differentiation between frequent heartburn and what we really call the treatment of GERD.

The specific question is in your population of individuals who had frequent heartburn of more than twice a weekly, do we know how many, what was the distribution of those individuals with regard to their actual frequency? Specifically how many had it three times, four times, five times a week?

DR. ZORICH: I would say that within our actual use trial which observed people for a three-month window, what we found is that at the three-month contact that 43 percent of them said they were not having frequent heartburn. Here is a group of individuals who stated they weren't having frequent heartburn. The vast majority of them then took 14 days of omeprazole and when contacted at three months said they no longer had frequent heartburn.

Right there you see almost half of the population saying they are not having frequent heartburn. If you extrapolate that group, even if they were to then the next day after you called them have another bout of frequent heartburn and this went on, that would be perhaps four times a year.

That's why I thought it was very important to bring in data that is more specific to your question like the publication by Bardhan which looked specifically at people who had a diagnosis of GERD.

They did have screening endoscopy at the entrance to the study. Only Grade IV was excluded from participation on the grounds of ethical reasons that these people need healing, to your point, Dr. Levine. What you saw there was that 75 percent of the people actually did well using intermittent.

That brings us to the question that you asked first, are we making a distinction between frequent heartburn and GERD. I think importantly it undoubtedly reflects the same bias that limits our ability to look broadly in the data.

Most of the clinical trials have looked at maintenance of remission and really maintenance therapies dealing with erosive esophagitis. There's only a few trials like Bardhan -- and there's another by Lindh -- that really allows people to elect to take treatment on the basis of symptom occurrence. The Lindh study is another one I didn't show but it shows very similar results.

What you see there is there are a population, and it turns out to be the majority of people, who have episodes -- well, that's not a good word in this setting because it means something else, but they have periods of time when their frequent heartburn is acting up and then is goes into a quiet phase and it may act up again in the future. Only about 25 percent of the people seem to be requiring more chronic therapy.

I think it's this 25 percent of the people who are those that end up in the clinical trials and the ones who are chronically seeking physician care for further intervention who have relapsing symptoms.

We're not targeting the therapy to them but it's a perfectly acceptable therapy for them if they are using it with their physician being knowledgeable about it.

DR. CRYER: Okay. So as you were responding, Dr. Camilleri actually pointed me, I guess, to the briefing document that was provided by the sponsor. From the efficacy trials the mean reported baseline of heartburn frequency was five days per week. Would that be an accurate statement?

DR. PEURA: In our efficacy trials 171 and 183 the average days of heartburn was about 75 percent of the days which would be five out of seven days. The severity was mild to moderate on a three point scale.

You had also asked about the population in our actual use study as well as the people. Did you ask that question?

DR. CRYER: No, I didn't have the question but I would be happy to hear the answer.

DR. PEURA: It's similar population range within that. Less than 1 percent of the people actually had infrequent heartburn less than one day a week and about a third of the people had it two to three days a week. The other third had it five to seven days a week.

DR. CRYER: If I might, just to follow up on the population and the actual use study. I believe you said a little earlier that the population in your actual use study is as close as you can get to a population that would be making a purchase decision at a pharmacy or a drug outlet store.

Given that you actual use population is fairly reflective of the actual population, I really want to get back to this issue of your low-literate population. Under low-literature population, if I understand it correctly, there was a 50 percent response rate in terms of label comprehension.

You've told us -- you've suggested that when the actual compliance over 14 days that their compliance was somewhat better. I think the description then was 50 percent but I never heard a quantification of how much better than 50 percent was their actual compliance over 14 days.

DR. PEURA: Over the 14 days the compliance was only 2 to 3 percent different than the "literate" population. This is in compliance for using the product appropriately, the one dose per day, one tablet per dose, and also over the 14-day dosing regimen period.

When we look at correct self-selection of those criteria, there was about a 10 percent difference. There was 70 percent for the low-literate group compared to 81 percent for the average group.

DR. CANTILENA: Okay. Thank you. I think we'll have some time as well this afternoon to ask more questions. I just have one which involves the actual use study. At what point in the study did you obtain informed consent?

DR. PEURA: We obtained informed consent after the people made the purchase decision to buy the product.

DR. CANTILENA: Okay. I was wondering if you had a copy of the informed consent document with you?

DR. PEURA: We don't have it with us.

DR. CANTILENA: Okay.

DR. PEURA: We cam probably get you a copy if you would like to see it.

DR. CANTILENA: Okay. If you could have that for after lunch, that would be great. What I would like to do now is -- we are just a little bit over -- take a 20-minute break. Come back at just after 11:00 a.m.

(Whereupon, at 10:45 a.m. off the record until 11:06 p.m.)

DR. CANTILENA: While we're waiting for people to come back, the final score Germany 1, USA 0. But it was a great run.

Our first speaker for the FDA is Dr. Mark Avigan.

DR. AVIGAN: Thank you. I'm a board certified gastroenterologist. Before coming to the FDA I served on the faculty at Georgetown University. There's been a longstanding interest by a number of sponsors to make treatments for the management of heartburn symptoms directly available to consumer in the OTC arena.

On the occasion of the first Joint Advisory Committee that discussed omeprazole magnesium on October 20, 2000, Dr. Larry Goldkind and I presented an overview of efficacy and safety issues related to the use of this product in an OTC setting.

At that time the sponsor was seeking approval for the following indications. First, the relief of heartburn, acid indigestion, sour stomach. Second, the prevention of these symptoms brought on by consuming food and beverages or other inciting events.

Third, the prevention of symptoms for 24 hours.

Both FDA reviewers and a majority of the Advisory Committee attendees concluded that results of the studies performed by the sponsor did not demonstrate efficacy for the first two listed indications.

In the case of treatment of episodic heartburn as a symptom, neither multi-center placebo controlled trials 092 or 095 revealed superiority of a single 20 milligram dose of omeprazole magnesium over placebo.

These studies contained over 600 subject with a history of heartburn occurring at least two days per week in each treatment arm. Although four hour prevention of meal-induced heartburn by single dose of omeprazole magnesium was demonstrated in a 1,200 subject multi-center double-blind placebo-controlled study. That study is 006.

This result was not replicated in a separate study, study 005, which was virtually identical in its design and execution. In contrast to the absence of efficacy in studies for the first two indications, results of studies 171 and 183 supported the third claim, the prevention of symptoms of 24 hours.

It is these two studies that the sponsor is now resubmitting for consideration of the newly proposed indication, the prevention of symptoms of frequent heartburn for 24 hours.

Results of clinical studies of omeprazole magnesium provided by the sponsor can be tied to the mechanism of action of all proton pump inhibitors including omeprazole. Normally omeprazole has a short half-life in the circulation; that is, between one and two hours, because the proton pump molecules, which are irreversibly targeted by omeprazole, may not all be simultaneously accessible to binding by the drug.

The dosing interval with omeprazole is only once per 24 hours and the degree of acid suppression after a single dose is low. In fact, consecutive daily treatment for a few days is required to build up to a maximum PD response.

This characteristic can be contrasted with that of antacids and H₂-receptor antagonists which achieve identical pharmacodynamic effects after each dose including the first.

It is FDA's concern that omeprazole's buildup effect of acid suppression over consecutive daily doses may reinforce continuous unsupervised usage by consumers seeking optimal relief of chronic heartburn.

In the remaining time that I have I will touch on the following areas. First, the safety profile of omeprazole with regards to short-term and long-term drug exposure.

Second, findings of previously submitted studies which reflect on the potential for long-term usage of this product in an OTC setting without physician supervision. Third, the results of pivotal studies 171 and 183 measuring the effects of omeprazole magnesium on symptoms.

Fourth, highlights of early symptom and drug usage patterns of the actual use study 007. Finally, some of the outstanding issues surrounding approval of the drug for OTC use that the advisory committee must address.

An analysis of the safety record of the drug is supplemented by an array of clinical studies in post-marketing surveillance data of the entire code of prescription formulation.

Short-term administration of omeprazole has been linked to a number of serious adverse events. Rarely these may be life threatening and include severe hypersensitivity reactions such as angioedema and anaphylaxis, toxic epidermal necrolysis, agranulocytosis, and clinically significant hepatic dysfunction.

Although the precise incidents of these adverse events cannot be determined from a voluntary post-marketing reporting system, it appears that they are quite rare and similar to serious adverse event rates associated with some other OTC approved products.

Omeprazole magnesium has also been associated with mechanisms that may lead to clinically significant drug-drug interactions. Competitive inhibition of CYP 2C19 metabolism and gastric acid neutralization by the drug are each known to affect pharmacokinetic profiles of certain agents.

The potential for critical omeprazole induced increases in circulating levels of some drugs such as warfarin, phenytoin, diazepam, digoxin is small but it can be further minimized by appropriate consumer labeling. Similarly, the more likely disruption of effective levels of antifungal such as ketoconazole can be minimized by labeled instructions to consumers.

A separate series of safety concerns that were raised during the first Joint Advisory Committee meeting are relevant only to long-term continuous or intermittent self-administration of omeprazole without physician supervision. These were discussed because of the pharmacological properties and potential for such usage which I just mentioned.

Safety concerns tied to long-term usage include the following. First, there is a potential of the drug to mass symptoms associated with underlying medical conditions that warrant early diagnosis and adequate treatment.

These include severe forms of erosive esophagitis, Barrett's, metaplasia and dysplasia, or cancer of the esophagus or stomach. In some individuals with these conditions, a significant delay and physician referral and patient evaluation may lead to worse outcomes.

A second safety issue related to long-term unsupervised administration of the drug is absence of prospective controlled trials to determine whether such exposure confers an increase in the absolute risk for the development of certain neoplasia in a large population of users.

All proton pump inhibitors induce increases in circulating gastrin concentrations which have trophic effects on some mucosal cells. In addition, these drugs may have genotoxic effects in a variety of cell types when in an activated form.

The concern about the potential for significant numbers of consumers to engage in long-term self-administration of omeprazole magnesium without physician supervision despite labeling for only short-term use was prompted because of the following points.

First, the drug is intended to prevent recurrent episodes of heartburn in individuals with frequent symptoms rather than effectively treat episodic symptoms.

Both in the clinical efficacy studies and actual use studies based on clinical characteristics, it was not possible to assert that many of the enrolled individuals with frequent symptoms did not have gastro esophageal reflux disease referred to as GERD which is a chronic and often long-term condition.

As described by Dr. Castel at the first Advisory Committee meeting, individuals with long-standing heartburn and the spectrum of complications of erosive GERD and the severity of mucosal changes cannot be consistently correlated with symptom severity or frequency.

Second, the recurrence rates of heartburn in individuals with GERD are high within a short period of time after cessation of acid suppression treatment. Third, in a national usage study that was presented at the first Advisory Committee meeting more than 60 percent of individuals using omeprazole magnesium for the prevention of heartburn exceeded 10 consecutive days of treatment despite a labeled instruction not to treat beyond that point.

This is not surprising since large percentages of individuals who have self-selected for OTC treatment with omeprazole magnesium in those actual use studies have had histories of long-standing heartburn that are consistent with the diagnosis of GERD.

It should be noted that at the first Advisory meeting the panel was split on the question of whether chronic heartburn or GERD is an acceptable OTC indication. The panel decided that sufficient evidence had not been provided to support either a favorable benefit risk assessment or approval for any of the three possible indications of acute symptomatic heartburn, prevention of episodic heartburn, or chronic heartburn.

In the current submission the sponsor has proposed an indication for prevention of symptoms of frequent heartburn for 24 hours and only for those who suffer heartburn two or more days a week.

Studies 171 and 183 were double-blind placebo-controlled two-week treatment studies which contained approximately 500 subjects in each treatment arm. Inclusion criteria included the presence of heartburn at least two days per week for one month prior to enrollment.

Although the primary efficacy variable was no heartburn over 24 hours between the first and second daily dose following randomization, heartburn free 24-hour periods over each subsequent treatment day and during the single-blind placebo follow-up phase were also measured.

These studies were associated with the following findings. First, there was a substantial proportion of studied subjects who experienced no heartburn despite treating with placebo both on day one of treatment, 32 percent, and on day 14, 43 percent.

Second, although statistically significant, the therapeutic gain of omeprazole magnesium 20 milligrams versus placebo was only 16 percent on day one but increased to 29 percent on day four confirming that the maximal pharmacodynamic benefit of treatment relies in consecutive daily dosing.

Finally, even after 14 days of treatment with the 20 milligram doses, almost 30 percent of subjects experienced break-through heartburn.

It is significant that the frequency of heartburn symptoms prior to treatment had a substantial impact on the rates of response to omeprazole magnesium and placebo.

In subjects with pretreatment heartburn that occurred less than half the days, 50 percent of the days, the difference between drug and placebo response rates referred to as the therapeutic gain was only 4 percent on day one of treatment.

This small difference was due to a placebo response rate that was over 65 percent. Even by day 14 of treatment the therapeutic gain in this group of subjects did not rise above 11 percent since the placebo response rate was over 70 percent.

In contrast, subjects with daily heartburn prior to treatment demonstrated more robust differences between drug and placebo in response rates. These differences reflected substantially lower response rates to placebo when compared to the group with less frequent heartburn. In the daily heartburn group on day one of treatment, the therapeutic gain was 18 percent and by day 14 it rose to 39 percent.

The conclusion that can be drawn are consistent with omeprazole's function as a potent inhibitor of gastric acid secretion and the important role that it can play in the management of severe forms of gastric esophageal reflux with associated frequent symptoms.

These conclusions can be summarized as follows: First, in subjects with low frequency heartburn at baseline the therapeutic gain from drug was small because of a high placebo response rate. Second, the therapeutic gain was greatest in subjects with daily heartburn because only a small percent responded to placebo.

Nonetheless, despite the higher therapeutic gain in these subjects there was a 40 percent break-thru rate of heartburn on the last day of treatment with the drug.

The American College of Gastroenterology has issued published guidelines for the diagnosis and treatment of GERD. These include the following: GERD is characterized by chronic symptoms or mucosal damage produced by the abnormal reflux of gastric contents into the esophagus. Furthermore, many, perhaps most patients, with GERD require long-term possibly life-long therapy.

Based on the distribution of frequency of heartburn prior to treatment in the sponsor studies, it is likely that many of the study subjects had GERD. Therefore, it is not surprising that in both studies after cessation of treatment with omeprazole magnesium the recurrence of heartburn was rapid. Within three days the apparent therapeutic gain compared to placebo disappeared and the daily percentage of subjects with heartburn over 24 hours rose to approximately 55 percent.

In addition to three label comprehension studies, the actual usage studies 007 that measured characteristics of individuals who chose to purchase this product and their usage of the product of a duration of eight to 12 weeks has been provided.

Most patients who self-selected for OTC treatment in that study who had GERD is supported by the following observations. First, among the treated population more than 90 experienced heartburn for more than one year and almost 50 percent for longer than five years.

Second, 57 percent of these subjects experienced heartburn four or more days per week. Third, a substantial percentage of subjects used other products or prescription medications to relieve heartburn when symptoms recurred after completion of the 14-day course of treatment with omeprazole magnesium.

It should be emphasized that the actual use study did not measure the potential for long-term intermittent usage of the product. More details about results of study 007 will be described by Dr. daiva Shetty.

In summary, omeprazole magnesium will be used by substantial number of individuals with GERD. In the proposed labeling the consumers were warned to "notify your doctor if you have had heartburn for three months or longer without talking to your doctor," and instructed to "stop use and a doctor if heartburn continues or returns after using this product everyday for 14 days." The consumer is also instructed, "Do not continue beyond 14 days unless directed by a doctor."

The advisory committee must address the following issues. First, whether a single two-week treatment course of omeprazole magnesium in an OTC setting meets the short-term and long-term needs for acid suppression of individuals who purchase this product.

Second, whether occasional courses of treatment in an OTC setting without physician consultation are consistent with the sponsor's proposal. Finally, whether limitation of usage to a single 14-day treatment course is an important feature to protect the safety of consumers.

If so, it must then be determined whether the sponsor has provided adequate information about the potential for long-term continuous or intermittent use of this product without physician supervision to ensure a favorable benefit risk assessment.

Thank you. Now I want to introduce Dr. Karen Lechter from CDER's FDA's Office of Drug Safety who will discuss label comprehension studies.

DR. LECHTER: I'm going to talk to you briefly about the purpose of label comprehension studies. I'll then discuss the two standard Prilosec label comprehension studies.

Excuse me. This mouse is not working. I'll have to use -- I'll just use the button. Thank you.

I'll focus primarily on the issues about which we have concerns where there is clear evidence of problems and where there is an adequate evidence to draw conclusions. I'll then discuss the implications of the results for the label.

The regulations state that OTC labels must be likely to be read and understood by the ordinary individual, including individuals of low comprehension, under customary conditions of purchase and use.

As one way to satisfy this requirement sponsors conduct label comprehension studies to test how well their proposed label communicates. Sometimes this i done as an iterative series of studies with the label changes being made after the study and then the label being retested.

In some cases this goes on for several rounds. For the Prilosec OTC product the sponsor conducted two standard label comprehension studies. The first was study 02255. There were 684 persons in this study, 43 percent male. There were five cohorts. 297 were in a general population.

Two cohorts were frequent heartburn sufferers. One was low literate which was 8th grade reading level or lower of which there were 162 members. Another was high literate and there were 155 in this group.

Frequent heartburn suffers were those who had heartburn two or more times a week or who were taking a prescription medicine for heartburn. The fourth cohort were 96 heartburn sufferers taking drugs listed on the label as requiring physician advice.

The fifth cohort was 42 pregnant or nursing heartburn suffers. These participants examined the label and answered questions about it with the label available for reference.

When asked about the product purpose, 39 percent of the general population were completely correct. They said "prevent frequent heartburn." The two literacy groups had similar percentages of completely correct responses.

As this was an open-ended question, we're not as concerned about getting a complete response as we would be for other types of questions. However, these responses may reflect problems in understanding all the aspects of the indication and we need to look at questions in which the label information is applied to learn if there is a problem understanding the indication.

All of the hypothetical scenario questions about use for episodic relief or prevention should have been answered that the product is inappropriate. However, only about half of the responses were correct about episodic use. About half answered that the product could be used for prevention or relief of individual heartburn episodes.

For the three questions about episodic relief, the correct scores in the general population ranged from 48 percent to 55 percent. An example of these questions was, "You ate chili for lunch. The chili gave you heartburn. You have not had heartburn before. You want to take something now to get rid of this episode of heartburn. Based on the label, is this product intended to be used for this situation of heartburn or not?"

For the two questions about episodic prevention in the general population, scores ranged from 54 percent to 61 percent. An example of this type of question is, "The food you brought for lunch today usually gives you heartburn. You would like to take something just for today before lunch so you don't get heartburn. Based on the label, is this product intended to be used for this situation of heartburn or not?"

Participants were asked if the product was intended for them personally to use. We call this the self-selection question. Frequent heartburn suffers with symptoms listed on the label as requiring physician consultation before using the product were correct only 41 percent of the time. Frequent heartburn suffers taking medications listed on the label as requiring physician consultation before use were correct only 50 percent of the time.

This rose to 82 percent after they were given a list of brand names that correspond to the generic names on the label. However, we believe that the 50 percent figure is more valid because consumers selecting OTC medicines in the store would not have a list of brand names.

Overall 67 percent of those who should not use the product or should consult a doctor first were correct in the self-selection question. The cohorts of non-heartburn sufferers, infrequent sufferers, those allergic to the product, and pregnant or nursing heartburn sufferers responded correctly at the rate of '76 percent or greater.

However, we are concerned that the self-selection responses suggest there is a problem in applying the label to one's self when one has symptoms listed on the label or is taking medications listed on the label.

On the other hand, scenario responses based on hypothetical situations about use with listed medications or medical conditions suggest good understanding with correct responses generally in the 90s.

But the high positive results may be due to an artifact of the study in which almost all questions required a response that the product should not be used or a doctor should be consulted. This could have created a nay-saying basis in which responses are likely to be influenced by this artifact rather than by knowledge of the label.

In conclusion, in study 02255, the tested label failed to convey adequately that Prilosec 1 is not for periodic use, that it is not for acute symptoms or for prevention of meal-induced heartburn.

This conclusion is supported by data I presented earlier that only 48 to 55 percent correctly said the product could not be used for relief and only 54 to 61 percent correctly said it should not be used for episodic prevention.

Also it is not clear that people can apply the label well to their own situation if they take any of the medicines listed on the label or have any of the health conditions listed on the label as requiring physician consultation before using the product.

Further testing would help determine if the proposed label works better than the one tested in this study. However, the proposed label does not address the prevention and episodic issues any differently than the tested label. Improvement on these issues is not likely.

Study 12179 was designed to see if people who should not use the product without a physician's advice due to medical conditions would understand that fact and if they understood the indication and understood the label information well enough to apply it to three hypothetical situations in which people should consult a doctor before use and in one situation which consultation is not necessary.

There were 145 study participants, 41 percent male. All had frequent heartburn. All had at least one condition mentioned on the label as requiring physician consultation. They were not taking medications listed on the label as needing physician consultation.

The label used was similar to the final proposed label but the tested label listed six medications requiring physician consultation rather than the three that were on the final proposed label.

We analyzed the results of this study differently than the sponsor did. We eliminated 40 of the 145 participants. There were two participants taking Prilosec but we were only to identify only one of those so the others still are in our analysis. We did remove the one that we could identify.

We removed those who should not have been in the study at all because they did not have a condition that required physician consultation before use according to the label. These included those with infrequent chest pain or infrequent wheezing. The label said those with frequent chest pain or wheezing should consult a physician.

We analyzed the results for the 105 remaining who should not use the product without consulting a physician. All should have said they would consult a physician before using the product.

The sponsor scored anyone who had ever discussed their condition listed on the label with a healthcare professional as okay to use the product. These conditions included frequent chest pain, chest pain with other specified symptoms, trouble swallowing food, frequent wheezing, and wheezing with heartburn and unexplained weight loss.

Unlike the sponsor we did not believe that having ever discussed a nonheartburn medical condition with a healthcare professional is a surrogate for getting approval to use Prilosec 1. There is no evidence that these participants ever got or would have received approval to use Prilosec 1. All of the 105 in our analysis should have said they should not use a product or should consult a doctor first.

More than half of the participants in our analysis answered incorrectly about whether they could use the product based on the label. Forty-five percent answered correctly. Of these 26 percent said they would ask a doctor and 19 percent said they would not use it.

When asked an open-ended question about the product purpose, about one-third of the participants gave the complete response "prevent frequent heartburn." A series of four questions asked about whether people with certain medical conditions could use the product, three of these conditions were listed on the label as requiring medical approval.

The other, headache, was not listed. Scores for these questions were in the 90s. however, one-third of participants said a doctor should be consulted if the person has headache. This suggest participants were very conservative in their responses and may not have been responding as they would in normal use. It suggests that the scores for the other conditions may have been inflated.

After these studies the label was not changed to improve communication about nonepisodic use and use only for prevention. However, the list of medications requiring a doctor's consultation was shortened in the proposed label from six to three.

The list of medical conditions in the proposed label is shorter and more bulletted than in the 02255 study and is the same as in the 12179 study. We do not have evidence that the proposed label communicates the problem messages any better than the labels tested.

These studies suggest that participants understand Prilosec 1 is for frequent heartburn. Do not use Prilosec 1 if you do not have heartburn, have infrequent heartburn, are allergic, or are pregnant r nursing.

I mentioned in the beginning of my presentation I would focus on areas that concern us. Therefore, I did not mention that there was evidence of good understanding of some other aspects of the label information. This included that the product should not be used if you have trouble swallowing, chest pain with other symptoms, chronic cough, black tarry stools, unexplained weight loss, you are under age 18, or you have heartburn that has become worse with nausea and vomiting.

Despite some of these good results, these studies do show that consumers believe Prilosec 1 can be used episodically for relief of acute heartburn symptoms or to prevent meal-induced heartburn. Further, it is not clear if consumers with medical conditions listed on the label were taking medications listed on the label would understand they should seek medical advice before use or decline to use the product.

The actual use study has similar results. Dr. Daiva Shetty from the OTC Division will now discuss the actual use study.

DR. SHETTY: My presentation briefly covers some highlights of the regulatory history of over-the-counter Prilosec program, the proposed label, target population, and the results of the actual use study 007.

As you have already heard Dr. Mark Avigan explain some aspects of the regulatory history of OTC omeprazole. I will summarize the differences between the original and the resubmitted NDA.

There were multiple changes made to the original NDA. The dose was increased from 10 to 20 milligrams. The target population from anybody above 12 years of age with heartburn symptoms was changed to 18 years and above with frequent heartburn symptoms two or more days a week.

The initial proposal had relief as well as prevention of heartburn claims. The current submission has only the prevention of frequent heartburn indication. The duration of treatment was extended from maximum of 10 days of intermittent use to 14 continuous days.

In support of the current submission, the sponsor has provided the results of one actual use study, three label comprehension studies, a new proposed label, and a safety update.

The study 17859 called deselection study was classified as label comprehension study. Actually it was a marketing study. Therefore, the data from the study will not be presented.

Now I'm going to talk about a proposed OTC label and the target population. The label used in the actual use study was very close to the label proposed for OTC marketing. The use section on the label states that Prilosec will prevent frequent heartburn for 24 hours in people who experience heartburn symptoms two or more days a week.

The directions also stated anyone who is 18 years or older should take this drug one tablet a day every day for 14 continuous days and directs to consult a doctor if symptoms return after this 14-day course of therapy.

There are multiple warnings listed on the proposed label. I would like to draw your attention to one of the warnings. It's called heartburn warning and it states, "Notify your doctor if you have heartburn symptoms for three months or longer without talking to your doctor." I will refer to this warning once again when I discuss the findings of the actual use study.

Now I'm going to present the results of actual use study 007. There are certain actual use issues for OTC Prilosec. First of all, are consumers able to self-select and deselect appropriately? Do they understand what precludes them from the use of Prilosec? Are consumers able to treat themselves to follow label use directions for duration of use and do they follow directions when to seek advice from a healthcare provider?

The actual use for the 007 was a three-month duration multi-center open-label, all-comers with minimal inclusion and exclusion criteria. It was intended to assess how consumers would use omeprazole in naturalistic OTC conditions following proposed labeling instructions. It did not address all the issues that the agency is concerned about.

On the next few slides I will try to walk you through the disposition of the study subject. A total of 1,301 subjects participated in self-selection interview. After looking at the package the majority of them, 1,251, stated that Prilosec is appropriate for them to use.

The purposes of this presentation, subjects who self-selected that the product is appropriate for them to use will be called self-selection population. Unlike the sponsor, we believe that this population should have been the primary population for analysis of self-selection behavior.

I will later point out what the sponsor's definition of the self-selection population was. Of those who self-selected that the drug is appropriate for them to use, 683 chose to participate in the study by agreeing to sign some consent to buy the drug, to fill up a diary, and return for end of study follow-up visit. Three hundred and 84 subjects elected not to participate in the study.

The reasons why 384 subjects stated that it is an appropriate drug for their use but chose not to participate are listed on the slide. One-third of them stated that it is inconvenient for them to participate in the study. More than a quarter of them, 104 subjects, stated that they would not try new medicine without a physician's approval.

These groups actually could use the product if it were to become freely available over the counter. Of those 863 subjects who elected to participate, 854 purchased the drug and received one or more diaries. Nine subjects did not meet inclusion criteria and were not allowed to enter the study. Four of those nine did not provide the consent, one was pregnant, and four previously participated in similar studies.

Seven hundred and 62 subjects completed the study by returning one or more diaries. Ninety-two subjects did not return diaries. Majority of them were lost to follow up. They were a minimum of five attempts by phone and at least one letter sent trying to locate these people.

Of the 762 subjects who returned diaries four returned blank diaries and 758 kept a record of the study drug use. Those who had the record of steady drug use will be called the treated population. I will focus on it talking about compliance with the label use directions.

Of the 758 treated subjects, 649 were available for a three-month follow-up. This final follow-up contact was done by phone and 109 subjects could not be reached.

If we are going into the results of self-selection behavior, I would like to show you the difference between our and the sponsor's primary population for self-selection objectives.

we believe that subjects who participated in the initial self-selection interview and stated that Prilosec is appropriate for their use, it's the actual over-the-counter population that would not be objected to further screening as it was done in the study.

In this presentation the sponsor called those subjects who decided to participate in the study as their self-selection population. However, in the background package to the agency, the sponsor's definition of self-selection population included treated subjects plus additional 12 subjects that were excluded from the study by the investigator.

These subjects not only selected a drug for their use but also had to sign a consent that they agreed to purchase the drug, fill up a diary, and return for a follow-up visit. Using sponsor's population for self-selection objectives have sufficiently increased correct self-selection rates.

Demographically the self-selection population was fairly representative of the overall U.S. population with 59 percent being female. The age of the participants range from 16 to 91 with a mean of 48 years. The majority were caucasian, 65 percent, and 18 percent were African American. Low-literacy group consisted of almost 10 percent of the self-selection population.

Looking at the heartburn study for follow-up with the self-selection population, you can see that majority of them had long-standing heartburn. Over 90 percent of the self-selection population had heartburn symptoms for over a year and almost half, 45 percent, over five years.

Most of them had frequent heartburn as defined by the sponsor, two or more days a week. However, 14 percent of the self-selection population had heartburn symptoms one day or less a week and, therefore, failed correct self-selection.

Analyzing self-selection behavior and compliance with the label used directions the sponsor incorporated one variable, the consultation for heartburn that the healthcare provides them.

I would like to point out what the sponsor's definition of the consultation with the healthcare provider was. It included advice from a physician or any healthcare professional, or the use of any prescription heartburn medication anytime in the past.

This contact was not verified by the study personnel. Therefore, we don't know what particulars were discussed or what advice was given by the healthcare provider.

As you recall, the label states to talk to your doctor if you have had heartburn symptoms for three months or longer. This pie chart shows that less than half of the self-selected population consulted healthcare provider for their heartburn within a year.

Additionally, 17 persons consulted a healthcare provider more than a year prior to the study. Thirty-seven percent did not speak to their healthcare provider about their heartburn at all.

There was a similar ratio of these subgroups for the treated population, those who purchased and used the drug.

The correct subselection was based on the sponsor's predefined criteria. The subject had to be 18 years or older with heartburn symptoms at least two days a week, not pregnant, not allergic to omeprazole, not having certain contraindicated conditions, and not taking contraindicated drugs listed on the label.

The correct self-selection was 76 percent for the self-selection population which included, as I mentioned, subjects who stated that Prilosec is appropriate for them to use. Two hundred and 90 subjects failed correct self-selection for the following reason, 169 experienced heartburn one day or less a week.

There were certain relative contraindications listed on the label, yet some consumers with those conditions are taking the list of drug-selected Prilosec for their use.

One hundred and thirty-four were having certain contraindicated symptoms that were listed on the label. Fifteen were using contraindicated medications. Three were less than 18 years of age and one was pregnant.

Of those 854 subjects who purchased the product the majority purchased only one carton of 14 tablets. There were a few that purchased more than one carton. Even though the subjects were allowed to purchase up to four cartons, the limit of 14 tablets in the package have impacted their pattern of use.

Over all compliance with the label use directions was achieved by 63 percent of the treated population. Those were the subjects who purchased the drug and used the drug and returned diaries with the record of use.

Unlike the sponsor, we believe that compliance subjects had to follow all three label use directions, take one table a dose, one dose a day for full course of therapy.

The compliance rate increased significantly from 63 to 79 percent when the sponsor changes the criteria for the compliance with 14-day regimen. The sponsor considered compliant dose who took 11 to 14 doses in an 11 to 17-day period. We believe that such an analysis increases the compliance rate.

The majority of noncompliant subjects took the drug less than 14 days. Nine percent took more than one dose per day. Four percent took more than one tablet per dose. three percent exceeded 14 consecutive days.

The response to the three-month follow-up questionnaire showed that more than half, 57 percent of the subjects available for follow-up, had their heartburn symptoms return. When these subjects were asked what they did after their heartburn returned, 20 percent stated that they talked to a healthcare provider or made an appointment to see one in the future.

Forty-six percent started using antacids, 27 percent switched to prescription heartburn medication, and 21 percent used over-the-counter acid reducer. This suggest that subjects who used Prilosec already had access to over-the-counter as well as prescription heartburn medications.

The study had several limitations. It was a relatively short duration total of three months. It did not address a question if Prilosec is likely to be used intermittently, a few courses over a year, and what the consequences of such a use would be.

It did not address the concomitant use of other heartburn medications. The methodology of the study did not allow us to assess if consumers understand that this drug is for relief of acute -- is not for relief of acute heartburn symptoms and what consumers would do if Prilosec does not relieve their symptoms.

Overall conclusions that can be drawn from this study of who and how would use over-the-counter Prilosec would be summarized as follows:

Most of the consumers who self-selected to use Prilosec had a long history of frequent heartburn. Even though the label stated to see a healthcare provider prior to the use of Prilosec more than a third of those subjects did not do so.

More than half of the treated population available for follow-up had their heartburn symptoms return. The majority of them switched to other prescription or over-the-counter heartburn medications. Twenty percent decided to seek advice from a healthcare provider.

It is unclear how the interaction with the healthcare provider prior to or after the use of Prilosec would have influenced consumer behavior. The study also showed that Prilosec is likely to be used by consumers with contraindicated symptoms and is likely to be used by consumers with infrequent heartburn.

This concludes my presentation and overall FDA presentations. Thank you.

DR. CANTILENA: Okay. Thank, Dr. Shetty. I would ask that Dr. Avigan and Dr. Lechter join Dr. Shetty for questions from the committee. We will now open the discussion, Brass, Johnson, and Goldstein to start.

DR. BRASS: I have three related questions. The first has to do with the issue of recognition of contraindicated medications on the label. This is certainly an issue the committee has struggled with time and time again. The issue of brand versus generic names is not unique to this particular NDA and is also an issue that has been discussed.

But as was alluded to in discussion earlier today, the evaluation of these questions really has to be linked to the consequences of not accepting the information or not processing the information properly.

I would like kind of an integrated assessment from the review team as to whether concomitant use of omeprazole at this dose with any of the medications listed would pose a serious safety problem.

DR. ALFANO: Well, let me try to shed a more clinical perspective on that. There are two ways of looking at that problem. You can look at a whole population and ask what is the incidence of a bad untoward drug-drug interaction and find that it is a low number. Or you can ask who might be in the population susceptible to such an interaction. It really is the second type of approach where there are some concerns.

An example would be in the class of PPIs there's a known interaction as I alluded to with warfarin. Usually that is an interaction which is not clinically very important but there are already some known postmarketing reports for various members of the class of individuals who might have developed increases in prothrombin time that have warranted reporting to the agency with the pages on chronic warfarin who then started a PPI.

In one or two cases, actually developed clinically significant bleeding. If you are asking a frequency question, the answer is generally these drug-drug interactions are not common for the group.

DR. BRASS: Yeah, but, again, has the clinical significance of the interaction between omeprazole 20 milligrams and warfarin been studied and what was the conclusion of such studies?

DR. ALFANO: It has been studied. Perhaps we might get some other comments on this but there have been studies in individuals where they have been challenged with single doses or who have been on one drug and then have been essentially tested with the other. Reassuringly in a small population of tested individuals there were no dramatic effects either on prothrombin time.

But the problem with that is the potential again in numbers of marketed -- if you market this to large number of people are the outliers and the confounding effects of new facts such as is the patient not only someone who is on these drugs but perhaps has also a problem with metabolism because of something else, an isoform difference. I think this is where it is very challenging.

DR. BRASS: Okay. A similar theme question has to do with the contrasting of the two difference definitions of compliance in the actual use study, the rigorous everyday 14 versus the range.

My question is given what we know about the pharmacodynamics and the time course of action of this specific drug, do you believe those differences in definition would translate into meaningful differences in risk benefit assessment?

DR. SHETTY: Probably not. We just took the more conservative approach to see how people followed all three directions on the label.

DR. ALFANO: Again, just a clinical perspective. I think it was already noted by a member of the panel that one of the criteria for compliance that was not a criteria for compliance but is on the labeling is if you've had heartburn for three months go see your doctor first. That was excluded from the criteria.

DR. BRASS: Yeah, I asked that question before because I think that's an example, quite frankly, of a warning that is not very meaningful because 100 percent of the population is going to qualify for it and already has not seen their doctor.

Finally, a question for Dr. Lechter. You identify quite appropriately a number of concerns and limitations in the label comprehension study.

My question to you is after seeing the actual use study and the difference in the results, and given all the methologic differences in those two trials whether you personally have any reassurance that your concerns from the label comprehension context are in anyway mitigated by the actual use context.

DR. LECHTER: I think in general we get better information from actual use studies but this study did have some limitations and we need to take that into account. We still have concerns about whether people understand the episodic use. Some things were not studied in this use study.

DR. CANTILENA: Okay. Dr. Johnson.

DR. JOHNSON: I have two questions that are directed towards Dr. Lechter.

You indicated that with respect to the contraindicated drugs that there was 50 percent comprehension with generics but it went to 82 percent when brand name was given. You suggested or implied that the 50 percent is the meaningful number indicating that brand names can't be put on the label. I'm wondering if that's what you mean and, if that's the case, why? Why can we not put brand names on the label?

DR. LECHTER: I'm not sure. That may be an FDA policy which I can answer. I think typically we don't put brand names on OTC products but perhaps someone else could answer that.

DR. CANTILENA: Dr. Ganley, do you want to take a shot at that?

DR. GANLEY: Sure. In the OTC labeling rule -- it's not in regulation but in the preamble they had not wanted to put in brand names into the drug facts labeling. I don't think it really addressed the issue of contra indicated medications and putting actually the generic and brand name in.

I think Doug Bierer may have noted earlier that they wanted to have some discussions with us based on this results where there is a dramatic improvement in comprehension if you actually put in the brand name. From my viewpoint, I think that is something that could be a consideration. I don't think there is a regulation that says that we cannot do it.

DR. JOHNSON: I think those data are not at all surprising and frankly I am surprised that 50 percent recognize generic names. I would have thought it would have been lower than 50 percent.

My second question relates to the drugs that are on the list. I can't remember whose section of the FDA packet this was in but there were data on itraconazole which had significant interaction. Not quite as significant as ketoconazole but I'm wondering -- I wanted to ask this question of the sponsor and didn't get a chance -- why itraconazole isn't on the list or why you were not recommending itraconazole to be on the list.

DR. ALFANO: Right. I think there are two approaches again. One is to have a comprehensive list and fit it in, as we heard before, in a relatively small service area and add another word because in reality -- I think your point is well taken -- there is an effect where the gastric acid neutralization has an effect on all those related antifungal compounds. To be comprehensive and complete one would then -- if that was the attack that one was taking, one would have to have a complete -- write a complete list.

DR. HOUN: I think we can ask the sponsor your question relating to the itraconazole. We could also ask them the other questions about the number of patients they formally studied on these various contraindicated drugs and the data they have on that. That would be important, too.

DR. CANTILENA: Yes. If you have that on a slide, that would be actually the most helpful.

DR. PEURA: Let me first address the itraconazole question. Itraconazole is listed as a drug-drug interaction in the Rx data package on that. However, for ketoconazole it is not listed on the Rx data package for that. We felt it was important to include ketoconazole in our labeling but not itraconazole.

DR. CANTILENA: Hold on just a second. Both drugs interact one slightly more positively than the other but now we're going to have on the Rx side one drug and on the OTC we're going to have another?

DR. ZORICH: It should be mentioned somewhere.

DR. CANTILENA: One or the other or both.

DR. ZORICH: Well, the important thing is communication to the patient. Since itraconazole does communicate it, they would be aware that they should not be taking omeprazole. When they were prescribed that, it is in that labeling. The question should be whether ketoconazole does, too, but it does not. Since it does not, we felt an obligation to include it on our label.

DR. CANTILENA: As opposed to having itraconazole and ketoconazole on the OTC label?

DR. ZORICH: To ensure that somewhere there is appropriate communication to the person who might be using both.

DR. CANTILENA: But the OTC is sort of a stand-alone.

DR. ZORICH: Yes.

DR. CANTILENA: On the shelf all by itself without -- okay.

Dr. Johnson, do you have another follow-up?

DR. JOHNSON: Yes. I guess I just have a comment. I understand there is limited space to put drugs. I guess my impression would be that it's much more likely you would have a clinically significant interaction with itraconazole and this drug than with, for example, warfarin and this drug.

If you feel there is only room for three drugs, I think there has to be a really critical assessment of what are the three most clinically significant drug interactions because I'm not sure those three are the three that are on the list.

DR. CANTILENA: How about if we do this? As we're going around with the questions, if someone has the actual slides that show the data for warfarin, ketoconazole, phenytoin, for example, that would be helpful for us to actually see that. If you don't have it handy, then we can start with that after lunch.

Dr. Goldstein.

DR. GOLDSTEIN: I don't have a question per se but I have a passing observation that I would like to make. The presentation on communication contained in it both a touch of irony and a touch of perhaps unfairness in the sense that the irony being the sponsor making a good faith effort to include various diseases for this heartburn medication on the label.

The unfairness perhaps is that it is the only one of this group, or any heartburn group. Neither the antacids nor the H₂-RA antagonists have been required to include the series of diseases to the best of my knowledge. I think that needs to be taken into consideration by the panel.

DR. CANTILENA: Okay. Dr. Geller and Dr. Cryer.

DR. GELLER: I have two questions about labeling. The first is in all the references to your doctor here. The verbs are "ask, discuss, and notify" and "see" is not used. I would think "see" is much stronger. I guess my question to the FDA is do you have a distinction about how strong the recommendation to contact the physician should be.

My second question --

DR. CANTILENA: How about if we hold on there and then we'll ask the second one after we hear that.

DR. ALFANO: Well, this in a sense highlights the clinical problem of management of patients with chronic heartburn generally and what the purpose of the labeling is. It really ends up being a rhetorical question for discussion.

Part of the context of that question really has to do with what is the optimal management for GERD and chronic heartburn. As perhaps will be raised later, there is in this algorithm many physicians do empirically treat individuals with a history prior to undertaking if they don't have alert symptoms and so on for a period of time prior to undertaking diagnostic studies.

I think that question should be to some extent asked to the sponsor what their intention is with regard to that wording. Is it to simulate management of patients who otherwise might have seen a physician? Or is the intention as a primer to get into the healthcare system?

DR. CANTILENA: Would the sponsor want to comment on this?

DR. PEURA: I think the purpose of the labeling is really to try to provide as clear a direction as possible to the consumer who might be using this product. In that regard, we do have testing that shows that the word "notify" is actually a more action provoking verb than the word "see." When you show those words to consumers, "notify" gets them to do more.

Since our intent here is to be sure that people who use this product understand that it is important to keep their doctor in the loop for this condition, that would be our choice of wording probably.

DR. GELLER: My other question concerns the process of deciding on what a label should say in an OTC setting. It seems to me that there should be an iterative process if you don't get it right the first time.

When you change the label, it seems to me you should satisfy -- attempt to satisfy all the conditions or questions that have been raised and then go and test it again. It might take more than two attempts. The label change, as I understand it, hasn't been tested again. Is my assessment of the process correct and my assessment of what's happened here correct?

DR. SHETTY: Usually it's not tested if we approve the drug. Now we know the label comprehension and actual use study and the proposal for marketing labels are already close. They are very similar so we know that these people will use the drug that's used in actual use study.

If the decision will be to approve this drug for over-the-counter marketing, unless the committee feels that we need to do another study or do like Phase IV commitment to test the new level before approval, we can request the sponsor to do that study.

DR. GELLER: So you're saying it's not usually an iterative procedure?

DR. SHETTY: No.

DR. GELLER: Okay. Thank you.

DR. LECHTER: Very frequently the sponsor will do a series of tests and change the label and retest. They don't always do that. In this case the last label used in the label comprehension study was, as Dr. Shetty mentioned, tested to some extent. '

Well, actually it wasn't the last label used. It was kind of a cross between the last label used and the label comprehension study and the new proposed label was used in the actual use study.

Ideally if there are concerns after the actual use study, perhaps the label should be looked at again, changed, and retested but that is an ideal situation. It isn't often done.

DR. CANTILENA: Dr. Cryer and then Dr. Uden.

DR. CRYER: So the data that Dr. Avigan reviewed for us were data that were directed towards the initially proposed indication for OTC omeprazole. I'm trying to place that data in the context of newly requested proposed indication which has changed since the previous review.

The question is how do your conclusions change in light of the revised proposed labeled indication?

DR. ALFANO: I don't think that -- I mean, you can discuss these slight nuisances in the proposed changing of wording, the for-24-hours insertion. I think that there are different ways of understanding what for 24 hours means and that becomes a point of language.

I think that the prevention concept in the study is applicable. Basically the difference between the first and the second meeting is that we have excised out the first two indications and we have come back with a focus on the third. I think that, in my view, it follows.

DR. UDEN: I would like to get back and follow up just a little bit on what Dr. Geller started here. I'm going to follow up on my question that I asked the sponsor earlier on. I was not completely satisfied by the answer that I received in terms of endpoints.

I don't know if this is the time to talk about it, and maybe we should talk more about it later, but I think the FDA or the sponsors need to set out some definable endpoints in terms of what is understandable.

When the FDA started their presentation, it started that labeling is likely to be read and understood. What does understood mean? Does it mean understood by 80 percent of the people, 90 percent of the people, 40 or 50 percent of the illiterate people and 90 percent of other people?

I think if sponsors went in with predefined, "This is what we want. We want 80 percent of the people to understand the label of all people," then we would be able to get back and design a label and only 50 percent don't understand it. You design another label and you change the wording. We're talking about these are minor words and these are not minor words.

I would argue that notify your physician is not understandable to somebody who has low literacy. Notify is not a great word for that group of people. Probably not. I think we may need to talk a little bit more about that later on.

One other comments here. When sponsor put up -- I don't see it in the label but when sponsor put up the supplemental educational materials they had three circles up there and comparing omeprazole with antacids and H₂-receptor antagonists and basically a marketing piece which antacids work for an hour or two and H₂-receptors will work for 12 hours. This drug will work for 24. Nowhere do I see in the label any statement that you will not see this medication work for one to two days.

There is nothing in there to tell people that if you take this for a day and you are expecting a response in six hours or 12 hours you're not going to see a response. I would like at some point in time us to discuss about the addition of what they should expect from this drug.

DR. CANTILENA: Thank you. I'm sure that will come up this afternoon in our discussions.

Dr. Alfano, do you have a question for the committee -- I mean, for the FDA? Not the committee.

DR. ALFANO: Yes. It's a question for Dr. Shetty. At one point, Dr. Shetty, you criticized the actual use study because the sponsor didn't contact physicians to confirm that, in fact, they had been contacted by the participants. I guess my question is if it's an all-comer study, how would you do that and not infringe on the doctor-patient confidentiality and violate HIPA and things like that.

DR. SHETTY: We've seen studies in the past when they wanted to check whether people really went to the physician. They asked who is their primary physician and asked permission to contact their physician to ask whether really that patient saw that physician and whether the physician approved of what was the decision made. Here there was no -- they didn't have to go to see their particular physician those subjects.

They could have asked anybody who is a healthcare professional, a friend or a relative, "I'm taking this medicine for my heartburn. Is it okay?" They would say okay and that was considered that they consulted a healthcare provider.

DR. ALFANO: So then you're suggesting that releases would be sent to whomever?

DR. SHETTY: I don't know the particular --

DR. ALFANO: -- contacted on an all-comers basis? I guess my point is it seems to be an unrealistic requirement.

DR. SHETTY: Maybe it's unrealistic but that would be perfect or realistic to know whether really physician approved that medication for that patient to use. It could be done at the end of the study after the study is completed and contact made to the physician.

DR. CANTILENA: Some sponsors have actually handled that in a different way on other applications.

Any other questions? Dr. Davidoff.

DR. DAVIDOFF: Yes. I have a question primarily for Dr. Lechter. It has to do with the wording of the label. It says, "Do not use with other acid reducers." I wondered if that statement is clear to you? Whether the meaning of that statement is clear to you? It's not clear to me because I think the intent is directed at H₂-receptor antagonists.

On the other hand, we've heard that the data either are missing on whether there is a reaction or that, in fact, H₂-RA errors are, in fact, make a difference because they diminish the efficacy of this drug.

We've also heard that people apparently took both H₂-RA antagonists and antacids during the course of the trial, although that -- well, I don't know about during the course of the trial because that wasn't asked for but that has apparently been true in some of the other data that was presented.

It seems like this is an ambiguous statement not just to me but perhaps to others. Do you have any notion of how clear that meaning is?

DR. LECHTER: That particular issue is not tested in the materials that I have received. However, I might note that I believe, and the OTC Division can correct me if I'm wrong, that all the over-the-counter products that are acid reducers will say acid reducer on the drug facts label.

If they are not an acid reducer, they might be called something different. Is that correct? So that if consumers look at the drug facts label for the other products they are taking it will say acid reducer if that is what it is. I agree in general the term is probably not clear to the lay people.

DR. DAVIDOFF: That is helpful because if it does say that on the HRA package, that is fine. If you just ask 100 people what they understand what is an acid reducer, I don't think that would be a highly germane point because not everybody reads the package of the H₂-RA.

DR. CANTILENA: Okay. Thank you. One more question, Dr. Camilleri.

DR. CAMILLERI: I would like to ask Dr. Shetty her advice with regard to the correct self-selection. I see from the table you have provided us that 134 of these 1,251 patients had contraindicated symptoms.

In the context of risk management, I would have thought that a much larger study would be helpful to understand whether people with contraindicated symptoms would deselect the option of using omeprazole.

I guess from a design perspective or from the numbers that we have, is this a sufficient number to reassure us that deselection is going to occur appropriately?

DR. SHETTY: Well, I don't know. We don't have any endpoints for what is acceptable or not acceptable failure on those subjectives. We know that some people deselected in this study also those who had contraindicated conditions and didn't buy the product or refused to participate for that reason.

This was around 10 percent of that population that had those contraindicated conditions. We can discuss about that more whether it is acceptable or not. Certain conditions are more serious than the others if they are not reported to the physician.

DR. CANTILENA: Okay. Dr. Zorich, did you want to make a comment? Either that or you have to leave the room.

DR. ZORICH: Well, maybe. Considering how confusing this is becoming, maybe leaving the room is good.

The reason you saw me kind of jump up is that this is an area that is confusing to me just how the sponsor should handle appropriately. These contraindicated symptoms have really nothing to do with omeprazole. They have to do with people misconstruing heartburn for something else.

Whether it's frequent heartburn or episodic heartburn, it's really -- I don't think that it's germane because if we are talking about people having anginal like symptoms, then the fact that is happening to them at that point when they are making a purchase decision at a Walgreens, it has nothing to do with the purchase decision of omeprazole or an acid reducer or an antacid.

We were trying -- now I see sometimes that no good deed goes unpunished. We were trying in a very responsible way to communicate to people that anytime you have heartburn, there should be this other constilation of symptoms that you are considering in making a purchase decision.

I would like to clarify that I do not believe that they are uniquely related to a purchase decision about omeprazole. They are instead the AGC warning signs which could be -- somebody could be experiencing whether or not they are having a frequent or infrequent heartburn.

DR. CANTILENA: Right. I understand your point but if it happens on your study, then you have it.

Why don't we -- actually, I would just like to go over sort of the homework assignment. I would propose that right after lunch just before the charge to the committee by Dr. Katz if we can get a copy of the ICD.

Then we would want to see the actual pharmacokinetic interaction data for warfarin, ketoconazole, and then the drug-food interaction data because that was a question that came up earlier this morning. These can just be slides with the curves to show us the effects. Does anyone else on the committee want to see any other pharmacokinetic data?

Dr. Brass, did I leave anything out?

DR. BRASS: No. I think you covered it but I'm sure when we see it there will be questions about its limitations.

DR. CANTILENA: Very good. Let's pause and we will actually start on time this afternoon at 1:30.

The committee is reminded during lunch not to discuss issues that are before the committee. Talk about the soccer game and see if you can catch a replay.

(Whereupon, at 12:26 p.m. off the record for lunch to reconvene at 1:30 p.m.)

A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N

(1:35 p.m.)

DR. CANTILENA: I will start the afternoon with some follow-up items that we listed just before the break. I'll turn it over to Dr. Triebwasser from Procter and Gamble. Can I have your attention, please? Thank you.

DR. TRIEBWASSER: We're going to present now the data, looking at the drug-drug interaction, the specific data on warfarin.

We're not on? Do I have it turned on? I have it turned on. Let's try it again. There we go. All right.

We're going to p resent now some data on the drug-drug interaction questions regarding warfarin and also the food interaction studies which were asked for earlier. Dr. Levine from AstraZeneca will present this data.

DR. LEVINE: Thank you. Can I have slide 58, please? We'll start with the data requested regarding drug-drug interaction studies involving

omeprazole and warfarin. This is from my slide set, please. Thank you.

These are data that were shown at the October 2,000 advisory committee meeting. I would like to refresh people's memory. Warfarin is a racemate that includes two optical isomers, the R and S form. It turns out that the anticoagulant effect delivered by the racemate is primarily through the S-isomer which does not share CYP-2C19 as the primary metabolic pathway with omeprazole.

The R-isomer, which is metabolized through 2C19 does not contribute nearly as much of the anticoagulant effect. These are group data, two studies, the first in healthy subjects. One can see if one looks at the S-isomer, which is clinically more important with regard to delivery of the anticoagulant effect, there is no change in serum concentration with a 20-milligram dose of omeprazole after 14 days. There is a very small clinically insignificant effect, mean effect, of about 12 percent.

I'm not going to use the pointer. Thank you.

The bottom study was performed in anticoagulated patients. I'll show you additional data and a couple of additional slides. Similar effects were seen. Again, no changes in the S-isomer concentration with omeprazole, in this case 20 milligrams over a 21-day period, whereas with the R-isomer there was almost a 10 percent change.

We have data looking at the pharmacokinetic effect which is of greater clinical importance. What I want to show you on the next slide, which is slide 59, I need to walk you through this.

This is a rather old study in which a coagulation test known as a thrombo test was used. This was a study conducted in Sweden and this is not a test that we have presentationality with so we don't have data using prothrombin times or INRs. The TT values were a clinically relevant means of following anticoagulation in patients treated with warfarin.

What you have on the X axis are the initial run-in values with patients who are treated with warfarin but they are not yet on omeprazole. Just to give you a guide, the therapeutic range that is aimed for using this test is with a value of between five and about 18. If you look carefully all the way to the right, there is one outlier with regard to, you know, just during the run-in whether or not they were within therapeutic range.

Now, what we had the opportunity to do in this study was have a couple of run-in values. What we have done on the Y axis is run-in value at week one and run-in value week 2 and looked at the difference just to give you the measurement variability in coagulation function just on warfarin. You can see that there is a very wide range. This has to do with the variability and warfarin effects and measurements of anticoagulation.

Now, keep this in mind. You can see that there's a range of plus or minus at least five to six points using the TT value but outliers that are even greater than that.

Now, in the next slide what I'm going to show you is the actual study slide. We had data on 28 patients. This was a randomized placebo controlled crossover study which in one period patients were randomly allocated to receive placebo or then were allocated to receive omeprazole obviously being maintained on what was thought to be their stable warfarin dose based on the run-in values.

Here what we have again like in the previous slide on the X axis below this is showing what the last run-in value was using the TT test. On the Y axis what we have here is the difference.

These are individual patients, the differences in the TT value on omeprazole with warfarin or on placebo with warfarin. Again, you can see that the nature of the variation is actually within measurement variability of the TT test if you recall the previous slide.

The other point that I would simply make and, again, I apologize if this is confusing. The lower the TT value the higher the -- the greater length of time it would take for coagulation to occur so there is a bit of an inverse value.

My point is if you were looking down closer to the four to eight range, if omeprazole was having significant interaction, you would be seeing the differences trail up into the left and you do not see this.

Our interpretation of these individual data are that, in fact, when you look at omeprazole effects on warfarin, the changes in the pharmacodynamic effect of the drug is actually just within measurement variability as when you are looking at warfarin alone.

Would you like me to proceed through the other drugs or take questions?

DR. CANTILENA: How about if you go through the rest of the data and then we'll do it all at once.

DR. TRIEBWASSER: Okay. Next slide 55. This is looking at phenytoin. Again, I don't want to take too much time. This is a slide also shown at the October 2000 Advisory Committee Meeting. Here we have three studies with healthy subjects and a fourth study in individuals with epilepsy that required phenytoin treatment.

What we showed with coadministration of omeprazole in the healthy subject's doses of 40 milligrams either at seven days or three days, or in the epileptic patients 20 milligrams of omeprazole for 21 days, we did not see any clinical significant changes in phenytoin levels.

Now, on the final study in epileptic patients there were eight patients. We have individual point values that we can show you on slide 57. I apologize because this is more raw data and a little bit difficult to see but the patients are one through eight down below.

If you read across on the top, baseline phenytoin levels were obtained at week zero, week one, and week two before omeprazole treatment was introduced. Omeprazole was then added during weeks three, four, and five, and then stopped. Then we have washout values off of omeprazole at week six or seven.

These are all phenytoin levels. The therapeutic range for phenytoin was approximately 40 to 80. What one can look at is really no significant movement of individual values on omeprazole that are clinically significant. I'll leave that up if you would like to look at it carefully or we can make it available later in a hardcopy.

Slides off just for a moment. Ketoconazole, I apologize, we do not have data. The data that were looked at with regard to the interaction between ketoconazole and omeprazole were actually not sponsor related studies.

In the prescription label we indicate -- this is essentially a concession that because of the known effects of acid, the requirement for acid for absorption of certain drugs, the prescriber is to take that into advisement.

Now, we are aware of published data where omeprazole was given to individuals with ketoconazole as part of the drug interaction study. The OUC levels of ketoconazole actually declined by about 80 percent.

We think that it is very important from a medical standpoint to know that if one is treated with acid suppression, the therapeutic value of the antibiotic, in this case, is not going to be very high. I'll defer to others to speak about the labeling contingencies for the OTC product. We recognize that omeprazole will significantly decrease the absorption of antifungal agents like ketoconazole.

Finally, you asked about food effects. If I could have slide 68, please. There are three curves here. This is a standard plasma concentration time curve for the use of omeprazole tablet at 20 milligrams in the dark squares and the omeprazole tablet 20 milligrams after food.

What is not relevant here is the third curve which is the omeprazole capsule. If one wants to understand the food effect on the tablet, if one looks at the very first curve on the left compared to the second curve, which is the fasted versus fed administration of 20 milligrams of omeprazole magnesium, one can see that the c-max declines at the time the c-max extends. But the area under the curve stays the same. I can show you on a table, new slide 247.

Again, if we are looking at the mean values for AUC, c-max and t-max for either the MUPS, omeprazole magnesium tablet administered in the fed state versus the fasting state. On the right column there is the ratio. What we show is that if you compare areas under the curve there is unity.

There is a difference for c-max so based on the way bioequivalence is interpreted, this may not be judged by equivalence. Our position is that AUC is a very good surrogate predictor of acid suppression and the fact that one sees no difference in the area under the curve. There is an element of equivalence here.

DR. CANTILENA: I just have one quick question if you can just go back one slide, SBU-68. The way in which the area under the curve remains the same is the slowing in the absorptive?

DR. TRIEBWASSER: That's correct. One would reasonably predict that in the fed state digital and gastric emptying by essentially the bioavailability based on AUC. There are different criteria that you are well aware of. Based on AUC they are the same.

DR. CANTILENA: All right. And so the title of your slide really is just referring to AUC then?

DR. TRIEBWASSER: Yes, but we also have other data that shows that the AUC is a good predictor of acid suppression.

DR. CANTILENA: Right. But there is an effect on c-max as well as t-max.

Now, just from a scientific standpoint, can you tell us what you think is most important in terms of the ultimate pharmacodynamic effect? Is it c-max or is it area under the curve?

DR. TRIEBWASSER: We think that the most relevant factor is the AUC based on the relationship to acid suppression.

DR. CANTILENA: Okay. Other questions?

MS. COHEN: Yes.

DR. CANTILENA: Go ahead.

MS. COHEN: I see that you have to take it in the morning. Now, what happens to people who don't take breakfast or people who just have coffee or people who do have to eat after it, before it?

The other question I have along with it, can you take it with, say, orange juice or grapefruit juice or should it be taken with water? I am concerned about consumers, whether they take it without having breakfast or they take it after breakfast or how they should take it because this only says in the morning and that doesn't mean anything.

DR. TRIEBWASSER: Not all the studies have been done to specifically address each of the contingencies that you addressed. Based on the bulk information we don't think that there is really a lot of difference whether or not the drug is taken with food or other beverages that you mentioned.

MS. COHEN: Suppose someone doesn't eat any breakfast at all and take it?

DR. TRIEBWASSER: We think the drug would still be effective.

MS. COHEN: What would it react with?

DR. TRIEBWASSER: I don't understand.

MS. COHEN: Isn't there something that --

DR. CANTILENA: I think on the slide the answer to your question would be that you would be looking at the fasting curve.

MS. COHEN: Fasting curve.

DR. CANTILENA: So he has that information. It's a higher c-max but the area under the curve doesn't change.

MS. COHEN: Thank you.

DR. CANTILENA: Dr. Brass.

DR. BRASS: Yeah. I would like to return to the focus of relating this information to the question posed by the reviewer as to the adequacy of the warning label and which of these drug interactions, in fact, need to be communicated effectively to avoid a public health problem.

First, I would like to thank the sponsor for providing the individual subject data to allow us to understand the outliers as well as the mean response which is quite helpful. It is clear in that relatively small population that with warfarin there wasn't any evidence of a clinically meaningful effect. The question is how does that small sample effect extrapolate to a large population and whether or not there are at risk populations that are identifiable.

At the same time coming back to the point that was raised, I'm a little bit concerned about over extrapolating spontaneous reports because we all know that in any cohort followed on warfarin there will be individuals who will go out of whack for no clear reason at various times.

If they happen to be on omeprazole and happen to be reported, there may be a link. What I'm trying to gauge in terms of whether or not -- I do believe any patient on warfarin should talk to their doctor before they take any medication.

In terms of the standard of effectiveness of the warfarin warning, is the review team comfortable that based on this data that this is not a large population concern or do they remain concerned that there are specific subpopulations or stronger data to suspect this is a risk.

DR. CANTILENA: Charlie, does someone from your team want to handle it?

DR. GANLEY: Yes. I guess the point that I would make is that we've seen a pattern, a cluster of such reports in the class, as I mentioned. In some cases some of the reports actually indicated a salutatory response to dechallenge so that basically in some cases not only is there a theoretical interaction based on the CYP-2C19 metabolism but in some cases empirically there was improvement after cessation of the proton pump inhibitor.

DR. BRASS: I mean, were there any rechallenge in any of those in terms of doing formal study?

DR. GANLEY: That I would have to go back and look at that. There was enough concern to decide to change the labeling in the prescription formulation. By the way, again, I think the other point is that we wouldn't necessarily in a small test population see it but for a variety of confounding reasons you have outliers in a large population of users.

DR. BRASS: No, I understand completely. What about phenytoin? Do you still believe that phenytoin requires a warning in the general population?

DR. GANLEY: If you just wait one second.

DR. SULEIMAN: I guess with regard to what Dr. Avigan was talking about phenytoin interaction what we know is based on limited numbers of subjects and based on, I believe about three studies that were conducted. The studies showed mainly an increase in plasma levels of phenytoin by about 15 to 20 percent.

There was one study, of course, in epileptic patients which did not show any significant pharmacodynamic adverse events associated with that increase in plasma levels which was conducted by the sponsor. Those are the only available data that we have at this moment.

DR. CANTILENA: Okay. Are there any other questions specifically about the pharacokinetic data that we saw? Dr. Davidoff.

DR. DAVIDOFF: I had a question about some pharmacokinetic data that we didn't see and that had to do with digitalis drugs because the therapeutic margin -- of course if their dig is relatively narrow. The patients taking it are often quite fragile and are electrically unstable. I just wondered if there are already data that would help us decide about if we should be thinking about that for the label or not.

DR. CANTILENA: Is the sponsor aware of any interactions with dig?

DR. TRIEBWASSER: Can I introduce Dr. Tommy Andersson? He's our pharmacology expert on omeprazole.

DR. ANDERSSON: I'm sorry. Can you repeat the question?

DR. DAVIDOFF: Yes. The question is about interactions with digitalis derivatives specifically because the therapeutic margin is so much smaller for dig than many other drugs.

DR. ANDERSSON: That is not an interaction on the metabolism level as was suggested in the presentation before. That's an absorption interaction. I mean, digoxin are degraded in the stomach before it's being absorbed by some bacteria degradation.

By increasing the pH that degradation prior to absorption does not happen. That's what we see here as an increased AUC or some 10 percent as an average value in the study we did. It's nothing to do with metabolism.

DR. DAVIDOFF: The mechanism isn't so important as the end result. If there really is a 10 percent increase, that could be quite substantial for some substantial number of patients. I wonder if that should be considered as a conservative thing to do to put that on the label?

DR. ANDERSSON: That's more of a clinical judgement, I guess.

DR. CANTILENA: Yes, Dr. Houn.

DR. HOUN: I'm just wondering if the company could comment on other drug interactions with diazepam

and clarithormycin.

DR. TRIEBWASSER: I was still under digoxin and I found a study.

DR. CANTILENA: Okay. I'll take notes of the other ones and we'll remind you.

DR. GANLEY: Could I just intercede with just a clinical point as you're looking just to raise this issue that patients with renal failure, for example, were not tested and, again, some patients were on digoxin and have other reasons to have reduced clearance of the drug which in combination with such a challenge might have a more exaggerated blood level response.

For drugs where there is a narrow therapeutic index, those are the kinds of issues that, again, I don't believe have really been directly tested. Those kinds of patients have not been stressed.

DR. TRIEBWASSER: Regarding clarithromycin there have been three studies that have been done. One study showed a positive interaction with a 15 percent change in clari levels and no changes in the others.

I would like to address part of the rationale with regard to which drugs perhaps should go onto the label had to do with worse case scenario what would be a clinically significant effect. Our feeling was that, first of all, for clari this would not be clinically significant.

Whereas with warfarin and phenytoin although we regard the risk as exceptionally low, the risks of extended prothrombin times of any cause obviously have fairly significant medical consequences. That was the rationale for choice.

I forgot the other drug in addition to clarithromycin.

DR. CANTILENA: Diazepam, digoxin.

DR. TRIEBWASSER: We have -- we're relying on our own data and other drug interaction studies where with diazepam one can see what would appear to be significant changes in clearance of the drug between 25 and 54 percent.

We referred in our submission to another drug interaction study that was done between diazepam and cimetadine which showed a similar level of interaction but no clinically significant effects with regard to CNS status. Again, based on the clinical effect study we do not regard the interaction as clinically significant.

DR. HOUN: That is a decrease in clearance. Correct?

DR. TRIEBWASSER: Yes. Let me show slide 53 just so there is clarity on this. I would ask you to look at the bottom two curves where we are looking at the usual extensive catalyzers, and you can see the difference. The bottom of this curve is when omeprazole is coadministered with diazepam.

Then the second curve up is with placebos. You are seeing a bit of a washout following the start of treatment. This is intravenous infused on diazepam.

There was a question on digoxin. I can show this slide rather than waste time but we do have evidence of a study where in a crossover study with placebo or omeprazole there is actually no difference seen in digoxin levels in that one study. It involved 22 subjects.

DR. CANTILENA: If you have that data, that would be good.

DR. TRIEBWASSER: I can -- could I have slide 64. I apologize. This may be difficult to see. It's the top most bar, digoxin. Elderly patients and of 22, baseline and concomitant treatment so there is a crossover study using digoxin doses of 0.125 to .25 milligrams daily. Omeprazole was administered 20 milligrams daily for 10 days and there was no affect seen on the serum digoxin levels.

DR. CANTILENA: You mean no affect as in statistically significant or whether outliers?

DR. TRIEBWASSER: We would have to dig that data up.

DR. CANTILENA: Yeah. I think that addresses some of the issues as best we can in the setting.

What I would like to do now is actually move to Dr. Katz to charge the committee and I think sort of begin to focus our discussions on the issues at hand. Then you will all have time to ask questions and to share your comments about sort of everything that has been discussed.

If I can now ask Dr. Katz to charge the committee.

DR. KATZ: Good afternoon. I feel like you've already been charged with the issues and this discussion has gone full speed ahead. But I'll take you back again a little bit and kind of go through some of the issues that we would like to -- that we talked about earlier and we would like to have you think about and focus on as you go through your deliberations for the rest of the afternoon.

What I would also like to do is to highlight a little bit about some of the products and where we are in the current armamentarium of OTC heartburn products. A little bit about the prescription to OTC switch process. Then finally touch on the issues that you will discuss this afternoon.

In the interest of time I will skip around a little bit from some of the slides since you do have all of the slides available to you. Let's kind of begin now with currently where we are in terms of the OTC marketplace.

As we've heard, there are currently two classes of products that are available OTC. The first ones that have been out there the longest are the antacids that are approved for the relief of heartburn, that are actually indicated for the relief of heartburn only.

The H₂-receptor antagonists, also known as the acid reducers, are approved for relief of heartburn and prevention of heartburn as related to a meal at different specified times depending on the nature of the product.

As we have also heard today, Prilosec 1 is not looking for these two indications but are looking for the indication as a prevention of frequent heartburn, frequent heartburn described as greater than two episodes per week for a 24-hour period of time.

When considering whether or not a product should go from the prescription to the OTC arena, there are a variety of different places where we stop to look in the decision making process.

We look for the benefit-risk. The consumer's ability to self-diagnose and self-treat the condition. The consumer's ability to understand labeling instructions including monitoring, follow-up care, and treatment.

The ability of the consumer to understand what the goal is that they should reach from the treatment and if they've attained it. And the ability to recognize any toxicity and, again, to understand what to do about it if toxicity does occur.

This is an old slide and I'll just kind of run through it very simply again to compare and to contrast the prescription versus the OTC marketplace. When we have prescription drugs what we think about are patients.

Patients have a disease or a condition that requires monitoring and requires perhaps prescription medication which the healthcare practitioner is the one who prescribes what is appropriate and follows the patient to make sure that no adverse effects occur.

On the other hand, we have the OTC drugs. In the OTC drugs the patient actually is the consumer and they are seeking to relieve some kind of symptoms when they go to purchase a product. No prescription is needed and the consumer may or may not have the benefit of somebody to give them advice at the point of purchase depending upon whether or not they buy the product.

So clearly the labeling must be understandable enough for a consumer to understand which product they might want to buy or choose to get the maximum benefit that they may achieve which would be the relief of their symptoms for very little cost which would be an adverse event.

We have also heard earlier today descriptions about the actual use and label comprehension trials. These are conducted on most products that originally switch from the prescription to the OTC world to be able to understand a little bit about consumers behavior for using these products.

These studies are aimed at looking at the consumer's ability to self-select, ability to use the correct dosage for the specified time on the label, the ability to identify when to see a physician, the ability to identify serious as well as any adverse events, and the ability to avoid any interacting drugs.

With that, I would now like to turn slightly to the issue -- to change gears a little bit and just try and talk a little bit about the issues for the discussion before you.

As you'll see, this is a brief summary of the questions that you have in your package. But these are some of the salient points again that we want you to focus on.

The first would be as the sponsor identified an appropriate target population? When answering this question, we would like you to think about the symptomatic overlap with GERD. What happens to consumers who have less than two episodes a week of heartburn use the product. The issue of relative contraindications.

The issue of recurrence of symptoms after discontinuing therapy. The issue of chronicity of therapy; that is, repeat dosings since the label indicates that the treatment is only for 14 days and symptoms may recur and consumers may need further advice or may choose to use the product again.

And what to do about the acute symptoms since people may again have acute pain that they want to have relieved.

Further, other issues that you will be asked to address would be has the sponsor demonstrated that the consumer can adequately self-select to use the product. Did consumers with recurrence understand how to use the product. And has the sponsor proposed an acceptable duration of therapy for OTC use remembering that the OTC proposed label is for 14 days and the currently proposed -- the currently approved prescription labeling is for 28 for treatment of GERD.

Also for issue is part of the discussion would be the short-term versus chronic intermittent use of the product. The issues which we discussed earlier about delayed diagnosis of a potential serious condition and concerns about rebound or recurrence after discontinuing treatment.

Finally, we will ask you to address the approvability. Has the sponsor provided sufficient information to support the approval of OTC Prilosec 1 for the treatment of frequent heartburn.

In answering this last question, if your answer is yes, we would also like you to address the possibility of if there is any additional information that you might feel might be needed to give that answer such as additional information from the sponsor, a Phase IV commitment, further labeling, modifications that would help you to arrive at that decision.

If the answer is no, we would also like you to give very succinct reasoning as to why and what kinds of things the sponsor might do to be able to eventually combat or to be able to further deliberate on this issue.

With that, I would like to turn the meeting now back over to Dr. Cantilena to begin the afternoon's discussion.

DR. CANTILENA: Okay. Thank you, Dr. Katz for that excellent summary and I think sort of the beginning to help us focus on our discussion.

What I would like to do now is open the sort of general discussion of the issues and we'll try it just going in an open fashion. If it looks like we're not getting anywhere, then we'll sort of go issue by issue using Dr. Katz' handout. I would elect not to do that to potentially inhibit individuals if we just start that way.

I would like to hear, for example, from some of the GI members about the issues that Dr. Katz has just talked about, but there are other important issues that can also be discussed. At this point let's just open it for general discussion. Would anyone like to start?

Are we ready to answer the questions? Just kidding. Dr. Johnson.

DR. JOHNSON: As a non-gastroenterologist I would like to hear what the definition or the difference is between frequent heartburn and GERD.

DR. CANTILENA: Are you a gastroenterologist, Dr. Brass?

DR. BRASS: Yes. No, but I thought about this question for two years in the context -- nonstop -- in the context of this particular switch and have concluded that Dr. Johnson's question is theologic and that there is, in fact, no meaningful answer, differentiation, and that focusing on the question becomes a distraction.

Clearly while our evolution in the past 15 years of understanding about upper GI pathology has led to both changes in our labeling of individual patients and our management of individual patients. From the patient's perspective nothing has changed.

They have this symptom of chronic heartburn and they frankly don't care what you call it. Also there is no doubt in my mind that patients who are in this cohort that we are talking about now, however, we label them, are currently being treated with OTC medications. We are not, in my opinion, opening up a vast new population.

We are shifting a population that is currently being treated with OTC and considering an option of another OTC paradigm for that same population whatever you label them. I don't think it matters. Nor do I think it matters to a primary care physician who has a patient come in with these sets of complaints. Again, they will treat that without differentiation.

DR. CANTILENA: Comments from other members of the committee? Dr. Cryer?

DR. CRYER: I take a slightly different view to the answer to that question. I think it is a matter that is something more than theologic. I think it is, in fact, a very practical issue. I think it is actually, in fact, the crux of the issue being discussed.

That is if you look at the currently recommended treatment guidelines for the management of the symptom of heartburn versus the treatment of a chronic disease such as GERD it is critical what we are actually -- how we are actually categorizing what's being described as frequent heartburn.

So it is an issue really in my mind of semantics but looking at the actual data, and this gets to the crux of the question that I asked earlier, what is the actual distribution of frequency of heartburn that was experienced in the patients who were evaluated in the trial.

We were provided data here. I guess this was in Dr. Shetty's discussion. In this patient population it looks like to me that 50 percent of the patient population or greater was experiencing heartburn four times per week or greater. In fact, 40 percent of the people have heartburn six or seven days a week. As a gastroenterologist I would call that GERD.

I think if you look at the current management guidelines as suggested, for example, by professional associations such as the ACG, that dictates a specific treatment course that is not one that is episodic or that is associated with short-term treatment.

DR. CANTILENA: Bryron, I think sort of the issue of duration of therapy is something that we are asked to comment on specifically. I'm sure that will be something that is very important.

Other comments? Yes, Dr. LaMont.

DR. LaMONT: I'm a gastroenterologist and I would like to side with the first speaker. I'm sorry, I can't read your name from this distance.

DR. BRASS: Since you agree I will identify myself as Eric Brass.

DR. LaMONT: Excuse me, Eric. I would like to quote from the previous meeting that was held here. Dr. Sid Cohen was a member of the group that deliberated on this. Perhaps that is where the idea about religion came in. He said, "Are you trying to make a distinction between GERD and heartburn?" I don't see how you can do it. I don't know a difference.

I think if you filled the room with a group of Talmudic scholars, they couldn't tell you either. There is no difference. I agree. I don't think we should spend a lot of time trying to tease this out because I don't think it's relevant to the discussion.

DR. CANTILENA: As you see it, is it sort of just a question of severity in terms of the number of episodes a week?

DR. LaMONT: It's likely that the vast majority, greater than 90 percent of patients that have the symptom of heartburn, have reflux. Then I guess we're going to argue about the word "disease." Is it a disease? Do they have end organ histologic changes in the esophagus?

I don't think it matters because whether or not they have erosions or not, or Barrett's or not, or whatever it is, relies on control of acid and that is precisely what the medication does. I don't see it as relevant.

DR. CANTILENA: Okay.

DR. CRYER: If I may have a rebuttal to my colleague's opinion. I agree that the definition, the semantics, really are not relevant. I guess what my position is is that the course of treatment really is dictated by the severity or the frequency of whatever it is that we're describing.

For individuals who have more frequent disease, that would likely require either more potent, more aggressive therapy or longer duration is essentially the point that I think we actually agree on.

DR. CANTILENA: Dr. Fogel.

DR. FOGEL: I think I agree with Dr. LaMont, although after the comment comes out you may think that I disagree with him.

The majority of patients who have reflux symptoms who have heartburn have what is called nonerosive reflux disease where there is no structural damage to the esophagus. Our professional society says that it is okay to treat people with symptoms actually for four to six weeks without doing any investigation. The concept of treating someone for two weeks, or even four weeks with an over-the-counter medication certainly is within the range of what is acceptable.

The concern that I have is that the use of this drug may remove the physicians from the care of patients with esophageal reflux. If you have a treatment that is available over the counter that removes your symptoms, there is no need to see a doctor. What we know is that a percentage of these people will have Barrett's Esophagus.

A percentage of these people will have erosive esophagitis which requires more aggressive treatment. We won't be able to identify and treat them appropriately because of the fact that their symptoms will be controlled with this over-the-counter medication.

The greater concern to me is not the two weeks of treatment which the sponsor has suggested but what happens to the people who take the medication more than twice. From the use data and the comprehension data, it appears that is a significant risk.

DR. CANTILENA: I have just a quick question for the GI doctors. A comment was made at the beginning of the sponsor talk about when you normally start someone on a PPI you go through some screening questions.

I guess my question to you as some specialist is will that interaction, will sort of the information that you gain in that sort of history, can that be substituted by the label? Are you comfortable with that substitution, the interaction that you have version on the box on the shelf?

DR. FOGEL: Most of the patients who come to the people in our practice, and I guess for most gastroenterology practices, are already on PPI. They are already receiving medication. These are drugs that are prescribed by the primary care physicians. I guess the question is whether the primary care doctors ask those questions.

DR. CANTILENA: Any other comments from GI? Then we'll go to primary care.

Go ahead, Ed.

DR. GILLIAM: My question goes to the actual use trial. It's at the three-month follow-up more than 58 percent had their heartburn return but only 20 percent went to their healthcare provider. My question to our GI folks is how much does that concern them and do we need to have stronger labeling for follow-up if your heartburn returns.

DR. CANTILENA: Dr. LaMont.

DR. LaMONT: Yeah, I can start. We have already discovered in discussions during the break that there is a big range in how frequently or how soon we decide to endoscope patients and look for disease.

There's one group that would say you don't have to start looking for Barrett's until you've had heartburn for 10 years which is what a lot of gastroenterologists do. That leaves a lot of time for starting and stopping medications.

Others feel, for example, Peter Karilosys, a noted esophagologist, he says any patient who requires continuous maintenance medical therapy should undergo endoscopy to rule our Barrett's Esophagus. You have a huge range here. I think the incidence of Barrett's is low.

The number of patients with Barrett's who get cancer is very low indeed. It only comprises a tiny wedge of the entire pie of adenocarcinomas of the esophagus. I think the danger could be easily overstated of delaying workup, if that's what your question is.

DR. GILLIAM: So then you wouldn't have a problem with these people who have heartburn that --

DR. LaMONT: Don't get studied?

DR. GILLIAM: -- don't get followed up and don't see their primary care provider, whoever that is.

DR. LaMONT: Unless they had some of the danger signals that are listed on the label.

DR. CANTILENA: Dr. Goldstein and Dr. Brass.

DR. GANLEY: It occurs to me in listening to this discussion that just as easily one could interpret the 14 days of therapy or even, let's say, in a certain percentage of instances a second course of therapy as a filter as those who are relieved, well, of course, by definition get relieved. Those who are not clearly will be directed -- apropos Dr. Fogel's earlier comment, will be directed not away from the medical establishment but into it truly needing it.

DR. CANTILENA: Dr. Brass.

DR. BRASS: Yeah. Again, I think it helped me to crystalize this problem in terms of putting it into another context. That is, I truly do wish everybody who had chronic uncontrolled symptoms or minimally controlled or required chronic therapy would, in fact, go see a healthcare professional for advice.

All the data continues to say those patients are already out there not getting advice. Look at the entry cohort. The entry cohort into the actual use study and other cohorts we've seen presented here and a number of other studies. There is no doubt in my mind these patients are out here.

To say that just because they happen to use this OTC product versus another, there is a different expectation. I don't think it's realistic. The fact that we are directing a portion of the cohort, though not as large a cohort, might be interpreted as some positive impact. I think again in my thinking the fact that these patients are out there right now self-treating is really critical to providing a context for what the impact is going to be.

DR. CANTILENA: Dr. Camilleri.

DR. CAMILLERI: Just to expand on that point but to give it a slightly different twist, I believe marketing studies have been done with regard to the prescription practices for PPIs. It certainly appears that at least 50 percent from my recollection -- and the sponsor's may have a more accurate assessment -- at least 50 percent of prescribed PPIs is still for the symptoms of heartburn and GERD.

To dismiss this as a nondoctor oriented issue that is dealt with entirely in the community with over-the-counter preparations I think would also be an error.

DR. CANTILENA: That's a very good point.

Other general comments or specific comments? Yes.

DR. NEILL: Richard Neill. I'm a family doctor at the University of Pennsylvania. I sat through the meeting two years ago when we were presented data about the 10 and 20 milligram dose and I appreciate the work that the sponsor has done to address some of the concerns that were raised then.

My remembrance is not as clear as the transcript in our book, but to summarize my memory this combined group felt that the 20 milligram dose was safe and effective, the patients could self-select. We asked for an indication that would fit.

It seems like you have come up with that in this frequent or recurrent heartburn. I think it's appropriate if we discuss some of the concerns about whether or not allowing a medication like this over-the-counter is going to result in patients who never come in to see the physician or a healthcare provider, and yet it's clear those patients are already there.

They are already taking other medicines. There are also obviously other contributors to Barrett's Esophagus and morbidities that face those patients which have never been discussed by the group and probably aren't the proper topic for us, but which, I think, put the risk of Prilosec in tiny, tiny perspective given the risks that they bring to the table.

I guess what I'm doing is building up to an overall conclusion of the things that I've heard. It certainly seems to me like something that ought to be approvable. I guess I would like to get on to answering the specific questions about the advice that FDA staff is looking at since much of the discussion that we're having now seems to remind me of what we talked about two years ago.

I don't think we're talking about safety and efficacy today. We did that two years ago. I think we're talking about patient selection for this indication and labeling and label comprehension for this indication.

DR. CANTILENA: I think that is exactly what we're talking about. We have to make sure that everyone is on the same page and everyone is comfortable with having an understanding so we can go forward.

Any other general comments before we start to address the questions? One more. Dr. Cryer.

DR. CRYER: Actually I'll follow-up on a point actually that was raised by both Dr. Goldstein and Dr. Brass. I guess it's the issue of how we see, or how maybe we might discuss the differences between kind of the acid reduction therapies, the currently available H₂-blockers and how would that differ with respect to OTC use and kind of my view on it.

I agree entirely with you. We're talking about a very safe and effective class of medication, specifically the PPIs. I think the major difference with respect to the consumer and OTC use is selecting out severity of symptoms or selecting out severity of disease, however we call it.

With the H₂-blockers clearly there is going to be a population of individuals who are not effectively treated by H₂-blockers. In many instances that would likely drive them to further evaluation.

In contrast with a proton pump inhibitor, for example, providing free access, open access to a consumer, there would likely be a greater population of individuals who would be more effectively treated and, therefore, we are selecting out more severe disease that might have otherwise have gone on for medical attention.

I guess one discussion point, at least for consideration, is what would be the consequence of having selected out that more severe disease population by having treated them with the proton pump inhibitor in an OTC fashion and are those consequences significant clinically with respect to other down-the-road consequences of not having involved a healthcare provider for more severe disease.

DR. CANTILENA: I think that is a very good point. Also the fact that these individuals will probably recurrently treat themselves inadequately possibly and what are the consequences of that which hasn't been studied obviously. We have to sort of estimate that with our expertise.

Dr. Brass and then Dr. Geller.

DR. BRASS: Yes. I think the proposition posed is a very fair one. I think it is part of the crux of the issue. Again, having thought about it, where I came out is that to some degree it's the ability of the individual consumer to replicate the empiric therapy that would be done as a first round therapy for a healthcare provider setting.

Though the duration might be different in a healthcare provider recommendation, that an empiric treatment with PPI would be done and if symptomatic benefit was taken over a period of time people would be happy and no further evaluation would be done. If that empiric therapy was ineffective, further evaluation would be done and that is certainly the intent of the labeling. Form a first-round decision I'm very comfortable.

The concern is whether or not the warning will be completely ignored and you'll get large percentages simply going on continuous OTC therapy and missing a cohort of unknown size both in terms of what percentage of patients will do that and what percentage of the cohort is at risk if they do that and what your judgement is on that relative size. I guess my bias is currently that is not a large product if you look at the two different groups.

DR. CANTILENA: We have Dr. Geller and then Dr. Levine.

DR. GELLER: A second concern aside from the continuous administration that I have is how to write the label clearly for distinguishing between relief and prevention. I don't think that the company has succeeded in that and I don't know how to do it.

DR. CANTILENA: Dr. Levine and then Dr. LaMont.

DR. LEVINE: I'd like to make the point that in my experience I have a lot of confidence in patients with chronic heartburn. They know their disease. They can almost tease out and tell the physician when they come in, "I tried H₂s. I tried one and I tried two pills, etc., of the pill you gave me."

One of the interesting things is the day-con score or the daily average consumption scores that people who take Prilosec for chronic disease. Some 27 million Americans have nocturnal classical reflux. Of that group a large proportion have classical GERD. Maybe erosive or nonerosive gastritis.

The average score was 1.6 for PPI so they are actually taking more than one a day. I do suspect realistically patients will decide if they have more severe symptoms they will try to relieve it with more than one pill a day necessarily.

I think we ought to give a little more confidence to the people out there with chronic heartburn because they do know their disease. Most of them -- many of them have been to doctors before.

I agree entirely with Dr. LaMont. I think the physician has to get into the loop but I'm not sure the physician has to get into the loop soon and with the advice and counsel of the patient we can probably do a pretty good job.

DR. CANTILENA: Thank you.

Dr. LaMont.

DR. LaMONT: I would just like to extend that a little bit. In fact, there is some data from the Bardhan study that was reviewed this morning that about half of the patients would have some kind of recurrence so it's a big number.

What happens when patients recur if they go to a healthcare provider be it a primary care physician or a specialist they would probably be put on omeprazole. In fact, the people that I would worry more about are those that don't respond. Maybe they have something else.

Maybe it's really gallstone disease or angipectis or something else. I don't think a lot is going to be lost by patients continuing this medication for a period of time because it's precisely what we're going to do anyway in the vast majority without a big workup.

DR. CANTILENA: Okay. I would like to move -- if there are no strenuous objections I would like to move to the questions. We will not read the big preamble under background and we'll jump to subject area No. 1 which is population.

The first question is, "Is it acceptable that some patients with GERD plus or minus erosive esophagitis self-treat with OTC medication?" What I would like to do is first get a show of hands and then we'll go around and have you explain the circumstances that it's either acceptable or not acceptable depending on your vote.

Let's say all in favor -- all in agreement that it is acceptable to self-treat with Prilosec -- excuse me, with an OTC medication for this problem raise your hand. All in favor? All opposed? Any abstentions?

Okay. The vote was 16 to 2 with the negative votes coming from Dr. Cryer and Ms. Cohen. If I can actually ask Dr. Cryer to comment on why he said no and then we'll ask everyone else.

DR. CRYER: There are two principle concerns. I guess part of it has to do with the duration of the therapy that is required for, and the question is specific, GERD plus or minus erosive esophagitis.

We've seen that and we know from experience that while the PPIs are very effective, the duration of therapy that would be required in an OTC setting is likely to be more than 14 days, particularly for someone who has erosive esophagitis.

Once these individuals come off of this therapy, the efficacy studies indicate that there will be a 75 percent recurrence of symptoms of frequent heartburn within three days after cessation of therapy.

So it's not so much the initiation of therapy but the expectation that there will be a durable response with only short-term, and the continued requirement for prolonged therapy.

Look at the population overall of the mean results than some aspects of the presentations today depending on whose presentation you listen to, either the FDA's or the sponsor's. For the group overall it looks like many people are able -- not all are able to self-medicate. That definitely changes in 10 percent of the population that was low literacy.

The thing that just sticks out in my mind is that there was a 50 percent response rate in terms of the label comprehension for individuals who fell into that category. I think from a public health perspective and a public health concern, I'm just not so sure that even though we are potentially providing greater access for them, that that would be the appropriate thing to do for that specific subsection of the population.

DR. CANTILENA: I agree with you on that and we will have an opportunity further down the questions to talk about that. I think that is a very important point and I agree with you in that regard.

Ms. Cohen, can I call on you to ask why you voted no?

MS. COHEN: Dr. Cryer is very eloquent and I have a feeling you're talking about people. We get a little esoteric and we talk about all the people seeing physicians. I would like to talk about all the people who are not, the 44 million who don't have health insurance.

I'm concerned about relief versus prevention. I'm concerned that people have serious problems and these might take care temporarily for 14 days, but they can't afford to go see a physician. I am concerned about direct advertising to consumers. What's going to be said to them? How much information is going to be given to them? How much are they going to know about relief or its periodic? I just feel that the labeling is inadequate. I'm just worried that we have to think about the consumers out there.

Dr. Cryer, I have a feeling you deal with people and I work in some areas in the community. It's not esoteric. It's really people just being able to manage and can't afford a lot of things. I want to make sure that people can take less expensive products that will take care of the problem.

I'm concerned about those who have serious problems. Not everybody knows what angina is or a lot of things. I don't think there's enough information to help our consumers.

I repeat again I am very concerned about direct advertising to consumers. We have to educate consumers about diet and prevention and maybe some of that will take care of needing to take any medication.

DR. CANTILENA: Okay. Thank you very much. Thank you for your comments.

The next question that we're asked to deal with is in the category of self-selection. The question is, "Has the sponsor demonstrated that consumers with heartburn can adequately self-select use of Prilosec 1?" I think with this question we'll also do similar style of voting. We'll vote by a show of hands and then we'll go around to get individual opinions if needed.

Let me call for the vote. The vote is all who agree that the sponsor has demonstrated that consumers can adequately self-select for the use of this product, please raise your hand.

All who feel that the sponsors have not adequately demonstrated that consumers with heartburn can adequately self-select for the use of the product, please raise your hand.

Okay. Any abstentions? Anyone abstain?

Dr. Fogel, I think we missed your hand.

DR. FOGEL: I thought that they had shown.

DR. CANTILENA: Okay. So he was a yes. So the final vote was three yes, 15 no. I think in this case maybe if we go around the table and just ask for individual opinions.

DR. TITUS: I want to enter into the record the three yes votes were Drs. Brass, Levine, and Fogel. The remaining votes are obvious from the record then.

DR. CANTILENA: Okay. Thank you. Why don't we start on this side of the table with Dr. Davidoff. If you could, explain your answer or your vote, please.

DR. DAVIDOFF: There are a couple of levels. One is that the actual use study was based on people who had actually read the label. But we also know from a number of pieces of information, some of which you've gotten today, that a very sizable proportion of people don't even read the label to start with which changes the denominator very substantially.

On top of which, although there is some argument about interpreting the data, it does look like a very substantial proportion of people select themselves inappropriately no matter how you look at the actual data of people who have read the label.

DR. CANTILENA: Dr. LaMont.

DR. LaMONT: Yes. My concern is about how many times they might use it in a year and it may not be appropriate for this particular question of self-selection. My concern is how many times it can be used over a period of time. I don't think the label tells us about that.

DR. CANTILENA: Dr. Patten.

DR. PATTEN: Yes. My understanding is that 24 percent of the self-selection population incorrectly self-selected themselves. To me that's a high percentage, unacceptably high.

DR. CANTILENA: Thank you.

Dr. Lam.

DR. LAM: I have the same concern as Dr. Patten that some of the data provided did not actually provide sufficient evidence that the consumer can adequately self-select the product.

DR. CANTILENA: Thank you.

Dr. Levine.

DR. LEVINE: My concern is that a lot of patients will not look at that label and we all recognize that whatever the label shows. The other day I looked at the Pepcid AC label just because I haven't seen an OTC label and I was very impressed with bold print, red strips. I think that is very important when you get into labeling.

I do think my vote was yes with the provisal and thought that there would have to be some type of educational campaign at the level of the pharmacist, etc. I gave credit to the fact that there is a group that won't even look at it but that we absolutely need some type of educational campaign.

DR. CANTILENA: Thank you.

Dr. Gilliam.

DR. GILLIAM: My comments reflect Dr. Davidoff. I do want to commend the sponsor on especially the packaging which I think is actually pretty good. It's just that we know that most people don't read the package inserts or, again, the cartons.

If we can do, again, more education to get people to actually read these and follow the directions on the label, then I would be in favor of it.

DR. CANTILENA: Ms. Cohen.

MS. COHEN: If I may read, please. This is on page 3. "Thirty-three percent took the drug for less than 14 days. Only 48 percent of them had spoken with their physician within the last year. Thirty-five percent had not spoken to a healthcare provider at all."

DR. CANTILENA: Dr. Neill.

DR. NEILL: I actually had a hard time voting no because while I agree that the data suggest from the actual use study that patients can't self-select, of those that selected inappropriately, the majority of those seemed to be patients who had heartburn less than once per week and simply are taking a very effective medicine for their not as severe condition.

The remainder that fall into that group, the majority of those appear to have conditions for which the consequences of incorrectly self-selecting are meaningless. While I voted no, that they can't self-select, it doesn't seem to matter much.

None of the things that I've heard so far are different for this product and this labeling than what I've heard related to other OTC products that require patient self-diagnosis of a condition that may mask or represent other important problems. This is a long way of saying while patients may not be able to self-select perfectly, it's acceptable to me.

DR. CANTILENA: Okay. Dr. Clapp.

DR. CLAPP: I find the package labeling ambiguous and a little confusing. I can see room for lots of error. First of all, notifying your doctor if you've had heartburn for three months or longer without talking to your doctor seems a little sequitis. I'm sure that's an area that can leave lots of question in terms of interpretation.

One of my other problems is I don't see a clearly stated relative contraindication of saying -- not contraindication but saying that there is no acute symptomatic relief with this medication.

Thirdly, I think that it should be stated on the panel that there is no expectation that the drug will work for you sooner than two or three days. With those things in mind, I don't think it helps the population adequately select for this drug.

DR. CANTILENA: Dr. Geller.

DR. GELLER: I thought that there were two issues that weren't addressed by the sponsors. One of them is repeat use because their study is only about using the drugs for 14 days. The other is whether patients expected relief -- immediate relief rather than just prevention. These were not addressed.

DR. CANTILENA: Dr. Uden.

DR. UDEN: I basically voted no on the principle that the sponsor doesn't go into doing their -- we haven't asked you do so that's why it's on principle -- go into their label comprehension studies with a benchmark that they are going to try to achieve in all the populations that they are studying.

I'm still concerned about the low literacy group in terms of their understanding the present label at 50 percent. My final concern, and Dr. Clapp touched on this, is that I think if we are concerned about the individuals who have episodic heartburn less than two episodes per week, that there has to be somewhere in the label some expectations for efficacy of the product, that it's not going to work for a day or two or three.

DR. CANTILENA: Dr. Williams.

DR. WILLIAMS: My vote no was purely on the basis that we know that people can identify heartburn but specifically for Prilosec 1 there is an indication for the prevention of heartburn. I think too many people will look at the box and miss their diagnosis of prevention. I think that the sponsor should provide a little bit more caution as to it's use, especially for prevention as its main focus.

DR. CANTILENA: Dr. Fogel.

DR. FOGEL: I voted yes for the same reasons that Dr. Neill voted no. Of the 290 people, I think it was 169 were taking the drug because they had heartburn less than one day a week. They were just wasting their money.

Among those who had contraindicated symptoms and contraindicated medications it's not really clear what the actual medical risk is. I agree with the other comments made around the table that the labeling could be substantially improved but I think that the sponsor is moving in the right direction.

DR. CANTILENA: Dr. Camilleri.

DR. CAMILLERI: Two points. One has already been made. One in four patients got it wrong in the self-selection process. The second is I still think that the deselection study is an important one in terms of risk management.

I think that the sample size was insufficient. Even probably the study setting in a kiosk in a mall probably didn't really address the question in the right study population.

I don't think that this is a show-stopper. This is the sort of study that can be done in the future in terms of making sure that risk can be managed because it is a very small quantitative risk.

DR. CANTILENA: Dr. Brass.

DR. BRASS: I voted yes. I was so impressed by the use of the word sepisodically that I just had to vote yes.

DR. CANTILENA: Thank you, Dr. Brass.

DR. BRASS: We were forced to vote black and white on a gray question. Clearly my yes vote does not reflect my opinion that the label is perfect and there are not things that can be done to improve it.

Key to my vote was the word "adequately." The implications of that word in terms of the decision making in this cohort. I think that we have to be very careful about our expectations about actual use studies and label comprehension studies.

Those of you who have not looked at lots of these studies and understand the kind of protocols and the mechanisms and the all-comers type of data, I mean, to pretend that these numbers have absolute significance, I think, is really overextrapolating.

My standard is always intimately related to an understanding of the consequences of getting it wrong. And as has been pointed out by several of my colleagues, that in terms of the self-selection criteria the errors would be of no clinical significance and would be auto-correcting in terms of subsequent decisions based on the experience of the incorrect selection and represent, therefore, no incremental risk from a public health or an individual subject perspective. Therefore, I voted yes.

DR. CANTILENA: You are sort of rolling question five into question two basically.

DR. BRASS: Everybody else was rolling the label into it and redosing.

DR. CANTILENA: That's all right. You don't have to apologize.

Dr. Cryer.

DR. CRYER: The reasons in support of my vote no have already been stated.

DR. CANTILENA: Dr. Johnson.

DR. JOHNSON: I voted no but it was, again, sort of a difficult no vote on a gray question. I felt that strictly speaking they hadn't adequately shown, but I agree with Dr. Neill and others that those who selected incorrectly is probably clinically irrelevant.

DR. CANTILENA: Comments from Dr. Alfano and Dr. Goldstein.

DR. ALFANO: I would have voted yes on the basis that people are already self-selecting products over the counter to manage this condition de facto. I thought there was a reasonable job done to achieve better education with the label. I do believe, though, as has been stated several times, the label could be improved.

DR. GOLDSTEIN: I would have voted as well yes, but I would like to make an observation to address directly something that Dr. Levine said and indirectly other members of the panel and which the sponsor cannot make.

It is widely acknowledged within our industry that they are probably the best at addressing the consumer. They have had long experience in a variety of fields. And I think in terms of education the concern of Dr. Levine and, indeed, of all of us I think they will develop the programs and are certainly able to and have done so in the past to address that particular issue.

DR. CANTILENA: Thank you. We are on to subject area number three.

DR. PEURA: (Inaudible.)

DR. CANTILENA: I'm sorry? Is there a comment you want to make?

DR. PEURA: A couple of comments that I have heard repeatedly around the table that I really would like to just respond to. One is the comment about 50 percent self-selection within the low-literate population in the label comprehension study.

DR. HOUN: I just want to make sure from the Advisory Committee rules is this okay to have an open discussion now? Is it going to influence further questions? I just don't want to have --

DR. CANTILENA: I understand. In fact, why don't we just say if you have a specific comment that would correct something that has been stated incorrectly, that would be good. Other than that, we have to go on with the questions. If there was something that was assumed that is not factually correct that you have data for, sure, that's fine. If we are going off --

DR. PEURA: Well, the statement that I was going to address is that the 50 percent comprehensional low-lit in the label comprehension is factually correct. I think we lose cite that in the actual use study the low-lit numbers were considerably higher than that.

Also the question of whether it's effective on the first day or not. The drug does begin to become effective on the first day. The first day is statistically significant. It's not like it doesn't work then.

DR. CANTILENA: Okay. The point is that the actual use data is a little bit different than the comprehension study.

DR. UDEN: I absolutely understand that it starts working the first day. When do they appreciate their symptom resolution, though? That was my concern.

DR. PEURA: We have data on symptom resolution on day one. They do appreciate it.

DR. CANTILENA: Okay. Thank you. Let's go ahead and get back to subject area three, actual use. The specific question is, "Did consumers who had a recurrence of heartburn symptoms respond appropriately?"

They want you to specifically comment on the likelihood that consumers will seek advice from a healthcare professional or the likelihood of the consumer using the product again without the advice of a healthcare professional.

DR. GELLER: Could I address the quality of the question?

DR. CANTILENA: Actually --

DR. GELLER: I don't think the studies were designed to answer this question.

DR. CANTILENA: I think here they are actually asking for advice based on your experience, your specialty and expertise. The questions have been revised, edited, thought about, and we are just going to try to answer the questions as written.

In the comments that I will solicit from everyone, you are certainly free to say, "If you had asked me X, I would have said Y." I think for the purposes of the committee, why don't we just stick with the program as they say.

DR. LEVINE: Mr. Chairman.

DR. CANTILENA: Who is talking?

DR. LEVINE: Dr. Levine here.

DR. CANTILENA: Okay.

DR. LEVINE: Could you possibly since this leads into number four, I think, directly, could you combine three and four? Some of our answers would complete be dependent on the duration of therapy.

DR. CANTILENA: I understand that but there is a specific reason why they want to separate these out internally with the divisions. We've had this conversation. That was my first suggestion as well and then the explanation, I think, was satisfactory. It will take a little bit more time but we are significantly ahead of schedule.

DR. FOGEL: Mr. Chairman, I also have a question. There is a great variation in treatment of people who have recurrent symptoms. What exactly is the appropriate treatment that they should have done?

DR. CANTILENA: I actually think that would be in your mind what you think they should have done. We would ask you to explain that as we go around the table because you are exactly right. Really sort of depending on your specific views, that certainly can change. There is no gold standard that is accepted.

I agree it's a difficult question but there are a lot of reasons why we will answer the question so we will answer the question.

Okay. So as you are thinking about this, have in mind sort of the specific data that you are referring to. Let me pose the question to the committee. All those who believe that the consumers did respond appropriately when they had a recurrence of heartburn symptoms, please indicate in the affirmative by raising your hand.

All those who feel that consumers did not respond appropriately, please raise your hand and kept them up for a minute. Sandy is straining.

Okay. Can we have the nos vote again, please. All those who feel the consumers did not, please raise your hand.

DR. GELLER: I'm having a little trouble finding exactly the data that I'm looking for here.

DR. CANTILENA: Okay. The recurrence data in the actual use study.

DR. PEURA: Dr. Cantilena, would you like the slide?

DR. CANTILENA: Actually, I think we have your slides. Correct?

DR. PEURA: Yes. Slide 45.

DR. CANTILENA: Slide 45.

DR. GELLER: That really doesn't tell us what they did.

DR. CANTILENA: What about the FDA slide?

DR. GELLER: Then there is --

DR. CANTILENA: 27 on Shetty's slide.

DR. GELLER: Thank you.

DR. CANTILENA: Our numbers did not add up so what we'll do is we'll revote here in just one minute. Slide No. 27 of Dr. Shetty's presentation. It lists exactly what people did when they had their heartburn return by percentage.

The question to the committee is, "In your opinion is that an appropriate response based on that study?" What we'll do is Dr. Titus will read the no votes and see if we missed anybody.

DR. TITUS: The nos are Dr. Camilleri, Dr. Cryer, Dr. Clapp, Davidoff, and Ms. Cohen.

Dr. Uden, are you a no?

DR. UDEN: I'm a yes.

DR. TITUS: Okay. So there are five nos. That means everybody else voted yes.

DR. CANTILENA: Hold it. There's -- okay. There's one abstention.

What's the final?

DR. HOUN: Sandy, you should just do this again. Yes, no, abstentions.

DR. CANTILENA: Start from the top. Take it from the top. All who believe that the consumers acted appropriately with a recurrence of their heartburn symptoms, please raise your hand indicating yes, that they acted appropriately. Keep your hands up, please.

Okay. All you feel that the consumers did not respond appropriately, please raise your hand and keep them up. Okay. All who have abstained. Okay. So we have 12 yes, five no.

DR. TITUS: The record needs to show that Dr. Geller was an abstention. The remaining 12 members on the committee voted yes.

DR. CANTILENA: Okay. Now what we would like to do is to go around the table and this time we'll start on this side. We'll start with the voting members and then we'll come back to Dr. Alfano and Dr. Goldstein after we complete the circle.

Dr. Johnson, if you could tell the FDA why you voted the way you did.

DR. JOHNSON: I voted yes. There are several reasons for that. One, I think that the responses of individuals probably wasn't perfect and didn't follow necessarily the package instructions because only 20 percent consulted a healthcare professional.

The majority did something. Most of them sought other therapy, in some cases prescription therapy. My sense is that many of them if this product was available OTC would seek another course of this therapy. I guess I'm not terribly concerned about that because my impression is if they saw a physician, the physician would probably give them a course of this therapy.

DR. CANTILENA: Dr. Cryer.

DR. CRYER: I voted no. I agree with several of the comments that Dr. Johnson made. The perspective I come from is I see the entire goal of this is if the patients, the consumers are not responsive to therapy, it's to get them to a healthcare provider. That's been a recurring theme in our discussion today.

Looking specifically at the proposed label, it states, "Do not continue beyond 14 days unless directed by your doctor. If you frequent heartburn continues or returns, it could be a sign of a more serious condition."

It's the specific statement with specific instructions to the consumer that recurrence could indicate a serious condition but, nevertheless, only 20 percent of those individuals went to a healthcare provider. I think, in my opinion, that constitutes a no.

DR. CANTILENA: Thank you.

Dr. Brass.

DR. BRASS: I voted yes because I rewrote the question in my own mind and I posed it to myself as if the results of the actual use study were replicated in the general population would you care? No, would you mind so I could vote yes. I won't mind. The point is, I don't care. It's fine.

DR. CANTILENA: What did you vote anyway? I forget now.

DR. BRASS: Because this profile if it occurred in the general population would be okay with me. That apropos of Dr. Cryer's point, I would have been very interested in the breakdown of a follow-up based on symptomatic relief during the primary treatment course.

In other words, did those 30 percent who didn't respond, were they more likely to have sought medical attention whether or not they recurred quickly after. In other words, how did the clinical response predict the segregation in terms of behaviors afterwards, I think, would have been interesting data to help address that point.

DR. CANTILENA: Dr. Camilleri.

DR. CAMILLERI: I voted no because I looked at the actual use study as a trial run of what would happen if and when this medication was approved for the OTC market. The answer to me is unequivocally that people did not respond appropriately in the context of the study and the question that was being asked.

The question that was being asked was, "You go and see a doctor if you don't respond at the end of these two weeks." That is the standard that we will need to maintain in the label in order to make sure that this can be used safely in the OTC situation. I voted no for that reason.

DR. CANTILENA: Thank you.

Dr. Fogel.

DR. FOGEL: I voted yes. I think that for people who have a mindset of self-medication, these people did the right thing. It would be nice to have a little bit more patient information but the study wasn't designed that way to tell us what was going on, severity of symptoms, so on and so forth.

To my way of thinking, it would be nice if these people came to doctors but a lot of people don't come to the doctor for these symptoms. The way they treated themselves is appropriate.

It's an unrealistic expectation to expect an over-the-counter drug to alter how people think about healthcare. You can't expect people to go to the doctor just because the over-the-counter medication caused them to have a recurrent -- because they had a recurrence of their symptoms when the drug effect disappeared.

DR. CANTILENA: Thank you.

Dr. Williams.

DR. WILLIAMS: I voted yes with the understanding that at least 20 percent of the individuals that received the product did come to the doctors and 20 percent that would have never gone. I'm more optimistic about what I'm seeing than what is negative.

DR. CANTILENA: Dr. Uden.

DR. UDEN: I voted yes because those who didn't respond treated themselves. I wish I could have seen more diet and exercise but this is America and that's not going to happen. End of statement.

DR. CANTILENA: Thank you, Dr. Uden. Dr. Geller.

DR. GELLER: I think I should probably change my vote to no because I just realized that the abstention is totally wasted so I think I'll change my vote to no. I do that because 20 percent said they consulted a healthcare provider but that number probably has a bit of an overestimate because if they did read the label, then they would have known that was what they were supposed to do but they may not have done it and just said they did it.

I guess some of the other issues raised about questions not answered are very disturbing to me. In particular, at the three-month follow-up period there was no opportunity to go and get your Prilosec 1 again over the counter so I would like to know if patients would have gone back to have a second trial of Prilosec 1. I guess I'm distressed that none of the studies the sponsor provides answers the question about compliance in that regard.

DR. BRASS: No, the actual use study did include the opportunity to go back and buy more.

DR. GELLER: But actually the period of time is too short because over that three-month period people are coming in.

DR. BRASS: I'm just correcting the statement that you said they didn't look at it.

DR. GELLER: That was inadequate because people coming in uniformly over the period.

DR. CANTILENA: Thank you.

Dr. Clapp, please.

DR. CLAPP: As the question is posed, I have to vote no because clearly 37 percent did not respond as the package label panel instructed them to, and that is to seek medical advice if the treatment time of 14 days was not sufficient.

DR. CANTILENA: Dr. Neill.

DR. NEILL: I voted yes looking at slide 27 from Dr. Shetty's presentation and considering the fact that we still haven't settled what is appropriate. Given that you are asking us to make a decision about whether consumers are doing the appropriate thing and none of us up here have reached any consensus on what is appropriate, I trust that staff is taking all of these comments with a large grain of salt.

Having said that, of the things that they did in the actual use study, all of those things strike me as being appropriate. I've got to tell you I can't get 10 percent of my patients to change their lifestyle related to anything.

The question and the one concern that I've heard seems to be that not enough of these patients have chosen to go back and see a healthcare provider who in many instances is simply going to tell them to do one of the things that they have already done here.

One of the choices that a consumer does not have but which may be appropriate given the large number that had heartburn for more than five years is doing an endoscopy.

Maybe that will be a meeting sometime in the future when we have endoscopy suites in the mall and patients can take a kit home and they will check their cholesterol on the way home and endoscope themself with a little virtual endoscopy device or something.

DR. CANTILENA: You'll probably be off the committee by then though.

DR. NEILL: Gladly.

DR. CANTILENA: Ms. Cohen, please.

MS. COHEN: Dr. Neill, will you take the endoscopy for me? I would like to quote again because it's the best way I can do it, the responses to the follow-up question here, three months after the study showed that 58 percent of the consumers available for the follow-up had their heartburn return.

Forty-six took an antacid heartburn medication. Twenty-seven took a prescription heartburn medication. It's here, 21 took an OTC acid reducer and 20 of those contacted their healthcare provider. There's more.

What was the final result? They did it but what was the final result? Did they go to a physician? Did they find that they had something serious? I feel it's inadequate information. It starts out and gives it but you don't know the end results. Pardon my Boston accent. I can't do any better.

DR. CANTILENA: Thank you.

Dr. Gilliam.

DR. GILLIAM: I voted yes mostly for the reasons Dr. Neill stated. Also the comments that he and Dr. Brass made on the previous question. I really changed by vote to the previous question in the way I vote this time.

DR. CANTILENA: Dr. Levine.

DR. LEVINE: I voted yes mainly for the reasons Dr. Johnson and Fogel pointed out and for my hope and confidence that with the next question when we change it, possibly the duration of treatment that 57 percent figure will plummet to 25 or 30 percent.

DR. CANTILENA: Dr. Lam.

DR. LAM: I think the question can go either way actually but I voted yes because, in my opinion, the patient did act appropriately by opting for alternative treatment.

DR. CANTILENA: Dr. Patten.

DR. PATTEN: I voted yes based on the information in Dr. Shetty's slide No. 27. As I take a look at that, of the 57 percent who had a recurrence, 27 percent took a prescription medication which says to me they are already under the care of a physician. Then an additional 20 percent consulted a healthcare professional.

DR. CAMILLERI: Excuse me. It's not additional.

DR. PATTEN: Pardon?

DR. CAMILLERI: Nowhere does it say that it is additional.

MS. PECK: Oh, I see. Okay.

DR. CANTILENA: It could be the same person.

MS. PECK: I see. Thank you for correcting me. At any rate, I would concur with Dr. Neill that to have 20 percent of a population opt to see a healthcare provider is rather impressive.

I would like to have seen a longer period of follow-up. I'm not sure when within that three month window the recurrence occurred so I don't know how close in time the recurrence was to a decision to change lifestyles, see a healthcare professional, and so on. It's clear to me that people are doing something.

DR. CANTILENA: Dr. LaMont.

DR. LaMONT: Yes. I agree. In fact, I was going to make the same point that possibly more than 20 percent consulted a physician. I would also like to clarify that we don't do endoscopy on patients who when they stop their PPI get heartburn. What we do is put them back on the PPI. We're not looking to scope a whole bunch of patients and look for cancer that have this complaint. This would be, in my view, inappropriate use of a resource so I voted yes because I think these behaviors are safe and appropriate.

DR. CANTILENA: Thank you.

Dr. Davidoff.

DR. DAVIDOFF: I voted no because I also focused primarily on the issue of whether people went to see a healthcare provider. My concerns there were two. One is the number of 20 percent which I think is a pretty squishy number for two reasons. One is there was no conformation that they did as has been mentioned.

The other is that there is no confidence interval so we don't have any idea what the behavior might be if this kind of behavior were generalized to the general population. This is in one small study.

But then the related question -- important question is does it matter whether they saw a healthcare provider. I agree that very likely many times healthcare providers would have done many of the same things that are on this list. But the healthcare providers might actually have taken a good history.

DR. CANTILENA: Comments from Dr. Goldstein and then Dr. Alfano and then, I think, Dr. Johnson had her hand up.

DR. GOLDSTEIN: The word "appropriate" gave me pause. It depends on how you want to interpret the word "appropriate." Appropriate for whom? For traditional medical orthodoxy or -- and I'm a physician -- or for themselves? Was the patient acting in their own best interest? If it's medical orthodoxy, then perhaps by our standards they did not react appropriately, but if it's for themselves, then it's my feeling that they acted very appropriately and very normally.

I should also point out and reinforce the fact that as Dr. Williams and others have pointed out, the fact that 75 more people were some how persuaded to see their physician who had not is a good on the whole public health outcome. I would have voted certainly yes under those circumstances.

DR. CANTILENA: Thank you.

Dr. Alfano.

DR. ALFANO: I, too, thought it was a good outcome because it was a better outcome than they had before they entered the study in the sense that they saw a healthcare provider or they consulted a healthcare provider at twice the rate that they did in the prior year. Then it does look like there was a modicum of lifestyle changes that occurred.

DR. CANTILENA: Thank you.

Dr. Johnson, did you have an additional comment?

DR. JOHNSON: One additional comment I wanted to make that I guess makes me optimistic is that I think there is a potential in the real world situation for this to be better. While I share many of Ms. Cohen's concerns about direct-to-consumer marketing, I think it can either be used sort of inappropriately or it can be used in a very educational way.

I think if the DTC marketing focuses on reenforcing the importance of follow-up, particularly television advertising where there is a verbal message, follow-up with their physician, I think there is the potential that there may be many, many more patients who eventually have care by their physician than do at present.

DR. CANTILENA: Yes, Dr. Fogel. You have an additional comment?

DR. FOGEL: Actually, can I ask a question to the sponsor about the data?

DR. CANTILENA: About the data, yes, please.

DR. FOGEL: Of the 57 percent that had return of their heartburn, what was the -- do you have any data about the response to these various treatments? Were they still symptomatic or did their symptoms go away with antacids, change in lifestyle, etc.?

DR. PEURA: We did not ask that question. We only asked if it returned and what they were doing about it.

DR. CANTILENA: Okay. Thank you. Let us now turn to the issue of the duration of therapy and repeat use. The next question is, "Given that the treatment for GERD with erosive esophagitis is a minimum of 28 days, is the proposed 14-day duration of therapy acceptable for this population?"

Then I guess if you answer no, we'll ask you in the comment period should the treatment be longer. Does everyone understand the question?

Dr. Johnson.

DR. JOHNSON: Would it be possible for us to have some discussion on this? I frankly am not comfortable voting on this without some discussion and hearing from the gastroenterologists, their feelings on this point.

DR. CANTILENA: Sure. If you would like to open the discussion, go ahead and then we can ask for their input.

DR. JOHNSON: Okay. I would like to hear the opinions of the gastroenterologists. And Dr. Brass, of course.

DR. CANTILENA: Are you going to set the board exam soon?

DR. BRASS: I think, again, in my mind here is how I thought about this. I have no doubt from reading the literature as well as sponsor's data that if you treat these patients for a 14-day course of treatment with omeprazole at 20 milligrams per day, a percentage of that cohort will be symptom free for some follow-up period afterwards. And that if you did endoscopy there would be endoscopic improvement in a subset of that population. Those would both be significantly different than placebo.

I also have no doubt that if you treated for four weeks those numbers would both be higher. If you did it for eight weeks, they would be higher still. I have no doubt that at two weeks there is benefit in terms of both a period of complete symptom relief and endoscopic healing.

At the same time, I am also aware of this being a philosophical shift in trying to get consumers to use an OTC product for a daily basis during which they are not symptomatic but need to complete the period to get benefit. We have all raised concerns about masking symptoms, appropriate compliance, patient's decision making.

That is only with a two-week course. It is unclear in the absence of data whether going behind two weeks would, in fact, be associated with the kind of improvement that has been observed in clinical trials and whether that would be offset by further deterioration in parameters that the committee has already expressed concerns about with respect to compliance, etc.

DR. CANTILENA: I just have one comment in that regard, and that is really you have to balance sort of that issue with the fact that you would be approving and basically endorsing substandard therapy. If the standard therapy is 28 days for this indication, then you are saying it's okay to treat half as long knowing that the failure rate will be higher.

DR. BRASS: We are dealing with a different cohort. Again, remember the cohort we recruited in those studies all had endoscopic inclusion criteria, or at least many of the studies had endoscopic inclusion criteria so the cohorts are not identical so that whether it's substandard care or whether a different care option for a population that selects it is open on the risk to benefit. I'll leave it at that.

DR. CANTILENA: Just one more point and then I'm just going to ask Dr. Ganley one question. If you look at the actual use and you have really more than half the people with recurrent symptoms, my question is what would that have been if we treated for 28 days? That is sort of the first issue in the back of my mind.

Then the question for Dr. Ganley's group is what are the data? Is the Rx label for GERD 28 days? That is the first question. If so, are there data where they looked at shorter courses for efficacy?

DR. GANLEY: The Rx labelling for GERD without erosive esophagitis indicates that -- recommends a dose of 20 milligrams up to four weeks of therapy. For erosive esophagitis the recommended duration of therapy is four to eight weeks.

DR. CANTILENA: Okay. And is there any information if you treat for shorter periods of time?

DR. GANLEY: Well, I believe the sponsor may have a comment. I'm going to -- I'm sorry. Repeat the question, please.

DR. CANTILENA: Yeah. I was just asking if you have any efficacy data and if the treatment period is half of that time for 14 days other than the actual use study here where we have 57 percent recurrence.

DR. RACZKOWSKI: We have seen that data but also with the original approval of the product there was some more data, yes.

DR. PEURA: Dr. Cantilena.

DR. CANTILENA: Yes.

DR. PEURA: I believe we do have data that speaks to shorter duration, two weeks.

DR. CANTILENA: I've seen the efficacy but are there larger studies, I guess, is the question. Are there larger studies that would add our confidence or increase our confidence?

DR. PEURA: Yes, there's the databases published by Bardhan that talks about period of time. There's the Castell study, 14 versus 28.

DR. CANTILENA: So if you treat for 28 is the recurrence rate significantly lower.

DR. ZORICH: I believe I can address that. May I use this? Thank you very much. I would like to specifically address this because the Bardhan study really adds some interesting perspective to this because when you look at the clinical trial data, of course, and even as I showed as healing, and Dr. Brass mentioned going longer is always better.

With the question relative to what actually you need for consumers who are electing to use this product, what the bardhan study actually showed was that the strongest indicator of a good outcome was symptom control at two weeks. In fact, that was more important interestingly enough that negrade of esophagitis at entry, duration of symptoms, body mass index, gender, or age.

So what Bardhan allowed to do, and I didn't really spend a lot of time during my presentation, but for those people who did not respond at 14 days, they immediately were given another 14 days. In that group of individuals, all it really pointed was that you have identified a group of people who will go on to need more chronic therapy. That is the enriched population who ends up being dosed on a more maintenance basis.

It does not, in fact, stopping at 14 days you do not, in fact, have a lesser affect in the majority of the people. In fact, what you're doing by letting those people go for another 14 days is just identifying the people who would be better served we think being directed to their physician after 14 days.

DR. PEURA: Dr. Cantilena, the other point here I think just to be very clear on this from the sponsor's point of view is that the indication that we are seeking to switch is for prevention of frequent heartburn. Although we understand there may be some GERD people within that population, we're not asking for that indication to be switched.

The second point I would make is that when you look at symptomatic relief 14 days is essentially as good as 28 days for symptomatic relief in this population. You get 90 percent of the symptom relief in 14 days so you don't need to keep going any longer than that.

The other thing I think I would say is that in moving this product into the OTC setting, one of the reasons that we looked at a 14-day label instead of the 28-day Rx label is that we think it's actually more prudent in moving this product into the OTC setting to give people instructions for shorter duration of therapy because we think overall that is a better course of therapy for the overall target population as opposed to trying to seek to treat some subset of that population just to be clear.

DR. CANTILENA: Okay. Thank you.

Dr. Geller.

DR. GELLER: I just wanted to say that I would hope the committee would not vote to change the number of days to 28 because we haven't seen any data in this population and it would be, I think, very unfortunate if we would make a decision based on what we surmise to be the case.

I think we have seen plenty of data on 14 days and I think there would have to be many additional studies in this population if they wanted to go to 28 days.

DR. CANTILENA: Dr. Neill.

DR. NEILL: I still haven't heard any of the gastroenterologists on the panel speak and I would like to.

DR. CANTILENA: Dr. Levine, Dr. LaMont, and then Dr. Cryer, and Dr. Fogel.

DR. LEVINE: A comment by the sponsor left me a little fuzzy because our organizations clearly show, as shown right up there, that 28 days of treatment be it intermittent versus maintenance.

Twenty-eight days of treatment with a PPI will markedly increase the healing of a subpopulation who will be taking this drug because of frequent heartburn. That is the group with GERD and that is particularly the group with either nonerosive or erosive esophagitis, etc.

If you, in fact, take this drug for 28 days, the population that will be effectively treated, the percentages will go down significantly. Perhaps 20, 25 percent. At that point if these patients do not have relief, it is much more significant -- albeit there are some exceptions -- it's much more significant to have a referral and subsequent follow-up by the individual physician, general physician, or specialized physician.

To me to negate the 20 percent, 25 percent patients who could use this drug makes no sense. I think it's so hard with symptoms to tell this heterogenous population that we have to treat all the patients with frequent heartburn. That is why I think 28 days is preferable and almost mandatory than 14 days.

DR. CANTILENA: Yes. There's a comment from the FDA first.

DR. RACZKOWSKI: I would like to give Dr. Hugo Gallo Torrez, who is our medical team leader in the Gastrointestinal Division an opportunity to address some of his issues. Thank you.

DR. GALLO TORREZ: This is a quick comment in support of 28 days and in support of what Dr. Levine has specifically just mentioned. According to the labeling erosive esophagitis is healing.

The healing at four weeks with 20 milligrams is 39 percent only, I should add. At two weeks it has to be in the late 20s, early 30s. There will be significant improvement in the healing if one administers the medication for 28 days rather than 14 days.

The early data from the labeling from Prilosec will be 74 percent healing in eight weeks. This is from the labeling.

MS. COHEN: For how long? How far has that been carried out to see how long that lasts, the efficacy?

DR. GALLO TORREZ: Oh, eight weeks treatment 74 percent. Four week treatment, 39 percent.

MS. COHEN: But were those people followed afterwards? It was 74 percent at that time but did that continue?

DR. GALLO TORREZ: After the eight weeks?

MS. COHEN: Yes.

DR. GALLO TORREZ: I don't know.

DR. CANTILENA: She's asking about the recurrence rate of symptoms.

MS. COHEN: Thank you.

DR. CANTILENA: Dr. LaMont, you had a question?

DR. LaMONT: Just a comment about in response to the request for GI input. I think 14 days is the right duration of treatment because it strikes a balance between what the ideal treatment would be at the hands of a gastroenterologist which might be six weeks or even longer and what we could expect with over-the-counter self-treatment. These are patients that are self-treating.

The goal here isn't to heal esophagitis because we don't know whether they have esophagitis or not. It's to heal a simple symptom of heartburn. It seems like if after 14 days that hasn't healed, then these other things would happen.

Notice that in this table on Dr. Shetty's data on slide 27 only six percent of the people that had frequent heartburn return did nothing. That means all the rest did something. They either took more medicine, which would help healing, or they went to a doctor or both.

I think 14 days is right because if we are worried about masking symptoms, and I've heard a lot of that, masking other diseases with this treatment, then this would be a good balance between efficacy for a simple symptom and avoidance of masking.

DR. CANTILENA: Dr. Cryer and then Dr. Fogel.

DR. CRYER: So I agree with Dr. LaMont's comments. It's clearly -- if we are speaking about GERD and erosive esophagitis, the data are very, very clear that 14 days would be inadequate for the treatment of GERD and erosive esophagitis. Thinking about applying that patient population to this potential OTC setting, my overriding goal is to get these people to a doctor for the ones that have severe disease.

So if it is an accurate statement, and it is definitely an accurate statement, that erosive esophagitis will clearly take 28 days of treatment, and the proposed duration is only for 14, what we are doing then is kind of selecting out the people who will have severe symptoms and get then potentially to a healthcare provider earlier even though it may only be 20 percent of those.

Also another comment I would like to make kind of in defense of a previous comment that Dr. comment that Dr. Triebwasser made is that I don't know that this is the appropriate question to ask of comparing this patient population of what we are considering to be frequent heartburn to those who have erosive esophagitis because the population under question is not necessarily -- will not 100 percent have erosive esophagitis.

I have kind of come around a little bit on this. I was looking at the actual use population. Looking at the data from two of the efficacy trials, after 14 days here, if I'm reading it correctly, 75 percent had recurrence of their symptoms within five days. Clearly for 75 percent of the people 14 days of therapy was inadequate. That would likely be the people hopefully who would be pushing to the healthcare provider for 25 percent it was adequate.

DR. CANTILENA: Thank you.

Dr. Fogel.

DR. FOGEL: The Bardhan data and a comment made by Dr. Goldstein previously used the term that this is a filter. A subset of people with heartburn will get better with a 14-day course of treatment. I think that we should accept that. We are not looking to treat reflux disease or chronic heartburn or erosive esophagitis. We just want to treat the symptom.

For those people who remain symptomatic since the drug is going to be over the counter, they may choose on their own to take a second two-week course or they may choose to see a doctor. They have a choice as to what they want to do. They can use other over-the-counter medications.

I think if we go towards a 28-day treatment regimen, I believe Dr. Geller's comments are correct that we have no data about it. We are going to change the locus of care from the physician to the patient with all the risks that are associated with that.

DR. CANTILENA: Dr. Camilleri.

DR. CAMILLERI: Well, you asked the question, Mr. Chairman. If you extend this to four weeks of treatment, what would be the recurrence rate? I would venture to suggest from clinical experience that if you take people with an average of five days of heartburn a week and you treat them for four weeks, you will keep them in remission a little longer but ultimately all will recur within 12 months.

DR. CANTILENA: Okay. Thank you.

Dr. Johnson, have you had enough input from the gastroenterologists? Okay. Very good. We will proceed to question area four. The question will be answered as follows:

All those who believe that a 14-day duration of therapy is acceptable for this population, please raise your hand. All those who feel it is not acceptable, please raise your hand.

DR. TITUS: I want to enter into the record that, if I watched hands correctly, there was one no which was by Dr. Levine. Everyone else voted yes. That means there were 17 yeses and one no.

DR. CANTILENA: Dr. Levine, would you like to add anything to the comments that you have already made on this?

DR. LEVINE: Yes. My point was I think at 28 days you would also have the same recurrence rate that you would have at 14 days. I think the window of getting the patient to somebody is very important for the alarm symptoms.

I think the alarm symptoms of dysphasia and weight loss, etc., are very important to be explained to the patient in anyway that we can educationally. I don't think significant changes in the long-term outcome will change from 14 to 28 days if the alarm symptoms would be lesser or greater by just making that one change.

DR. CANTILENA: Okay. Thank you. Would anyone else like to comment? I'm not going to go around the table unless we have a burning desire. Not heartburn but a burning desire.

Okay. Let's get onto the ultimate question, question five, in the subject area of approvability. Has the sponsor provided sufficient information to support the approval of Prilosec 1 for the prevention of frequent heartburn? Any discussion required?

DR. BRASS: Clarification.

DR. CANTILENA: Yes.

DR. BRASS: If we feel that the label isn't yet perfect but we thought it was probably doable, how would you like us to vote?

DR. CANTILENA: I would prefer that in that case you vote in the affirmative because underneath that we are going to ask you about specific things that you want.

DR. BRASS: Thank you.

DR. CANTILENA: Dr. Ganley, is that correct? Okay. Any other clarifications?

DR. GELLER: Even if you think that there should be another -- that the label should be significantly rewritten and there should be an additional comprehension study with the new label, then you still should vote yes?

DR. CANTILENA: I think so. If you think the condition is something that can be treated that the consumers can adequately self-select ultimately but it needs more work, you'll have the opportunity to specify what that work is. Is that on track with the FDA?

DR. GANLEY: The third bullet under that where it says, "Are there any additional labeling or marketing suggestions." We are trying to make it really clear cut.

DR. HOUN: I think if the panelists feel strongly that something should be done premarket, that should be stated clearly. If you think it could be done postmarket, that should be stated clearly.

DR. GELLER: To clarify, if you think something should be done premarket, should you still vote? And you think it can be done.

DR. HOUN: If you think ultimately it can be approved but you want something done premarket, you can vote that it ultimately can be approved. It is approvable but you would like to see X, Y, or Z premarket. If you think that it should not be approved now at all, then you would be voting no. If you think it can be approved but some of the studies can be done after market, you could state that.

DR. CANTILENA: So really the question is approvability and on your judgement considering really all the data that you've seen and all the discussions that have taken place.

Let me pose the question to the committee. All those who feel that the sponsor has provided sufficient information to support the approval as recently now clarified as an approvable condition and proposal, please raise your hand in the affirmative. Keep your hands up, please.

All those who feel like it's not approvable, raise your hand. Ms. Cohen and Dr. Davidoff negative. Any abstentions?

Did the math check out?

Okay. For this part what I would like to do is as we go around the table, for those of you who recommended approval, we'll ask you to comment on each one of the areas that are listed under the section if the committee recommends approval so that we capture your comments right here and now.

For those two who recommended that it should not be approved, please specify in your comments what additional information should be provided to ultimately support the approval.

We are due to start over on this slide with Dr. Davidoff. You were in the negative so if you can start, please.

DR. DAVIDOFF: Yeah. This is a really difficult vote for me to decide on because I realize that there are potentially some very substantial potential benefits. They are, however, in my view moderate. The number needed to treat in terms of efficacy is in the range of five to 15 which is pretty good but not fantastic.

The gain would be if people actually read and followed the label, the availability of a single course of 14 days of treatment before they went to see a physician and dealt with the issue in another way. This would get more people to physicians, I agree.

On the other hand, the risks, it seem to me, would be a sudden major increase in essentially unsupervised exposure to a drug for which there are, in my view, a number of very major unanswered questions.

The one that concerns me particularly for which I feel there is not evidence that at least I have seen or that I have not seen as being convincing is the issue of potential oncotgenecity. We are not talking about exposure of small numbers of people. We are talking about the exposure of millions of people over long periods of time potentially.

I feel, therefore, it's premature to approve this for over-the-counter use when, in fact, patients can continue to get the drug by going to see their doctor but they are doing it under a somewhat more supervised, followed, and monitored kind of situation.

Finally, I agree there are some labeling issues. I think those probably can be dealt with but they are substantial.

DR. CANTILENA: Thank you.

Dr. LaMont.

DR. LaMONT: Yes. I think the repeatability is the issue that I am most concerned about regarding the labeling. I would suggest that wording to the effect that if the symptoms of heartburn recur within four months of finishing the course of Prilosec 1 or however it's going to be identified, that you should contact your physician.

That would allow them by strict definition a person could take this two or three times a year which I think would be safe. That is, could take 14 days of this medication two or three times a year.

DR. CANTILENA: Okay. So the number of courses then should really be no more than three per year and you are comfortable with the 14 days. Are there specific issues with the label other than for the number of courses that you would like addressed?

DR. LaMONT: No.

DR. CANTILENA: Okay. Dr. Patten. I'm sorry, Dr. Katz had a question.

DR. KATZ: Yes. When people are giving their answers, can you also ask if the answers are things that need to be done prior to or after so where it would lie in terms of additional information requests.

DR. CANTILENA: Thank you.

Dr. LaMont, would you like to go back over that?

DR. LaMONT: This would be a labeling thing so it should be done pre.

DR. CANTILENA: So everything that Dr. LaMont had said would be prior to marketing.

Dr. Patten.

DR. PATTEN: Yes. I have a question regarding the third bulleted point here, "Are there any additional labeling or marketing suggestions?" Does that translate into requirement?

DR. CANTILENA: Yes.

DR. PATTEN: Suggestion is a requirement?

DR. CANTILENA: Yes.

DR. PATTEN: Okay. I'll revisit this.

DR. CANTILENA: So you would like us to come back to you after we go through? Okay. You'll have, I think, probably plenty of time by the time we get around the table.

I'm going to just go out of order here very briefly to Dr. Camilleri who has to leave early so we're just going to skip over here to Dr. Camilleri and then we'll go back.

DR. CAMILLERI: Thank you very much, Mr. Chairman. I completely agree with Dr. LaMont that the issue of repeatability is of paramount importance in my mind. I had actually come to the same conclusions that about once every four months and two to three courses of 14 days each per year would be appropriate. Therefore, under bullet three the unsupervised use of this medication would be two to three courses per year in my mind.

Under bullet four in Phase IV commitments I still think that the deselect study in patients with alarm symptoms, perhaps in a clinic, would be quite important to do just to make sure that the comprehension is also at a very high level for the package insert, for the labeling in the context of a high risk group. Thank you very much.

DR. CANTILENA: That would be something that you would want after approval obviously, Phase IV.

DR. CAMILLERI: Yes. Thank you.

All right. Let's go back here to Dr. Lam.

DR. LAM: On the second bullet I actually put down the limit on the number of calls to about two so basically that is in agreement with two of the GI specialists. As far as the time period, it will be up to about four months or so.

I think actually there are a lot of labeling that need to be worked on in terms of instruction and making it clear to the consumer such as prevention with relief of symptoms and some of the information on the drug interaction. Do not use with the other acid reducer that we have actually have visited before.

Rather than saying that it had to require faithful commitment, I'm thinking in terms of if we do revise the label significantly, that the sponsor should actually do a new label comprehension study before we actually approve it.

DR. CANTILENA: Right. So your specific comments are that we need to improve the labeling on those areas that you mentioned. That should be done prior to approval. I'm sorry. What did you say for the number of courses?

DR. LAM: Two.

DR. CANTILENA: Two courses per year?

DR. LAM: Yes.

DR. CANTILENA: Per year. Okay. Very good.

Dr. Levine.

DR. LEVINE: I'm more concerned about alarm symptoms more than anything else. I think even more than Barrett's in the long run. I think somehow the expertise of the OTC committee and those who have had experience with it to make it headlines, bold, red type, etc., I think that would be the most reassuring thing that I would recommend.

The other thing I would recommend is, again, with techniques like communication techniques, educationally-wise or on the print, that the healthcare provider be contacted. I concur with two or three recurrent episodes.

DR. CANTILENA: Thank you.

Dr. Gilliam.

DR. GILLIAM: I would agree with the alert symptoms. It would be nice if they could -- on the package insert they have, "Keep your doctor in the loop," and if they could somehow bring that out on the carton I think would be good.

Then also bring out more of the tips for managing the heartburn, the lifestyle modification which we know patients have trouble doing. It would be good if that would be brought out more. I would agree with the two to three courses per year.

DR. CANTILENA: Thank you.

Ms. Cohen.

MS. COHEN: I have a question for the FDA. If the advisory committee feels that further work needs to be done in terms of labeling, is there a requirement that they have to do it within a specific period of time?

DR. GANLEY: It depends on whether we think it's a prerequisite for them to get the drug approved. They could get an approvable letter that states that they have to fix the labeling and they have to do a label comprehension study.

MS. COHEN: Within a certain period of time.

DR. GANLEY: Well, no. If they get an approvable letter, they can't mark it until they do the study so I'm sure they would do it rather quickly. The other option is that we would suggest labeling changes and then as a Phase IV commitment. Now Phase IV commitment has a timeframe in it of when they have to have that study completed and submitted to the agency. In previous years that was not the case but now that is a requirement so they have to do that.

MS. COHEN: You can tell I come from a consumer protection background. I feel the follow-up studies are not complete by any means. I think it's premature. I don't know about repeatability I think that the labeling is totally inadequate. I am very concerned about consumer advertising on television.

I think it has moderate efficacy and there might be other things out in the market that will work just as well for consumers. I certainly think we need to give consumers a lot more consumer education and I'm concerned that people will not see a physician.

DR. CANTILENA: Thank you.

Dr. Neill.

DR. NEILL: I agree with Dr. Gilliam that on the package insert it would be nice to put, "Keep the doctor in the loop" on the outside. However, I'm anxious to know what physician you've ever been involved with that allows you to say it never hurts to make a phone call to the doctor. It hasn't been my office. Don't answer that question.

The other comment that I wanted to direct to staff related to the package insert has to do with the graph that is on the top that very clearly to me implies that the advantage of this class of medicine

-- this particular medicine is going to be that it begins to work and lasts for 24 hours and that you ought to take this, not the other two.

It doesn't adequately convey that this medicine works differently than the other two. I would hate to see direct consumer advertising that used that same type of representation because I think that it's misleading.

It's not that it's a bad medicine. It's a different medicine which in some respects is better for what it carries an indication for. That's the only thing I would add there. I don't have any other useful comments.

DR. CANTILENA: Thank you, Dr. Neill.

Dr. Clapp.

DR. CLAPP: I concur with Dr. Lam on his opinions on the duration of treatment and frequency of four months and two courses per year, as well as the concerns that he expressed about labeling.

I really would like the labeling to address in a very clearly stated manner that this is not for acute relief of symptoms. I think that is an important use of disclaimer to have on the label packet. Also I think it would be useful to guide consumers as to when to expect maximum relief of symptoms.

I think there is some ambiguity with saying that it's the prevention of the symptoms of frequent heartburn for 24 hours which then implies that if you use it, within 24 hours you will have relief of symptoms. As I am guided by the pharmacologists, it seems as if that is not the case.

DR. CANTILENA: So just to clarify then, you would want the changes in the label and then another comprehension test or just the changes in the label?

DR. CLAPP: I think the comprehension test is essential. I do have very great concern for low literacy. I see that their studies imply that they were not as competent at self-selection but they were more attentive to taking the medication properly so then we have a group of people who will be then victims of circumstance because they don't understand the use. They will be more likely to purchase and then use unnecessarily.

DR. CANTILENA: Thank you.

Dr. Geller.

DR. GELLER: I agree in large part with what Dr. Lam and Dr. Clapp have said. I am very concerned about the label as written. I think it's written in poor English and it's redundant and needs to be redone. It needs to be written more on a level comprehensible by a population that doesn't read quite so well. That includes use of the phrase "acid reducers." I'm not so sure about wheezing. The sentence, "Notify your doctor if you have had heartburn for three months or longer without talking to your doctor." This is gobblety gook. "For prevention of the symptoms of frequent heartburn for 24 hours." This, too, needs to be rewritten so that a normal person, or even one who is educated can understand it. So the use of the word "frequent" in the label is not good because it means different things to different people. "Not to expect a response in a day," and "Not to use it for symptom relief," are not clear.

I think that once the new label is written there should be a comprehension study prior to marketing this drug OTC. One thing the company should do is have benchmarks for what is acceptable levels of comprehension, preset levels.

As far as a limit on the number of courses a consumer should take over a period of time, I think it should say that we don't have -- data was not given for taking it on more than one occasion of a 14-day period in this population. Rather than extrapolate from other populations, I think it should be -- we should base our information on what has been shown to us.

I guess regarding the Phase IV commitments, I guess some of the phrases used which are informal phrases, "Keep your doctor in the loop," I would like to see an educational campaign that assures that people are more aware that this condition may not be just a passing thing.

DR. CANTILENA: Dr. Geller, can you just clarify on your comment about the number of courses that you can take? Are you saying that it should just be one course without submission of new data?

DR. GELLER: Well, I'm saying that there hasn't been any other data shown. Anything else is extrapolation.

DR. CANTILENA: Okay.

Dr. Uden.

DR. CRYER: I was wondering if I just might make a comment in response to Dr. Geller's concerns. I was wondering whether or not the Bardhan data might actually help you in that assessment. That was a 12-month evaluation. In that evaluation 68 percent of the individuals requested three or fewer recurrent treatment regimens with Prilosec. Courses I should say.

DR. GELLER: I actually wouldn't feel comfortable using that as evidence without reading the study. I was a little bit concerned about again this slippage in the denominator somehow. I noticed that in the presentation but it wasn't anything to ask about. I'm just not familiar enough with that study. I would need to be more familiar with it to be able to use that as evidence.

DR. CANTILENA: Dr. Uden.

DR. UDEN: I'm a recommender of two treatments within two months because if they need two treatments within two months, they should see a physician. I'm a little different in that aspect for others.

Looking at the label I do like -- even though Dr. Zorich apologized for putting in, "Ask your doctor if you have frequent chest pain," etc., I do like that section of the label. I think it is mandatory that it be kept.

I do believe that there have been many suggestions for improving the label and again including the one I would like to see that, "The symptoms will last for 24 hours. You may not see symptom relief for six to eight," or whatever number you want to use that will be supported by literature should be in there as an educational piece.

I do believe, Dr. Brass, that the rigor of these studies can be designed and evaluated better than, I think, what they are. Sometimes I think the information is presented to us not necessarily by this sponsor but by sponsors that we will give them something and they will approve it anyway. I don't like to be in that position.

Lastly, I would like to offer to the sponsors the Mall of America in Minneapolis is a great mall to do your next label comprehension study. Thank you.

DR. CANTILENA: Dr. Uden, are you conflictive with the Mall of America?

Dr. Williams.

DR. WILLIAMS: I endorse the Mall of America. I've been there. It's a great place. My concern on voting yes is two fold. I think that the repetition of two cycles is adequate. No more than three in a year.

The other concern that I have that's not really been talked about is the delay in the onset of action. I hear two things. One was that the product should be taken before breakfast on an empty stomach with a glass of water or any other beverage. I don't know if that's been clarified.

The second thing is that the product has a time delay of about one hour before it's useful. I'm concerned that those two things be evaluated as we go to the marketing of this product and that the labeling that we talked about certainly be put in place.

DR. CANTILENA: And then should that label be tested prior to approval for marketing?

DR. WILLIAMS: Yes, I agree. Prior to marketing.

DR. CANTILENA: Okay. Dr. Fogel.

DR. FOGEL: I voted yes with regard to subsequent courses of medication. The literature indicates that 80 percent of people will have recurrence of either erosive esophagitis or symptoms within a number of months.

I agree with the suggestion of two, three courses a year. I think that is fine but I think there has to be an emphasis in the information that goes with this drug that if symptoms are not better after two courses of if you require more than three courses of the drug in a year you should see a physician.

The other specific point I want to make is that on the label where it says ask a doctor before use if you have any of the following and it talks about chest pain, wheezing, etc., that's really not strong enough. The labeling needs to say that, "The following symptoms may be indicative of series diseases other than heartburn. If you have the following symptoms, see your doctor. I think that needs to be made much strong. I think in general the labeling needs to be improved substantially and I think it should be retested before the drug is released,.

The final point I wanted to make is I think that the direct-to-consumer advertising that goes with this drug needs to be scrutinized carefully. This is a major change in how we treat people with esophageal reflux. I think close supervision of the information that is disseminated is very important.

DR. CANTILENA: I would certainly agree with that. I just have one sort of follow-up question for Dr. Ganley regarding chest pain warnings. Is that on the other heartburn drugs that are out there now? Why not?

DR. GANLEY: Did you want class labeling then for them?

DR. CANTILENA: Well, I mean, I think if we are treating the same symptom we have the same concerns. It should be on the other products. Otherwise, it's illogical.

DR. GANLEY: I think the -- you know, I wasn't around when the other ones came over and some of the folks here, Eric, I think, were. I don't know if that was discussed then.

I think this is sort of a different situation where I don't think they envisioned the use that would be occurring with these products that occurs now which some people just take them all the time. Clearly if that is the situation, it may be something to consider. I know Eric may be able to add some historical --

DR. BRASS: Well, I won't rehash or even attempt to rehash. I think it's fair to say we are learning all the time and we are learning more about the population that use it.

It think early on one of the concerns were whether these messages could be effectively communicated to a degree that would justify taking valuable label space. Again, every square centimeter on a label is very, very valuable and you prioritize. I think at that time there was concern about the effectiveness of communicating those messages and what the benefit would be.

DR. CANTILENA: Okay. I think it's your turn, Dr. Brass.

DR. BRASS: All my comments are related to the label and it would be my preference if they were addressed premarketing to the agency's satisfaction, whatever that took whether that was another study or your own wisdom.

I have a minor point that actually starts at the very beginning. Under uses it reads, "For prevention of the symptoms of frequent heartburn for 24 hours." I found the phrase "for 24 hours" dangling at the end of that to be very confusing how that would be interpreted.

I would put that under the paradigm of communicating the expectations about when benefit would occur more clearly that you are really taking a two-week course and how that overall message gets communicated.

As was mentioned, I agree that a whole lot of more appropriate bolding and coloring and highlighting for the more significant messages would be important.

I already alluded to that I'm not sure the message under warnings, "Notify your doctor if you've had heartburn for three months or longer without talking to your doctor," has any added benefit against this whole matrix.

I agree completely with the intent of that message and if it's felt to contribute to patients contacting their doctor, that's fine, but we've actually not seen the impact of that statement evaluated and, again, it takes up a lot of space.

"Do not use with other acid reducers." Again, the complete rationale for that wasn't entirely whether it was because they were concerned it would interfere with efficacy or that you would be using redundant medications. If so, whether or not you need to name more specifically what medications.

Under the drug interactions, I would be in favor of leaving warfarin even though there is not much data only because I believe any patients on warfarin should talk to their doctor before they take anything as a general rule. If putting a brand name next to that improved that, I would be in favor of that.

In contrast, I'm not sure if phenytoin has an adequate basis and would lean towards removing that. The differentiation between ketoconazole and itraconazole I confess to have been totally confused by the rationale, why one was on the Rx and the other was on the OTC.

Perhaps in that class whether some phraseology like a prescription medicine to treat a fungal infection or something like that might communicate that whole group more efficiently and more effectively than the generic names which might not be recognized.

Under, "Stop. Use a doctor if..." it says, "Heartburn continues or returns after using this product everyday for 14 days." That means you could have continued pain for the entire 14 days before you would call somebody.

I'm not sure if the curve of benefit suggest that is the optimal time point if you are still having symptoms to call or whether an earlier time point might be better.

I agree that the label should contain something about repeat courses but I actually don't know how to do that. I think a number like three is right but I don't think we have very much experience in communicating. That is a relatively abstract concept at the time of purchase to communicate to a consumer.

I'm not sure we have any experience in communicating a message like that. That would be an example where if that was felt to be important for the safe use of this drug over-the-counter some serious thought about what syntax would effectively communicate that concept would be required.

I also agree with the comments about strengthening the "don't use" under various circumstances should be more dogmatically "don't use" as opposed to down at the bottom, "Notify your doctor," or "Ask a doctor before use." To my perspective, and this is a general issue we've talked about before in the labeling, "ask a doctor before you use" as opposed to "don't use."

Same with the pregnancy warning. It implies that it's probably going to be okay but just check as opposed to "you should not do this unless instructed by." And whether the strengths or the warnings for the warning symptoms and the pregnancy can be made more definitive with stronger language would be another concern I would have. Thank you.

DR. CANTILENA: Thank you, Dr. Brass.

Dr. Cryer.

DR. CRYER: Sir, my principal two issues with regard to OTC usage as well as labeling are comprehension and getting the consumer to the physician early in the event of inadequate symptom relief. Given that, I would agree with all of the comments that have been stated that the label needs to be rewritten in a way that is more comprehensible at a lower grade level along the lines of some of the specifics that have been discussed.

It's my understanding that the label in its current format actually has not been tested. Given that it was likely to be rewritten, I think it would be very important to test comprehension of the rewritten label prior to actual introduction to the market.

With regard to getting people to the doctor earlier, we have here a comment that says, "Stop use and ask a doctor if your heartburn returns after 14 days of this product," which I think is appropriate. I think you might want to consider also including that specifically under the directions.

It doesn't specifically state in the directions box that you might want to see your doctor if the heartburn returns after 14 days. You might want to either move it to that box or state it in two places.

Then with regard to this issue of these drug-drug interactions, I actually agree with the comments that Dr. Brass made. The impressive data that I saw was that the comprehension with respect to the drug-drug interaction significantly increased from 50 percent to 80 percent when comparing a generic -- when specifically listing the generic versus the trade name.

This may be one instance in which you may want to break your precedent with regard to the inclusion of trade names, specifically with respect to this drug-drug interaction. I think that is actually a public health benefit.

DR. CANTILENA: Thank you very much, Dr. Cryer.

Dr. Johnson.

DR. JOHNSON: I probably won't say much that hasn't already been said but with the thought that repeated messages give strength, then I'll repeat them.

I think two to three courses per year is reasonable. In terms of labeling, again, I think there are a lot of changes needed. I find the "uses for prevention" statement confusing as many others have. I find the "notify your doctor" statement very confusing.

Under, "Do not use with acid reducers," I'm not sure that if you ask a lot of health professionals what exactly is an acid reducer -- that's not what we call them. I think there needs to be some way to make that message clearer to patients what that actually means.

I agree that the doctor statements are important and probably should be strengthened. As sort of a sidebar in terms of class labeling, I think one could make an argument that kind of labeling for antacids and H₂-RAs is even more important than it is for this product because those products are used for treatment of acute symptoms.

If you consider that the thing that we are probably most concerned about being confused is acute MI and you look at the data on acute MI and how many patients have taken an antacid prior to coming to the ER, I think the evidence is even more compelling that it should be on those products. That is sort of a sidebar.

In terms of the drug interactions, I really think it's critical that brand names are on there. I'm not sure why the policy is such that it precludes that. I concur with Dr. Brass that warfarin probably is justified to stay on and phenytoin may not be. I think there is no logic to having ketoconazole but not itraconazole.

I think the message needs to be strengthened that this product is not for immediate relief of symptoms and is not for episodic use. I think that's about it for the labeling.

In terms of Phase IV, in terms of an assessment I'm not sure I have a recommendation but I feel very strongly about the direct-to-consumer marketing and feel that there really should be some mandatory aspects in their direct-to-consumer marketing.

I think it needs to encourage lifestyle changes that would increase the response or decrease the need for drugs. There needs to be a strong message in that marketing about referral to physicians or when they need to see their physician.

I think it may be also very important to clarify in that message that they are different than other heartburn medications. I guess that could be a selling point but specifically that they are not for acute symptoms so there is not confusion about use of this product compared to the other heartburn products.

DR. CANTILENA: Thank you, Dr. Johnson.

Dr. Alfano.

DR. ALFANO: Yes. I also agree two to three times a year is appropriate. I think it should be two to three because I think we all agree that it is an empirical decision as opposed to fixing a number as what's appropriate.

I've already commented that I think there should be some label changes along the lines of what everyone has suggested here and I won't repeat that other than to make a comment that I thought Dr. Johnson had a particularly cogent point about the number of people who get an MI and will take an antacid. The fact those other products are not labeled is actually a little disconcerting.

Also, I don't see a need for Phase IV trial but would agree with some of the other panelist that suggest that the FDA ought to be able to make a determination as to how the labeling should be modified and whether or not it should be studied again and how.

DR. CANTILENA: Dr. Goldstein. I believe labeling always needs to be internally consistent and memorable where possible. In this case the caution, warnings I should say, about if you've had heartburn for three months or longer might well be linked to no more than two courses in a three-month period or three courses in a year, three being the link.

The other thing is I've spent considerable time looking at that "notify your doctor if you had heartburn for three months or long without talking to your doctor," and wondering how you notify your doctor without talking to him.

Of course, there are intermediaries. Nonetheless, I truly believe that the sponsor and the agency together will be able to with their resources and their viewpoints work out appropriate labeling that will make these committees and all of us satisfied.

As for Phase IV requirements, perhaps the sponsor may have more to say on that. I'll leave that to them.

DR. CANTILENA: Okay. Thank you. My comment is not to restate what has already been stated but just in general to the agency that approval of this would be significant new ground for the agency. I think that you've all heard fairly consistency that the label needs a lot of help and revision and validation.

I think all that needs to be done prior to approval. We want to get this as close to right as possible. Then I think it would be helpful for a limited focused Phase IV study to test the effectiveness of the label. It's sort of the ultimate test but we have a lot of work to do on it for all the reasons that have been mentioned. I see it as an approvable proposition for the reasons stated.

Dr. Davidoff. I'm sorry, Dr. Bull.

DR. BULL: One clarification on your comments. Do you see the labeling comprehension as needing to be done before the drug is put on the market?

DR. CANTILENA: Oh, absolutely.

DR. BULL: Okay. That would not be Phase IV.

DR. CANTILENA: Absolutely. The label as presented is in great need. Significant changes have to be made. Then it has to be validated with a comprehensive study. All the comments about subpopulations that need to be tested, I think, are important.

I'm sorry, Dr. Patten. I told you it would be a while but it was so long I actually forgot. My apologies.

DR. PATTEN: That's fine. Many of the things that I had in mind about label suggestions have already been made but there are a couple of additional points that I would like to make. It's very clear that this will be marketed as a preventive. I think people will be looking for something that they might do in addition to prevention to give them very quick relief.

It's noted here that it is not to be used with other acid reducers. I wonder if it might be wise to tell people what they can use for symptom relief. If it can be used with an antacid, for example, perhaps that should be stated.

I think the matter of comprehensibility when you are letting people know how many repetitions they can do in a year is going to be a very serious issue to be tested. I think that will be very challenging and very different than saying you can take eight tablets in a 24-hour period. I think that has to be attended to very carefully.

I think that something should be said about the relationship of the morning dose to food intake. Is there an optimal lapse of time between taking the tablet and eating your breakfast, for example. I think people should be given some guidelines if that is at all possible.

Then finally just a very picky point under "other information." Maybe this is even a theological issue. Is this a carton or is this a package? We are told, "Keep the carton and the package insert."

I gather that both of these terms refer to the box that this is going to come in. Just to be consistent with the terminology, I think it would be useful and would improve comprehensibility to a broader spectrum of the population. That's it. Thank you.

DR. CANTILENA: Thank you for those comments.

Are there any other issues? Dr. Davidoff and then for the FDA.

DR. DAVIDOFF: Yes. Since I voted no, I realized that I didn't meet my obligation to say anything about this question for those who did not recommend approval, namely what additional information needs to be provided.

Actually, since it looks like the recommendation is for approval, I should convert that into what should the agency require as any Phase IV commitments because essentially I think what I have in mind is the same thing. The drug, as I understand it, has been on the market as a prescription drug for 13 years.

That's a pretty long time, but it may not be long enough to detect some of the longer-term effects that I raise as concerns. It might take 15 or 18 or 20 or 25 years for substantial population-wide effects both in terms of masking of neoplastic disease or potentially induction to appear.

I would, therefore, strongly encourage that particularly if the drug is approved for over the counter some form of continuing monitoring, very close looks at the data from a sophisticated epidemiological point of view. I mean, it's one thing to hear about

-- it's important to hear about the experience of individual practices, but I don't think that answers the question. I would strongly encourage that that happen.

On the issue of -- I also didn't comment on the label and I won't repeat what other people have said but there are a couple of things that I thought might be useful to add.

One was that in terms of wording that might help understand what the purpose of this drug is, prevention of the return of symptoms strikes me as a kind of wording that might actually be more helpful than prevention for 24 hours.

When you think about it, that is really what it's aimed to do. It's not to prevent -- it's not to relieve what's happening now. It's to prevent it from coming back. The word return or recurrence or something might be useful.

My final comment on the label has to do with this issue of digoxin because that keeps slipping off the table. I can't understand why because we've heard two things about it. One is that it's a very small study, 22 patients with no confidence intervals presented showed no difference. We have no idea how negative that study was because we didn't see the confidence intervals.

We also heard there was a 10 percent increase in levels. Digitalis is a dangerous toxic drug and I would suggest either that there be more data collected -- that's what ideally I would like to have happen -- before a decision is made about labeling on digoxin. Or, pending that, that the conservative thing would be to put some kind of wording that does at least make a caution about people on digoxin, particularly if they are in renal failure.

DR. CANTILENA: I think that is an excellent point. Just a short of word for future sponsors is that it's very helpful to the committee to see individual data and/or confidence intervals as opposed to just an average without anything. That's not very helpful and it probably does more harm than good.

Comments from the FDA? Dr. Houn.

DR. HOUN: I just want to summarize a lot of you gave labelling suggestions. Some of them very extensive dealing with the indications, the warnings, new information such as limited course two or three times a year and no more than maybe every four months or so, new messages.

Most of you except for, I guess, Dr. Brass as voting members are wanting to see them tested out in terms of their level of comprehension by a naturalistic public setting prior to approving the final label. Is that a correct summary?

DR. CANTILENA: I believe it is.

DR. HOUN: Then I also heard this new information about limiting frequency to two to three courses in a year. Is that something you want tested to show that, in fact, people are not taking this continuously or are you saying this is nice to have on the label but you don't need to have the data to show that people are, in fact, complying with that aspect?

DR. CANTILENA: We didn't actually answer that. I guess -- well, I'm not sure. How many of the committee would want to see that specific statement actually tested as opposed to just being there? All in favor of seeing it actually tested?

DR. BRASS: Tested on a comprehension basis or a one-year longitudinal trial for use?

DR. CANTILENA: Tested in a comprehension. Is that what you're asking?

DR. HOUN: Well, I would ask both. I'm asking both.

DR. CANTILENA: Okay. How many would like to see the new label including that specific message tested in a one-year or longer actual use study? All in favor, raise your hand. One hand. I don't know if you're recording this.

DR. GELLER: Can I just justify that? I guess I don't like the idea of putting number of doses permitted or recommended on the label without any data. The alternative is to say that only one dose was tested.

I mean, people -- it's obvious from what has been said here that we expect people to use it more than once. Perhaps some caution should be urged but to say two or three times is okay without data is, I think, really an extrapolation that we shouldn't be making.

DR. CANTILENA: Okay. I guess just to follow through, how many would like to see that message that you should not have more than two or three courses in a year tested in a label comprehension setting? Hands? Okay.

Sandy, do you have them all?

How many of you don't want to see that tested at all or don't think it's required? Don't think it's required.

DR. GOLDSTEIN: Mr. Chairman.

DR. CANTILENA: Yes, sir.

DR. GOLDSTEIN: If I may, all of this, two, three courses of presumably the original 14 days has been in the last few minutes talked about with the background of 28 days over and over again. A little while ago we heard that 28 days was more effective. To now limit it in this fashion may be not entirely appropriate.

DR. CANTILENA: I just think that we're asking if that message should be tested that we can effectively translate that or communicate that. I think that is all we are saying. I think the job of the Advisory Committee is to advise the FDA on whether or not there should be a limit and, if so, what it should be. A lot of what we do we don't have the specific study to answer the question. If that study existed, they probably wouldn't ask us the question because they would have the answer.

Are there other issues from the FDA that you would like addressed? Any issues -- I'm sorry. Dr. Clapp.

DR. CLAPP: I have concerns about the methodology of the testing. Of course, all testing has to be contrived to a degree, but in that the participants were given a diary and the diary was rather -- well, we didn't hear specifically what was requested of the diary but it seemed like a rather nebulous vehicle for putting your thoughts and impressions of the medication rather than asking specifics as to, "What did you take instead of the medicine when you missed it?"

My other concern is that there was a great deal of artificial incentive in that and writing in the diary would make you more likely to take your medication just by the fact that you knew that was an obligation for which you had to resubmit your diary to get your 100 dollars and get the 20 dollars at the onset of the study. I would see that there could be other mechanisms by testing whether or not people were compliant that wouldn't involve such an inherent incentive.

Perhaps having a nurse go to the home at the end of the 14 days and see how many blister packs are left. There could be other methods that would increase the reliability of the data and not make it such an incentive driven solution.

DR. CANTILENA: Okay. Other questions or comments from FDA? Do you have one?

DR. HOUN: I hate to ask this. Are you interested in seeing this again?

DR. BRASS: No.

DR. CANTILENA: Does anyone else want to come back to Bethesda? I guess why don't we just vote on it for whoever is left.

DR. HOUN: Or you could just leave it to us.

DR. CANTILENA: Why don't we -- yeah. If it's an issue where it's a hard call, then it's obviously something you should look at but I think all the issues are addressed. I think we've probably talked about it enough.

Are there any other issues from FDA? Any issues from the sponsor? Thank you for your patience in having us talk about your drug all day.

DR. PEURA: No, not at this time. I want to thank the committee very much.

DR. CANTILENA: Okay. There are just two announcements from Dr. Titus and then we will close.

DR. TITUS: Thank you all. There are taxis downstairs going to National for those that are leaving immediately. If you want your material returned to you, put your name tag on top of the material and we will mail it to you. Otherwise, we will take care of it here. Thank you.

DR. CANTILENA: Okay. Thank you, everyone. The meeting is adjourned.

(Whereupon, at 4:35 p.m. the meeting was adjourned.)