UNITED STATES FOOD AND DRUG ADMINISTRATION

P R O C E E D I N G S 8:05 AM

MR. JEHN: Okay, we are going to go ahead and get started here. Mr. Chairperson, members of the Committee, invited guests, consultants and public participants, I would like to invite you all to the 84th meeting of the Blood Products Advisory Committee.

I am Donald Jehn, the Executive Secretary for this meeting. The BPAC meeting is unusual in this case because it is a cross-center meeting. We have a CDER topic to discuss in the morning here and again, we welcome our CDER folks.

At this time I would like to introduce the individuals seated at the head tables for today's first session. On my right, going around the table we have Dr. Allen, our Chair. We have Dr. Davis on the Committee here and Dr. Kato is a temporary voting member. He is not present at the moment, Dr. Harvey Klein, Dr. Brittenham, Dr.Whittaker, Dr. Doppelt, our patient rep, Susan Winner, our consumer rep, Judith Baker, Marc Ghany, a temporary voting member, Dr. Quirolo, Dr. Liana Harvath, Dr.Schreiber and our industry rep, Dr. Maldonado and then FDA folks sitting at the table are again, Dr. Shashaty, Dr. Kathy Robie-Suh, Dr. George Jills, Dr. Karen Weiss and Dr. Richard Pazdur.

I have got a COI statement, rather lengthy that I need to read now and again, also, what we have linked by teleconference is Dr. Portman from Houston, Texas. He was sick earlier. So, he couldn't travel, but he has agreed to be on with us today via teleconference.

The Food and Drug Administration is convening today's meeting of the Blood Products Advisory Committee under the authority of the Federal Advisory Committee Act of 1972. With the exception of the industry rep, all members and consultants of the Committee are special government employees or regular federal employees from other agencies and are subject to federal conflict of interest laws and regulations.

The following information on the status of this Advisory Committee's compliance with federal ethics and conflict of interest laws including but not limited to 18 US Code, Section 208, 21 US Code, Section 355 and 4 is being provided to participants today in today's meeting and to the public. FDA has determined that members of the Advisory Committee and consultants of the Committee are in compliance with federal ethics and conflict of interest laws including but not limited to 18 US Code, Section 208 and 21 US Code, Section 355 and 4.

Under 18 US Code, Section 208 applicable to all government agencies and 21 US Code, Section 355 and 4 applicable to certain FDA committees Congress has authorized FDA to grant waivers to special government employees who have financial conflicts when it is determined that the agency's need for particular individual services outweighs his or her potential conflict of interest, Section 208 and when participation is necessary to afford essential expertise, Section 355.

Members and consultants of the Committee who are special government employees at today's meeting include special government employees appointed as temporary voting members. They have been screened for potential financial conflicts of interest of their own as well as those imputed to them including those of their employer, spouse or minor child related to the discussions of a new drug application, NDA 21-882, proposed trade name Exjade, deferasirox tablets for oral suspension, Novartis Pharmaceutical Corporation proposed for the indication of the treatment of chronic iron overload due to blood transfusions, transfusional hemosiderosis. Also, the Committee will discuss the research programs of the Laboratory of Hemostasis and the Laboratory of Plasma Derivatives, Division of Hematology, Office of Blood Research and Review, Center for Biologics Evaluation and Research.

In closed session the Committee will discuss the report from the laboratory site visit of February 25, 2005.

These interests may include investments, consulting, expert witness testimony, contracts, grants, CRDAs, teaching, speaking, writing, patents and royalties and primary employment.

Today's agenda includes topic one, review, discussion and recommendations on the safety of Exjade, deferasirox tables for oral suspension for the treatment of chronic iron overload due to blood transfusions manufactured by Novartis Pharmaceuticals.

For topic two the Committee will review and discuss the research programs of the Laboratory of Hemostasis and Laboratory of Plasma Derivatives, Division of Hematology, Office of Blood Research and Review.

In accordance with 18 US Code, Section 208, B3 waivers have been granted to the following topic one participants. Please note that all interests are in firms that could potentially be affected by the Committee's discussions, Dr. Klein for ownership of stock in the sponsor currently valued between $5001 and $25,000, Dr. Portman who is participating via teleconference for consulting with the sponsor on unrelated matters for which he receives less than $10,001 for negotiation consulting with sponsor on unrelated matters and for serving as a site investigator for unrelated study supported by the sponsor where one patient was enrolled at less than $10,001. A copy of the written waivers statements may be obtained by submitting a written request to the agency's Freedom of Information Office, Room 12A30 of the Parklawn Building.

This conflict of interest statement and all acknowledgement and consent to disclosure statements will be available for review at the Registration Table.

Dr. Samuel Maldonado is serving as the industry representative acting on behalf of all related industry and is employed by Johnson & Johnson. Industry representatives are not special government employees and do not vote.

Based on the agenda for topic two it has been determined that the Committee discussions present no actual appearance of conflict of interest for today's meetings.

We would like to remind members and consultants that if the discussions involve any other products and firms not already on the agenda for which an FDA participant has a personal or imputed financial interest the participants need to exclude themselves from such involvement and their exclusion will be noted for the record.

FDA encourages all other participants to advise the Committee of any financial relationships that you may have with the sponsor, its product or if known the direct competitors.

Thank you.

After reading that very lengthy COI statement I just wanted to before I turn it over to our Chair, Dr. Allen, I just want to make sure everybody has their cell phones either muted or off and again, welcome to the meeting.

DR. ALLEN: Thank you, Don. As Chairman, I would like to welcome you all to the discussions today. This is a very important consideration. We have got one topic that is going to take up most of the day and I would just like to remind each of the speakers, please adhere to your time limits that are given to you and that will be clocked over here.

Given the number of presentations and the lengthy deliberations that the Committee is being asked to undertake I would really like to ask everybody please to adhere to the time limits, to give the Committee members a chance to ask questions and to participate in the necessary discussion.

For our FDA colleagues over here the equipment is sort of blocking you over there. So, if you have a comment or want to be recognized please ask one of your colleagues down towards the left of your side to wave a flag or something because I might otherwise not be able to see you and recognize you.

Okay, we will begin our morning's presentations with an introduction from the FDA perspective by Dr. George Mills, Director of the Division of Medical Imaging and Hematology Products.

DR. MILLS: Good morning. Thank you. I would like to take this opportunity to thank the Blood Products Advisory Committee this morning as well as the attendants from the public today as we bring Exjade first oral iron chelator to our review and discussion this morning of the safety and efficacy elements.

We will have presentations and then following that we have a number of very interesting questions. We will address many of the safety and efficacy issues that you will be hearing about this morning.

Without further information or need, we will proceed from here.

Thank you, Mr. Chairman.

DR. ALLEN: We will move right into our sponsor presentations by Novartis Pharmaceutical Corporation, first an introduction by Dr. Narang, Vice President, Global Head Drug Regulatory Affairs, Oncology Business Unit, Novartis.

Welcome.

DR. NARANG: Thank you, Mr. Chairman. Good morning. Thank you, Mr. Secretary, Mr. Jehn and other distinguished members of the Advisory Committee.

My name is P. K, Narang and I am the Global Head for Drug Regulatory Affairs for Oncology at Novartis Pharmaceuticals.

It I my pleasure to be here today and have the opportunity to present our new and exciting oral iron chelator submitted under NDA 21-882.

Our agenda this morning is as follows: After my brief introduction Dr. Porter from University College of London will present the clinical complications of iron overload, current treatment options and the need for therapy.

He is an internationally recognized expert in the field of chelation therapy and hemoglobinopathies and has treated iron overload patients with beta-thalassemia, sickle cell disease and myelodysplastic syndrome. He was, also, an investigator in our trials. He will be followed by Dr. Peter Marks, Senior Director in Clinical Development at Novartis who will present the efficacy and safety data from our large clinical profile.

Then we will have Dr. Elliott Vichinsky from Children's Hospital in Oakland who will provide his perspective on the benefit/risk profile for ICL and its place in therapy of iron overload. He is a world renowned expert in the field of chelation therapy and is currently the Chairman of the Medical Advisory Board of Cooley's Anemia Foundation.

He was, also, an investigator in our clinical trial program.

Last but not least I will ask Dr. Marks who is presenting the safety and efficacy data for us to moderate the Q&A session after our sponsor's presentation.

Iron overload is recognized as a serious outcome in patients with anemias who routinely receive life-saving blood transfusions.

If patients do not receive adequate chelation therapy iron accumulates in various endocrine organs and the heart, eventually causing end organ failure, significant morbidity and early mortality.

Desferal is the only approved chelation therapy in the US. It is a parenteral product, also, from Novartis and has been on the market for over 40 years.

While Desferal is deemed safe and effective its dosing is very inconvenient and leads to poor compliance. Its administration requires 8-to-12-hour subcutaneous infusion using pumps daily for 5 to 7 days per week.

As you can imagine this must be difficult for young children, adolescents and their parents who often risk potential complications of iron overload rather than deal with the cumbersome dosing.

So, suffice it to say there exists a significant unmet medical need for a safe and effective chelator with a positive benefit/risk profile.

ICL670, the one we are discussing today represents a new class of tridentate iron chelator. Its generic name as was previously mentioned was deferasirox and the brand name is Exjade.

This compound is the result of 14 years of research from Novartis for an orally bioavailable yet safe and effective iron chelator. Its drug kinetics, half life estimates, etc., support once daily dosing for this product and the data we will present today will demonstrate that ICL670 is efficacious and has an acceptable safety profile both in adults and in children.

The proposed indication in the NDA is for the treatment of chronic iron overload due to blood transfusions in adults and pediatric patients 2 years of age or older.

As you must realize the chronic iron overload population for ICL use meets the regulatory definition of rare diseases. That means prevalence in the US is 200,000 patients or less.

Just to briefly give you a regulatory history for developing ICL Novartis conducted a very large global clinical program. Three years after we submitted the IND ICL use in transfusion iron overload as I mentioned, a rare disease setting, was granted an orphan designation and a fast track status a year later.

Fast track programs are designed to facilitate development and expedite the review of promising agents to treat serious and life-threatening conditions. Agents with this designation are deemed to exhibit a potential to address a significant unmet medical need.

Using the special protocol assessment provision prior to initiation of our clinical trials in 2003, Novartis sought the agency feedback on issues related to these trials being deemed adequate. That means we discussed issues related to design of studies, conduct of studies and aspects of analysis plans.

ICL670 NDA was accepted into a continuous marketing program in January of this year. We submitted the final data in May and in June we learned that the agency had granted ICL NDA a priority review.

With these regulatory provisions FDA has provided us with timely guidance and an ongoing recognition for ICL and its potential to meet a significant unmet medical need.

Novartis' development and registration strategy focused on beta-thalassemia as a model disease to document drug effectiveness. This was supported by the agency as mortality and morbidity due to iron overload are well-documented in this population.

It is worth noting that the studies in the ICL program are the largest prospective trials every conducted for an iron chelator especially considering the rare disease setting.

Based on the minutes of the formal meetings and the SPA feedback we had from the agency we also believe that the findings from beta-thalassemia were generalizable to other iron overload settings. In fact, FDA's intent is captured by the following feedback we received, that the conclusion of efficacy from ICL studies 107 and 108 that we will share with you today may be applicable to other conditions of transfusional iron overload and anemias including sickle cell disease.

With that belief about beta-thalassemia disease population other studies such as in sickle cell disease were designed with a focus on safety and today we will share those data with you as well.

We,also, recognize the potential use of ICL in children. So, we strategically planned and prospectively included a large number of pediatric patients in our key pivotal trials, almost 45 percent.

These are the three key trials that we will be discussing today that are the basis for the assessment of safety and efficacy of ICL670.

Study 107, 108 and 109; 107 is the pivotal controlled trial versus deferoxamine, DFO in beta-thalassemia patients.

Study 108 is the supportive trial in beta-thalassemia patients which also included patients with rate anemias. These two primary efficacy trials enrolled 770 patients of whom 480 patients received ICL.

Other than 107 and 108 the safety population includes patients from Study 109, a Phase II supported safety study in sickle cell disease and in this study we also collected secondary endpoints related to the efficacy measures and we also include Study 106 which is shown at the bottom as a footnote, a safety and kinetics study in which 40 pediatric beta-thalassemia patients were enrolled and these are also included in our overall safety population.

So, overall safety population for us analyzed includes 652 patients treated with ICL of whom 292 were pediatric. All these studies were of 1-year duration.

I would like to take a minute and note at this point something that may be on many minds here today. Why are we here even though our Studies 107, 108 did not meet the protocol specified statistical boundary for the primary endpoint? Let me assure you that today we will share with you clear and convincing evidence of efficacy of ICL based on meaningful secondary analysis.

These analyses are supported by consistent dose response seen in several secondary endpoints. This principle of documenting clinical effectiveness of new agents based on dose response is well recognized within the guidances that have been issued by the agency.

Dr. Marks will cover these in his presentation. In addition to Drs. Porter and Vichinsky, two experts as I mentioned in the field of chelation therapy we also are pleased to have a number of other experts who are here to assist us and the Committee with any questions.

Dr. Alan Cohen from Children's Hospital of Philadelphia is an expert in hemoglobinopathies. Dr. Richard Larson from the University of Chicago has expertise in the MDS arena. Dr. Raymond Hirschberg, from Harbor-UCLA Medical Center is a nephrologist who was a member of our external safety monitoring board for our clinical program and we have Dr.Lloyd Fisher from University of Washington who helped clarify any specific statistical issues that were brought up.

Now, I would like to invite Dr. Porter to come and present to you about iron overload, its complications and need for therapy.

Dr. Porter?

DR. ALLEN: We have a question first.

DR.BRITTENHAM: Before you begin I would like to ask another question that I think may be on many minds. That is where did the name Exjade come from?

DR. NANANG: Maybe I should ask somebody else who was looking at this to give you an idea about Exjade. Peter, do you want to comment?

DR.MARKS: Peter Marks from Novartis. I would say that this was a name that was come up with I believe and I did not come up with it myself, I believe because jade is a green mineral, ex taking the iron out of the green mineral. So, I think that is how it became Exjade. I apologize if there is some mistake in that but that is the best of my knowledge.

DR. ALLEN: Dr.Narang, one quick clarifying question. On my agenda I have got Dr.Marks speaking after Dr. Porter and then Dr. Vichinsky as the last speaker. Is that the order?

DR. NARANG: That is the order.

DR. ALLEN: All right, Dr. Porter?

DR. PORTER: Thank you very much. Good morning. I am going to talk about iron overload, its complications and the need for treatment. This is a scheme for iron distribution in a healthy human.

You can see that the majority of the iron is found in the erythron and that is to say circulating red cells and the bone marrow. This is present as hemoglobin.

Some is present in macrophages namely as ferritin at about .6 grams and the liver is a major source of storage iron as ferritin and hemosiderin.

A small amount of iron is present in other cells in the body in parenchyma particularly as myoglobin and also in key metalloenzymes such as ribonuclear(?) reductase and others.

Transferrin which is the major conduit of iron turnover in the body contains a very small amount of iron, about 3 milligrams but is essential for iron turnover. In fact, about 20 milligrams of iron is delivered to the erythron every day and a similar amount is destroyed by macrophages as red cells senesce. This iron is released back onto transferrin and this is the major iron turnover pathway.

A small amount of iron is absorbed from the gut about 2 milligrams each day and a similar amount is lost through the gut and through skin desquamation.

About 10 percent of all the iron turnover is directed to the liver which will take up iron in times of plenty and release it in times of lack of iron. Now, what happens with blood transfusions, each unit of blood contains about 200 milligrams of iron so that if we imagine the normal adult or the iron content of 3 to 4 grams when we start transfusing there is no way of excreting this iron. So, iron inevitably builds up at a rate which has been calculated between .3 and .7 milligrams per kilogram per day in transfusion-dependent patients.

This is equivalent to 4 to 10 grams per year in an adult patient. So, iron accumulation is inevitable with repeated blood transfusion. So, this is the scheme here. Here we see red cells being transfused at a rate of about 20 to 40 milligrams a day or .3 to .7 milligrams per kilogram per day, and the senescent red cells are broken down in macrophages. The iron is then released at a very much increased rate.

Transferrin then becomes saturated and once the transferrin is saturated there is nowhere for the iron to bind and iron is present in the blood as non-transferrin-bound iron.

This iron has a different pattern of uptake into different tissues from transferrin-mediated iron, and it is this iron which is responsible for iron loading hepatocytes predominantly but also importantly in other parenchymal cells.

So, we can see here parts of the body where iron is importantly distributed to, the anterior pituitary gland, the heart, the liver, pancreas, the gonads and also other endocrine glands such as thyroid and the parathyroid glands. The increased iron uptake from non-transferrin iron is taken into these tissues at increased rates and the increased amount of labile iron within these cells is potentially available to generate a number of free radicals including hydroxyl radicals and it is this which leads to tissue damage.

So, the complications of iron overload most importantly are cardiomyopathy with heart failure which in the pre-chelation era typically developed after the age of 10, between the ages of 10 and 20 so that in the pre-chelation era thalassemia patients typically died from cardiomyopathy by the age of 20.

Another important complication if iron overload is not controlled or treated is hepatic cirrhosis. Fifteen to 20 percent of patients develop diabetes mellitus and the infiltration of iron in the anterior pituitary leads to hypogonadotropic hypogonadism which leads to infertility and poor growth and sexual development.

What conditions are associated with transfusion iron overload? Beta-thalassemia major is the best described because on a worldwide basis this is numerically the most important group, less so in the United States.

So, most of our experience with the effects of iron overload have been gained with beta-thalassemia major. Transfusion typically begins within a year of birth and therefore the effects of the iron load are very important with respect to the anterior pituitary load.

Other types of chronic transfusion dependent anemias which begin in childhood have a similar range of effects in terms of iron loading. These include the very rare Fanconi's anemia, Diamond-Blackfan anemia and some congenital dyserythropoietic anemias and from what we can see there is no difference fundamentally in the effects of the iron loading in these conditions from beta-thalassemia major.

In the United States sickle cell disease is a very important group numerically and although transfusion is not required to maintain life it has become increasingly important to treat the complications and prevent them and Dr. Vichinsky will be talking more about that later.

Aplastic anemia and mild dysplasia are important transfusion-dependent causes of anemia which develop typically later in life and again develop important complications through this process.

Now, what is the current practice with respect to initiation of therapy in iron overload? When should it begin? As I have already said with repeated blood transfusions iron rapidly develops in the body and chelation therapy is typically begun after 10 to 20 blood transfusions or when the serum ferritin exceeds 1000 micrograms per liter.

In more difficult cases where transfusion has been sporadic and it is difficult to get an accurate transfusion record then measuring the liver iron concentration can be a useful measure to estimate body iron loading and in fact liver iron concentration is a very accurate way of measuring total body iron stores and this is elegantly shown by the study of Angelucci and coworkers who undertook quantitative phlebotomy in patients who had had bone marrow transplantations of beta-thalassemia major and they looked at the relationship between liver iron concentration by biopsy and quantitative iron stores, and you can see that provided adequate samples are taken by liver biopsy there is a very good correlation between these two variables and in fact you can use a mathematical formula to measure body iron stores from the measured LIC, liver iron concentration.

Now, what are the consequences of iron overloading in terms of the rate of iron loading and what does that mean in terms of when we begin treatment? On the horizontal axis here we see patients' age with thalassemia. On the vertical axis we see hepatic iron concentration and I will concentrate on the axis on the right.

Normal liver iron concentration is up to about 1.6 milligrams per gram dry weight and we know that from the pre-chelation era that there is a very rapid accumulation of iron in the body so that by the age of about 2, 7 milligrams per gram dry weight is achieved and within a year or 2 after that the threshold of 15 is reached and by the age of 10 about 30 milligrams per gram dry weight concentration is found in the liver, in the pre-chelation era.

Now, what are safe levels of iron in the body? We know that heterozygotes for hemochromatosis can sometimes reach levels as high as 7 milligrams per gram dry weight by the age of 50 and that these patients do not get complications of iron loading. They don't get hepatic fibrosis. They don't get diabetes. So, in principle this may be a safe level of iron or one to aim for.

We, also, know from a number of studies that patients who exceed levels of around 15 milligrams per gram dry weight have increased risk of complications. Studies show increased risk of cardiac disease and there is a study showing decreased survival in patients who have liver irons above this value.

So, in general the goal would be to keep levels below 7 at all times and if patients exceed 15 to try to reduce the liver iron quickly.

Now, liver iron concentration measurement is not always a convenient test and generally around the world the plasma ferritin has been the most frequently used measure or estimate of body iron loading and as you can see from the graph here there is a correlation between liver iron and plasma ferritin both in sickle and transfused thalassemia patients but that there is a considerable scatter and this scatter relates to a number of things such as inflammation which falsely raises serum ferritin. Ascorbate deficiency is falsely decreased and so on. So, there is a correlation. There is a scatter but there still is use in measuring plasma ferritin or serum ferritin.

Firstly the trend of serum ferritin over time reflects broadly the change in liver iron concentration and therefore sequential evaluation of ferritin provides a good index of whether the patient is taking that treatment. There are, also, a number of studies itemized here, A2E which show that maintenance of serum ferritin below 2500 significantly correlates with cardiac disease-free survival.

I will show one of these here. This shows cardiac disease and percentage of time with serum ferritin above 2500 and essentially if serum ferritins are kept below 2-1/2 thousand and survival is significantly better than if that goal is not achieved.

Now, when it comes to iron chelation therapy itself or the choice of the chelator what are the ideal properties we need? The first thing is to control body iron, to prevent body iron building up in the first place or if it has already developed then to reduce the iron burden, so-called "negative" iron balance.

Clearly if you start treatment early enough then iron balance studies, the balance of what goes in and what goes out would be sufficient. We want to prevent iron-mediate organ toxicity. We want the chelator to be simple and easy to use for the patient. As we will see later this is one of the drawbacks of deferoxamine. Once daily oral administration would be ideal. We would like a drug that has a sufficiently long plasma half life to justify its use once a day.

We would like the drug to be effective by itself. We don't have to have to give a cocktail of drugs to achieve our therapeutic dose and then we want the drug to have an acceptable toxicity profile. I will just expand a little bit on this. The experience with deferoxamine has been that in patients who are less iron overloaded and in patients who receive relatively high doses of the drug, relatively iron loading that the susceptibility to iron chelated mediated toxicity is greater and the target organs with deferoxamine have been growth, decreased growth in adolescents and children and retinopathy and hearing problems and also experience with animal models has shown that the general property of the chelator will be in less overloaded patients or animal species an increased propensity to chelator mediated toxicity.

So, this is a challenge in developing iron chelators. Finally, we want the chelator to be really relatively simple and easy to monitor. We don't want a drug where we have to do a blood count every week in order to assure its safety.

Now, the current therapy for iron overload in the United States is deferoxamine, the only drug available. Because of its short half life of only 20 minutes it must be given by continuous infusion and it needs to be given 8 to 12 hours every day at least 5 days a week and ideally 7.

This is very demanding for the patient. They have to put a needle in their abdomen. They have to hitch up to a infuser device for these periods of time and as we will see some patients fail to comply with this demanding regime and therefore an easier-to-take treatment is desirable.

This slide shows the relationship between survival and the ability of a patient to take their treatment a different number of days per year and it works out that if a patient takes their treatment 5 or more days a week their survival is excellent as shown by the yellow and white lines but if they fail to achieve this goal survival falls off very significantly and there are other papers clearly showing relationship between compliance with therapy and survival.

So, to summarize the medical need and transfusional iron overload is inevitable with repeated blood transfusions whatever the disease process, currently the only approved therapy for iron overload in the United States is deferoxamine which requires subcutaneous infusion for 8 to 12 hours 5 to 7 nights a week.

Compliance is a significant issue and patients in different centers may be not adequately treated. Inadequately treated iron overload leads to organ toxicity and this is related to lack of control of reactive iron and deposition of iron in key tissues.

This leads to developmental endocrine dysfunction and ultimately cardiac dysfunction resulting in early death.

Therefore, the need for an effective, well-tolerated and convenient oral iron chelator is clear.

I would now like to introduce Dr. Peter Marks from Novartis who will present the efficacy and safety of ICL670.

DR. MARKS: Thank you, Professor Porter. I am Peter Marks from Novartis and it is now my pleasure to present the efficacy and safety data on ICL670.

In my presentation on ICL670 I will describe the compound and go on to review the dose-finding studies, the large efficacy trials and other supportive studies that were conducted.

As you have heard from Professor Porter untreated iron overload causes significant morbidity and mortality. Novartis introduced the current standard of care, deferoxamine which has led to a reduction in the morbidity from iron overload. However, the regimen of slow subcutaneous infusion required for deferoxamine to be effective has proven quite challenging to patients. It has resulted in relatively poor compliance with therapy.

In order to address the limitations of deferoxamine Novartis has been working on the development of a safe and effective oral iron chelator. Of many compounds screened ICL670 was found to have the most promising efficacy and safety profile.

In preclinical studies it was found to enter cells and remove iron and to remove more iron per chelating unit than deferoxamine.

In terms of pharmacologic properties ICL670 is orally bioavailable and this facilitates administration as a disbursible tablet to patients of all ages.

In contrast to the 20-to-30-minute half life of deferoxamine the half life of ICL670 in humans is 8 to 16 hours and this facilitates once daily administration. The iron bound to ICL670 is excreted in the bile and eliminated in the feces.

In this diagram the chelation of iron with ICL670 is illustrated. Two molecules of ICL670 bind to 1 atom of iron. The iron chelator complex then acts as a vehicle for elimination of iron from the body through its excretion.

In this respect the principle of action of ICL670 is more straightforward than many other pharmacologic therapies. The goal is to administer a sufficient amount of the drug to remove a given amount of iron.

I will now briefly review the dose finding studies for ICL670. After a single dose pharmacology study was performed and determined the dose-limiting toxicity to be nausea at 80 milligrams per kilogram study 0104, a multiple-dose pharmacology study was conducted.

Individuals with beta-thalassemia major and iron overload as documented by a serum ferritin level consistently above 1000 were admitted to a clinical research unit. Patients were then randomized to receive ICL670 or placebo at ICL670 doses of 10, 20, or 40 milligrams per kilogram for 12 days.

Iron intake and excretion were carefully assessed by chemical measurement of the amount of iron in the diet and in bodily excretions.

For study 0104 here is the net iron excretion in milligrams per kilogram per day plotted on the Y axis against the dose of ICL670 plotted on the X axis.

The light blue shaded area represents the amount of iron excretion necessary to balance the iron intake of a regular transfusion regimen.

Throughout this presentation whenever I speak of regular blood transfusions I am referring to the equivalent of two to four units of packed red blood cells administered per month to an adult.

So, with placebo iron excretion was 0.038 milligrams per kilogram per day. This is equivalent to approximately 2 milligrams per day for an adult and this is consistent with published literature. ICL670 at doses of 10, 20 and 40 milligrams per kilogram were found to produce iron excretion that was proportional to the dose administered.

The dose of 20 milligrams per kilogram was the minimum dose necessary to remove the amount of iron administered for regular blood transfusions.

Next, a longer-term randomized dose finding trial was conducted. In this study patients with beta-thalassemia major and iron overload who received regular blood transfusions were randomized to receive ICL670 at 10 and 20 milligrams per kilogram daily or to receive a standard regimen of deferoxamine at 40 milligrams per kilogram 5 days a week.

In order to assess the effects of ICL670 on removal of iron repeatedly at intervals during the trial liver iron concentration was measured using the non-invasive technique of SQUID. This technology measures very small magnetic fields related to the presence of iron in tissues such as the liver.

The baseline characteristics of the three groups were similar and the median baseline liver iron concentration as measured by SQUID was 8 milligrams of iron per gram dry weight with a range of 5 to 15.

Iron intake during the study was the same in all three groups and was close to 3 units of blood per month. At the end of 12 months at the early dose of 10 milligrams of ICL670 it was able to maintain a stable liver iron concentration similar to baseline.

The daily dose of 20 milligrams per kilogram was able to reduce liver iron concentration to a similar extent as deferoxamine 40 milligrams per kilogram.

In the dose-finding trials it was found that ICL670 produced dose-dependent iron excretion. Comparable pharmacodynamic effect was observed when deferoxamine and ICL670 were administered at a ratio of two to one.

Although not shown here the two-to-one ratio was also supported by preclinical studies that were conducted. The ratio will be important as we go on to further discuss the clinical development registration program.

In the registration program beta-thalassemia major was used as the model disease for the demonstration of efficacy because this regularly transfused patient population showed the well-documented pattern of complications of iron overload.

Patients with other types of congenital or acquired anemia requiring blood transfusions were also enrolled primarily to assess the safety of the compound. Extrapolation of efficacy was felt to be reasonable given the fact that the mechanism of action of removal of iron is the same regardless of the underlying disease state.

One challenge for the clinical development program was the standardized endpoints for determining the efficacy of iron chelation therapy have yet to be established.

Serum ferritin is the parameter most commonly used in clinical practice for initiating and monitoring iron chelation therapy. However, in agreement with health authorities the evaluation of liver iron concentration by biopsy was to be more reliable as an indicator of total body iron burden in the context of a clinical trial.

As an alternative to liver biopsy in some pediatric patients and in adults with contraindications to liver biopsy SQUID was used. A validation substudy that was conducted in parallel as part of study 0107 compared the values of liver iron concentration determined by SQUID and by biopsy.

This study revealed that SQUID values were approximately half of those determined by biopsy and concluded that SQUID was most useful as a relative measure of change in liver iron concentration over time in our studies.

Study 0107, 0108 and 0109 comprised the three large studies submitted in the new drug application in support of the efficacy and safety of ICL670 in pediatric and adult patients.

Study 0107 was a randomized reference therapy controlled trial conducted to demonstrate the non-inferiority of ICL670 to deferoxamine in regularly transfused pediatric and adult patients with beta-thalassemia major.

Study 0108 was a supportive non-comparative trial conducted to assess efficacy and safety in a special patient population with beta-thalassemia and in patients with rare anemias and study 0109 was a randomized trial conducted primarily to assess safety of ICL670 in patients with sickle cell disease.

Now, the goal of the ICL670 development program was to treat patients across the full spectrum of iron overload. ICL670 doses were selected to minimize the potential for excessive chelation with a new agent whereas the deferoxamine doses were selected based upon those that were known to be effective.

Patients were assigned by the investigator to receive ICL670 doses of 5, 10, 20 or 30 milligrams per kilogram or corresponding deferoxamine doses in the range of 20 to 60 milligrams per kilogram. Patients with liver iron concentration values less than or equal to 7, randomized to receive deferoxamine were permitted to remain on the pre-study doses even if they were higher than those specified.

Such patients were already deemed to be receiving deferoxamine doses that were safe and effective.

Now, turning to the results of the registration studies I will review the one large well-controlled randomized trial comparing ICL670 to deferoxamine in patients with beta-thalassemia, study 0107.

Study 0107 is the large reference therapy controlled trial that was conducted to examine the safety and efficacy of ICL670. Indeed it is the largest prospectively conducted randomized clinical trial examining the efficacy of an iron chelator that has ever been performed.

Regularly transfused patients with beta-thalassemia major were randomized and then assigned to one of four dosing groups to receive ICL670 or deferoxamine.

Patient accrual took place in 12 countries at 65 sites from March through November 2003. Liver iron concentration was measured at baseline and after 1 year of treatment.

Liver biopsy was used to evaluate liver iron concentration in 84 percent of patients and SQUID was used in 16 percent of patients.

During the study regular blood transfusions continued and serum ferritin and other laboratory safety parameters were measured monthly.

In the absence of established guidelines for evaluating the success of iron chelation therapy external experts were consulted in order to develop such criteria.

There was agreement that patients with different baseline liver iron concentrations should have different therapeutic goals for iron chelation therapy over the course of 1 year.

The goal for patients with baseline liver iron concentrations less than 7 was to maintain liver iron concentration and for those with higher liver iron concentration the goal was to reduce liver iron concentration.

Reduction of liver iron concentration to less than 1 was felt to represent excessive chelation and such outcomes were therefore to be considered treatment failures.

Such non-parametric success criteria were used for definition of the primary endpoint which was the treatment success rate at 1 year.

A non-inferiority design was chosen for the comparison of ICL670 to deferoxamine with a margin of minus 15 percent for the lower limit of the 95 percent confidence interval in order to demonstrate that there was no clinically relevant loss of effect.

The primary efficacy analysis was to be performed on the protocol one population which included patients completing the study and patients discontinuing for safety or considered treatment failures for the purposes of success rate analysis.

The main protocol specified secondary efficacy analysis was designed to demonstrate the reduction of liver iron concentration in patients with baseline values greater than or equal to 7.

In addition the change in liver iron concentration and the change in serum ferritin values in the entire population were analyzed descriptively. Based upon an emerging understanding that patients with liver iron concentration values less than 7 may have been underdosed with ICL670 relative to deferoxamine a post hoc subgroup analysis of the primary end point was added, that of non-inferiority in success rate in the predefined population of patients with baseline liver iron concentration values greater than or equal to 7.

Five hundred and ninety-one patients were randomized in study 0107 and 586 received treatment on study. For the purpose of the primary efficacy analysis as already mentioned the per protocol 1 population of 553 patients was used and included patients completing the study and patients discontinuing for safety.

The secondary efficacy analyses looking at the change in liver iron concentration were evaluated in the protocol two population which included only patients completing the study since the end of study liver iron concentration values were required.

The secondary analysis for change in serum ferritin level was evaluated in the safety population. Now, turning to the data from the trial, the two treatment groups were relatively well balanced in terms of gender, race, age, liver iron concentration and serum ferritin. Of note, despite the fact that 97 percent of the patients in both arms that were receiving treatment with deferoxamine for at least 1 month prior to the study, the median baseline liver iron concentration was 11.

Such a high value indicates that many patients were at risk for complications at baseline. During the 1-year study patients on both arms of study 0107 received a similar number of blood transfusions and therefore had a mean iron intake of about 0.4 milligrams per kilogram per day.

This is the transfusional equivalent of about 39 units per year or 3-1/4 units of packed red blood cells per month for an average adult patient with beta-thalassemia.

Following randomization patients were assigned to ICL670 or deferoxamine doses according to the baseline liver iron concentration. About 60 percent of patients with baseline liver iron concentrations less than 7 were maintained on the pre-study deferoxamine doses as allowed by the protocol. These same patients received relatively conservative dosing of ICL670. This led to a ratio of deferoxamine to ICL670 doses in patients with liver iron concentrations less than 7 of 4 to 1, whereas for patients with liver iron concentrations of at least 7 the ratio was 2 to 1.

As previously mentioned the effective ratio of deferoxamine to ICL670 doses was found to be 2 to 1. This was not achieved in patients with the lower liver iron concentrations in study 0107.

Turning to the results the non-inferiority endpoint was not met in the overall population. The lower limit of the difference in the 95 percent confidence interval was less than the pre-specified margin. Factors that may have affected the outcome include the conservative dosing of patients with low baseline liver iron concentration values of 5 and 10 milligrams per kilogram and the maintenance of effective pre-study deferoxamine doses in the same group.

Looking at the change in liver iron concentration as specified in the original analysis plan a statistically significant reduction of minus 5.3 milligrams of iron per gram dry weight was observed in patients with baseline liver iron concentrations greater than or equal to 7. That is in those individuals receiving ICL670 at 20 and 30 milligram per kilogram doses this reduction was comparable in magnitude to that observed with deferoxamine.

When one looks at the change in liver iron concentration on the Y axis by dose of ICL670 or deferoxamine on the X axis a dose effect relationship is apparent. ICL670 doses of 5 and 10 milligrams per kilogram were associated with increases in liver iron concentration. Twenty milligrams per kilogram was associated with maintenance and 30 milligrams per kilogram was associated with a reduction comparable to the deferoxamine.

A similar dose response was not seen with deferoxamine likely because the doses administered to all but the highest dose group were similar.

Changes in serum ferritin levels paralleled the changes in liver iron concentration for both ICL670 and deferoxamine. Five and 10 milligrams per kilogram ICL670 doses were associated with increased serum ferritin levels. Twenty milligrams per kilogram was associated with essentially no change and 30 milligrams per kilogram was associated with reduced serum ferritin.

In addition to the protocol specified secondary analyses, a post hoc subgroup analysis of the primary endpoint was performed to look at the success rate in the 65 percent of patients with liver iron concentration values of at least 7.

These individuals were treated with 20 and 30 milligrams per kilogram ICL670 doses and the ratio of deferoxamine to ICL670 doses was 2 to 1.

As you see the success rates were similar in this population of 381 patients and in fact the pre-specified non-inferiority boundary for the overall patient population was achieved in this subgroup.

This was true regardless of whether the per protocol 1 or the intent to treat analysis was used. Now, we turn to the supportive studies.

Study 0108 was a non-comparative trial performed in transfused patients with beta-thalassemia who were intolerant or non-compliant with deferoxamine and in transfused patients with rare congenital and acquired anemias other than sickle cell disease such as myelodysplastic syndromes or Diamond-Blackfan anemia. Intolerance of deferoxamine was defined by adverse events such as allergic reaction and non-compliance was defined by documentation in the medical record and baseline liver iron concentrations greater than 14.

The general design of the trial was otherwise quite similar to study 0107 with the exception that there was no comparator arm.

In addition the percentage of patients assessed with SQUID in this trial was higher and a greater number of patients had contraindications to biopsy necessitating this.

The same primary efficacy endpoint was used for study 0108 as was used for 0107. There were no historical data on success rates with deferoxamine in this patient population. Therefore after consultation was experts a success rate of 50 percent was felt to be adequate for the demonstration of efficacy. The primary efficacy analysis was performed on the intent to treat population.

The secondary endpoints for study 0108 were the same as for study 0107 including reduction of liver iron concentration in patients with liver iron concentration greater than or equal to 7, change in liver iron concentration and change in serum ferritin levels. The same additional subgroup analysis of a primary endpoint performed in study 0107 was also performed in this trial.

The baseline characteristics, the beta-thalassemia patient population enrolled in study 0108 revealed that these individuals were on average older and had higher baseline liver iron concentration and ferritin values placing them at a higher risk of complications from iron overload.

The median liver iron concentration value of 18 milligrams of iron per gram dry weight can be compared to the median value of 11 in study 0107. This is consistent with a population of patients who had not received adequate chelation therapy. The most notable demographic feature of the rare anemia population in study 0108 was age.

Particularly notable was the median age of patients with myelodysplastic syndromes at 66 years. This accounts in part for the older median age of the overall rare anemia population.

As for blood transfusion data during the study although there were some differences by disease category as a group these individuals received close to as much iron intake in the way of blood transfusion as the thalassemia population.

This was the equivalent of about 3 units of packed red blood cells per month for an adult. The range of average daily doses administered to patients in study 0108 was similar to those administered to the patients on the ICL670 arm of study 0107.

The primary efficacy endpoint in study 0108 in the intent to treat population was not met as the lower limit of the 95 percent confidence interval was less than 50 percent. The same was true in the prospectively defined analysis in the per protocol 1 population consisting of patients who completed the study and those who discontinued due to safety.

This was probably due both to the conservative dosing of ICL670 and to the use of SQUID to determine liver iron concentration in 35 percent of these patients.

In our study SQUID led to under dosing of patients assessed with this methodology. The success rates observed in this study are similar to those observed in study 0107.

Looking at the change in liver iron concentration a statistically significant reduction of 5-1/2 milligrams was observed in the population of patients with liver iron concentration values greater than 7 completing both baseline and end-of-study assessments. This reduction was similar to the reduction observed in study 0107.

When one looks at the change in liver iron concentration by dose of ICL670 administered in the population of patients with beta-thalassemia or rare anemias a dose effect relationship is also apparent.

For beta-thalassemia patients this relationship is quite similar to that observed in study 0107. In contrast in patients with the other anemic category there was maintenance of liver iron concentration with 10 milligrams per kilogram of ICL670 and a reduction in liver iron concentration with 20 milligrams per kilogram.

This was possibly because the rare anemia patients received about 10 percent less iron in the form of blood transfusions when compared to the beta-thalassemia population.

Changes in serum ferritin generally paralleled the changes in liver iron concentration. Five and 10 milligram per kilogram ICL670 doses were insufficient to prevent increases in ferritin levels. Twenty milligrams per kilogram of ICL670 was associated with a mild increase in ferritin in the beta-thalassemia population and a mild decrease in ferritin in the rare anemia population and the 30 milligrams per kilogram dose was associated with reduced ferritin in both groups.

To summarize then study 0107 and 0108 were consistent. Five and 10 milligrams per kilogram ICL doses were insufficient in regularly transfused patients to maintain the liver iron concentration of ferritin. Twenty milligrams per kilogram of ICL670 led to maintenance of stable liver iron concentration and ferritin values and 30 milligrams per kilogram led to decreases in liver iron concentration and ferritin.

If one looks at the post-hoc analysis similar to that in study 0107 for the patient population with liver iron concentration values greater than or equal to 7 at baseline the protocol specified success criteria were not met in this subgroup of the intent to treat population. However, the success criteria were met in the per protocol 1 patient population consisting of patients who completed the study and those who discontinued due to safety.

Next we will review study 0109, the trial conducted in patients with sickle cell disease. Study 0109 was a randomized trial conducted primarily to assess the safety of ICL670 in 203 patients with sickle cell disease. The study design was quite similar to that of study 0107 but one difference was that for every two patients randomized to ICL670 one patient was randomized to deferoxamine.

In addition based upon data that became available from study 0108 and 0107 approximately halfway through the study the 5 and 10 milligram doses of ICL670 were eliminated. Most patients were then dosed with ICL670 at 20 or 30 milligrams starting about 6 months into the trial. The assessment of liver iron concentration was performed non-invasively using SQUID. Although study 0109 was primarily a safety study several secondary efficacy parameters were also evaluated including the change in liver iron concentration and change in serum ferritin.

I will present the 24-week interim data on these parameters that were included in the original NDA. The 1-year efficacy and safety data from the study were included and the 120-day safety update to the NDA. Patients randomized in study 0109 to the ICL670 and to the deferoxamine groups were relatively well matched in terms of the baseline characteristics.

About 90 percent of the patients in this trial were black and 35 percent of the patients enrolled were chelation naive. Study 0109 included both regularly transfused and intermittently transfused patients with sickle cell disease.

On average patients with sickle cell disease in study 0109 received fewer blood transfusions during the study than patients in study 0107 and 0108 resulting in an iron intake of 0.2 milligrams of iron per day. This is equivalent to about 2-1/2 units of packed red blood cells per month for an average size adult with sickle cell disease.

The average daily doses of ICL670 and deferoxamine administered during the first 24 weeks of the study are noted on this slide. With the exception of the 10 patients in the lowest dose group these patients were dosed more closely to the optimal deferoxamine to ICL670 ratio of two to one.

Now, the objective of study 0109 was to examine the safety of ICL670 in patients with sickle cell disease. The efficacy assessments were secondary. The interim efficacy data at 24 weeks did reveal similar trends to study 0107 and 0108.

In the left panel the LIC values are shown and in the right panel ferritin values are shown. There was maintenance or reduction in liver iron concentration and ferritin at the 20 and 30 milligram per kilogram doses of ICL670.

In summary the following efficacy conclusions can be made for study 0107, the randomized trial in 586 beta-thalassemia patients. Although the primary end point was not met, the treatment success rate with ICL670 was similar to deferoxamine in patients with liver iron concentrations greater than or equal to 7 treated with 20 or 30 milligrams per kilogram of ICL670.

Looking at the change in liver iron concentration and change in serum ferritin the 20 and 30 milligram doses were effective in maintaining or reducing existing body iron burden in patients receiving regular blood transfusions.

ICL670 was shown to produce dose-dependent changes in liver iron concentration and serum ferritin and finally the optimal ratio of deferoxamine to ICL670 doses was confirmed to be two to one.

In study 108 there was a statistically significant reduction in LIC in patients with LICs greater than or equal to 7 treated with 20 to 30 milligrams per kilogram. There was a dose dependent reduction in LIC in ferritin and treatment effects were generally consistent with study 0107.

I will now turn to review of the safety data for ICL670 and will review patient exposure, adverse events and laboratory abnormalities.

Six hundred and fifty-two patients were included in efficacy and safety studies of a median of 1 year or more in duration and are included in the pooled safety analyses that I will present.

In addition, study 0105E2 which is still ongoing includes 51 patients treated for up to nearly 3 years with ICL670. Study 0106 was a study in 40 patients age 2 to 17 years with beta-thalassemia that provided safety data in this population.

For the purpose of providing an integrated overview of the safety profile of ICL670 the data from the 1-year core studies will be presented. Beta-thalassemia patients from study 0106, 0107 and 0108 were pooled resulting in a total of 421 patients in this group treated with ICL670. Therefore along with those with sickle cell and other rare anemias a total 652 patients were treated with ICL670 on these four studies.

Forty-five percent of the patients were under the age of 16 including 8 percent of patients less than 6 years of age. Here are the data on study participation withdrawals and serious adverse events for the two randomized registration trials 0107 and 0109. The number of patient withdrawals in study 0107 and 0109 was similar although the number of withdrawals for safety was somewhat higher for ICL670 than deferoxamine.

The percentage of patients who experienced serious adverse events in the ICL670 and deferoxamine arms of study 0107 and 0109 were similar.

The high percentage of serious adverse events in both arms for the sickle cell population reflected sickle cell crises which constituted about 23 percent of the serious adverse events.

The percentage of patients withdrawing from or having serious adverse events in the pediatric beta-thalassemia study 0106 was similar to study 0107.

The percentage of patients withdrawing from or having serious adverse events in the 1-year beta-thalassemia and rare anemia study 0108 was somewhat higher largely related to the inclusion of older patients with myelodysplastic syndromes who had disease progression or complications.

In study 0107 one asplenic patient died at home of a sudden death on the ICL670 arm and this event was reported by the investigator to be related to its administration though no autopsy was performed. So, we don't have additional data.

The causes of death in the three patients on deferoxamine were sepsis, convulsions and intraventricular thrombus. In study 0108 four of the five patients had MDS and one had Diamond-Blackfan anemia. All five deaths were reported to be unrelated to the administration of ICL670.

The overall percentage of patients experiencing adverse events in study 0107 was similar in patients treated with ICL670 and deferoxamine. Several adverse events were seen more commonly with ICL670 than with deferoxamine irrespective of their relationship to study drug in the clinical program and these include several gastrointestinal symptoms and skin rash.

In study 0107 although abdominal pain, diarrhea, nausea and skin rash were more commonly observed in patients receiving ICL670 vomiting was reported with similar frequency in patients treated with ICL670 and deferoxamine.

If we look at these same five events in patients with sickle cell disease in study 0109 transient diarrhea occurred more commonly than in the beta-thalassemia patients treated with ICL670 in study 0107.

Although nausea and vomiting were also more commonly reported than in study 0107 the incidence was higher in both the ICL670 and deferoxamine arms.

The most notable differences between the rare anemia population and the other populations treated with ICL670 was a higher incidence of diarrhea. Other adverse events were reported with similar frequency.

Looking at adverse events in the program that were more commonly observed with ICL670 the four gastrointestinal symptoms were most often mild-to-moderate in severity and lasted on average less than a week.

Maculopapular skin rash was most often mild to moderate in severity. It often spontaneous resolved even without discontinuing ICL670 and with continued ICL670 dosing.

Rash only led to discontinuation in about 1/2 of 1 percent of patients. The incidence of gastrointestinal adverse events and rash were similar across the pediatric age range with the exception of diarrhea which had a somewhat higher incidence and rash which had a somewhat lower incidence in the youngest children.

Turning to laboratory abnormalities overall in clinical studies 36 percent of patients receiving ICL670 experienced at least two consecutive increases in serum creatinine of 33 percent that never exceeded twice the upper limit of normal on two consecutive occasions.

For comparison 16 percent of patients receiving deferoxamine on study 0107 and 0109 had similar increases. Only a small percentage of patients had creatinine values that exceeded the upper limit of normal. This was most notable in the rare anemia group who were generally older and therefore a number of these individuals had serum creatinine values at the upper limit of normal at baseline.

In order to proactively address this laboratory abnormality with the help of an independent program safety board a dose reduction guideline was instituted for increases in serum creatinine.

This guideline specified dose reductions for adults with two consecutive increases in serum creatinine of greater than or equal to 33 percent in adults and for increases of greater than 33 percent and over the upper limit of normal in children under the age of 15 years.

Using this dose reduction algorithm 13 percent of patients receiving ICL670 had dose reductions for two consecutive increases in serum creatinine greater than 33 percent.

The remainder of the individuals were either pediatric patients with increases to less than the upper limit of normal or patients in whom the serum creatinine spontaneously returned toward baseline.

All patients who underwent dose reduction for serum creatinine had stabilization and normalization of serum creatinine. In addition to these data safety information on serum creatinine is available from the 51 patients treated for up to 3 years in study 0105E2.

These individuals have had no progressive increases in serum creatinine. Aside from serum creatinine the only other laboratory abnormality of note was elevated serum transaminases. About 6 percent of patients receiving ICL670 had two ALT values greater than 5 times the upper limit of normal during the study. The majority of these patients had at least one abnormal ALT value at baseline suggesting pre-existing liver function abnormalities.

In two cases the ALT values were related to, apparently related to drug and that was due to the temporal association. The increases in ALT likely represented drug-induced hepatitis on the basis of the temporal association to drug rechallenge.

In both cases discontinuation of ICL670 resulted in resolution of the abnormalities.

Regarding safety parameters identified as issues for other iron chelating agents these were rigorously assessed in the clinical program. No significant differences between ICL670 and deferoxamine were identified particularly in the comparative studies.

So, the most common adverse events with ICL670 were transient gastrointestinal symptoms and rash that were manageable.

These rarely led to study discontinuations and mild increases in serum creatinine and increases in liver function tests occurred but were successfully managed in a large majority of patients allowing them to continue to receive ICL670.

Changes in other parameters were similar in frequency to deferoxamine and over a 1-year interval ICL670 had no adverse effects on pediatric growth and development.

I would now like to review the proposed dosing and monitoring for ICL670. Iron chelation therapy should not be initiated prior to sufficient iron loading of the body. The recommendation is therefore that evidence of iron overload should be clearly apparent on the basis of readily available clinical parameters prior to the initiation of therapy.

The recommended initial dose would be 20 milligrams per kilogram with a high of 30 milligrams per kilogram dose considered for patients with more severe iron overload as documented for example by a serum ferritin value greater than 2500 which roughly corresponds to a liver iron concentration value of 14.

These recommendations for the use of serum ferritin for initiation of chelation therapy are consistent with current clinical practice. Serum ferritin levels are also routinely used for monitoring patients on chelation therapy. For the monitoring of patients on therapy with ICL670 use of the trend in serum ferritin level over a 3 to 6 month interval would be recommended in order to determine the potential need for dose adjustment.

In terms of safety monitoring we will recommend monthly monitoring of serum creatinine and liver function tests. Dose reductions for confirmed increases in serum creatinine are recommended. Discontinuation of ICL670 is recommended for patients with progressively elevated liver function tests in the absence of an alternative etiology.

Now, while we have a very large clinical trial program for an iron chelator we recognize that unanswered questions remain. About 900 patients therefore have been enrolled in non-comparative extension trials that will examine the safety of ICL670 as well as its efficacy for up to 3 years.

Among other indications ongoing studies are looking at the use of ICL670 in patients with sickle cell disease taking hydroxyurea and in patients with myelodysplastic syndromes. Planned trials include studies examining the mechanism of the changes in renal function and a large study examining the dosing and monitoring of ICL670 based on serum ferritin.

These studies among others will provide important additional information on the compound.

So, in conclusion our studies have consistently demonstrated that ICL670 removes iron from the body in a dose-dependent manner. Patients treated with 20 and 30 milligrams per kilogram doses of ICL670 had maintenance or reduction in body iron burden similar to that achieved with deferoxamine.

Rash and the gastrointestinal effects are manageable in the large majority of patients. Increases in serum creatinine and liver function tests seen in clinical studies can be successfully managed with routine monitoring. ICL670 is a once-daily oral iron chelator that has demonstrated a favorable benefit-to-risk profile in the treatment of patients with iron overload due to chronic transfusions.

I would now like to introduce Dr. Elliott Vichinsky who will provide further perspective on the benefit-to-risk of ICL670.

DR. ALLEN: Thank you, Dr. Marks. Could I ask you to stay there for 1 second? This was a long and complex presentation. Let us just take a moment for any clarifying questions before we move on.

Let me ask you just one question. I believe you said in study 109 partway through the lower dose patients at 5 and 10 milligrams you increased the dose. Was that associated with any additional adverse events or side effects that were noted with that increase?

DR. MARKS: It was not.

DR. ALLEN: Thank you. Any other clarifying questions?

DR. MARKS: I am sorry, I should actually clarify that. There were, when the patients were increased in terms of the gastrointestinal adverse events that is the case. In terms of a percentage of patients with an increased number of serum creatinine values greater than or equal to 33 percent there was a slightly higher incidence.

DR. ALLEN: Can we recall these slides when we have discussion this afternoon?

DR. MARKS: Absolutely.

DR. SCHREIBER: Is there a disease relationship with the ferritin? I noticed that you mentioned that at 2500 ferritin it was associated with a liver iron concentration of 14 but in your sickle cell it looks like the median is 8.6 and the ferritin is 3500. So, it seems that you would be starting treatment at a different point.

DR. MARKS: To answer your question there are two issues there. First of all it is true that sickle cell patients have inherently more variability in serum ferritin levels. However, the sickle cell study was conducted using the SQUID methodology. So, the LIC values you see there need to be multiplied by a factor of about two. So, that means that those liver iron concentrations of about 8 really are about 16, again, not exact but rough.

DR. QUIROLO: So, the data that you present here hasn't been recalculated for that difference that you just mentioned?

DR. MARKS: We have not because for study 0109 we felt it would be most appropriate to provide the data as it came in without any recalculation and since the relative change is really what was most important.

With that I will introduce Dr. Vichinsky.

DR. VICHINSKY: Good morning. My role here for brief minutes is first I feel a need to give you a brief overview of sickle cell disease and its relationship to iron burden. Sickle cell disease is the most common disease in the United States requiring transfusions and there is less information known and yet these people would be very strong candidates for this medication.

Secondly, I would like to spend a little bit of time just going over the whole issue of compliance which is an important variable in chronic illness and when you assess intent to treat efficacy and finally, to give you my perspective on the risk/benefit and recommendations.

This slide summarizes through a questionnaire I did and it is a few years old about transfusion therapy in sickle cell disease and I will try to, in North America, Canada and the United States, and what it is illustrating is an important concept that I will give you an overview of.

The life span of sickle cell patients has dramatically changed in the last 25 years.

A pediatric disease primarily associated with early death has now become a chronic disease with mean median life spans in the fifties and approaching sixties. With increasing survival has come increasing morbidity that requires therapy and the medical community has identified two important roles. By screening asymptomatic children early and adults they have identified patients who are going to have problems and the primary intervention for all the new screening technology has become transfusions.

So, the National Institutes of Health let out a census release saying that all children should be screened with a transcranial doppler and anyone with abnormal doppler should be put on chronic transfusions. Those studies have been ongoing and have shown so far once on transfusions they can't be stopped and about one-third of the children when screened will have an abnormal doppler.

In addition, lower levels are associated with stroke, too, but at a slightly lower rate. So, that has already become an initiated standard of care. Recently the NIH has embarked on 10 years of MRI neuropsych follow-up and has found progressive neurocognitive abnormalities in children who don't have a stroke that results in significant executive function disorders that lead to, has a higher rate of stroke.

This has finally led to an ongoing trial now that is randomizing these patients to chronic transfusion as the model of the STOP(?) study has.

On top of that is a clear understanding that acute chest syndrome and lung injury is the No. 1 cause of death in sickle cell and that transfusions can increase oxygenation and improve outcome.

So, overall the pediatric population has increasingly been chronically transfused. Despite interventions with gene therapy attempts on a Phase I trial bone marrow transplantation or hydroxyurea the risk of transfusion therapy is growing.

In fact, as of now 91 percent of adults with sickle cell anemia have received one or more transfusions or two times during their life and their indications are rapidly increasing because they are living longer and with improved standard of care comes interventions to manage the chronic complications including brain, lung, renal, cardiac and chronic pain and arthritis.

So, the net effect is with expanded life spans in sickle cell and without the hope of immediate cure on the horizon transfusion is becoming the mainstay of therapy for a majority of patients.

Published data has demonstrated that these transfusions not only lower the frequency of those events but they basically can lower the progression of organ injury and a whole risk of variables that I have listed here, brain injury can be prevented. Lung injury can be prevented, pulmonary hypertension, growth and surgical complications and more importantly or almost as importantly is the general well-improved quality of life in chronically transfused patients, and this data was reported by the STOP study, myself and others. So, there is a wide belief that transfusion has benefitted patients.

The risks of transfusion have been debated because in contrast to our sister disease thalassemia there have not been the long-term trials evaluating iron morbidity. So, most of the information is extrapolated from the chronically transfused thal community. Five years ago an NIH trial was embarked with me as the principal investigator which followed 250 sickle cell patients, now 150 thalassemia patients and all chronically transfused and matched them with 44 sickle cell patients who were not transfused.

You could see that in this trial they are all matched for the same age and that the transfused had dramatically elevated liver irons and the control group was almost within the normal limits.

What was a surprise to us and actually very important was that we have started to identify in the first year; this is the first year of the study that 9 percent of the sickle cell patients had iron induced cardiomyopathies.

Now, this excludes cardiomyopathies from anemia or infarcts and stuff. There is an adjudication committee with John Woods in which each case is reviewed. So,this is separate from the high rate or other associated cardiac disease.

So, we are now identifying I think for the first time a clear cardiac myopathy that was iron related in sickle cell. This trial is going on and we will then determine the frequency but now we know they do have these problems.

In the first year of the study 13 deaths occurred in the sickle cell population and 3 in the thal and what we noted in the deaths of the sickle cell patients they were heavily iron overloaded. They had an abnormal left ventricular ejection fraction and they were somewhat older.

We presented this last year and when we looked for analysis of these groups we found that the deaths were really correlated with a cardiomyopathy, an iron-induced cardiomyopathy and so this preliminary data is quite supportive and it is an ongoing trial and we want to look at long term that sickle cell patients so far are expected to follow the same trail at least in cardiac disease as thalassemia raising the flag that iron chelation is a critical issue in this under chelated population.

Now, deferoxamine is now the standard of care for treatment of iron overload. In fact, it is a very efficient chelator that is clearly efficacious in inducing negative iron balance and reducing morbidity and mortality from iron overload in patients who use it.

The problems with it are in the intent to treat. It is a very difficult treatment not only because it requires daily prolonged subcutaneous infusions but in actuality those of you who run programs you have to have an infrastructural program to increase compliance, a program that has a home care program, a child nurse, a nurse who specializes in the skin lesions and a complex management team of ophthalmologists, audiologists, cardiologists, growth and everyone who is really familiar with deferoxamine.

Otherwise you really have limited ability to pick up complications and prevent morbidity. So, failures are complex. They are not the patient alone. It is the failure of the team and the invasiveness of the therapy.

in fact, this correlates extremely well with death and in this recent study if you look at compliance with deferoxamine you can look at survival.

Now, this is with thalassemia and you can see as patients become non-compliant by groups their survival has dramatically dropped and in fact if you are not taking the drug and in the 20 percent compliance rate it pretty well overlaps with the previous data published in the seventies of the natural history of transfused thalassemia patients without chelation.

So, clearly Desferal is effective but if you don't use it you are going to die. So, the reasons for the failure I want to just stress to you are not just the patient. Obviously getting needle sticks and all the other things are important and they impair quality of life, but we are really looking at a holistic health care system that has to be able to provide for the family an infrastructure. Take a pump of software monitoring to a patient and health care in the United States has decentralized to the extent that this is not readily available to the majority of patients. Given this we have a very high need for an easy-to-use, safe and effective oral chelator for the treatment of transfusion iron overload.

Now, you may ask,"Where is your data on this?" and there is some data. In our program Dr. Treadwell and colleagues try to evaluate adherence to deferoxamine protocol and this was published in Pediatric and Blood in May of this year and basically 40 patients with sickle cell who were chronically transfused went through a very complex and I won't go through the methods technique to evaluate compliance using diaries, home visits and a whole bunch of other kind of scales and techniques, and they were assessed and basically the simplest one to show here was only 60 percent of the patients actually met standard criteria to even pick up the drug at the pharmacy or have it delivered and the shocking thing to me was that the average median number of days during the last time they actually used the Desferal was 8.7 days. It ranged from 2 days to 22 days.So, patients were not using the Desferal the way it is supposed to be used and finally a problem that is ubiquitous even within the sickle cell and thal centers that are transfusing patients the mean number of hours infused was only 4 hours and as you have seen or maybe Dr. Porter mentioned the efficacy of Desferal is directly related to the duration of infusion and its toxicity is inversely related to it dose.

So, what I have here is that patients are being under treated with the number of days and actually too few hours demonstrating they are just not getting effective therapy.

Now, we looked at satisfaction recently and this was an Ash(?) abstract in the 109 study in which we assessed sickle cell patients on deferoxamine and about to be randomized onto the trial of Desferal or ICL670 and we asked them a series of questions about satisfaction and criteria were developed from life scale and by and large 20 percent were satisfied with the Desferal treatment. After randomization and at the end of the trial patients were offered to go back onto their therapy with deferoxamine or be able to continue the ICL670 and 84 percent of the patients chose to continue ICL670 for many reasons I could outline to you which were obvious, needle sticks, compliance, ease of use.

Now, the preclinical safety data is equally important. We have a lot of extensive data on the animal efficacy and toxicity of this drug. It is highly selective for iron. It has no effect on growth and development. It is non-carcinogenic or teratogenic. It has a very long half life which remarkably enables these patients to take a medication only once a day which should have a marked effect on compliance.

In addition there has been a huge trial undertaken, the largest trial for any chelator ever in which 700 patients were treated with this drug for at least 1 year and not only in thalassemia, the drug has expanded into studying sickle cell disease and other groups that will eventually get the drug.

So, ICL670 removes iron from the body proportional to the amount of drug administered. At doses of 20 and 30 it is able to maintain or reduce liver iron concentration and serum ferritin and at these doses it is equal or comparable to deferoxamine.

In general ICL670 is well tolerated. It has side effects which were mild to moderate GI upset and rash. There was some mild increase in creatinine that appeared to be potentially related to excessive rapid chelation but showed no progression on the patients who were kept on extended studies.

Finally there were no cases of agranulocytosis, growth failure or bone abnormalities. So, it appears that ICL670 is convenient, well tolerated, available as only a once-a-day pill oral mixture for the treatment of iron overload in both children and adults and is likely to increase compliance and therefore decrease the iron burden of patients.

Now, what does this mean if I had it translated to use as a clinician? The reality from my standpoint is with effective therapy and taking into account intention to treat we don't have therapy now for patients that prevents death from iron overload. Most thals today despite the most aggressive programs die from iron overload. So, we don't have a therapy now that prevents the death from the underlying disease. It does modulate it but they are all dying from that.

So, we don't have a drug that affects it. We cannot prevent death. We do have a drug that can modify it.

I believe therefore that the addition of a pill that has oral efficacy and can increase compliance with similar dose, similar responses should go a long way to improving the death rates that are still predominantly from iron overload and so, I think this drug should have significant improvement for the quality of life and long term-wise the survival rates of patients.

Thank you.

I am going to ask Dr. Marks to chair the questions.

DR. ALLEN: Thank you, Dr. Vichinsky. We will now have a period of time for questions and answers of the last three presenters on their data.

Yes?

DR. MALDONADO: Just a quick question, Dr. Marks. You said that you considered treatment failures those who were over chelated in the study I believe 0107. I may have missed it but there was a comparison on the safety about over chelation between the two drugs?

DR. MARKS: In terms of patients who got down to that level of one I believe there were only two patients on clinical trials that got down to a level below one. So, it was a very small number of patients.

So, there were very few failures due to that criteria.

DR. KLEIN: Just to start are there any data directly correlating liver iron content with cardiac function since most of these patients die from cardiac disease and have you got any data on cardiac function in any of the patients?

DR. MARKS: We don't have a systematic percentage of ejection fraction on these patients in the trials. We performed routine cardiac monitoring, electrocardiograms and patients with a history of cardiac impairment had echo cardiograms performed while they were on study to be sure there weren't any gross changes.

In terms of prospectively looking at cardiac iron reduction, that was not a part of this program because the assumption was that reducing body iron burden would reduce all of the organs.

What I will do though is ask Dr. Porter to come up and make some comments. He performed a substudy at his site using T2 star MRI methodology to look at iron removal from the heart and so I can ask him to come up and comment on that.

DR. PORTER: Thank you. As part of our routine monitoring of our patients in London since about 2000 we have been fortunate in having the opportunity to have MRI measured at the Brompton(?) Hospital where the T2 star method was developed. This is something we have been doing in our patients for a while. So, we continued to do this in patients who we then sent to the study. All this wasn't a prospective randomized study. We just had some information on this.

So, what I am going to show you here are 22 patients on ICL670 on studies 107 and 108 and 16 with transfusion-dependent beta-thalassemia, 6 with other iron overload conditions and 2 with pyrite kinase, 2 cytoblastic, 2 Diamond-Blackfan. Mean age is shown there, follow-up of 13 months and also there were 7 patients randomized to Desferal as part of our observations and you can see here the cardiac T2 star on the vertical axis before and after this period of follow up.

The yellow lines are the thalassemia patients and the pink are the other anemias and you can see generally speaking it is thought that a cardiac T2 star of less than 20 milliseconds is associated with increased risk of a fall in the ejection fraction.

This is from studies from the Brompton Hospital. If we accept this you can see that the proportion of patients below 20 is significant before starting. The average cardiac T2 star is 18 milliseconds, geometric mean and then after rises significantly to 23 milliseconds.

So, this we think is encouraging although it was not an intended trial, this is encouraging in our patients. I would say that we looked at ejection fractions in the same group of patients and there was no significant change in those patients.

DR. KLEIN: Could I just follow up by asking a question on reversibility either of the cardiac issue or of the endocrine overloading once you have organ deficiency how likely are you to get reversal?

DR. PORTER: With respect to cardiac dysfunction there are many papers showing that with intensive continuous deferoxamine you get reversal. With respect to endocrinopathies the data really aren't there. There are a few case reports of possible improvement in diabetes but really the message is once that has developed it is too late.

MS. BAKER: Could one of the panelists please comment about the experience of oral chelators in Europe or Canada?

DR. MARKS: Perhaps I will call on Professor Porter to speak about that since he is from the United Kingdom.

DR. PORTER: So, in Europe but not in Canada deferaprin(?) has been raised as a second line agent for people who can't tolerate, adults who can't tolerate deferoxamine and this is used variably in different clinics. We use it in some patients but the vast majority of our patients still take deferoxamine.

With respect to how it compares with ICL670 of course there aren't any head-on-head studies but the drug has to be taken three times a day. It is associated with agranulocytosis in a proportion of patients and also arthropathy and so on.

So, it is not a perfect drug and I, personally, feel in my patients we need a better investigated and better tolerated, more efficacious agent.

MS. BAKER: So, you mentioned it is only used in adults, correct?

DR. PORTER: Yes. Well, there are people who have used it in children but in terms of data sets showing its safety and efficacy I am not aware of those.

MS. BAKER: And is the data primarily in thal patients?

DR. PORTER: The data on deferaprin in the literature is primarily on thalassemia patients.

DR.MARKS: Just to clarify, to my knowledge I believe the age that the European registration may be down to an age of 6, but I cannot confirm that. Can you confirm?

Thank you.

MS. WINNER: In addition to a question I just would like to comment that as a thal patient myself and as a representative of all the patients in this country this is a monumental event here today because we have been waiting a very, very long time for an oral chelator and part of the reason why I have remained compliant is so that I could live to see this day.

I have been receiving blood transfusions for 43 years and I have been doing Desferal for 28 and the slides can tell you how difficult it is but it is extremely high maintenance. It is very difficult to remain compliant because it is extremely painful. I was just commenting to Judith that I infused last night and removed it this morning and the infusion site is quite painful right now.

So, in light of all that I think it is extremely important that we look at this drug and realize that it is going to be extremely beneficial to patients who are non-compliant first of all because they are not compliant with Desferal and their life expectancy is affected tremendously and those of us who have been compliant because we do need a break. It is extremely difficult to continue doing it. I am going to be 46 years old and it is very tiring, high maintenance.

My question is in relation to the older patients that you did on the study you said that there were some severe adverse effects and I was curious to know if that was in relationship to their condition as far as whether they have hepatitis or diabetes and do you think it is dose related or do you think it is because of their health condition?

DR.MARKS: The differences in studies, we have to I think differentiate here between beta-thalassemia and the rare anemia population and the beta-thalassemia patient population which we studied, we studied patients from 2 to about 50 years old. The adverse event rate was relatively similar across the spectrum. It was in the population of patients with other rare anemias which included patients with myelodysplastic syndromes and myelofibrosis, some of those patients were as old as 80 years old and they are the ones in which we saw a somewhat higher frequency of serious adverse events.

MS. WINNER: I just would like to clarify I believe that in the reading there was a mention of some hepatic situation. Was that in the rare diseases or with thalassemia?

DR. MARKS: There were hepatic, I probably will call Dr. Ford here to make comments on this but there was a patient who had the drug-induced hepatitis that was a thalassemic patient.

DR. ALLEN: Dr. Ford, would you please identify yourself for the recorder?

DR. FORD: John Ford, Novartis. The transaminase elevations were scattered across all populations. They weren't limited either to thalassemia or the rare anemic categories.

DR. GHANY: If I could just follow up on that question, I was actually interested in the two patients who had ALT elevations greater than five times upper limit of normal. Did that occur in patients with higher underlying total body irons? Did they have underlying liver disease and did the total serum bilirubin rise in those two patients/

DR. MARKS: First of all the total serum bilirubin did not rise significantly. In the patient with beta-thalassemia that had drug-induced, what we would associate which could have been drug-induced hepatitis that patient had a higher baseline liver iron burden. They were placed on a starting dose of ICL670 of 30 milligrams per kilogram because they had a liver iron concentration greater than 14 at baseline and they had a history of hepatitis B and C which was reportedly inactive at study entry with I believe a negative PCR, RNA PCR.

There was another patient. The other patient was a sickle cell disease patient.

DR. GHANY: Do you have any idea of what the underlying histology was on these patients or was biopsy not performed in these patients?

DR. MARKS: In the thalassemia patient a biopsy was performed and it was consistent with drug-induced hepatitis.

DR. GHANY: I just have one more follow-up question. What I am getting at here is many of the patients with thalassemia prior to a screening test for hepatitis C are infected with hepatitis C and they were excluded from the registration trials. I wonder if you could comment on how you would recommend using this drug in patients with underlying known liver disease.

DR. MARKS: Actually we did not exclude patients with hepatitis B or hepatitis C from the registration trials. We only excluded those with evidence of active hepatitis such as an elevated RNA PCR.

So, if a patient was hepatitis C antibody positive they could be enrolled in the clinical trials and about 25 to 30 percent of patients on study had hepatitis B or C infection.

DR. GHANY: Active or inactive?

DR.MARKS: Not active, inactive.

DR. GHANY: How about patients with active hepatitis because many of the patients are still viremic.

DR. MARKS: Those would require additional studies.

DR. BRITTENHAM: I would first like to recognize what Ms. Winner has very eloquently expressed that this is really an important and historic time, that Desferal was really a serendipitous gift or blessing that came over 40 years ago and that has kept many of the individuals in this room alive and it deserves to be recognized for that. It has given us experience in how to use chelation, but as Dr. Vichinsky described it has many, many shortcomings and so we are anxious to find a replacement.

ICL670 though is not deferoxamine and we need to be acutely aware of the differences between the drugs as a guide to the safety and so I would just like to briefly ask a couple of questions to bring some of those differences into relief.

So, first, could we look at CE7, the slide CE7 again? It is the one of the dose-response curve that was done and I think the person who did it, Dr. Grady is in the audience because if we look at this it is true as you described that the mean values for a dose at 20 milligrams a day in these particular patients was at the lower limit of what is needed but since it is the mean it means that half the people are not getting enough and if you look at the 40 milligram dose there is still considerable scatter.

Now, it is always difficult to interpret these sorts of plots because of the relationship between the body iron burden and excretion. At a given dose of a chelator I think virtually all chelators the higher the body iron the more iron excretion there will be and since these patients surely had different levels of body iron it is hard to separate these things out but first given that qualification the people at 5 or 10 really had very little use, had levels of iron excretion that certainly for transfusion dependent patients would be inadequate, and here you gave a dose of 40 but the highest dose that is mentioned in the materials you provided is 30.

Could you clarify both the use of the 5 and 10 and the non-use of the 40?

DR. MARKS; If I could have the core efficacy slide that shows the dosing that was used in clinical studies?

When we started these studies we wanted to be relatively conservative about the dosing. Remember this was a 12-day study, study 104 and our core, the studies lasting 1 year obviously we are looking at safety and efficacy over a much longer interval.

So, we were concerned over longer intervals particularly on the basis of the data we had from study 105 which was somewhat different from that in study 0104. It seemed to suggest that the 10-milligram dose would be somewhat more effective or have somewhat more effect in maintaining liver iron concentrations stable. So, we picked a 10-milligram dose here because the 3 to 7 range was a group that we felt would benefit from having stable liver iron concentrations. The 20-milligram dose was used because of study 0105 because that was a dose that reduced liver iron concentration to a moderate extent and the highest dose, the 30-milligram dose was used in patients with the very high iron burdens with the provision in the protocol that they could have been dose increased during the study to as high as 40 milligrams per kilogram.

In practice that did not occur in the core studies although in the extension trials to these studies we have now in a few patients dose escalated to 40 milligrams per kilogram.

DR. BRITTENHAM: If you can leave that there, I would like to make one comment. Because of what I just mentioned, that the higher the body iron burden the more excretion with a given dose of chelator the assumption that patients with lower body iron loads need less chelator isn't necessarily true. It really depends on the rate of iron entering the system about what needs to be gotten out and so I am sure there are patients here who were very disappointed when they had high irons and they worked very hard to get them down hoping, well, maybe I can use fewer days or a lower dose and there is very little change really that occurs, and could we go to CE26?

This has to do with the comparison of the doses of the ICO670 with the deferoxamine and if you look here you can see that the deferoxamine at least maintained balance and got rid of iron at higher doses all across the board.

Now, this is somewhat confounded because the assigned amounts of DFO are quite different from this. I doubt that there is any statistical difference between the amounts of DFO in any of the groups but what is informative here and is striking is about how little the doses of 5 and milligrams per kilogram of the ICL670 do again.

Now, partly DFO is a compound. One molecule wraps around the iron and activates it and carries it out of the body, but for the triazo(?) you need two and so if you go down in the dose then you have a real effect on efficacy. So, the idea that there is a fixed relationship between the dose of DFO and ICL670 can't be true because of at least this difference in chelation.

The other point is that it is really a very striking decline in efficiency when you get to these lower doses and there may be other mechanisms that are involved in that. There might be preferential binding sites on the albumin for example that could be involved. I don't know of any information about that, but I think that is important in trying to think about the doses that may be used in patients.

DR.MARKS; Okay, just to respond to a couple of points first of all we actually looked at the efficiency of chelation quite carefully precisely because of the excellent points that you brought up here.

The efficiency in animal studies was relatively constant across dose and the efficiency in study 104 if you actually look at the efficiency of the doses that were administered ranged from 15 to 22 percent.

Now, when one looks actually at the efficiency of the chelator in proportion and relative to the amount of dose that was administered and the amount of iron that was taken out one also then sees that across the different doses the chelation efficiency was about 25 to 27 percent. That is consistent with the animal studies which showed an efficiency of about 29 percent.

This is obviously measured by the liver biopsy method in terms of how we determined what was being taken out of the body. What went into the body was measured by carefully recording the amount of transfusional iron intake in terms of milligrams of iron. So, the amount of iron intake here was measured, the amount of excretion and we came to the fact that there was a relatively constant efficiency across the doses.

DR. ALLEN: Thank you for those questions. I think that helped clarify. I have got just one follow-up question. I am mindful of the fact that we are at break. Given the initial dosing information that you got from study 104 and the results that you found in 107 with the 5 and 10 milligram doses in retrospect how would you redesign 107? Would you have redesigned 107 to take that result information into account?

DR. MARKS: Right. We have already started and we have a trial ongoing in sickle cell disease where we have taken that into account and that means that we understand now that in order to keep someone's iron balance stable one simply needs to give enough of the chelator. So, we will give in patients who are receiving regular transfusions we understand to maintain iron balance we need to give at least 20 milligrams per kilogram on average. The dose obviously needs to be tailored to the individual and that is why we will recommend dose adjustment as patients are monitored on therapy and progress to either have decrease or increase in ferritins.

DR. ALLEN: And do you have an upper dosage range at this point? You mentioned 30 as your recommendation summary.

DR. MARKS: In our NDA we are proposing the highest dose to be 30 milligrams per kilogram because those are the doses that we have studied well.

DR. ALLEN: Thank you.

Dr. Schreiber?

DR. SCHREIBER: I was trying to ascertain the success rate at the different dose levels to pick up on what Gary was saying. It looked to me that the overall success rate was about 50 percent so that this would not be a panacea for half the patients. Half the patients would either have to be treated with a much higher dose because even at the 40 it looked like it was about 50 maybe maximum of 60 percent or they would have to have a secondary treatment. Do you have a breakdown of success rate at the different dosing levels?

DR. MARKS: Could we bring up our success rate by dose, please?

The issue here is the success rate. The choice of our success criteria here, remember were somewhat arbitrary and so our success rate here for the patients with liver iron concentrations greater than 7 was they had to have a 3-milligram-of-iron-per-gram reduction in the value and the other patients had to have maintenance.

So, if you look in the population here you can see that there is the dose-dependent effect here. We have it up here by success rate. You see the 40 percent success rate in study 107 at the 5 and 10 milligram doses and the higher success rate at the higher doses and particularly at the 30-milligram-per-kilogram dose you see the 74 percent success rate.

We anticipate that as patients are dosed with this drug that they would have adjustment of their therapy. So, patients who were started on 20 milligrams per kilogram would potentially have an increase in dose as they progressed on therapy with adjustment of the dose after 3 to 6 months based according to the individual response in order to address the individual patient's needs.

DR. ALLEN: Okay, quickly before we go into break Dr. Portman, thank you for joining us. Do you have any questions at this point?

DR. PORTMAN: I do and I apologize that I can't be there. This is less than optimal and since I can't see any response slides from the sponsor you can refer them to me in the supporting data that I have here or perhaps fax me the slides.

I think the first question I have is related to the issue of creatinine. Creatinine is not an optimal way to look at renal function. I think that is well established and in calculating your GFR which there is very little data on GFI in fact in the supporting information for adults or you said for all patients you used the Kochgraf-Galt(?) formula for calculating GFR which is not ideal for adults as I think we agree. The MDRD formula is a better measure but it is certainly no appropriate for children which compose 45 percent of your population which is usually done by the Schwartz formula. Do you have that data or did you use the Schwartz formula for children?

DR. MARKS; We did us that formula and we do have those data. I am going to ask Dr. Hirschberg to be ready to comment on the serum creatinine increases that we observed.

I would like to have the slide SK3 please?

DR. ALLEN: Dr. Hirschberg, would you identify yourself please for the record?

DR. MARKS: I will just bring this up first for a moment just so I can describe it to Dr. Portman. So, what I have shown here or put up on the screen is a slide that looks at the serum creatinine increases greater than or equal to 33 percent from baseline at two consecutive visits in the studies 0106, 7, 8 and 9. On one side are ICL670. On the other side are deferoxamine.

The pooled beta-thalassemia, sickle cell and rare anemia populations are shown for ICL670, the deferoxamine, for deferoxamine if you compare the trials you see beta-thalassemia and sickle cell disease and what one actually sees here is that not only do these mild stable increases in serum creatinine occur with ICL670, they occur in a percentage of patients on deferoxamine and granted they appear to be somewhat higher with ICL670 when one looks particularly at the sickle cell population these increases are quite similar according to dose here.

Now, the dose for the sickle cell populations is confounded by the fact that the 10 milligram patients were increased many of them to 20 milligrams per kilogram during the study. So, you see there it is similar in the ICL but I put this up to show that there are also increases in deferoxamine and I will ask Dr. Hirschberg to come up, identify himself and speak about this.

DR. HIRSCHBERG: Hirschberg, UCLA. I am a nephrologist and member of the Renal Advisory Board for the development of the drug.

To Dr. Portman's questions, No. 1 I completely agree. Creatinine serum measurements are not the most accurate and optimal way to assess renal function but, Dr. Portman you will agree it is the only practical one and you are certainly also right that the Kochgraf-Galt formula cannot be used in children. We did not use it in children. We had advice from two pediatric nephrologists pointing out exactly that issue and we used the Schwartz formula for the children population within the trial.

Since creatinine serum levels are not the most optimum measure of function there is hardly any alternative. No. 1 exact measurements using creatinine, using glomerular filtration rate measurements with true filtration compounds such as indulin clearances or iosolamate(?) clearances are absolutely impractical to be done in a large clinical trial series such as here and would be an impossible burden to the patients participating.

Last but not least we did make a great effort to in addition of serum creatinine measurements also use the most accurate serum concentration available to assess renal function which is cistatin(?) C. Perhaps you would agree that cistatin C is the absolute most optimum measure that is theoretically possible at the moment in a large clinical trial.

Those levels were not presented because they were virtually identical in their changes and in their overall behavior to the serum creatinine levels. The cistatin C measurements are not widely routinely available for monitoring in the general health care arena because the measurements are rather expensive and technically somewhat complicated. This is why I advised and I would continue to advise with all the limitations but to continue using serum creatinine for the monitoring of treatment.

DR. PORTMAN: Okay, then along those lines for a follow-up it is good that you used the Schwartz formula. I think that cistatin C is coming along as a marker of glomerular filtration rate from a clinical standpoint but I don't believe it is there yet personally, certainly not in general use and in children in any case.

So, you may have to fax me this but do you have a slide looking directly at the calculated GFR changes?

DR. MARKS: We have some data looking at calculated GFR and I can put up that slide for you.

DR. PORTMAN: I can't see it.

DR. MARKS: I will describe it to you. I need to see it in order to be able to describe it to you.

MR. JEHN: Dr. Portman,those slides should have been faxed to you but these are not the ones in the presentation.

DR. MARKS: These are not in the presentation. They are in the briefing materials.

DR. PORTMAN: Do you know what page they are on?

DR. MARKS: SK57

DR. HIRSCHBERG: This slide shows the creatinine clearance changes in the overall population. It is study 0107 and as you can see they just show a mean of change, a slight decrease in creatinine clearance by about between 5 and 8 milliliters per minute occurring at the first measurement after 1 month and being maintained throughout the 1 year of study. This is exactly a mirror image of the serum creatinine which would show a similar slight rise above the baseline that is maintained throughout the study.

I think I can make a very brief important point indicating that these changes, the decrease in calculated creatinine clearance or the rise in serum creatinine occur at the first measurement which is just 1 month into the study by a very small amount of around 8 percent and is maintained throughout the study indicating a functional change in the glomerular hemodynamics leading to this very small decrease in renal function that is not progressive and thereby quite reassuring to me.

DR. PORTMAN: Well, perhaps reassuring to me as well, but another question and I could ask my pediatric colleague there for the kids particularly who have the beta-thalassemia I don't care for them personally but my understanding in looking at the literature is that they are relatively short and may have a decrease in bone mineralization as well as muscle mass. Is that your feeling?

DR. HIRSCHBERG: I am sorry, I didn't quite understand the question. The acoustics are quite bad for this transmission.

DR. ALLEN: In the pediatric patients a decrease in bone mineralization and muscle mass.

DR. MARKS: So, I guess the effect on creatinine of that effect?

DR. HIRSCHBERG: I can certainly briefly comment on it. Of course, since creatinine is derived exclusively from myocytes muscle tissue plays an important role in the serum levels of serum creatinine. There are two arguments to that. No. 1 the calculated creatinine clearance did not show anything different from the creatinine measurement.

No. 2, over a period of several months or up to a year you would not expect such a drastic change in muscle mass even in these children that this would affect serum creatinine to an extremely meaningful extent but over the long run that might be important but last but not least as I indicated we have been studying measurements which show identical changes in renal function and are independent of muscle mass and would thereby be suitable in those children but at the moment as you will agree are not widely enough available to be used as a routine screening tool.

DR. PORTMAN: Thank you for those responses.

Let us move briefly to proteinuria at this time. We did not mention that in the initial presentations. Do you have a slide about the incidence of proteinuria with this medication?

DR. MARKS: We do and it will just take us a moment to bring it up. Here we have incidence. What I have brought up onto the screen here is a slide that looks at the percentage of patients with increases in the total urinary protein to creatinine ratio in a randomized study 0107 and I have the ratio, urinary protein to creatinine ratio divided by groups in rows going down two to four, four to six and greater than six and then I have in the columns ICL670, deferoxamine and all patients and when you look down the columns perhaps with the exception of the bottom column of the urinary protein greater than 6 milligrams which I will ask Dr. Hirschberg to comment on the lower columns are very similar.

So, perhaps you would like to comment on this, Dr.Hirschberg?

DR. HIRSCHBERG: Yes, I could comment on these as follows. Very small amounts of proteinuria are apparently quite common in patients with these diseases that we are dealing with.

This is also true in beta-thalassemia where there is rather little literature but very scanty literature that there is shows some baseline proteinuria suggestive to be associated with and probably caused by the very modest iron overloading that may also occur in the kidney but plays otherwise a rather little role.

There was a moderate incidence of increases in urinary protein over creatinine ratio which was how we assessed proteinuria in patients on both drugs, a little bit higher in the ICL-treated group. Those were very sporadic and transient and did not appear to trace in any way with blips in serum creatinine and again they were transient and disappeared rather quickly in most patients.

We did not observe which we were actually very careful evaluating any occurrence of a nephrotic syndrome which would be of course a particular issue in patients with sickle cell disease.

DR. PORTMAN: You did have three patients who left the study because of proteinuria. Is that correct?

DR. MARKS: Yes, there were three patients. There were patients who left the study because of proteinuria. I will ask Dr. Ford to come up and comment on those patients.

In the meantime just to show you here this is not a slide that you probably need to see perhaps to understand what it shows. The urinary total protein to creatinine ratio measured over time, it was measured actually monthly on the spot urine samples and in the chemistry samples we had and for the ICL620 and the deferoxamine arms on study 0107 for the population they are superimposable and now I will ask Dr. Ford to comment on your question.

DR. FORD; I would like to stress that these protein elevations are very, very transient and the whole picture consists of peaks and troughs and with very occasional peaks and very long troughs in fact. I am aware across the clinical program of only two discontinuations due to proteinuria although I know that the figure of three does occur in the FDA documentation.

One individual discontinued quite early in the course of the study when there was a peak of beta-2 macroglobulin and we were uncertain of the significance of this finding at that time and we thought it prudent to take the patient off study in discussion with the safety board.

The second individual had peaks, was a diabetic and had peaks of albuminuria. Now, albuminuria is the commonest lesion I understand in patients with diabetes and the renal safety board told us that with this individual they probably would not have recommended taking the patient off study. So, those are the only two individuals that I know of in the study.

DR. PORTMAN: Okay, it is good that you bring up data to macroglobulin because that is where I was going next. Clearly the animal data is variable depending on whether you are looking at rodents or marmosets and so forth but it is concerning that there is tubular toxicity in many of the animal models and clearly that can lead to some increases in low molecular weight proteinuria and I noticed that at least in the 105 study you looked at several different tubular proteins but I would like a full description of what you found.

DR. MARKS: I will ask Dr. Ford to come up here because we did look extensively at various renal tubular parameters. We are not prepared to show all of them here at this time but Dr. Ford can describe them.

DR. FORD: This slide shows the urinary albumin creatinine ratio and you can see that as with the previous slide which showed the urinary total protein creatine ratio again the albumin shows no progressive increase over the course of the study and again of course that is a box showing the averages but the review of individual patient data showed no increases in any individual case. In fact, we measured beta-2 macroglobulin retinal binding protein, cistatin C in the urine, various proteins and all of them have essentially this exact same appearance, a totally flat curve with occasional spikes in individual patients which sets them back to normal.

DR. PORTMAN: You said that you were doing a study upcoming looking at the mechanisms of renal abnormalities. Do you feel free in describing what you are planning to do there?

DR. MARKS: I can describe in broad terms that because of the hypothesis that this creatinine change may occur early on we intend to look early on after the first, second, third dose of administration of ICL670 at renal parameters including the rigorous evaluation of glomerular filtration rate using research methodologies in order to assess the effect of ICL670.

DR. ALLEN: Thank you very much, Dr. Portman. We will now have a break. We will reconvene and I would like to have people back here in the room at ten-thirty-five, please by the wall clock over there and we will start 2 minutes after that.

(Brief recess.)

DR. ALLEN: Our first speaker on behalf of the Food and Drug Administration is Dr. George Shashaty.

DR. SHASHATY: Thank you, Mr. Chairman and members of the Committee. My name is George Shashaty from the Division of Medical Imaging and Hematology Products. I am the primary reviewer for Exjade and will be presenting the FDA's review of the application.

Now, in preface to my remarks I would like to congratulate the Novartis group for an excellent presentation and in reviewing my own slides of course we don't make these slides together, I realized that there was going to be a considerable duplication of slides.

So, I think in the interests of time what I would like to do is to point out the differences in our interpretations and review and emphasize certain points which we think would be important.

You have heard from the sponsor that this is the proposed indication and the only comment I would like to make about this indication is that I would like to point out that the indication carries no restrictions with regard to populations.

I don't think we need to look much at this slide either. You all know about deferoxamine but I would like to make two points. Although based almost entirely on retrospective studies the efficacy of deferoxamine in reducing morbidity and mortality in this condition is almost universally accepted.

Second, the long-term safety of deferoxamine is known from 40 years of experience and except for adverse effects on the eye and the ear and local reactions at the site of administration appears to be free from long-term important sequelae when used as approved and because of this desoxamine was used as a comparator drug in the control trials because it was believed unethical that patients not be left untreated for an entire year.

You have heard some comments about the application for Exjade. There were promising indications at the early phases and later phases of Exjade and therefore Novartis was granted a number of requests to move the product along and the FDA agreed.

If I might just go back to that slide for a minute, the last thing that I believe was not mentioned was that Novartis requested a treatment protocol to allow patients to have access to Exjade prior to potential marketing approval and that was granted as well.

I am going to be reviewing pretty much the same kind of information that was reviewed so well by Dr. Marks. We will be including some information about efficacy, safety. We will talk about the proposed dosing. We will have a summary and some information about risk/benefit assessment.

This is a slide that shows the studies that were done in support of the application of Exjade and I want to point out several things. First of all, the most important of the trials 0107 was a controlled trial. It included patients only with beta-thalassemia, It was large. The efficacy endpoint was liver iron concentration at baseline and at 48 weeks and the dosing was 5 to 30 milligrams per kilogram per day.

The control drug was deferoxamine. The other trials, the other major supportive trial, 0108 was non-comparative. About half or less than half, 85 patients with beta-thalassemia were enrolled in this trial and 99 patients with other chronic anemias were enrolled in this trial. There was the same efficacy endpoint and the doses were the same.

The other trial I would just point to 0109 was done in patients with sickle syndromes. It is moderately large. This was the only one. This was done and it was a two-to-one ratio of randomization, two patients to Exjade to every one patient treated with deferoxamine. It was primarily a safety trial and was reviewed by us virtually exclusively for safety.

The other earlier trials which were sort of briefly mentioned were the study 0105 which was basically a dose-finding study and then 0106 which involved 40 patients and was a pediatric study to determine tolerability and safety in the pediatric population.

Now, as you have heard the sponsor's pivotal trial was 0107. It was a randomized open-label parallel group, multi-institution study employing a non-inferiority design using deferoxamine as the comparator.

Five hundred and eighty-six patients with beta-thalassemia and transfusional hemosiderosis were enrolled and treatment planned for 48 weeks. Two hundred and ninety-six patients received Exjade and 290 received deferoxamine. The dose of each drug was determined by liver iron concentration at baseline and the efficacy was based on the difference in success rate between the Exjade and the deferoxamine arm. That is important to remember.

Now, the measurement of liver iron concentration was performed by liver biopsy in about 85 percent of the patients enrolled in the study. Liver biopsy is considered to be the gold standard for this but I would just want you to know that there are some problems in interpretation of liver iron concentration even by liver biopsy but basically it is the standard. It was agreed to that that would be the measure that would be monitored in the studies for efficacy.

The SQUID as you heard about was used in approximately 16 percent of the enrolles and somewhat more commonly in children in beta-thalassemia and in the patients with the rare anemias it was often somewhat in the, more often used in the elderly population because of problems with biopsy in that group of patients.

The SQUID was not validated as a measure of liver iron concentration and the sponsor's own studies measuring liver iron concentration by SQUID compared to liver biopsy indicated that SQUID underestimated the liver iron concentration by a factor of two, and I think you have heard that from the sponsor already.

Interestingly enough however, there were three centers that performed liver iron concentration by SQUID and there was not a very close correlation amongst the three centers for the ratio.

So, I think we have a lot of problem with the use of SQUID in this study. Remember though the SQUID population in the beta-thalassemias represented only 16 percent of the patients.

You have seen this slide. This is the slide for the dosing of both Exjade and deferoxamine. I won't go into it too much more except to say that 97.4 percent of patients recruited into the trial had already been receiving deferoxamine prior to enrollment. The sponsor permitted patients who were assigned to deferoxamine to reinstitute deferoxamine at a higher dose after a washout and run-in period if that higher dose had appeared to be effective prior to enrollment regardless of the liver iron concentration and you will see how important that is. It has already been described by the sponsor.

There were 15 deferoxamine naive patients. Seven were assigned to Exjade and eight were assigned to deferoxamine, so, equivalent distribution.

You have also seen the following slide which is the determination of success and this was a discussion that was held with the agency prior to the studies being performed.

Normal liver iron concentration is approximately 1 milligram per gram of iron per gram of dry weight and levels below that value were believed to suggest over chelation and that is the reason why the one is in there. As you have heard liver iron concentrations up to 7 milligrams of iron per gram of dry weight are usually not associated with the development of organ toxicity. That data is primarily from patients with hemochromatosis rather than from patients who have transfusion related. So, one could say, "Well, does it have a relationship here?" I don't want to speculate.

One of the problems we had with the study was it was based on a non-inferiority analysis and the efficacy was to be shown if the lower bound of the 95 percent confidence interval for the difference in success rates was greater than minus 15 percent. I hardly understood what that meant, but I learned about it, but if you have a little difficulty understanding it don't worry about it.

This is a statistical analysis that had been gone over with the agency and there was an agreement that what would happen is if the patients had entered into the post-treatment iron concentration, liver iron concentration from where they had begun they would be considered to be successful. Anyone else would be considered to be a failure.

My own bias is that that provides some information but there is probably more information that can be obtained from other analyses.

Now, it should be noted and this is one of the problems that the data to determine the effect of deferoxamine on liver iron concentration at 1 year of treatment is almost unavailable and in non-inferiority trials one has to demonstrate that the control agent has a certain efficacy and I in my review of the literature and the literature provided by the sponsor there was almost no data in which liver iron concentrations were measured at the beginning and were measured 1 year after administration of deferoxamine.

Now, all kinds of other endpoints were measured but not liver biopsy, liver iron concentrations. Therefore the evaluation of this study as a non-inferiority trial to support the efficacy would be somewhat difficult.

These are the analysis populations. It has already been shown to you. I will not dwell on it. The demographics of the patients assigned to the Exjade population and to the deferoxamine-treated arm were quite similar.

Now, the dose cohorts I think that this slides bears a little bit of looking at and it has already been alluded to by the sponsor. If one took the baseline liver iron concentrations and put the patients into the category to which he or she belonged what one would notice was that now remember these are patients who have been assessed for liver iron concentration both with SQUID and with liver biopsy, 16 percent in the SQUID, 84 percent in the liver biopsy and what one sees is that the distribution the 5 to 4.8, 26, 27, etc., are reasonably close, but what happens is when the patients are then assigned a certain dose of drug to take in the patients who were given deferoxamine the ranges are quite comparable, as a matter of fact exact for Exjade I should say.

However, in patients assigned to the deferoxamine arm you will notice that in the patients who were in the lower group in the lower LIC group fewer of them were assigned to what would have been considered the protocol specified deferoxamine dose and more of them were in the higher doses for deferoxamine. That was not the case with Exjade and patients generally remained in approximately Exjade arm, Exjade dose to which they were initially assigned and it is higher. Generally speaking it is at the upper limits of what would have been expected with the dose of deferoxamine or a little bit higher. That has already been discussed but this is a very important issue to understand.

Here is a slide that shows the disposition of the patients. Ninety-four point three percent of the patients completed the trial in the Exjade arm, 95.9 of the patients in the deferoxamine arm completed the trial and there were discontinuations of 17 in the Exjade-treated arm, 12 in the deferoxamine-treated arm.

The discontinuations were primarily related to a somewhat greater incidence of adverse events in the Exjade arm and surprisingly to me anyway a little bit higher death rate in the deferoxamine-treated arm.

This was what failed in the trial. The primary efficacy endpoint was success as previously defined. The primary efficacy population was the per protocol 1 population. These patients all had to have liver biopsies and it included patients who were discontinued from the trial for adverse events and iron overload death none of which occurred, at least iron overload death.

Now, the point estimate for Exjade was 52.9 percent and the point estimate for success for deferoxamine was 66 percent and when you look at these statistics the difference is minus 13.5 if you remember that Exjade was to demonstrate that it was no worse than 15 percent worse than deferoxamine.

Unfortunately although the difference, the point estimate in a way is acceptable when you look at the confidence intervals and statistically that is the important thing the lower bound of the confidence interval was exceeded.

So, instead of the probability of remaining above minus 15 was 95 percent or greater here, it could easily be exceeded. There was a minus 21.6 lower bound of the confidence interval.

Because non-inferiority of Exjade compared to deferoxamine could not be demonstrated in the entire primary efficacy population the sponsor performed a post hoc analysis for non-inferiority in the subgroup of per protocol 1 patients whose liver iron concentration at baseline was equal to or greater than 7 milligrams of iron per gram of dry weight determined by either liver biopsy or SQUID.

This subgroup analysis was reviewed but is not included in my presentation because the agency believes that results of subgroup analyses are useful for hypothesis generation and for further studies but cannot be used to provide evidence for efficacy.

In study 0107 the sponsor pre-specified a secondary endpoint which was the change in liver iron concentration from baseline to the end of study in patients in the per protocol 2 population.

Now, in these patients the mean change in liver iron concentration from beginning to end of study was minus 2.4 milligrams of iron per gram of dry weight. If one looks at the patients treated with deferoxamine the mean change is minus 2.9 milligrams of iron per gram of dry weight.

In patients if you then take this group and you subgroup it and we are not saying that we would permit that but this is in a review of the information that is available to us because we think it is important in the subgroup of patients whose liver iron concentration who received 20 or 30 milligrams per kilogram per day of Exjade; these would have been patients who had liver iron concentrations greater than 7 milligrams of iron per gram of dry weight, if one looks at that subgroup you will note that the difference is greater and Dr. Brittenham has already pointed this out that it is likely to be greater and the mean change over 48 weeks was minus 5.3 grams per gram of dry weight.

Now, to put this into some kind of numbers I sort of made a quick calculation. Based on body weight for example, in a 60-kilogram person -- why did I take 60 kilograms or 132 pounds? Well, there were a lot of children and there were older people. So, I think it is probably lower than the average American, but if you took a 60 kilogram person this fall in liver iron concentration would represent the excretion of iron contained in approximately 18 units of packed red cells over the 48-week period.

Now, the sponsor has indicated that based on what they have it is likely to be responsible for the excretion of somewhat more but if one looks at this group of patients minus 5.3 milligrams of iron per gram of dry weight is a substantial reduction.

Now, this is over a 1-year period. Potentially patients treated for longer periods would have a similar fall over the next 48 months. In study 0107 the sponsor pre-specified a secondary endpoint which was the change in serum ferritin from baseline to end of study.

At low doses of Exjade serum transferrin rises over when the dose is 5 or 10 milligrams and the corresponding, you have seen this slide, the corresponding levels and this has been pointed out by Dr. Brittenham, the corresponding changes in transferrin are basically zero at low doses of deferoxamine but you will note that at higher doses of Exjade, 30 milligrams and higher doses of deferoxamine there is a decrease; there is a significant decrease, about 1000 micrograms per liter in patients who are receiving 30 milligrams compared to 50 or more milligrams of deferoxamine.

Now, the supportive study for Exjade was called 0108. This was a single arm study. So, there is no control. It was multi-institutional. It involved the administration of Exjade for 48 weeks and in this study there was sort of an amalgamation of patients. There were 85 patients with beta-thalassemia. These were said to be persons who were either unable or non-compliant or appeared not to respond well to deferoxamine and there were an additional 99 patients recruited into the study who had what was said to be rare anemias and hemosiderosis related to chronic iron transfusion who were thought to be capable of benefitting from Exjade.

In the latter group of 99 patients there were 47 patients with myelodysplastic syndrome, 30 patients with either Diamond-Blackfan syndrome or Blackfan-Diamond syndrome and 22 patients with other diverse types of transfusion dependent anemias.

The dose of Exjade initially administered was based on liver iron concentration as in study 0107. In this study as mentioned the liver iron concentration was measured by biopsy in about 65 percent and by SQUID in approximately 35 percent and the analysis of efficacy was based on the change in liver iron concentration from baseline to end of study.

Now, the sponsor did call for evidence of efficacy if 50 percent or greater of the patients enrolled in the trial were successful, again, based on what liver iron concentration was at outset, what liver iron concentration was at the end of 48 weeks of trial.

Now, there was nothing here to compare it with because this was a single arm trial. Deferoxamine was not used. So, what we wanted to do was sort of to concentrate on the change in liver iron concentration first of all in all patients and now this is the PP2 population which is different from the intent to treat population because a number of patients fell off the study but you will note that liver iron concentrations in the entire population, PP2 population of 147 fell from 18 to 13.8 a fall of 4.2 milligrams of iron per gram of dry weight.

If one takes the patients, separates out the 126 patients who received either 20 or 30 believed to be more effective dosing there was a fall in liver iron concentration of 5.5 milligrams per gram of dry weight.

So, again this seems to indicate that there is an effect of Exjade on liver iron concentration and from a clinical point of view I would think one would say that it is significant.

Now, if one looks at the patients in 0108 who have beta-thalassemia again all of the patients with beta-thalassemia a drop of 4.7 milligrams. In patients who received what we would perhaps consider appropriate therapy and that was the majority, there were 67 out of 76. The fall was minus 6.1.

In study 0108 the sponsor pre-specified a secondary endpoint which was the change in serum ferritin from baseline to the end of the study and here you can see that at doses of Exjade of 5 to 10 milligrams per kilogram per day the serum ferritin rises over the 48 week period but the number of patients treated at that dose was quite small. At a dose of 20 milligrams there is a lesser increase perhaps in serum ferritin in patients with beta-thalassemia and a fall in serum ferritin in patients with other chronic anemias and then at higher doses there is a mean reduction in serum ferritin of about 1000 micrograms and 645 micrograms per liter in patients with beta-thalassemia and rare anemias respectively.

In summary the single pivotal study 0107 failed to demonstrate non-inferiority of Exjade to deferoxamine based on the protocol specified analysis. However, in study 0107 treatment with either Exjade or deferoxamine reduced liver iron concentration from baseline. This is consistent with a biologic effect and a reduction of the iron burden.

This reduction in iron burden it should be remembered occurred in the face of continuing transfusion requirements. I am quite certain that if the patients were not administered either drug the liver iron concentrations at the end of 48 weeks would have been higher than they would have been at the baseline.

Study 0108 results show a decrease in liver iron concentration from baseline again suggestive of a treatment effect of Exjade.

Now, I should now like to turn to our evaluation of safety and I want to provide data on the safety database and the safety findings particularly from study 0107. I then want to review some organ-specific toxicity suggested from 0107 and the other studies as well as a review of the deaths of patients in all studies.

There were if you add them 700 patients exposed to Exjade in clinical trials most of whom had beta-thalassemia. I will take this back. Some of you are actually going to add them up. When you add them up you are going to find that if you add these numbers it comes out to 703. I have an explanation but I am not going to go into it. It was very complex. It took me a week or to figure out where they were but just make the assumption that there were 700 patients who were exposed to Exjade in the clinical trials.

Most of these patients had beta-thalassemia. Now, in addition to this 421 we have to also put in 50, actually there were, you see this is where the difficulty comes. There were 48 patients who were in this original 0105 trial and then they went on to an extension of 0105 and there were three patients who were down here who then were put into 0105.

So, that is where the difference in numbers comes about. At any rate there were about 99 patients as we said who had rare anemias, 132 patients in the Exjade arm. Notice the two-to-one ratio with deferoxamine. There are extension studies 0105 and others I will discuss and in clinical pharmacology studies for things like maximum tolerated dose absorption and distribution, metabolism, excretion, PK, PD they did an extensive QT study and so on. An additional 237 patients were exposed to Exjade but most of them only got one dose or something like that.

So, the 237 patients include both healthy volunteers for some of the pharmacokinetics and the QT trials and so on and some of those patients also had beta-thalassemia. So, it is sort of a combination.

These are the results in 0107. First of all if one looks at the deaths in 0107 as mentioned there was one death in a 3-year-old splenectomized thalassemic male on the 84th day of treatment he was receiving 31 milligrams per kilogram per day. You say that is a peculiar dose but there are reasons why it wasn't 30. It was 31. This child lived in Tunisia and he lived a great distance from the treatment center and the description in the case report form and the laboratory studies that were provided he seemed to have been doing quite well and then he woke up from sleep. He appeared to be somewhat lethargic. He was not febrile at the time and it simply states that he died in his mother's arms and there was no medical attention, there were no medical people attending to him and there was no further information. So, we have no idea why that particular patient died.

Now, there were three patients who died in the deferoxamine-treated arm. One of these patients died from a convulsion, one from sepsis and one from an intracardiac thrombus on days 180 through days 367 of the trials.

What were the discontinuations due to adverse events? Adverse events led to the discontinuation of Exjade treatment in nine patients treated with Exjade and in four patients treated with deferoxamine.

I would just like to call your attention to the fact that this number of nine is different from the number of eight that was if you remember it on the slide before and the reason for this is that there was one patient who discontinued in the following manner. The patient was receiving Exjade. I can give you all the information. I have it on a separate sheet but this patient developed what appeared to be hepatitis. The patient was then removed temporarily from the Exjade. The Exjade was reinstituted. There was a recrudescence of what appeared to be drug-induced hepatitis. There was a liver biopsy which was very compatible apparently with drug hepatitis and the patient decided to stop the medication on her own.

That was said to have been the cause then for discontinuation. It was then said to have been patient withdrew consent. I interpreted this to be related to an adverse event.

So, of the discontinuations due to adverse events there was one because of death in the Exjade arm, three from deaths in the deferoxamine-treated arm. There were four drug discontinuations because of hepatic adverse events and these included an increase in serum transaminases of a significant degree in two patients and then in two patients who had pretty classical for me drug-induced hepatitis one of whom had a rechallenge and both of whom had liver biopsies.

Additional reasons for discontinuation of the drug included one for drug-induced fever, one for cataract, one for skin rash and one child interestingly who on several challenges developed hyperactivity and insomnia.

Serious adverse events were seen in 27 patients on the Exjade arm and 25 patients with deferoxamine treated arm and generally I would say the distribution of serious adverse events was reasonably similar in both arms of the trial.

Most of the serious adverse events appeared to the investigator and I am reading the reports and so on not to be related to study drug but some were.

Here is a listing of adverse events occurring in more than 5 percent of the patients in study 0107; 85.8 percent of patients receiving Exjade and virtually the same percentage in patients treated with deferoxamine experienced at least one adverse event during the trial and if you can't read this I will read them for you.

The frequency of fever, headache, cough, pharyngeal-laryngeal pain, back pain and urticaria was somewhat greater in deferoxamine-treated patients. The frequency of abdominal pain, diarrhea, nausea, rash, arthralgia, ear infections and an increase in serum creatinine were greater in the Exjade-treated patients.

Now, what we did was to try to break them down a little bit to indicate what we thought were notable differences in adverse events for patients treated with Exjade versus deferoxamine and the differences were primarily in gastrointestinal symptoms, skin rash, increases in serum creatinine. This is a greater increase in serum creatinine than previously described and in 11.1 percent it was seen in 11.1 percent of patients on Exjade and 0 percent of patients with deferoxamine.

Transaminases were increased in 5.7 percent of patients treated with Exjade and 1.7 percent of patients treated with deferoxamine.

Another important thing and when we talk about organ-specific it is true that at lower, that there were equal numbers or equal percentages of patients who had sort of low-level proteinuria whether they received Exjade or whether they received deferoxamine but when patients were separated out by what I would think would be fairly heavy proteinuria and I calculated this to be about between 800 and 900, correct me if I am wrong, 800 and 900 milligrams of protein in what I would consider to be a 24-hour urine, these were not spot urines. The frequency of heavier proteinuria greater than 0.6 milligrams per milligram was seen in 18 plus percent of patients on Exjade and only in 7 percent of patients on deferoxamine.

There is also some increase in hepatobiliary. This included some of the patients with transaminases but Exjade is excreted through the biliary tract and there were some patients who developed hepatobiliary symptomatology which may have been related to that mechanism.

Preclinical studies indicated that Exjade had toxic effects on the kidney. These were primarily tubular in location but rare glomerular changes were also seen.

Because of this patients were excluded from the trials if the serum creatinine was raised. In addition the protocol stated that an increase in serum creatinine greater than 33 percent above baseline on two consecutive measurements 1 month apart would lead to dose reduction or interruption.

An increase in serum creatinine triggering dose reduction or interruption occurred in 11.1 percent of Exjade-treated patients and no patients who were receiving deferoxamine.

This increase appeared to be dose dependent. For instance at a dose of 10 milligrams per kilogram per day 2.6 percent had that increase which called for dose reduction, etc. At 20 milligrams it was 8.3 percent and at 30 milligrams it was 20.2 percent.

I want you to understand that I don't think that patients if the drug is approved, I don't think that patients are going to receive 5 milligrams per kilogram nor will they receive 10 milligrams per kilogram. They are much more likely to receive 20 milligrams or 30 milligrams per kilogram and therefore the safety, the percentages and so on that you read, the global percentages are not likely to be those that will be seen in the treated population.

In addition, you saw that there was what I called heavy proteinuria in 18.6 percent Exjade patients compared to 7.2 of the deferoxamine-treated patients. The sponsor states that there were no cases of renal failure and I did not see any in my review.

The other supportive studies provided the same sort of findings except as mentioned patients with rare anemias in whom an increase in serum creatinine led to dose reduction or interruption in about 19 percent of the patients treated in that group.

Adverse events. Exjade and Exjade iron complex are excreted through the bile. Preclinical studies indicated that Exjade had toxic effects upon the gallbladder and bile ducts in marmoset monkeys.

Patients were excluded from the trials if the transaminases at entry were greater than five times the upper limits of normal. So, anybody who started with a significantly irritated liver was not included in the trial.

Increased transaminases defined as equal to or greater than five times the upper limit of normal occurred in 5.7 percent of Exjade-treated patients and in 1.7 percent of deferoxamine-treated patients.

I mentioned before there were two cases that in my reading clearly demonstrated drug-related hepatitis, I mean as best as can be determined. I know that there are no specifics but for instance in one patient whose transaminases rose removed from the drug, transaminases fell, rechallenged transaminases rose. Stopped the drug. In fact, in one of the patients as of the last follow-up date which was about 3 months I believe following the cessation of the medication this was a patient who had normal transaminases at entry into the study, that patient's ALT was still about 130. The other patient appeared to have resolved the liver function abnormalities.

So, this is two cases of what one could only consider to be drug-induced hepatitis out of a total population of 700 or 703 patients.

Now, this did not appear to be dose related. There were increased transaminases that led to discontinuation of Exjade in two patients. Increased transaminases led to dose adjustment or interruption in three Exjade-treated patients. Interestingly and the question was asked before bilirubin levels up and down, up and down; there is a hemolytic process that they do have other causes for liver disease and so on but the patients who had these abnormalities did not tend to have elevated bilirubin levels. That has been a focus of drug-related hepatitis whether or not the bilirubin levels were raised and I would say that there was no correlation.

In other studies performed by the sponsor similar findings to these were noted and we will go through them in just a little bit. Now, preclinical studies indicated that Exjade had toxic effect on the lens of the eye and in patients receiving deferoxamine it is well known that they may suffer adverse events on the eye and ear and therefore attention was paid to this aspect, and I would just make mention of the fact that in study 0107 cataract occurred in one Exjade-treated patient at age 16. The drug was discontinued and in two patients on deferoxamine ages 18 and 36 the drug was continued.

There was diminished hearing mostly on the basis of doing otological examinations, 8 patients receiving Exjade, 7 patients receiving deferoxamine. No change was made in this group of patients but the drug was interrupted in two patients here.

One patient on Exjade developed vertigo. No intervention was made and in the other studies presented by the sponsor generally similar findings and in addition in study 0108 there was one patient described who developed a retinal hole and there were three patients in study 0108 who developed cataracts. Remember that 0108 there were a lot of elderly folks and in study 0109 which is the sickle cell population there were retinal vascular changes that occurred.

There were four in patients receiving Exjade and two DFO but remember the ratio in this trial was two to one. So, one would say they are approximately equivalent.

This is a listing of all the events that occurred during the clinical trials. There was one patient who died, we mentioned before in the beta-thalassemia trial. In 0108 there were six deaths. They all occurred in the patients with rare anemias. There were no deaths in the beta-thalassemia population.

I could give you a description of what each of the patients died from but I would just sort of aggregate them and say that they probably died of -- five of the patients had myelodysplastic syndrome. They died probably of progression of myelodysplastic syndrome.

One patient with Blackfan-Diamond syndrome died. That patient died of sepsis. That patient entered the trial with a low white count. The white count went down further and the patient developed sepsis and died.

I would just like to say that one of these patients died as a result of liver injury when the liver biopsy was performed to determine the end-of-study liver iron concentration. The patient had to be operated upon and she died of intra-abdominal sepsis following the surgery.

These are what we thought were notable adverse events that occurred during the extensions of the trials. Remember that many of the trials the patients when they completed the 48 weeks were invited to remain on Exjade. They did, a number of them did and this is what we found from these extension trials.

First of all there are 426 patient altogether, 360 with beta-thalassemia, 66 with rare anemias. They were mostly on the trials for a total of 72 to 96 weeks. There was one death of a beta-thalassemia patient from congestive failure and four deaths from myelodysplastic syndrome.

These three are additional to the six who died in the 0108. This patient is included because she actually died although here end-of-study liver biopsy was done within the 48-week time frame or shortly thereafter. She died about a month and one-half after her liver biopsy.

In addition drug discontinuations were seen in five patients in the extensions. That included steatosis with increased transaminases and one glycosuria, colitis and increased creatinine and in one very dramatic patient who developed Henoch-Schonlein purpura. It was a classical case of Henoch-Schonlein purpura that occurred in the 13th month of administration of the drug. It led to problems with the kidney where glomerulonephritis was present. The patient had purpura. The patient excreted approximately 4 grams of protein in a 24-hour sample and as of the last report that patient had not recovered from Henoch-Schonlein purpura.

Serious adverse events occurred in 7 patients in the extension trials, one DVT, one pulmonary embolism, atypical tuberculosis and increased transaminases and then cholelithiasis.

From a safety point of view our concerns are mostly for the kidney and the liver and the frequency of the adverse events associated with Exjade is greater than with deferoxamine in these organs. Adverse effects on the eye and the ear appear to be similar in frequency to those with DFO but the numbers are small. Gastrointestinal and dermatological adverse events exceed the frequency of those, adverse events associated with Exjade exceed those in patients receiving deferoxamine but are manageable.

The frequency of uncommon, important adverse events is not known because the safety population is small and the indefinite use of drug and the likely greater dose of drug to be administered to patients will most likely be associated with more frequent and different adverse events.

You have seen this slide. It is the sponsor's proposed dosing schedule and our comments on it are as follows. In clinical trials dosing was based entirely on liver iron concentration. I don't think anybody in the world thinks that patients will need to have a liver biopsy to determine whether or not they should be started on or continued on a drug which eliminates iron and this is a very important question that we will be asking of the Committee.

The transfusion of 100 milliliters per kilogram as a point to begin therapy with Exjade would be expected by my calculations to increase the liver iron concentration from normal to about 7 to 10 milligrams of iron per gram of dry weight and finally the sponsor's recommendation of a ferritin of greater than 1000 there is a great deal of scatter and I think that the correlation that the sponsor has provided, the correlation is about .63 between ferritin and liver iron concentration, but if you have a person who is regularly transfused and you do a bunch of ferritins on him or her it isn't going to be due to another confounding cause.

In summary the problems that we had with the studies that were performed are as follows. First, wee believe that an adequate dose-response curve was not established prior to the beginning of the pivotal trial.

Second, there is little information from the literature to quantitate the effect of deferoxamine on the reduction in liver iron concentration after 1 year of therapy.

Therefore it is very difficult to evaluate a non-inferiority trial to establish the efficacy of Exjade.

Third the data for Exjade have been provided mostly in patients with beta-thalassemia. Experience in other populations is limited.

The efficacy of the drug is based on an assessment provided by a single randomized trial and one supportive single-arm trial. Generally the agency requires two adequate and well-controlled trials. In addition the randomized trial did not meet the pre-specified non-inferiority statistical criteria for success.

Nevertheless a treatment effect was observed in both trials examining the change in liver iron concentration from baseline even with continuing transfusions. Again, this is an important point for the Committee to comment.

From a safety point of view there is evidence for kidney and liver adverse events. We believe that patients have demonstrated adverse events on the eye and on the ear and we have no idea what is going to happen after 20 years of treatment.

We have a pretty good idea of what deferoxamine does after 20 years of treatment. It was approved by the FDA in 1965 and a treatment that is going to be provided lifelong we cannot predict whether or not significant, particularly hepatic and renal damage will occur.

From our own assessment of benefit and risk we believe patients treated with an Exjade dose of 20 to 30 milligrams per kilogram per day will demonstrate a decrease in liver iron concentration as seen during the study with patients continuing to receive transfusions.

Exjade has shown clinical adverse events and laboratory abnormalities, we believe mostly non-serious in these clinical trials. However, the safety database is small and the length of treatment in the clinical trials comes nowhere near the length of treatment that patients will have.

In addition, the clinical benefits of Exjade on morbidity and mortality which is what a clinician thinks of, you know, I don't care whether the person's liver iron concentration goes down or up or sideways. I want to know did the patient die sooner or later. Did the patient develop congestive heart failure? Did the patient develop cirrhosis? Did the patient develop hepatoma, etc., things that are important?

We have some idea of that in deferoxamine even though most of it is retrospective. We don't know whether or not treatment with Exjade causing the liver iron concentration to fall is going to be met with clinical morbidity and mortality benefits.

I would like to acknowledge the other members of the review team. Most of them are sitting in the front and to the left. This is a group effort as you probably know at the FDA and I was just selected to be the representative for this team.

I am now open for questions.

DR. BRITTENHAM: Could I suggest that we defer the questions and discussion until after lunch.

DR. ALLEN: Put your mike on, please?

DR. BRITTENHAM: Could I suggest that we defer the discussion of this which will be lengthy until this afternoon. I think it is important that the patients be heard.

DR. WEISS: Dr. Allen, that would be acceptable to us if it meets everybody else's schedule. That would be fine because we do have this afternoon built in to have some additional discussion an some of the discussions will probably take place centered around the questions that we have posed to the agency.

In fact, we would like to have discussion associated with these questions.

DR. ALLEN: Dr. Portman, is that satisfactory to you? Are you planning to be on this afternoon, also?

DR. PORTMAN: I am. I will be.

DR. ALLEN: All right. Thank you Are there any questions with regard to the slides and the technical points of clarification that people would like to raise at this point?

DR. SCHREIBER: In your analysis did you do anything that was age related since the indication or prescribed indication is anybody up to the age of 2? Did you do anything with the younger age groups to see if there was the same effect of treatment?

DR. SHASHATY: Actually and I would turn that question over to the sponsors, but in the patients under the age of 6 in the comparisons in that subgroup it appeared that deferoxamine was more effective than Exjade but in the subgroups anything above 6 it appeared to be reasonably similar.

DR. ALLEN: Okay, thank you very much for your presentation. We will now move to the open public hearing. We have 12 scheduled speakers. I would like to ask that any of the speakers who have, I will use the old term "slides" or PowerPoint presentations, please bring them for loading now.

If you can bring them please to our Committee Secretary we will get them loaded for you. I would like to ask you please to adhere to the time limits that are given to you.

Mr. Jehn will introduce each one of you. I need to read first of all the open public hearing announcement for particular matters. Both the Food and Drug Administration and the public believe in a transparent process for information gathering and decision making. To ensure such transparency at the open public hearing session of the Advisory Committee meeting the FDA believes that it is important to understand the context of an individual's presentation.

For this reason FDA encourages you, the open public hearing speaker at the beginning of your written or oral statement to advise the Committee of any financial relationship that you may have with the sponsor, its product and if known its direct competitors. For example, this financial information may include the sponsor's payment of your travel, lodging or other expenses in connection with your attendance at the meeting.

Likewise FDA encourages you at the beginning of your statement to advise the Committee if you do not have any such financial relationships. If you choose not to address this issue of financial relationships at the beginning of your statement it will not preclude you from speaking.

We will move into the hearing.

MR. JEHN: Okay, thank you,Mr. Chairman. We have received at this time 14 written submissions for the record which will be posted on the web after the meeting once they have been cleared. Also, there was a petition that was circulated and has 1800 names identified as signing the petition and that will, also, be put up on the web. We have received at this time 12 requests for oral presentations. We are going to limit them to 3 minutes to make sure that everybody has an opportunity to speak.

We have a microphone there off to the left here that the open public hearing speakers can come up to or if they prefer the podium they are certainly welcome to do that also.

I am sort of listing the names of speakers as they come up by order. We are trying to get the patients and some of the younger children out of here earlier as was requested of me.

So, the first speaker will be Amy Celento and she also has her son, James Stamteris.

MS. CELENTO: Good morning. Thank you very much for the opportunity to speak with you today and I do not have any financial relationship to the sponsor of the study.

My name is Amy Celento and this is my son, James. He is 7 years old and James likes pretty much what other boys like, video games, pizza, soccer, running outdoors, running indoors but he dislikes things like, hugs from girls, homework, eating all his vegetables, Monday mornings but one of the other things he dislikes that most other 7-year-old boys don't have to deal with is having to get a needle stuck in him in order to do his Desferal.

He is a great kid and even though the whole process is difficult and painful he is usually pretty good about using his pump due to the fact that he is pretty sharp. He has put two and two together and he has heard about shortened life and about chelation and longer life with chelation.

So, I am luckier than most moms because of that but even so it is a stressful situation for both of us and of course even on good nights there is a degree of stress underneath the surface.

As I said, James is a great kid and I think that on the whole he has been more compliant with his treatment than many other kids for several years, but I also know that talking with parents of other patients that the best patients tend to lapse when they get into their teen years. Sometimes the lapses are fairly infrequent, maybe a day or two here or there but sometimes they can last for weeks, stretching into months and in some cases patients just stop altogether.

They are teenagers and like every other teenager they think that they are going to live forever. Unfortunately and tragically too many of them find out the truth too soon, often within a few years.

So, those 20 years of treatment may not be a reality due to the challenges of compliance with Desferal.

In addition to being the mother of a child with thalassemia I am also a very active board member of the Board of the Cooley's Anemia Foundation and every so often, more often than I would like I check my e-mails and find one from the foundation with a subject line of sad news. It takes me a few moments to work up the courage to actually open that e-mail because I know that I will find out that someone, a board member, a beautiful daughter that I met at a parent or patient conference or a stranger I have never met but whose name I recognize has passed away.

The e-mail will announce that the cause of death was often heart failure or liver disease or something else but we know that the real reason the patient has died has nothing to do with the heart or liver or whatever else the e-mail says. It is because they could not, not would not but could not comply with the current treatment.

As a mother this is something that scares me more than I an possibly express to you here today with my son standing next to me.

So, all of us who have children with thalassemia try to teach our kids how important compliance is but the reality is that the current therapy is simply too big a burden for almost any patient.

The drug is very good, but the difficulty of administering it severely reduces its effectiveness and the result is that too many continue to die much too young.

That is why I am here to offer my support of the new drug application for Exjade. Children with thalassemia desperately need alternatives to the current treatment. They need drugs that can be administered in a way that offers a reasonable chance for proper compliance.

For decades the thalassemia community has been told that an alternative is right around the corner. I ask you to end the waiting and to let us at long last turn that corner.

Thank you very much.

MR. JEHN: Thank you. Okay, next on the list we have Hank and Kathy Backer.

MS. BACKER: Thank you for the opportunity to allow me to say my piece. I have no financial connection with the supporters of this medication.

Our son Todd died on June 14, of last year from iron myopic(?) congestive heart failure at the age of 29. The reason for this, he was not compliant enough with Deferal, using the Deferal pump. Unfortunately he is not alone. Desferal only works if you use it. It sounds easy enough to stick yourself with a needle and remain attached to the pump for 10 hours a night but when you have to do it you tire of it quickly and find yourself making excuses to free yourself from this mechanical appendage. It causes discomfort at the needle site which can last for many days. The medicine also appears to cause agitation in the mood of some young men in addition to other known side effects.

The information which I have garnered about the oral chelator appears to be less severe than the known side effects of the currently available infusion chelator. Every time Todd heard there was hope for an oral chelator being made available soon he would faithfully use Desferal for about 6 months. Then he would read or hear yet another delay and he would quit once again for up to a year.

As a result the iron stores overloaded his heart. The oral chelator is coming too late for Todd and for many others but there are many more. We have a younger child also with thalassemia who has been showing resistance to the daily infusion therapy. He, like his older brother complies only for the hope that soon an oral chelator will be available. Please don't kill his hopes and those of all the other patients. We don't want to have to bury another child.

Thank you very much.

MR. JEHN: Thank you. Next on the list we have Otto Szanto.

MR. SZANTO: Good morning and thank you for the opportunity. My name is Otto Szanto. I am an MDS patient with Regg(?) sideroblasts and no deformities of the chromosomes. I have only been on Desferal for 15 months. I will say it is a challenge, nothing I can't handle but it does change your entire life. Anytime I want to do something we have to consider where we are going, where I am going to put the needle in and when I am going to take it type of thing. I usually do it at night. There is some soreness at the area where the needle goes in and it is a little bit of an irritant. I really don't get a good night's sleep because as I roll over and I notice as you get older you have a tendency to roll from one side to the other and I used to sleep on the right side all the time when I was young but I roll from one side to the other and I have to literally pick up the pump and move it to the other side and so consequently I am not sleeping as well as I normally would. I have no connection with the pharmaceutical company, no financial basis or anything. I would say that I am looking forward to an oral drug being placed on the market. That sure would beat having the injection of a needle or an infusion I should say and my infusions last almost 13 hours, 5 days a week.

I hope it gets on the market.

Thank you very much.

MR. JEHN: Thank you. I will call a couple of names up so we can start queuing so we can minimize the time in between.

The next group will be Kathy Jo Olfano Langan and followed by Dr. Zahra Pakbaz.

MS. LANGAN: Good afternoon. I am a thalassemia patient. I have been transfused regularly since the age of 3 when I was diagnosed with thalassemia. I began on Desferal at the age of 12 and was able to maintain a regular treatment schedule for several years. However, when I went to college I faced such a tremendous amount of pressure from all the assignments, activities and having a roommate without the privacy that I needed to do Desferal and without the time to set up the infusion. The process had become difficult and I became non-compliant. While in college I became very ill due to all of the iron in my body and had to return home. At 24 I had a portacath(?) installed. We went to the option, but the only choice available to me if I wanted to use my Desferal which I desperately needed by then.

The port worked well for 6 years until I developed a clot and the port hooked into my jugular, causing a severe problem necessitating it to be removed. I was really crushed by the development. I tried to once again force myself to use the Desferal but it was no use. I simply couldn't do it. It is not that I was lazy. It was just too difficult. I simply physically couldn't go through with it. I would push and push with all my might to get this needle into my stomach and it would never penetrate causing much frustration, much aggravation and ending with me sitting on my floor crying in pure desperation.

I cannot tell you how hopeless I felt. I was on a rough, rocky, painful road leading to cirrhosis of the liver, heart failure or both.

Luckily I was able to get on the program for Exjade, ICL670. I cannot tell you how much this has changed my life. One year and 11 months and 19 days ago is when I started the study. I won't be overly dramatic when I say any of this, what I am saying here today.

Before I started Exjade my ferritin levels were up to about 10,000, in excess, I can't even begin to tell you how much over the limit of what my ferritin level should be. My skin was simply green. I was in horrible shape all over. Today my ferritins are at 3386. My liver iron measurement from a biopsy has come down 13.6 points. I no longer have children pointing at me and saying, "Mom, that lady's skin is green. Is she an alien?" I feel healthier than I have in years and everybody I know is amazed at the tremendous change in me.

This drug has changed my life and could make a positive impact on many others. Children could go to sleepovers without fear of ridicule and their parents could let them go not being afraid that they are going to skip their medicine. Young adults would feel more confident and accepted at college. We could have less anger and frustration that we can't get rid of right now. We would have free lives, something we don't have as long as we are tethered to a pump.

For this trip I did not have any financial ties to Novartis. I am in their study of course.

So, thank you for your time.

MR. JEHN: Thank you.

DR. PAKBAZ: I am Dr. Zahra Bakbaz. I am appearing as a patient and as a physician who works with thalassemia patients. I have thalassemia. I have been transfused since I was 8 years old and I am 32 years old and I have been on Desferal since I was 9 years old and I was born and raised in Iran and I have three pages to read for you. I have 10 minutes.

So, I work at Children's Hospital, Oakland as a research associate and our center is one of the centers for the Novartis study, but let me make clear I don't have any kind of ties to Novartis.

So, the reason I am here is because I do research. I work with thalassemia patients. I am a researcher. Because I have been involved in many thalassemia associations like the Thalassemia Association in our country and Thalassemia International Federation. So, I know the problems that thalassemia patients face in different countries across the world and one of the issues that patients talk about is the time that they have to spend for preparing Desferal and stuff like that but the main issue is that I have been using Desferal for 22 years and after you have to use it for 2 or 3 years that is fine but using it for 22 years it causes scars and usually doctors can tell their patients that you have to rotate so that you don't get so many scars but because we don't have many spots to rotate since we are not an average person we are thin. So, there is nothing in locations to start and so after 22 years you get scars and you can't do anything about that.

So, for myself and many other patients that Susan also mentioned I continue to do the Desferal and it looks impossible and as you see many of us many patients now are really active in the society and we would like to live longer than 40 years or 50 years and in the United States there is no patient older than 50 years old on transfusion and Desferal and this doesn't mean that Desferal is not good but Desferal kept me alive for 22 years. I am 32 years old but now we need something that can keep us here maybe longer.

Thank you for listening.

MR. JEHN: Thank you. Dr. Sophie Lanzkron will be next followed by Gina Cioffi.

DR. LANZKRON; I have nothing to disclose. I am Dr.Sophie Lanzkron. I was asked to speak. I take care of adult sickle cell patients at Johns Hopkins and I was asked to speak to emphasize the need for an oral iron chelator for our population of patients.

I have about 200 patients that we take care of at our center and about 25 of those are on chronic transfusion therapy and they are on chronic transfusion therapy mostly because they have had strokes but I also have patients who are on it because they failed hydroxyurea therapy and get frequent acute chest syndrome. They have pulmonary hypertension or they have leg ulcers and a couple of other reasons patients are on chronic transfusion therapy.

As with any decision-making process in medicine we have to weigh the risks and benefits of this therapy and so if we take for example an individual who has had a stroke and lots of the people that I take care of have had strokes as children and so I have met them while they have been on chronic transfusion therapies their whole lives. They now present as adults and really want to come off of chronic transfusion therapy and so we talk about the risks and benefits and clearly the benefits are that they won't hopefully have a recurrent stroke and that has been shown that that is true at least in pediatric patients, that the risks of continuing chronic transfusion therapy include that of aluminization(?) getting infectious but also obviously the issue of iron chelation and so a patient who decides to go on chronic transfusion therapy or stay on chronic transfusion therapy has to commit to chelation therapy.

So, of those 25 individuals who are on chronic transfusion therapy I probably have three who are truly compliant with their therapy and as was mentioned deferoxamine has been shown to have efficacy in decreasing iron but no one has looked at the effectiveness of this therapy outside of the setting of a clinical trial and I think that the reason I am here is because my personal experience is that the effectiveness is quite low because nobody uses it and we have to go through this process of monthly checking ferritins and chasing down patients and having them take their Desferal and then they will do it for a couple of months as you have heard from other individuals here today and then their ferritins continue to climb up as they become less compliant and then finally I just want to mention that of our patient population 50 percent are on medical assistance and 25 percent are on Medicare and so if this medication or any oral iron chelator is not available through those venues then the cost of the medication will become an insurmountable barrier to patients receiving this therapy.

Thanks.

MR. JEHN: Thank you very much. Is Gina Cioffi not here? Okay, great.

MS. CIOFFI: Good morning. I have no financial conflicts. My name is Gina Cioffi and I have the privilege of serving as the Executive Director for the Cooley's Anemia Foundation and our foundation has been around for 50 years and we are the leading advocate for the thalassemia community in the United States and during the five decades that we have been around our mission can be really summarized in two words which is care and cure.

That is really what we have been focused on and our ultimate goal is the cure but we are continuing to focus on improved treatments and so the standard of care took that first leap forward so many decades ago with Desferal and I had a friend and he had one of the first pumps. It was like this around his body and now it is something that patients can hold in the palm of their hands but we haven't been able to eliminate the pain associated with using that pump either of the pumps that are available now and the problem is that eventually they are going to run out of a site where they can stick themselves without pain and endure that long process of using Desferal.

So, in the years since we have had Desferal we have seen an increase in the life span but it is very rare that a patient reaches 40 years old and too often they are not making it into their thirties.

When we first learned that a date had been set for this hearing we put out a petition drive through the Internet and we got 1800 signatures in a matter of days which shows that there is just an overwhelming need. People are just really desperate for alternative therapies.

So, we also received a number of e-mails from friends and families and I just want to read from one of the e-mails. "My friend passed away in June for not being compliant in using his pump. He dies from heart failure at the age of 29. Every time he would hear that there was hope for a normal chelator being made available he would use his Desferal and then maybe he would hear that there was a delay or it wasn't going to happen yet, then he would quit. My prayers are with you today that there will be an oral chelator."

This next message really paints a very clear picture of why we need an oral chelator. In most instances things get easier over time. I can personally attest that this is not the case with Desferal. The pain and complications you endure from Desferal worsen each day and maintaining a consistent level of compliance grows harder and harder.

For the past 18 months I have taken Exjade. Exjade has provided me with a quality of life that I have only dreamed about and have never experienced. Most important Exjade has given me hope and has provided me with renewed confidence in being compliant with my therapy and my overall health.

We know what a milestone this meeting represents. The reality is that our patients desperately need an alternative therapy. This is a moment that so many people have worked and waited so hard for.

A positive recommendation from this Committee will mean that you understand the enormous need to provide an alternative treatment for these patients and so on behalf of the Board and our patients we really urge you to recommend approval for this drug.

DR. ALLEN: Thank you very much.

As was mentioned a copy of the petition is available at the table and will be posted on the web site as well as other e-mails and letters that have come in to the committee.

Next will be Gargi Pahuja followed by Dr. Willada Edwards.

MS. PAHUJA: Good afternoon. My name is Gargi Pahuja and I would like to start by thanking the members of the Blood Products Advisory Committee for giving me this opportunity to express how I feel about this new chelator, Exjade.

I am 30-year-old woman with beta-thalassemia major and I am also the President of the Thalassemia Action Group or TAG which is the patient support arm of the Cooley's Anemia Foundation.

I represent some 700 individuals in the United States with thalassemia. In addition I was diagnosed with diabetes at the age of 15. I have a variety of endocrine disorders and I am unable to have children.

I wish I could say that I am unique but unfortunately in the thalassemia world my story is all too common. We are susceptible to a wide range of complications most of which are directly related to iron overload, from osteoporosis to cardiac failure.

Like most people with thalassemia I have mixed feelings about Desferal. I appreciate it for enabling me to live a longer life but I, also, resent it for being so difficult to take.

This difficulty has trapped me into a constant seesaw of compliance and non-compliance. There is never an easy answer of yes, I will do my Desferal today or no, I won't.

To blame myself for developing diabetes and other endocrine problems will do me no good but it doesn't stop me from wondering if I were more compliant or if Desferal were easier to do would I have less complications today.

It is true that the choice of whether or not to do my Desferal each night is mine. There are many times when I have had to make difficult decisions such as whether I should stay up late studying for an important exam or take the time to chelate and I make that choice willingly but no delivery system used to treat a chronic illness should be so burdensome that such a decision has to be made on a daily basis.

The prospect of an alternate treatment such as Exjade excites me tremendously and I sincerely hope that this Committee recommends in its favor today.

I do feel, however, that as the President of TAG I have a responsibility to raise some issues that need to be considered.

Foremost amongst these facts is that each and every thalassemia patient has different needs in terms of therapy. Any chelator whether it is Desferal or Exjade or another drug will serve different patients in different ways. Even if Exjade is approved there will be some patients who will need to use Desferal solely or in combination with Exjade. So, it is important to realize that Exjade should be available but as part of a larger menu of chelating options from which doctors may use to tailor treatment to individual thalassemia patients.

I encourage the FDA to approve Exjade but I believe for patient safety's sake it is critical that the FDA fulfill its responsibilities to carefully monitor the use of this drug, the incidence of side effects and any unanticipated adverse events.

That said I am impressed by the studies to date and feel grateful to Novartis first for making the enormous effort required to develop Desferal and then for expending so many resources in pursuit an oral alternative.

I look forward to the day when my doctor and I have the chance to discuss both treatment options and hope that all of you here will take the first step in making that possible for me and other patients alike in order to lead a better, longer and more fuller life, and I have received no financial remuneration from Novartis.

Thank you.

DR. ALLEN; Thank you very much.

DR. EDWARDS: Thank you. I am Willarda Edwards, President and COO for the Sickle Cell Disease Association of America and I am here on behalf of the 70,000 patients with sickle cell disease who are looking for your approval of Exjade as a medication to treat the iron overload side effect from transfusion therapy in patients with sickle cell disease.

Now, given the advances of modern medicine and the improvement in the treatment of patients with sickle cell disease we are really happy to be able to say that many now are living longer lives. In the seventies it was unusual to find a patient older than in their twenties. Now, at our last Sickle Cell Disease Association of America convention that was held just 3 weeks ago in Baltimore, Maryland we had some of our patients along with their physicians coming up to the microphone and practically bragging about the fact that they were in their forties, their fifties and sixties. I have never seen a larger group of old people talking about how happy they are.

All of this comes on the heels of better understanding of the needs for hydration, antibiotic therapy, the use of hydroxyurea and blood transfusions.

Many of you are aware of the early ending of the Stop 2 trial, the stroke prevention trial 2 study last year at this time.

The study looked at whether chidden with sickle cell anemia who receive blood transfusions to reduce stroke risk could be weaned off of transfusion. The study was halted when it was discovered that in patients in whom transfusions were stopped after a minimum of 30 months a significant number of children reverted to the high risk of stroke.

What this study set out to determine is how much clearer there is a link between the discontinuation of transfusion in children and the increase in stroke risk. We now know that transfusion is in fact a critical part of the optimal medical care for young sickle cell patients. Therefore we are encouraging the medical community and patients alike to make transfusion therapy a part of the treatment plan that they discuss with their doctors.

At the same time we know that the major side effect of transfusion is as stated here today iron overload. Our patients with sickle cell disease who have needed iron chelating therapy tell us of the painful and time intensive process that you have heard about today that the current regimen contains.

They have sent me here because they are clamoring for an opportunity to have an oral medication that will simplify their therapy.

The patients with sickle cell disease very frequently also as a result of sick days in schools and out of their employment have to use medical assistance as their only means of health care coverage. Therefore we find that the least able to afford the regular treatment and the purchase of medications that will improve their quality of care sometimes are not going to be able to benefit from the drug. This is a drug that if it is not approved for use in sickle cell disease and the cost to over 70,000 patients who would benefit from it is astronomical, because this medication would never reach -- because of this this medication would never reach the level of benefit it was developed to provide for so many that need it the most.

We need this drug to be approved and on the formulary for medical assistance drug coverage. So, I ask again that you give your support to approving this new product which may help us to achieve a goal that SCDAA has set as part of its mission to promote finding a universal cure for sickle cell disease while improving the quality of life for individuals and families where sickle cell related conditions exist.

Thank you.

DR. ALLEN: Thank you very much.

MR. JEHN: Next will be Dr. Ohene-Frempong to be followed by Dr. Janet Kwiatkowski.

DR. OHENE-FREMPONG: Thank you very much. My name is Kwaku Ohene-Frempong. I am a professor of pediatrics at the University of Pennsylvania and Director of the Sickle Cell Center at the Children's Hospital, Philadelphia. The Sickle Cell Center at Children's Hospital in Philadelphia is one of the 10 NIH designated comprehensive sickle cell centers conducting a wide range of research studies in sickle cell disease.

In my regular work I am also part of a team of clinicians that takes care of hundreds of children with sickle cell disease, severe thalassemia and other childhood blood diseases. I also represent a 33-year-old young man who has sickle cell disease.

Today I am representing the Sickle Cell Disease Association of America. We just heard from the President and I am the Chairman of the National Board of Directors.

There are approximately 70,000 people with sickle cell disease in the United States and roughly 50,000 of them have the most common and most severe form of the disease called sickle cell disease SS. The disease is characterized by chronic hemolytic anemia and episodic attacks of acute anemia with occlusive complications such as pain, stroke, AKHS(?) syndrome and priapism and serious infection. One of the most important therapies of sickle cell disease and perhaps the best non-curative treatment available today is blood transfusion. Blood transfusions are used episodically or on a chronic basis to manage or prevent complications of the disease.

From 1978 to 1998, the National Heart, Lung and Blood Institute carried out a multicenter study on the clinical course of sickle cell disease. That study was called the Cooperative Study of Sickle Cell Disease. Altogether 4000 patients were enrolled in that study. At enrollment into the study 50 percent of all the patients and 60 percent of the 2600 who were SS patients had received at least one blood transfusion in their lives and 5 percent of them were on chronic transfusion therapy.

The use of chronic transfusion therapy in the care of people with sickle cell disease continues to increase. Currently at the Children's Hospital of Philadelphia at our Sickle Cell Center 18 percent of our 460 children with sickle cell disease SS are on chronic long-term transfusion therapy.

Nationally this could translate to as many as 9000 people with sickle cell disease on chronic and perhaps lifelong transfusion therapy.

A major drawback of blood transfusion and the most common reason of chronic blood transfusion, the most common reason why doctors are either reluctant to start or continue chronic transfusion therapy in sickle cell disease is iron overload. Experience in the management of severe thalassemia and studies in transfused sickle cell disease patients clearly showed that iron accumulation rates in both diseases are equivalent to the amount of blood transfused over time and that sickle cell disease patients with iron overload have died under circumstances similar to those of iron overloaded thalassemia patients.

On a positive note physiologic response in terms of getting rid of iron through iron chelation therapy has been similar in both diseases. However, iron chelation therapy using deferoxamine delivered as a slow injection for 8 to 10 hours daily as we have heard has not been very successful in sickle cell disease. The treatment has been difficult for patients particularly adolescents to adhere to. Many children who have been on chronic transfusion therapy to prevent stroke have had their treatment discontinued when they became young adults because of accumulated iron and their inability to carry out iron chelation therapy by injection.

Unfortunately when transfusions are discontinued because of the fear of the consequences of continued iron accumulation many of these young adult patients develop stroke and other complications that are often fatal.

The availability of an effective and safe oral iron chelator will have a major impact on the management of sickle cell disease. It is likely that chronic transfusion therapy would be made available to many more patients and their compliance with iron chelation therapy will improve.

In the end many more people with sickle cell disease will lead healthier and longer lives free of stroke and other major complications of the disease.

I have no direct financial link to the manufacturer but I am part of the team that has participated in the research studies conducted for this drug.

Thank you.

DR. ALLEN: Thank you. Dr. Ohene-Frempong let me ask you one quick question. The majority of the patients studied so far are thalassemia patients. Do you have any statement about additional studies that might be needed in patients with sickle cell disease or follow-up recommendations that you would make?

DR. OHENE-FREMPONG: I think in general we have been encouraged in thalassemia patients and we have no reason to believe that sickle cell disease patients will not gain the same benefit. I think that more studies in sickle cell patients may reveal some special circumstances that may be interested in renal toxicity and things that may be slightly different in sickle cell disease patients but in terms of the ability to eliminate iron I think we are encouraged and we expect to see the same in sickle cell patients. So, continued studies and of course long-term studies to look at the long-term effects would be important but I think that this drug could be very useful in our patients also.

MR. JEHN: Next?

DR. KWIATKOWSKI: Good afternoon. I am Janet Kwiatkowski, an attending physician, hematologist also at the Children's Hospital, Philadelphia and I was asked today to speak on behalf of the Cooley's Anemia Foundation.

I, also, participated in the Novartis studies at the site but have nothing else to disclose. As a physician I have regular contact with two patient populations affected by transfusional iron overload, patients with sickle cell disease and patients with thalassemia.

You have heard the testimony today of many patients who have stated about the extraordinary difficult of complying with prescribed iron chelation therapy via subcutaneous infusion of deferoxamine.

As someone who works with these patient populations on a daily basis I can confirm that what they say is unfortunately true.

At the Children's Hospital of Philadelphia we have dedicated thalassemia and sickle cell treatment centers and within these centers teams of hematologists, cardiologists, endocrinologists, other specialists, nurses and social workers work together to provide specialized care to patients with thalassemia and sickle cell disease.

Even with these resources we continue to have patients who suffer from the effects of iron overload. Despite careful monitoring, clear and precise instructions about treatment and warnings about the dangers that result from excess iron we see patients whose serum ferritin levels and liver iron levels continue to rise, patients who develop diabetes and liver failure and most importantly patients who exhibit dangerous signs of cardiac disease.

When we are confronted with these cases we do everything we can to convince the patients to commit themselves to their treatment with Desferal but even with our encouragement the problems with this drug often lead to continued poor compliance.

Because there is at this time only one approved chelator in the United States our patients find themselves in circumstances that can only result in a disintegration of their health and unfortunately as health care providers we continue to see our patients die from a preventable and treatable problem.

You have before you the first new chelator since deferoxamine to be submitted for approval. The data indicate that it is safe and well tolerated and that it could prove to be an effective option for some patients especially those for whom Desferal treatment is inadequate due to inability to comply with its administration.

I hope that after careful review of the data you will recommend that Exjade be approved for use in the United States. As physicians we need to be able to prescribe therapies that meet the specific needs of our patients. Expanding the slate of chelators will better enable us to meet those needs and to better serve our patients.

Thank you.

DR. ALLEN: Thank you.

MR.JEHN: And finally we have Dr. Vasili Bardoukas who has a slide presentation.

DR. BARDOUKAS: I have been working in this field for over 30 years and first published on an oral chelating drug in 1979. As a specialist in the field of thalassemia I have managed patients both in Australia and Greece for decades and have had experience with both deferoxamine and Desferal brand but have had no direct experience with ICL670 to date though I was involved in the early advisory committee meetings on the compound organized by Novartis and I do have a confidentiality agreement with Novartis, but with no financial reimbursement.

My comments today are based on what I perceive to be the clinical needs of patients who have transfusion iron overload, particularly those with thalassemia.

My understanding of those needs has been fortified by my recent experience as the coordinator of an international randomized study comparing deferaprin the other oral chelator that is available in Europe with deferoxamine and this trial was sponsored by Apotex(?) with whom I do have a financial arrangement.

It compared the relative efficacy of either drug in removing excess cardiac iron by measuring cardiac T2 star as Dr. Porter, Professor Porter showed earlier in transfusion dependent thalassemia patients and this research has allowed me to understand the potential benefits of an oral chelator, of an iron chelator and the importance of its ability to decrease cardiac iron burden and decrease the risk of iron-induced cardiac disease and therefore these experiences have led me to speak to you today in the hope that my views may be considered by the Committee in assessing the potential of this new treatment.

Even at 40 milligrams per kilogram per day less than 50 percent of patients are in negative iron balance as I think that was our trial 105, Elizabeth Brown and others. Yet the maximum tolerable dose seems to be 30 milligrams per kilogram per day and I think this is also borne out in the liver assessments.

The cardiac iron loading results in iron-induced heart disease can now be assessed by MRI technology. Most studies have been reported with Exjade to demonstrate its efficacy at removing cardiac iron.

Professor Porter showed some slides on patients who have had cardiac T2 star and if you looked at that slide carefully one patient with thalassemia went from the loaded range to not being loaded.

The other six showed no change whatsoever and I am concerned that the primary endpoint of liver iron concentration is not adequate to keep the patients who are going to be transferred to this drug safe.

Liver iron concentration is not predictive of cardiac iron risks and most people will agree with this today. When we have looked at magnetic resonance imaging with both the liver and the heart we found that there is no predictive value in the liver iron concentration to tell us what the cardiac risks are.

We have also found that a significant number of patients with low hepatic iron have significant cardiac iron loading.

The reduction in liver iron concentration is therefore inadequate to provide assurance of reduced cardiac morbidity and mortality and this just demonstrates the slide on the left, the picture on the left shows a patient with no iron in the heart but with heavily iron overloaded liver and the other picture shows a patient who has very heavy cardiac iron load with practically nothing in the liver. If this patient on the right were to have reduced chelation therapy that patient would be at serious risk, we believe will be at serious risk of developing cardiac problems.

Now, the other study, the 108 shows that 3 out of 85 patients developed cardiac problems while on Exjade in 1 of the 99 patients with the rare anemia developed cardiac problems. Does this mean lack of efficacy? I am posing this question.

In addition there were 7 out of 297 patients on Exjade in the 107 that showed worsening of ECG abnormalities.

With respect to the safety I won't talk about the renal studies that have been discussed at length. However, I wish to point out that Dr. Heider who is one of the experts in iron chelation biochemistry has pointed out that tridentates have the potential to polymerase. So, this remains a concern. Other concerns have been that the molecular size of this compound once it chelates, two molecules of that chelate with iron it is possible that it might not be able to get out of cells.

The initial animal studies in iron loaded animals demonstrated impaired renal function. Elevations in creatinine were seen in the clinical trials. We do not know what happens in patients who have impaired renal function and patients with diabetes.

So, in summary published and presented data indicate that Exjade eliminates iron from the body but non-inferiority to deferoxamine has not been demonstrated despite potentially better compliance.

Since there is no data showing Exjade can remove cardiac iron I am and many others scientists are concerned that deferoxamine compliant patients may be switched to Exjade for convenience resulting in a decline in survival.

Additional data pertaining to both safety and efficacy are needed if Exjade is to be considered as first line therapy for patients with transfusion iron overload.

Over the past 5 years that deferaprin that somebody asked about earlier has been licensed in Europe and not only for patients who cannot tolerate deferoxamine since May of last year it has been licensed where treatment with Desferal has been regarded as being inadequate. We have seen remarkable changes in survival and reduction in cardiac death in patients who have been on deferaprin and also the study that we did clearly showed that deferaprin was superior in removing iron from the heart as compared to deferoxamine. I believe that yes we need more chelator drugs in our armamentarium but I am concerned. I would not like this one to be first line therapy unless we have further studies.

Thank you.

DR. ALLEN: Thank you very much. That is our last?

MR. JEHN: That is the last speaker that I am aware of.

DR. ALLEN: Is there anyone else who wishes to speak briefly during the open public hearing?

Will you introduce yourself, please? You have 2 minutes.

MS.WALKER: Hello. My name is Brandy and this is my sister,Tiffany.

DR. ALLEN: Can you give your last name, please?

MS. WALKER: My name is Brandy Walker.

MS.REYNOLDS: I am Tiffany Reynolds.

MS. WALKER: We are two of the 70,000 patients with sickle cell anemia. I will just briefly tell my story. When I was a child I was on Desferal and I used to pray every night for an oral chelation therapy because in the morning when I would wake up from my 8 hours or 10 hours or whatever being attached to the pump every night for 6 nights a week I couldn't move. If I had stuck myself in the stomach then I wasn't able to get up out of the bed because I would have to lay flat on the bed on my back and I couldn't go to school.I couldn't get up to go to the bathroom by myself. I couldn't do anything. I couldn't put my clothes on by myself. I needed assistance from my mother to do everything. If I had stuck myself in the legs then again I wouldn't be able to walk the next day and I could not stick myself in my arms because sickle cell patients commonly are smaller than their peers. So, I didn't have enough subcutaneous tissue to stick myself in the arms.

It was a living hell and for a year or so I was compliant with the therapy until our doctors at MCB approached us and asked would we be interested in receiving a portacath placement and we agreed. My sister and I both had portacaths placed in our arms and we have very vivid memories of being strapped down to the table and given topical anesthesia and them cutting our arms open and inserting the portacath in order to do the Desferal treatment and both of ours got infected and had to come out and 3 years after that I was in high school and I had another portacath placed and it was good for 5 years until I got to college, my last semester of college. I incurred a sepsis in my bloodstream as a result of something going wrong with the port or the medication and I almost died. If I didn't get to the emergency room I could have died within that week because I mistook it for flu symptoms and it wasn't and so there was a long drawn-out process and eventually that led to my non-compliance with Desferal and I am not on it today but I am so glad that there is finally an oral chelator therapy available because it answers every prayer that I prayed when I was a child and I need to be chelated because my iron is still high and I know that and I am just thankful for any sort of approval of this drug, whether it is in thalassemia patients, whether or not it gets approved for sickle cell patients, I am thankful that anybody could be able to use an oral chelator but I would just like to encourage you and implore you to approve it for sickle cell patients as well as thalassemia patients because there is a lot of people suffering and it would not be right if there was an end to the suffering and nobody did anything about it.

Thank you.

MS. REYNOLDS: Mine is going to be very brief. I can relate to the lady who said that she went to college and was having a lot of difficulties administering the drug because you know your social and all that kind of, what happens when you go to school.

I had a similar situation where mine got infected. I had it placed in my left arm. It got infected because the medication that I had was not, it was faulty.

I took it anyway. I got very sick, blood clots around my heart and had to have it taken out. I had another one placed like my sister said, had another one placed. Mine never worked. So, I was not able to take it that way. I had to have it removed. So, I think both of us have that drive or you know, feeling of this is an excellent opportunity for sickle cell patients. It is an excellent opportunity for thalassemia patients because I know students who have thalassemia and have to have treatment for it as well and we are just asking that the approval be granted so that there wouldn't be people like us who have to have this kind of treatment growing up for whatever your age is. I am 32. She is 26, blood transfusions all of our lives. Mine was from since the age of 6. Brandy's started at the age of 3. So, we have had to deal with this for our entire lives and we are just thankful that there is a Committee who would approve it and hopefully it will be approved.

DR. ALLEN: Thank you very much.

This ends the open public hearing. I am sorry we have got one more speaker. Okay, we are already an hour behind. Please, you have got 2 minutes but please restrict yourself to that time.

Introduce yourself?

MR. GRADY: Right, we will make it brief. Bob Grady from Cornell. I think that Exjade should probably be approved and made available to patients as an option but the key word here is that it is an option. It should not be a first line drug. It is not a replacement for Desferal .Patients taking Desferal, all patients can be put into net negative iron balance with Desferal. In the case of the results that have been presented thus far with Exjade at 20 to 30 milligrams per kilogram 10 to 40 percent of the patients cannot be brought in to net negative balance even if the drug is used 7 days a week. We can't ignore this fact and we can't ignore these patients.

Thank you.

DR. ALLEN: Thank you very much. I think this has been very informative and instructive. I would like to thank particularly all of the patients and their families who came to speak during the open public hearing sa well as the physicians and others who have taken time away from their clinics and their patients to present data and discuss the importance of this.

We are going to break for lunch at this point. There is a cafe across the street, the Gallery Cafe. I am not certain how rapid the service is. There is a table reserved for the Committee. I would like to have us back here and at 25 after one and ready to reconvene as quickly as possible at that time.

Thank you.

(Thereupon at 12:38 p.m., a recess was taken until 1:31 p.m., the same day.)

AFTERNOON SESSION 1:31 PM

DR. ALLEN; Okay, people can take their seats, please, and we will move first of all into our general questioning session and what I would like to do is to ask that we not get into general discussion but that we focus in initially on questions and answers for any of the presentations. Then we will move into general discussion and consideration of the specific questions that are before us.

So, the session is open for questions.

DR. MALDONADO: I have a question for Dr. Shashaty on the renal findings in proteinuria. I am not a nephrologist and not a pediatric nephrologist but in the pediatric literature about 2 or 3 years ago and I wish I could quote the article but I believe it is in the Journal of Pediatrics there was a publication about proteinuria, the assay itself because a lot of these assays according to that information were developed and FDA approved for diagnosis and follow-up of patients with nephrotic syndrome.

So, the message in the article was that these assays are very good with good linearity and predictability between one, two, three and four grams of protein but when you go to milligrams there is a lot of variability. The linearity is lost and then they assays become less reliable. When I saw the data, I believe and I don't have a good sense of how the spread of the data but we were talking about milligrams. So, I don't know how comfortable you are that the assay is really a good tool to follow the patients and I am not talking about spot urine or necessarily 24-hour urine because the assay itself has problems with reliability, so I don't know exactly how to interpret it.

DR. SHASHATY: If I can make a quick response and then if the sponsor has anything different to say I would ask them to contribute as well.

I think the problem was as I went through the review the sponsor was originally going to do 24-hour urines but it became impossible to do, to have all these urines coming in and so on. So, what the decision was made then too do sort of a relative study you would assay the urine for, spot urine for protein and assay it for creatinine and then develop a milligram per milligram ratio and there was then a dividing, there were several categories. There were those who had .2 to .4, .4 to .6 and now this is milligrams of protein per milligram of creatinine and the highest one was the category of greater than .6 milligram of protein to milligram of creatinine.

So, here was the way I sort of calculated it. If you figure that you put out about, and correct me if I am wrong, if you put out about 1000 milligrams of creatinine, you know it is 1440 minutes in a day. You have 1 milligram creatinine. You have 100 cc's. It comes out to be about that much.

Now, if you take .6 and multiply it by 1440 you come up with about 850 milligrams 24 hours. To me 850 milligrams in 24 hours is significant proteinuria and that was where the distribution was different between Exjade which I think was 18 percent and I think DFO was like 7 percent or something like that. I am sure the sponsor has the exact figures.

DR. MALDONADO: Actually I am not disputing the methodology you used. I am actually just questioning the actual assay itself meaning the lab tests because if you have spot urine and there is error there and then you multiply that error then you may end up with, I mean there is a lot of noise and variability in these assays when you measure milligrams. There is a lot less variability when you measure grams and maybe a pediatric nephrologist is best because again I just read it and I became very wary about making conclusions on milligrams of urine measuring those effects.

DR. ALLEN: Could we ask Dr. Portman to comment?

DR. PORTMAN: Yes, I am happy to respond to that. It is less than perfect but the technique is the standard in both clinical and research venues and it is very accurate, in fact was developed in pediatrics and adopted by adults later because it is so difficult to do 24-hour urines in children. The difference really lies in the interpretation where 850 milligrams excretion in a day for an adult may be considered significant proteinuria. One thousand milligrams per day in a kid is considered nephrotic range.

So, if you have actually 40 milligrams per meter squared per hour it is also considered nephrotic range. So, that amount of proteinuria is very, very significant.

DR. MALDONADO: Can I ask a follow-up question on that? What is the error of margin because I mean if you say that it is accurate still I mean every time that there is a small error in those spot urines that error gets multiplied too and then when you talk about a whole day you are multiplying there the error and I just don't have a sense of what that error is.

DR. PORTMAN: I don't have the variability in front of me either but it is something where you never rely on a single measurement. We generally look at only first morning urines in pediatrics because the prevalence of postural proteinuria is so high. So, we don't do random urines and we eliminate some of the variability because of that and then we do it on a repeated basis in order to get a better handle on the reliability or the repeatability of the measurement. So, you are correct but it is the best thing that we currently have available to us.

DR. BRITTENHAM: Would you consider the protein creatinine ratio the best available screening technique for renal damage in these circumstances?

DR. PORTMAN: I do from a clinical standpoint. Obviously from a research standpoint you know to do timed urines or 24-hour urines you know would probably be more accurate but trying to simulate what we would do on a clinical basis I think it is the best that we have and the National Kidney Foundation has a statement in their guidelines that this is what should be used for all ages, children as well as adults. So, I do think that it is the most accurate that we currently have available.

DR. ALLEN: Thank you. Dr. Portman, as we move through the discussion I will try to remember periodically to invite you specifically into the conversation but I think you can try to talk over us if you need to or want to weigh in at any point with a question or a comment.

DR. PORTMAN: Okay, I would appreciate that. I do have at some point some questions to make. Since I have your attention are we allowed to talk to the sponsor still or not?

DR. ALLEN: Yes, my understanding is that we can ask questions of the sponsor at any time. This is an open session throughout and we can ask for clarification or explication of their material or if they have got additional data that we haven't seen.

DR. PORTMAN: Then let me ask two questions, please? The first is related to blood pressure and the data that they have on blood pressure particularly that which they obtained in children. Do they use the most recent guidelines from the working group to determine hypertension and did they find any hypertension in either children or adults and the second question is what is their plan since they didn't study anyone who had chronic kidney disease to address this for the 20 million people in this country who have chronic kidney disease and might require this medication.

DR. MARKS: Peter Marks, Novartis. So, your first question was about the blood pressure in children which we recorded in our studies but we do not have a data analysis to give to you on blood pressure in the children.

Regarding your question on what we will do for patients with abnormal renal function, at this time we felt that a stepwise approach was appropriate. We would like to first complete our renal mechanistic study to understand what is going on here and then once we understand based on the fact that we do have some hypotheses about what is going on once we confirm our refute those hypotheses then we would proceed with studying ICL670 in a renal insufficiency population.

DR. PORTMAN: I assume you are going to go back and look at those blood pressures. That is very important particularly with the cardiac problems that these kids run into.

DR. MARKS: Dr.Ford, would you like to comment?

DR. FORD: John Ford, Novartis. We have analyzed the blood pressures and there were no significant findings but the blood pressures were measured just in a totally standard way in the clinics. There was no special procedure performed, no significant findings in the analyses of blood pressure.

DR. PORTMAN: If they weren't analyzed by the appropriate standards then you wouldn't necessarily find anything. That is my point. A 6-year-old child who has a blood pressure of 113/80 may look perfectly benign to you but it may be hypertensive based on the guidelines. So, that needs to be analyzed. That is the point.

DR. FORD: Right. We obviously need to go back and do that.

DR. WEISS: Dr. Allen, may I just ask in terms of the hypertension is it not correct that many of these children were studied in pediatric centers and I would hope that those physicians taking care of the children would be very familiar with blood pressure values corrected towards age but I take Dr. Portman's question. It is a very important question and I think it is important to make sure that there weren't any potential signals in those studies but if that is correct that gives me at least some level of comfort that people were attuned to hypertension in children.

DR. MARKS: Correct. These studies were conducted in, many of these were large thalassemia centers that were used to taking care of children with thalassemia and by the way actually patients with thalassemia have a renal lesion. I am not an expert in that renal lesion. We did have a pediatric nephrologist who was more expert about the renal lesion in thalassemia as a consultant.

DR. ALLEN: Yes,Dr. Klein?

DR. KLEIN: Dr.Shashaty I am not sure that I heard the agency position or at least their view on the pediatric group here which I am very concerned about because this is a key target population and the iron studies were based on SQUID and I guess we will hear a lot more about that later on but the agency apparently feels this wasn't validated. So, I would like to know in what way it was validated and how we can rely upon the data if that is all we had for the pediatric population and sort of the follow up to that is if we then really under dosed the pediatric population, a key population what can we say about the toxicity of the drug in that population if it was so far lower than what might be used in practice?

DR. SHASHATY: I think that that was one of the concerns we expressed, that is if the studies enrolled patients on 5 to 30 milligrams per kilogram per day which was what they did but then patients if the drug is approved are treated basically only with 20 or 30 because that is where the effectiveness appears to lie in the 20 to 30, then I would expect that there is going to be a greater frequency of some of the dose-related and I think that renal toxicity is probably dose related. Hepatic I am not sure. Gastrointestinal I think is probably dose related but is not that important. So, I think that what is going to happen is if we migrate from 5 to 30 to only 20 to 30 we are going to see more adverse events and we are going to see new, we may see new adverse events.

So, that is the answer. That would be my response to your second question. My response to your first question is that although proportionally there were more children studied by SQUID than by biopsy there were a number of children who were studied by biopsy and if I am not mistaken the majority of children had biopsies. Is that correct?

DR. MARKS: Yes, that is correct and if you would like I have the numbers for you. So, you can see them.

DR. KLEIN: Yes, I think it would be helpful for us to see the numbers.

DR. MARKS: So, do you see that breakdown by pediatric age category? So you see across the age range. It is a busy slide but if you look at the third row from the bottom you see the percentage of liver iron concentration by biopsy and you see it is roughly somewhere between 50 and 60 percent in the pediatric population.

DR. ALLEN: Other questions for the FDA in particular or the sponsor?

All right, why don't we move into our general discussion. We have got if we stay on time we have got an hour and one-half for consideration of six major questions, some with subsections. On all of them I think there is an issue that can be formally voted on. It would be helpful for the FDA to have that formal vote recorded and in addition in virtually all of the questions there also is room for discussion and comments if you will and sometimes the comments will be the very significant part of the answer that is given.

I think it might be helpful, however, to start out with a, before we get into the specific questions, into a general discussion of the data that we have heard, the c

concerns that have been expressed. Obviously we need to take into account considerations such as the fact that this is once-a-day oral dosing versus the hours-long injection therapy every night that otherwise is required. So, those considerations certainly are relevant.

We have got issues of dosing ranges, toxicity, pediatric, very important pediatric population to be served. So, it may be that people would want to just have some general comments and discussion before we get into the specific questions.

Dr. Brittenham, would you like to begin?

DR. BRITTENHAM: Yes. I would like to make a general recommendation for the Committee's consideration which is that the Exjade the ICL670 should be approved but that we should require substantial reanalysis of the data that has been accumulated both with respect to efficacy and safety to give us a clearer idea of both the efficacy and safety and their balance and that the recommendations for both dosing and the suggested labeling need considerable thought and revision as well.

Let me say first of all with regard to the efficacy as you heard earlier the data that was collected with the SQUID apparatus varied from center to center and overall gave values that were only about 1/2 of what was there. All the data that you have seen though has just taken these and used them as reported without making any adjustments.

My own sense is that all the SQUID data is unusable, that we don't really have a way of post hoc making corrections so that we can make a judgment from it. I don't think that we really understand the reason for the intrinsic variability for such a glaring miscalibration and so I don't see how we can really use that.

The second point with respect to efficacy is that in the midst of the trial the sponsor asked for a change in what had been the agreed upon biopsy size for analysis.

Previously it was a milligram and it was changed in 2004 to be half a milligram. The reason that is significant is that at least in the study that was done in patients with thalassemia major who had been cured by liver transplantation that if you included these lower specimens that their relationship to the body iron is only about as good as the serum ferritin. It is not really at the same level as if you have an adequate specimen from the liver. So, I suspect that the reason that this was done is there was difficulty in getting sufficient, in this study in getting sufficient sample and so they didn't want to repeat biopsies or go back and so I think both the SQUID data and the other data are fine for consideration for safety for looking at that experience but I don't really see how it can be used to give the most accurate view of efficacy and so it would be, we don't have any information about -- I haven't seen any analysis of the efficacy of the treatments that is based solely on the liver biopsies with sufficient sample.

Further when this passes into clinical use what is going to be used is neither SQUID nor liver biopsy. There is hope that in the future there will be MRI methods but this is still, there isn't a standardized way of going about this. So, in practice it is going to be serum ferritin and I think it would be very helpful to see comparisons of the serum ferritins with the absolute liver biopsies, with the liver biopsies of adequate size so we can make some judgment of what sort of result that is going to have in how the drug is prescribed.

Typically most of the things that affect serum ferritin tend to raise it rather than lower it meaning that patients will get more drug rather than what they would do from their true liver iron concentration and that may raise the risk of liver toxicity. So, that is why I think it is an important point.

For safety I think you have to view all the percentages that you have heard because of this background with some concern because a substantial proportion of the patients who have been examined have been under dosed. It is much less. It is just as Dr. Shashaty had pointed out. The dose that they have been given in this trial is much less than they may receive in clinical practice and so I think that there should be consideration given to requiring spot urine samples at each time that a patient undergoes transfusion as a way of screening, as giving you sort of an early warning about if there is a problem.

Remember these are all patients who have been looked after with exquisite laboratory monitoring and cared for by the best clinicians in thalassemia in the world and we really have to take consideration about what happens when this moves to another arena and I think it is really the responsibility to try to protect the patients against the risk of renal damage.

So, I just wanted to present this view as a background for these discussions because I think it may not have been obvious from the presentation that there are these concerns.

DR. ALLEN: Thank you.

Other general comments or concerns? Dr. Quirolo, anything else you want to add about SQUID or --

DR. QUIROLO: I think everybody has addressed it. They have the same concerns that I have. Another concern is that their exclusionary criteria to put people into the study eliminated a lot of people who potentially won't be eliminated if this drug is made available. So, you have to again wonder if the incidence of side effects is going to be multiplied when you have people who aren't screened like they did for the study and you have these other problems that Dr. Brittenham addressed.

DR. ALLEN: Dr.Portman, do you have any general comments you want to make with regard to well, anything but certainly renal complications?

DR. PORTMAN: I do and I can make them now. This is obviously potentially a very important drug and I certainly feel for the patients who have to take deferoxamine. I just think we need to be very careful not to convert them from a deferoxamine needle to a dialysis needle.

Now, there is certainly no evidence yet that that is at a possibility but I think the bottom line is that we just don't know and I am very concerned about some of the early signs that we have seen in just a year's worth of follow-up in most of these patients.

I think it is reassuring that there is no progressive renal failure and that there seems to be a response to a dose adjustment but again the follow up is really too short and it has not been well studied. We need the subsequent study that is being planned or being done to get some better answers on this.

I think that I, too, am concerned that the dosages that are going to be used are going to be on the higher end which is the same end that we see the greater amount of creatinine elevations and perhaps protein creatinine elevations as well. I mean to have 500 to 1000 milligrams of protein excretion a day is very concerning. It means that there are some significant capillary problems going on. There is significant renal damage going on and we know from the earlier studies that there is tubular, potential tubular toxicity from other animal studies although we don't have biopsies in humans to back that up and some of the changes that we see with low molecular proteins are also concerning.

One can also look at the creatinine clearance changes in one of these slides that was faxed to me and you see some small declines with deferoxamine but even more significant decreases with the current study drug and that is only after 1 year.

Now, granted the changes happened early and may be hemodynamically mediated and seem to stabilize but it is only 1 year of follow-up and I think that if this drug gets approved and I certainly have a bent toward that approval it is going to need extremely close follow-up and for each and every patient with a very specific protocol for following renal damage with the proteinuria, blood pressure and GFR and particularly for those patients who are going to have chronic kidney disease as mentioned earlier I think the drug is not going to be withheld from them as it was you know in the study.

So, there would need to be some sort of mandatory studies and maybe the one that is currently being done will fill the bill but to get a better handle on this renal toxicity which is quite concerning to me.

DR. ALLEN: Thank you. Those are very helpful comments.

Yes?

DR. QUIROLO: Just to go a little further with the renal toxicity in sickle cell disease about 20 percent of adults have renal disease. It is a common finding in sickle cell disease and about 6 percent of pediatric patients with sickle cell disease develop renal disease.

So, in that population although you know the drug is necessary because if you don't chelate you have other problems but I think it is compounded in sickle cell disease beyond thalassemia where there isn't a significant renal component to that disease.

DR. ALLEN: Comment, please?

DR. PAZDUR: The issue that I have is potential efficacy on cardiac endpoints, okay? And obviously if this drug becomes available and I understand truly the need for this drug as far as ease of administration, obviously clinicians are going to have to make a decision about deferoxamine versus this drug and obviously there are many selling points to this drug, ease of use and I think that is extremely important but we really have not had a very careful examination of the efficacy of the effect on, efficacy on cardiac iron overload.

I would like some discussion on that because if the drug becomes available and people are using it should we pay since we don't have data on this some specific attention to cardiac monitoring? For example if somebody to get a T2 star MRI baseline and then periodically to monitor people because truly if they were not receiving benefit from this one drug somebody would have to make a conscious decision whether they should be switched over to deferoxamine. Here again one would hope that that is not the situation but one would need some information on that.

This could also be a Phase IV commitment to study this population. I know the company is planning on doing that but I would like some discussion on that what people's thoughts are.

DR. ALLEN: I think that is a very important question particularly given the mortality is largely cardiac related and the fact that it was not well addressed in the --

DR. PAZDUR: And here again we don't know the relationship definitively on the LIC to cardiac iron content and changes, how that would be.

DR. BRITTENHAM: I would first like to comment that I think that at least for those patients who are able to use the ICL670 successfully and I do have to agree, we don't really know what proportion of the whole population that is but it certainly is a proportion and it is likely to be a majority or certainly a substantial proportion.

The drug is going to make the disease of transfusional overload very different from what we have seen with deferoxamine because in patients who begin on this drug they can be controlled and never develop iron overload.

In patients who are now well controlled with deferoxamine the hope is that they can be changed to this drug and continue to be, and then there is the category of patients who haven't been well controlled with deferoxamine and whether this will provide the same protection.

The information we do have is that again in patients who have been transplanted, who have been cured of their thalassemia and then had their iron removed the iron comes out of the heart. So, eventually although it has very different kinetics the iron is removed from the heart with good control of body iron.

It is quite true as Dr. Rudigus pointed out that you can see patients who have low liver irons and iron in their heart but to the best of my knowledge this is only sen in patients who have gone through a period where they have been out of control; there has been lots of non-transferrin bound to iron that has been deposited in the heart and then they have undergone a period of intense chelation. The iron comes out of the liver much more readily than it does out of the heart and that is how that situation is produced but the ICL670 at least has the promise of providing help for those patients. There may be well some who need to continue to use the deferoxamine to get rid of this backlog that is there from not having been able to use the drug earlier, but that it really has the potential to convert to change the disease into something where it is prevented and the iron overload is really prevented and well controlled, and it is going to enormously expand the indications for chelation because of its much greater ease of use and I do think it should be approved for transfusional iron overload as such even though as Dr. Shashaty has pointed out it hasn't really been studied in all the groups to which it could be applied but I think there is enough understanding of how chelators work that it can be effectively used.

DR.MILLS: I would like also for the Committee to expand somewhat in terms of how this is going to be monitored in the clinical area in terms of the comment about using serum ferritin, the relationship to the amount of iron within the liver and how fast that is going to be relieved in terms of this load as well as the potential for the iron within the cardiac area to relate those areas together because from the standpoint here we heard earlier comments in terms of the lack of applicability of liver biopsy for general routine use.

DR. BRITTENHAM: I am not sure I have a real list of the questions. If you could --

DR. MILLS: Let me elaborate? In other words if you are going to anticipate that these patients will be followed with serum ferritin clinically rather than liver biopsy and how should that be studied potentially in future trials to be able to relate that marker, that surrogate to the potential clearance from the liver as well as from the cardiac muscle.

DR. BRITTENHAM: As you have heard and as you have seen in the data in a general way the ferritin does provide an indication of the total iron overload. It is just that there is so much dispersion around that relationship it makes it hard to rely on this measure in individual patients, but one must recognize it is how Desferal has been managed for most of the time that we have used it, so at least we are no worse off. I think the real answer for this will come from MRI studies because MRI lets us look not only in the liver but also in the heart, in the anterior pituitary and all the endocrine organs that are targets and so, I think that we should give consideration to encouraging the sponsor to undertake such studies to be sure that these expectations are borne out.

DR. MILLS: And therefore you would anticipate using the MRI study to monitor the cardiac function and level of iron to be able to relate that to the serum ferritin to determine how much of a leaving or lagging indicator the serum ferritin may be?

DR. BRITTENHAM: The serum ferritin is never going to be very -- I think we know the serum ferritin is never going to be very good. The trouble about the MRI studies of the heart is that at present although the T2 star has gotten I think most of the attention there isn't a general agreement about the best way. There is not certainty about how to interpret it and a very important point is that even with 24-hour-a-day intravenous continuous iron therapy the T2 star changes very little in the course of a year and the accuracy with which it can be measured in different institutions and so on is something which is being examined but I couldn't in my hospital today send somebody down and say, "Tell me the T2 star." The worst thing would be if they did.

(Laughter.)

DR. KLEIN: From a practical standpoint ferritin is the best you have and it is going to be used and as far as the other non-invasive studies of heart and liver which I think are critically important I think the data need to be developed. I hope the sponsor will help do that but I don't think we have the data out there to really recommend that those be used in terms of monitoring these patients.

DR. KATO: I think we could learn more from the data that we do have though to do what I think Dr. Brittenham was proposing that we reanalyze or the sponsor could reanalyze all of the data as they have developed with landmarks of liver iron content to instead different gradations, different ranges of serum ferritin in order to understand if there are certain serum ferritin numbers that indicate risk for certain dose ranges.

DR. QUIROLO: Clinically I take care of a lot of patients who are iron overloaded and I think the reason we do or the reason why I do a liver biopsy is the ferritin is so high I have no ideas what the liver iron is. So, if somebody's ferritin is 900 I am not likely to do a liver biopsy for iron reasons. If it is 4000 or 5000 I have no idea what their total body iron really is. So, if you go into the future a few years and this is an effective drug with few side effects you may not really need to be doing as many liver biopsies as you do now.

In the short term to determine the efficacy you may want to do that but I don't think in a clinical world if this works as well as it might you would be needing to do liver biopsies hopefully.

DR. WEISS: Could I just ask you for a quick clarification. If you have a patients that has a serum ferritin of four or five thousand even though you don't have a good sense in a quantitative way what their liver iron concentration is can you be pretty sure that they still have iron in their liver that is maybe above the magic seven or I mean are you pretty certain that there is a potential problem that needs chelation or is that not -- in somebody that you know is getting chronic transfusion therapy, not somebody who has got inflammatory disease for instance?

DR. BRITTENHAM: I would like to say that is why

the data that the sponsors have accumulated would be very helpful. It is not necessarily the case that you see the same relations with deferoxamine as you do with Exjade. I don't know of data one way or the other, but it would be helpful to see. They have an enormous amount of information about this. It needs I think to be compared against adequate liver samples is the major consideration.

DR. GHANY: I would just like to chime in here on I think the potential role of liver biopsy. I agree with everything that has been said. I don't think in the practical sense that liver biopsy will be a good way to assess response or to manage patients. As you saw here there was one patient who died from a liver biopsy. One death is already too much but having said that I still think that liver biopsy does play a useful role. What we haven't talked about here is that many of these patients also succumb to liver disease and biopsy can help you in assessing the degree of liver damage which is related in most of these patients to the degree of iron overload and it is helpful in managing patients who are sometimes often on other polypharmacy how aggressive you can be. So, I think a baseline liver biopsy in patients who have elevated ferritin levels is probably warranted.

DR. ALLEN: I think what I hear everybody saying, however, in addition to the need to reanalyze data to answer specific questions to the extent that the data are there there is very much a need to continue looking for good non-invasive ways in which to measure and follow patients and to the extent that we don't have technology that is applicable today or we don't know how to interpret it or you can stick a patient in front of a machine and get certain measurements but if you don't know how to interpret it not much help to you, but I would anticipate that in the next few years as these questions are specifically answered and technology continues to develop one would hope that we would get the ability to track much better and much more specifically the measurements that we need.

DR. BRITTENHAM: I think there has been enormous progress in this in the past years and that that is a reasonable expectation.

DR. ALLEN: Other general comments? If not I would like to move on.

Yes, Dr. Kato?

DR. KATO: I would just like to clarify one point that was raised by Dr. Brittenham about the switch by the sponsor from a liver biopsy size of 1 milligram to 1/2 milligram. In my reading of the chronology of the regulatory activity that was requested but not granted. I just wonder if either Dr. Shashaty or the sponsor could clarify that for us.

DR. SHASHATY: In these kinds of things the FDA doesn't grant. It notes and makes comments and then the sponsor has the decision to make and after reviewing our comments.

Now, when the sponsor says that they are going to do it we make a statement that says that based on our knowledge of the literature, actually this is from Dr. Brittenham anyway, you really require a specimen that is larger than you know 1/2 milligram. So, we responded to the sponsor in that manner and that was what we could do and they can use that that way. I would like to ask Peter, however, to indicate; he has got the information, the percentage of patients who had greater than one and less than one, please, so we can sort of address the issue. It may be partially a non-issue.

DR. MARKS: Actually I thank you, Peter Marks, Novartis. We actually took seriously the comment that the reason why we couldn't go back and get -- the problem is for getting a dry weight biopsy. It is very difficult to know in advance whether you are getting a large enough sample and as you have heard there are serious complications of liver biopsy. So, when patients had a specimen and then the dry weight turned out to be less than one, we felt that it was not ethical to delete those samples unless they were less than the critical value in which case we did not include them in the analysis.

We do have the data on the number of biopsies and we had about two-thirds of the biopsy weights were above 1 gram. If you look at the baseline samples it is about 80 percent and when you look at both samples it is two-thirds.

So, I have the data by success rate that I can show you per Dr. Brittenham if you would like.

Okay, if you can bring up that slide for me?

So, you can see here that the success rate, these are in all biopsy samples. On the top row are samples greater than 1 gram, the second row down and samples less than 1 gram on the bottom row. You see the N's in the various groups. Again, I remind you these are where we had both samples that were larger than this size and you see the success rate if anything the samples greater than 1 gram you see the ICL success rate a bit higher. I wouldn't make any broad comments there except to say that it is similar.

DR. ALLEN: Other general comments? Yes, Ms. Winner?

MS. WINNER: Based on some of the safety studies I only read one thing in reference to drug interaction and it was with digoxin, I believe and I am just curious if any of the patients that were on the trial if they were allowed to continue their medications and if there were anything checked out with that safety-wise.

DR. ALLEN: Does Novartis want to respond to that/

DR. MARKS; Peter Marks, Novartis. Yes, all patients that were on digoxin and other cardiac medications, beta blockers were allowed to continue on these drugs.

DR. SHASHATY: I would like to introduce the biopharm person and ask if there are any studies which are going to be recommended from a biopharm point of view.

DR. AL-FAYOUMI: Suliman Al-Fayoumi, pharmacology reviewer for Exjade. The sponsor only conducted one drug interaction study which was with digoxin and did not see any effect on the metabolism with digoxin there. In vitro studies did show that there was some potential for Exjade or ICL670 to inhibit the metabolism of several Cytochrome P450 enzymes and we as a result we are considering right now whether to request additional studies from the sponsor looking into the potential inhibition of cytochrome P450 enzymes by ICL670.

DR. ALLEN: Other questions?

Yes?

DR. ROBIE-SUH: I just wanted to ask a general sort of question about over chelation. I know especially we are thinking about looking for the higher doses and maybe folks with lower iron burdens might be put on the drug or in terms of thinking about maintenance possibilities and monitoring. Do you have any feeling of use of ferritins or anything to avoid over chelation? I think you mentioned something about maybe some of the renal findings you thought were related to over chelation. If you could make some comment?

DR. MARKS: Peter Marks, Novartis. We have looked. Obviously what I will presented to you here is our exploratory analyses that were conducted to try to understand the increase in serum creatinine. Just to look at the slide that I showed you before when you look here at patients that were treated with increasing doses of ICL670 there seemed to be an increasing number of patients who had two consecutive increases in serum creatinine. A similar scenario was seen with deferoxamine, perhaps not to the same dose dependence but you also see here as was mentioned earlier many of them received similar doses. There also were these 33 percent increases on two consecutive occasions for deferoxamine and in an analysis that we conducted it appeared that patients who had greater decreases in serum ferritin over a given interval tended to have these 33 percent increases and reduction of a dose did not affect the overall outcome.

DR. ROBIE-SUH: And was that related to the rate of reduction and not to absolute levels of ferritin?

DR. MARKS: I think all we can say is that we know what happened over the interval of the 1-year period. I don't know whether you can say it is whether it is due to the amount that it was reduced over that 1-year period or whether it was due to the rate of its reduction.

DR. GHANY: I wonder if I could just continue on this point because do you think it is related to the quantity of iron and do you think that we could expect to see more toxicity with long-term therapy because in the animal models you said that the animals that were most sensitive to the renal tubular effect of Exjade were the ones who had normal iron levels and you could correct it by replacing iron?

DR. MARKS: I may ask Dr. Hirschberg to come up here and just discuss our alternative hypotheses in a moment but I would just like to start by trying to answer your question and say that yes, I guess in terms of the hypothesis we have, we have two different ones which are not mutually exclusive, one that it could be a glomerular effect that happens early on with a drug or one is that it could be an over chelation effect.

That is why we are recommending routine monthly monitoring of serum creatinine because it would appear that when the creatinine starts to rise either with deferoxamine or with ICL670 it is an indication that chelation may be and this is again exploratory, may be proceeding at too rapid a rate.

If you look at the patients in our studies who had dose reductions for serum creatinine increases and here you see the ones who had dose reductions no increases in creatinine and then the ones who had 33 percent increases divided by those who had no dose reduction or dose reduction, you see those who had the dose reductions because they had the sustained increases of creatinine had the greatest decrease in liver iron concentration despite the fact that they were dose reduced and those who had no dose reduction, that is because they either were pediatric patients or because they had spontaneously returned to normal during that time or towards baseline had a reduction of minus 4.8.

So, this would tend to support this hypothesis and I think Dr. Hirschberg can comment a little bit more on that.

DR. HIRSCHBERG: Yes, thank you, Dr. Marks. We had in our renal advisory board where we extensively discussed this issue two hypotheses. Hypothesis No. 1 is the sort descriptive based on our creatinine and creatinine clearance findings, highly indicative of a hemodynamic function response. There is a small drop in glomerular filtration rate leading to a small rise in serum creatinine which is sustained. This is essentially identical as you will observe with use of ACE inhibitors or angiotensin receptor blocking drugs. It is identical to that.

I am not saying that we believe ICL or chelation therapy blocks these enzymes that are involved in the RAS system meaning renin angiotensin-1, angiotensin-2 because those are really not enzymes that appear to be regulated by metal-dependent mechanisms.

However, there are other enzymes that participate heavily in the regulation of glomerular route of filtration on a physiologic level. For example, nitric oxide, a certain isoform of an isoenzyme is produced at the glomerular vascular pole and regulates in situ at this site glomerular hemodynamics.

Nitric oxide is an enzyme complex that has five prosthetic groups. Two of them methyl dependent and there are a variety of electron transporting enzymes. Nitric oxide synthase is an electron transporting enzyme. All of these electron transporting enzymes are metal related enzymes and I would just like to summarize that we believe the potential that both hypotheses, chelation of a metal away from an enzyme that has some functional contribution to glomerular hemodynamics and this observed change in glomerular hemodynamics with chelation therapy may just be the functional conclusion of the two. We don't have proof. It is a mechanism to be explored.

If the Chairman allows me, I would like to make one additional comment to an earlier discussion within the Committee brought up by Dr. Portman on proteinuria.

DR. ALLEN: Very briefly, please?

DR.HIRSCHBERG: This comment I completely agree with Dr. Portman that we have to look for proteinuria at a nephrotic range level, but I can assure you that our observations here do not include patients with nephrotic syndrome even in the pediatric definition because Dr. Portman defined for the pediatric patients a nephrotic syndrome as at about 1 gram of proteinuria but I would have to add that this needs to be glomerular hematuria and we did not observe glomerular hematuria at anything close of this level.

Much of this proteinuria is proteinuria of small molecular proteins that may reflect a functional reduction in tubular absorption of those proteins.

Thank you, Mr. Chairman.

DR. ALLEN: Thank you. Dr. Portman do you have any additional comments, general comments at this point?

DR. PORTMAN: I do. I need to respond to that because quite honestly it didn't quite ring true to me. First off we are having the argument which I don't think is something that you can say. While the curve looks similar I agree that you have with an ACE inhibitor you have a mild decrease in your GFR acutely and then it remains stable but to say that there is any kind of association is a real stretch.

Secondly, if you want to say that the effect of this drug is related to glomerular hemodynamics then you go into talking about proteinuria and say that the proteinuria is not glomerular but it is tubular; I mean you have more going on here in the kidney than we know. This is my point from my earlier comments. We just don't know what is going on and until we do I think I have grave concerns about what is actually going on with this drug and it needs to be very carefully monitored and studied.

DR. ALLEN: Thank you.

Yes, Ms. Baker?

MS. BAKER: Thank you. The need for a nationwide prospective surveillance of the complications of diseases and new therapies is something that many of us have stated and as we think about that I would hope that we might look to the model that has been in place for other inherited bleeding disorders such as hemophilia and von Willebrand's disease.

There is a nationwide network of 130 plus federally supported hemophilia treatment centers supported partially by the CDC and the Maternal and Child Health Bureau and since 1998, the CDC funding has been for the prevention of, for a system to monitor the complications. It is mandatory. We have 83 percent of the hemophilia population at these centers who are enrolled over 12,000 people since 1998, over 45,000 visits and both CDC and the MCHB have pilot programs now, the CDC in thalassemia and the Maternal and Child Health Bureau also in sickle cell that are looking to somewhat replicate the surveillance programs. So, it is not just a responsibility, I think of the sponsors but also on the government agencies and to expand this existing model we are now looking to increased involvement of the sponsors. So, I think a partnership is something that could and should be explored.

DR. ALLEN: Thank you. I think those are very pertinent comments.

Dr.Brittenham?

DR. BRITTENHAM: Yes, the only thing I would like to add about the chelators and the renal damage is it is an effect that is seen in other chelators, for the desfriviasins(?) for example it is the principal toxicity and it is tubular damage and I think that we don't really understand just what it is but it can be an effect of the chelator on the tubules themselves.

DR. ALLEN: At this point I would like to move on to consideration of the specific questions. Do you know if they are on the slides?

We will have the specific questions brought up one by one on the slides. I would like to read the preamble, however, which I suspect is not on the -- okay, let me just quickly read the preamble to question No. 1.

The standard measure of body iron stores is by assay of liver iron content, LIC. Reduction in LIC was used as the efficacy end point in study 0107 and study 0108.

However, LIC may not represent an accurate state of the total body iron burden or the distribution of iron in other target organs.

So, the question now for us is does the Committee believe that a reduction in liver iron content is an acceptable efficacy endpoint for approval of ICL670? If not, what efficacy endpoint would the Committee recommend?

We are open for discussion of this specific question.

DR. QUIROLO: I think it is a reasonable endpoint and the reason why I say that is the other endpoints that we would talk about. heart iron or whatever, those modalities aren't available and when the study was designed this was the gold standard as they say but the liver iron should be measured by a method that is reproducible and validated.

DR. ALLEN: Certainly one would like to see endpoints that are very clinically apparent. I think it is not reasonable given in particular that you are not measuring against a placebo agent in this case; you are using the existing treatment modality. The vast majority of the patients were in fact switched over from the existing treatment, a relatively small number of naive patients coming into the clinical trials. So, I absolutely agree with what you just said that this is at this point the best reasonable one. It is far from perfect or complete in my opinion and I think indicates the need for additional continuing studies in the future but it is certainly not an unreasonable one at this point.

Other comments or opinions?

Is the Committee ready to vote on this one question?

Go ahead and call the roll?

MR. JEHN: Dr. Schreiber, yes or no, and you may make a brief comment if you want?

DR. SCHREIBER: Yes, but I am concerned about the ability of the -- we are doing the first item?

MR. JEHN: Yes.

DR. SCHREIBER: I am concerned with the lack of correlation between the pre- and post-measures of the ferritin and the iron store. If I remember right the correlation was only about .6 would change and that means that it is only explaining about 36 percent of the difference which is not a very accurate indicator and I would be very hesitant to redose on changes of that measure.

MR. JEHN: Dr. Harvath?

DR. HARVATH: Yes.

MR. JEHN: Dr. Quirolo?

DR. QUIROLO: Yes.

MR. JEHN: Dr. Ghany?

DR. GHANY: Yes, but I would also like to make a comment that I think for the study it is an acceptable endpoint. I would encourage the sponsors to go back and try to examine what data they have collected to develop better surrogate markers that correlate with the long-term outcome. I realize that they don't have this from the registration trials but their long-term trials.

MR. JEHN: Ms. Baker?

MS. BAKER: Yes.

MR. JEHN: Ms. Winner?

MS. WINNER: Yes.

MR. JEHN: Dr. Doppelt?

DR. DOPPELT: Yes and if at all possible I would like to see some cardiac measures at some point if we have the technology.

MR. JEHN: Dr.Whittaker?

DR. WHITTAKER: Yes.

MR. JEHN: Dr. Brittenham?

DR. BRITTENHAM: Yes.

MR. JEHN: Dr. Klein?

DR. KLEIN: Yes.

MR. JEHN: Dr. Kato?

DR. KATO: Yes.

MR. JEHN: Dr. Davis?

DR. DAVIS: Yes.

MR. JEHN: Dr. Allen?

DR. ALLEN: Yes.

MR. JEHN: Dr. Portman?

DR. PORTMAN: Yes.

MR. JEHN: So, all yeses.

DR. ALLEN: Any other comments that the Committee would like to make on this one question?

We will move on to the second question.

The sponsor acknowledges that the efficacy results in study 0107 did not meet the prospective primary efficacy objective non-inferiority analysis to DFO. Although the trial 0107 failed to achieve its primary endpoint when analysis of all of the patients were done, the sponsor claims efficacy in a non-inferiority analysis in a patient subset, patients with an LIC of greater than or equal to 7 milligrams of iron per gram of dry weight.

The agency has generally viewed claims of efficacy based on subset analyses as exploratory and hypothesis generating. However, there is a finding of a mean decrease from baseline in LIC with Exjade in study 0107, minus 2.4 milligrams of iron per gram of dry weight and in a single arm study 0108, minus 4.2 milligrams of iron per gram of iron per gram of dry weight.

In both studies patients continued to receive regular transfusion therapy. That is question 3. We are trying to get two. All right, so the actual question 2 is does the demonstrated ability of Exjade to lower LIC in patients from their baseline to their end-of-study value provide evidence for clinical efficacy?

Discussion on the general question, evidence for clinical efficacy?

DR. WEISS: Dr. Allen, is this stumping people because we put the word "clinical" in there and would it be easier if we just eliminated that word and just put in efficacy? I don't know if that would make it any easier.

DR. ALLEN: If you take clinical out it is not much different from question No. 1.

DR. PAZDUR: I guess is it of benefit to the patient.

DR. ALLEN: And we are taking apart, you know we are not asking a question about patient comfort or compliance with treatment regimens and so on. Those are obviously separate, very important but separate considerations. We are talking about in comparison with existing treatment or no treatment.

DR. BRITTENHAM: I will make a comment about this. Yes, it shows that the drug can effectively reduce the body iron load and I would emphasize that the effectiveness of the deferoxamine in the control group in the study should not be confused with the effectiveness of deferoxamine in the general patient population and so if that were the comparison it is very convincing evidence of efficacy, of clinical efficacy.

DR. PAZDUR: We are not asking a comparison to deferoxamine. We are asking does in an individual patient obviously a decrease in liver iron doesn't just occur, it is a therapeutic effect of the drug.

DR. BRITTENHAM: Oh, absolutely.

DR. PAZDUR: And that is what we are talking about.

DR. BRITTENHAM: I think all that we know says that the more iron you have the worse off you are.

DR. PAZDUR: Of this magnitude that we are seeing here is it of benefit to the patient?

DR. BRITTENHAM: Absolutely.

DR. PAZDUR: That is all we are asking,

DR. KLEIN: If this is a clinical judgment based on the experience we have had over 40 years and based on the medical literature. The answer is yes. Based on the studies that we have before us today we wouldn't be able to say.

DR. ALLEN: I would certainly strongly agree with Dr. Klein's comment. A 1-year study in which you look at the LIC levels, we know based on the decades of experience with existing therapy what the answer is but I guess my sense is looking at this question very strictly I don't see evidence of clinical efficacy.

DR. KLEIN: But I think you are asking this overall, not just based on what we have seen today, and my answer to that would be yes.

DR. ALLEN: Any further instructions from FDA?

DR. PAZDUR: To have a drug approved you have to have substantial evidence of safety and efficacy. You don't need comparative safety and efficacy. You need safety and efficacy. So, I guess what we are asking here is given the available information that is brought to us by the sponsor in our review do you view that this lowering of liver iron in an individual patient or in the mean group of patients here, does this represent efficacy of the drug.

DR. WEISS: And you don't necessarily need direct measure of clinical benefit if you have enough I guess prior knowledge about iron and you know to make some kind of assumptions about this.

DR. BRITTENHAM: Unequivocally yes. I think we are reading too much in to the question here to be honest with you.

DR. SHASHATY: I would just like to ask the question that is already in the statement that precedes the question. What would you think would be the change in liver iron concentration from the baseline day to 48 weeks had the patient not received any therapy for iron based on let us assume 3 units a month times 12, 36 units of packed red cells?

DR. BRITTENHAM: We can calculate that. I mean I can't do it in my head but with a calculator I can tell you more or less reasonably exactly how much that would be and so it would be probably in rough terms it would have increased by about as much in the best results as it decreased, very rough but --

DR. SHASHATY: By my calculation it would have risen more.

DR. BRITTENHAM: It could be.

DR. ALLEN: I will make one additional comment and this is with regard to the preamble when you talk about the subset analysis. I understand the historical basis of the FDA's looking at total analysis based on what was initially proposed. Given the findings and the way in which the subset analysis was conducted I have no problems with that and think that it is very helpful in understanding. It certainly raises the questions that have been raised by the Committee earlier in our discussions as to one, if there was under dosing what does that say about the potential toxicity if you in fact gave the patients a correct dose to have reduced iron level effectively and I still have the question given some of the initial dosing data that were available and I am not sure that I still fully understand all of the reasons why such a low dose was initially chosen but that is immaterial.

DR. GHANY: I wonder if I could just clarify something or have the FDA clarify something for me. Are we trying to establish whether this drug acts as a chelator because that is what you are asking us? Yes. Clinical efficacy I don't think from a subgroup analysis that wasn't even pre-specified, we can't make any statements based on that.

DR. PAZDUR: Here again I think the non-inferiority analysis of this is a very difficult situation to be in and we didn't even go in that direction in much of our presentation because I think there is more evidence here than a non-inferiority analysis.

Basically given a patient that is getting transfusions you would not expect that patient to have a decrease in serum iron content. So, it is kind of obvious here that there is a therapeutic effect.

When we get into a non-inferiority analysis there are many things that come into play here and trying to make some comparison here might be difficult. Here again the fact that many of these patients in the deferoxamine arm were already on the treatment known to be benefitting from deferoxamine had the dose selected already for them really is an enrichment technique that presents a great bias in this study.

So, trying to make a non-inferiority analysis here on the fact that they won or lost on a non-inferiority analysis is a difficult situation and something I don't think we even need to get into to make a decision here on the therapeutic efficacy of this drug.

Basically what we are asking here to be broad, okay, does this drug have a therapeutic effect? You have basically patients that are getting transfusions. You would expect the iron to be going up in the liver. It is going down here. Of the magnitude that you are seeing here it is somewhat similar. We can't say that it is definitively from a statistical perspective non-inferior to deferoxamine but most people would consider it of benefit to have a decrease in their liver content and that is basically what we are asking here. I think you are reading much too much into the question.

DR. ALLEN: Okay, thank you for the explication. Is the Committee ready to vote?

MR. JEHN: Okay, Dr. Schreiber?

DR. SCHREIBER: Yes, but only at the higher dose levels.

MR. JEHN: Dr. Harvath?

DR. HARVATH: Yes.

MR. JEHN: Dr. Quirolo?

DR. QUIROLO: Yes.

MR. JEHN: Dr. Ghany?

DR. GHANY; Yes.

MR. JEHN: Ms. Baker?

MS. BAKER: Yes.

MR. JEHN: Ms. Winner?

MS. WINNER: Yes.

MR. JEHN: Dr. Doppelt?

DR. DOPPELT: Yes.

MR. JEHN: Dr. Whittaker?

DR. WHITTAKER: Yes.

MR. JEHN: Dr.Brittenham?

DR. BRITTENHAM: Yes.

MR. JEHN: Dr. Klein?

DR. KLEIN: Yes.

MR. JEHN: Dr. Kato?

DR. KATO: Yes.

MR. JEHN: Dr.Davis?

DR. DAVIS: Yes.

MR. JEHN: Dr. Portman?

DR. PORTMAN: Yes.

MR. JEHN: Dr. Allen?

DR. ALLEN: Yes.

Question No. 3, the sponsor is recommending an Exjade starting dose of 20 milligrams per kilogram per day. The safety and efficacy of this dose in patients whose baseline LIC was less than 7 milligrams of iron per gram dry weight has not been evaluated. Those patients received 5 or 10 milligrams per kilogram per day in the trials.

The dose-response data suggest these doses do not effectively LIC or serum ferritin. So, the question is can safety and efficacy the population of patients with LIC less than 7 milligrams of iron per gram dry weight be extrapolated from patients with LIC greater than or equal to 7 milligrams of iron per gram of dry weight who were treated with doses of Exjade at 20 or 30 milligrams per kilogram per day. If not, should these patients be further studied?

General discussion?

DR. BRITTENHAM: I would like to add to the question if I may because the study really has been done primarily in patients with beta-thalassemia and what is suggested is that the drug be approved for transfusional iron overload in general and if that is the case then there is a great heterogeneity and variety in the patients who would receive doses who would have very different amounts of, very different transfusion amounts in and so just would be needed will be out and I think we have made clear that it is difficult to know the exact balance between safety and efficacy in these patients who by and large were under treated.

DR. KLEIN: I am particularly concerned about safety in that subgroup because I don't believe we have any data on that at all.

DR. ALLEN: Are you concerned about, I mean but there is a subgroup of patients certainly those with more serious iron overload who have been treated. We have got the same safety data relatively.

DR. KLEIN: I guess what I am saying is that the way this question is written given these doses to the group that has the lower body iron or liver iron at this point I haven't seen any data on that unless I have missed something today.

DR. ALLEN: And certainly those would be the ones where you would be concerned about the potential for over treatment.

DR. BRITTENHAM: Does the sponsor have data relevant to this?

DR. MARKS: Peter Marks, Novartis. We have data not on patients dosed with doses of 20 and 30 milligrams by liver iron concentration but by serum ferritin values in terms of analysis that we have conducted. So, because of the intended labeling based on serum ferritin we conducted an analysis to look at the reduction in liver iron concentration by various dose groups of serum ferritin, for instance less than 1000, 1000 to 2500 and 2500 or greater.

DR. QUIROLO: Do you have the liver biopsy data that goes with those ferritins as well or just the ferritins in those patients?

DR. MARKS: We have the liver biopsy values and we have the, I am not sure I can show them to you in that way. I have the safety data if you would like to see it by the ferritins in the study. Would that be of use?

DR. QUIROLO: I would really like the liver iron correlated. I mean I think you guys have a lot of data here that you could mine and I think that is one of them. Without the liver iron concentration to me and maybe Dr. Brittenham feels differently but the ferritins aren't that helpful.

DR. PAZDUR: One other thing, if you do not feel that you could basically make a relationship between the serum ferritin and the liver biopsy here and if you do say that you are unable to really make a determination of safety and efficacy in the group that has an LIC less than 7, okay, if you cannot make that really what you are saying is you are predicating that we would advocate a liver biopsy be done to make that distinction. So, I just want to bring that point up. If you can't make that decision and you are saying no, you really have to understand the implications of what you are saying here.

DR. BRITTENHAM: No, if I understand it this is data which exists but it hasn't been presented and so we can't make a judgment about data we haven't seen.

DR. PAZDUR: I just wanted to bring that up because it is an important point that if you feel uncomfortable with the serum ferritin to the liver biopsy and you are saying you really don't feel you have enough data regarding dosing in the low iron cohort we really would have to do a liver biopsy in individuals to make that distinction.

DR. BRITTENHAM: But I think they were done and the other thing is just think these are the very patients who have struggled through to comply with deferoxamine and they also are the ones who don't have this backlog that has to be eliminated. So, they are the ones who really can have a benefit of this and it is just we need a better view of how safe it is or what toxicity to watch out for to protect them.

DR. MARKS: Peter Marks, Novartis. I can show you this is the change in liver iron concentration according to the different ferritin groups.

Dr. Quirolo, is that what you were looking for?

DR. QUIROLO: Right, but to me what I would really like to see would be that data with over the top of the ferritin the liver iron and then toxicity, the urine and the toxicity data along with that. That is I think what we are really looking for.

DR. MARKS: I can show you on the next slide the safety data that goes along with this analysis. So, you see the change in LIC in the group with 1000 to 2500 and then we are getting you the safety slide here which I believe is SP54.

DR. QUIROLO: So, I am looking at this. It looks like if you are less than 7 there was no change. These are with the lower doses.

DR.MARKS: Those were patients dosed with 5 and 10. What I would like to show you is that if you look at the patients, if you take the patients there were actually 57 of these patients who were dosed with 20 milligrams per kilogram and you see in the column with the 1000 to 2500 that the adverse event right here is there for you to see in comparison with the rest of the population.

DR. SHASHATY: But this is only information with baseline ferritins. I think that what we would need to really look at would be the relationship between ferritins as the trial progresses and the development of -- there are several, some of the patients who had adverse events the ferritin levels were mentioned in regard to what was happening. It seems to me reading the case report forms that there didn't appear to be a relationship between the adverse event and whether or not the serum ferritin was falling. That was my interpretation. I think that that is what we would like to know.

Is there a relationship between adverse event and the either the rate of fall or the level to which the serum ferritin fell?

DR. QUIROLO: To me the adverse events that you list I am not that impressed. I am really more worried about the renal toxicity. I mean the adverse effects like if you have diarrhea or whatever, that is not going to bother me as much as if their creatinine is going up and they are starting to spill protein.

DR. MARKS: So in the analysis that we conducted the patients who had the most rapid falls in serum ferritin or for that matter liver iron concentration had the highest incidence of the elevations in serum creatinine. Those are patients who were dose reduced then with normalization or at least no further increase in serum creatinine and they maintained efficacy. In other words they continued until the end of study and then had, in fact those were some of the patients who had some of the most dramatic reductions in either liver iron concentration or serum ferritin.

DR. ALLEN: Dr. Portman, do you have any comments or questions?

(No response.)

DR. ALLEN: Others?

Are you ready to vote on this one?

DR. DAVIS; May I ask a question? Now, if we vote no on this are we then as you suggested saying these patients should have liver biopsies?

DR. PAZDUR: I think that is something that you need to discuss. Okay, we have several options here. One would be to discuss looking at safety and efficacy here and maybe people would want to discuss efficacy versus safety, you know do we not have the safety data but we know that we might expect a similar efficacy here and to have a commitment toward safety, examining the safety in a group but one of the problems that we have here and the dilemma here is usually we have our labeling for our drug and our prescription for how the drug should be used is how the drug was studied and here we have kind of a disjunctive situation because we are looking at serum ferritin to guide our dosing of this drug and whether or not the drug should be initiated and at what does, etc., and that wasn't the way it was studied.

So, we have a peculiar situation here with a degree of disconnect here from how the study was done. Here again we don't necessarily want to advocate making it more complicated and would add basically you know potential deaths to patients by giving liver biopsies or morbidity and mortality here.

One option would be to recommend a serum ferritin level where a patient should be treated and then have a long-term commitment toward toxicity studies.

DR. BRITTENHAM: Let me see if I can disentangle some of this. First, I don't think there is any doubt that giving 20 or 30 milligrams to patients who have liver irons below 7 will produce iron excretion. Because their iron is lower it will produce less. So, I don't think there is a question really about efficacy. What we are bothered by is that in the trial that has been done there is little at best information about how safe that is and we are concerned that it is just these patients who treated with higher amounts of the drug may develop renal complications and that this is especially potentially hazardous in clinical use because they are not going to have all the information, the careful monitoring and so on that they did during this trial and so I think what the concern is is just we don't have much information about that. So, there needs to be some mechanism from a very close follow-up of that in this particular group, that the rate in the ones higher seems manageable but we really, that this is a very important population and we really have very little information about how safe it is and we don't want to produce renal failure.

DR. WEISS: And the problem is that and that kind of gets into question 4, if people subsequent to if this is marketed are not followed by measurements other than ferritin we are not going to have a handle on this population again to know unless you know there is a recommendation to do biopsies which isn't going to be probably very viable and so I mean there may be other ways clinically to try to manage this group but you will never know whether or not, you know if you start seeing toxicity, you need to have some dose modifications. You will never know whether or not it is the LIC less than 7 because you won't be able to identify those people in the future. So, it is kind of a very difficult question that we asked you to answer.

DR. BRITTENHAM: No, but it is important nonetheless to try to find an answer for this and so I think there are two things. One, first we would really like to know what is the experience in terms of the liver iron because that is what you can really depend on to give you a sense of what the body iron load is and then is our concern that patients with a lower body iron treated with these higher doses which may well be needed whether there will be more toxicity there and if so, then there is the next question in exactly those patients where we have the liver irons what are the ferritins and is there anything about the ferritin where we could make a judgment? Is there any cutoff? I am not so optimistic about that but it should be examined.

DR. KLEIN: I would agree with that and I would just like to make two points. The first is that I think you may have some of these data but I don't think we can dig them out in the next 15 minutes or whatever we have here but I think they may exist. I think the sponsor may have them.

The other thing is we have talked about renal issues and to be quite frank I don't know about hepatic toxicity or other toxicity as well because we don't have data on that and maybe that is going to be an issue as well. I just don't know and I am reluctant to vote yes without any data.

DR. BRITTENHAM: And especially because the population was very restricted with regard to the liver function of the patients who were entered and in clinical use it is going to be much broader and so there needs to be some sort of special monitoring to try to protect those patients as well.

DR. ALLEN: I think that these are extremely important points because especially if I am looking down the road given an oral agent that seems to be reasonably well tolerated it would be logical to identify as early as possible that you can start treatment to avoid a buildup of iron load so that this subcategory of patients I think for future clinical care is going to become at some point the largest population group that will be treated. So, it is very critical to answer these questions well. I think the sponsor probably does have data that hasn't been analyzed or at least hasn't been presented here and can't be this afternoon but certainly needs to be looked at very carefully and then additional follow-up and I am assuming that there will be approval for marketing, postmarketing follow-up is going to be very important.

DR. BRITTENHAM: Our goal is to get everybody in this group.

DR. PAZDUR: Would people feel uncomfortable if for example there was a postmarketing commitment that looked specifically at liver biopsies of patients that identified this group of patients, okay, that this was definitely a postmarketing commitment?

DR. DAVIS: I would as somebody that does liver biopsies and somebody that operates on other people's liver biopsies especially with abnormal livers I think we need to be very careful about recommending that.

DR. PAZDUR: No, no, what I am saying is a study that was done and having the drug out there to look at really marketed in the sense based on a serum ferritin level but obviously what I am sensing here is some concern about the subgroup of patients that the safety and efficacy has not been demonstrated and the only way really to get a handle on that would be a, well a study of that patient population. I am not talking about requesting it for everybody before they go out, you know, receive the drug because we have this situation. If you are telling us that you feel uncomfortable with this subgroup with the low serum iron the only way to identify that population obviously is to do a liver biopsy and here again I agree with you. We don't want to subject the whole population of patients that would be getting this drug to liver biopsies but here again there is some concern raised here that the people do not have an adequate handle on risk/benefit. So, this could be a postmarketing study.

DR. BRITTENHAM: Right, but in the context of the study it would be possible to do this with some non-invasive calibrated and validated method.

DR. PAZDUR: Such as?

DR. BRITTENHAM: One, I think Squibb does do this. It wasn't in this case but nonetheless but much more widely available is MRI that there are a variety of methods that would be for this purpose useful.

DR. QUIROLO: But there is also a precedent for just using ferritins. A study was done showing that people who had ferritins less than 2500 had less cardiac disease. I think as a clinician I would accept their data if they told me that all the patients who had a liver iron of less than 7 no one ever had a ferritin over 1500 or something like that. Then I would have a handle on that but just from the data that has been presented I just feel like I don't rally know what is going on.

DR. MILLS: So, as I am hearing this comment that I am hearing that the serum ferritin, MRI and liver iron concentration all need to be looked at carefully and tried to correlate and relate these values?

DR. GHANY: I don't see how else you are going to identify these questions. I mean that is the issue that we really need to reflect these are biomarkers and we are trying to relate them to understand which one is a surrogate for which.

DR. ALLEN: That is on the efficacy side as well as potential toxicity sides but certainly the toxicity, the renal needs to be looked at very carefully.

DR. MILLS: And with that the creatinine clearance would be related to those and therefore all of those would be an applicable postmarketing commitment.

DR. BRITTENHAM: But let me point out deferoxamine is also toxic and what people have used is the famous Porter formula for estimating this and perhaps it may be possible -- this is to avoid deafness and blindness. So, there is precedent for using ferritin for this although I am hardly a ferritin enthusiast I have to confess.

DR. QUIROLO: And you have to remember these are the patients if you keep them on 30 per kilo their iron is going to go down, down, down, down, down. At some point you want to know when can I back off or stop or change the dose.

MS. WINNER: May I just make a comment with reference to liver biopsies from a patient's perspective? It is not that we don't want to do it. I have done it seven times and I believe that that there is a way to correlate the ferritin using the SQUID just for this particular study that you are referring to. I think you may have more participants for one thing and you may get your same results.

DR. PAZDUR: We revealed that.

MS. WINNER: Yes, because liver biopsies are very risky especially for people who have had numerous ones.

DR. PAZDUR: Plus we would want to make sure we could extrapolate that to another population, also.

DR. WEISS: I was just wondering if there are centers, I mean people do get liver biopsies for other reasons, I mean it would certainly be helpful to grab that information if you are going to get a liver biopsy anyway for your hepatitis or other things that have been mentioned even though it would have some limitations. There may be ways potentially to get data if one is going to have to have an invasive test done for other reasons and at least to try to use the information if you can get it.

DR. QUIROLO: You know the trouble is if someone gets a liver biopsy for some other disease like hepatitis their ferritin could very well be really elevated. So, I don't know that you could use that population, but I really think the sponsor probably has the data already and they just haven't really put it together.

DR. WEISS: It would be so much using it for correlations but at least to be able to get handles on liver iron concentrations as more of a quantitative thing.

DR. ALLEN: All right, are we ready?

DR. DOPPELT: I just wanted to add one quick point. We seem to be grasping at other modalities such as ferritin and SQUID and MR and all that and I just want to come back to the point that whatever is used has to be something that is validated and just because you can measure it doesn't make it the right measure.

DR. ALLEN: Point well taken.

Okay, are we ready to --

DR. PAZDUR: We could in fact because of the discussion here I think we brought things together and perhaps we don't need to vote on this one because I think we have had an indication that there is sufficient confusion here. The Committee would be glad to think about this and discuss it further.

DR. ALLEN: All right, we will not vote on question No. 3 and move to question No. 4 and the preamble for that is patients who were entered in the trials were required to have a liver biopsy to determine LIC. In clinical practice serum ferritin levels have conventionally been used to select and follow patients for iron chelation therapy.

The question, is the available information sufficient to direct initial and maintenance dosing; if yes, what does regimen and monitoring would you recommend? Should a liver biopsy to determine LIC be recommended as part of the selection and monitoring criteria?

DR. PAZDUR: I think we have discussed this.

DR. ALLEN: Yes, I am delighted you are agreeing with that. Does anybody want to make any additional points with regard to question No. 4?

DR. QUIROLO: I don't think that the liver biopsy would be a recommendation for selection. I think that is kind of over the top.

DR. BRITTENHAM: Many of the patients who are going to be treated with this drug will already have been on a dose of deferoxamine and I think if there is anything to the trial that tells us that perhaps it would be useful to use the data that has been gathered already to make a plot of the iron excretion versus the dose for both the deferoxamine and for the ICO and use that to make the equivalents rather than the estimates of what these doses would be. The current DFO dose could be used as a guide for the initial one.

DR. WEISS: Would you anticipate that there might be then potentially different dose recommendations given conversion versus de novo Exjade if this was marketed? I guess we would have to look at the information and see how it comes.

DR. BRITTENHAM: Yes, I mean if you have somebody with sickle cell disease who has not been treated at all I think it is tough but I think it would be very helpful for everyone who has already been treated and it would solve a problem for a lot of people.

DR. PAZDUR: There would probably have to be a caveat about an upper level dose because obviously you can't recommend a dose that has never been given or studied.

DR. GHANY: I would just like to make a comment about liver biopsy. I understand what all the problems are with suggesting a liver biopsy to initiate therapy but we use liver biopsy to manage patients with chronic liver disease all the time and actually it is part of the, one of the recommendations for assessing patients with hereditary hemochromatosis.

So, I would say that in selected groups of people with iron overload that perhaps a biopsy should be recommended and it is useful especially for those who have concomitant underlying liver disease not just iron overload because it can provide a lot of useful information to the clinician and if you are going to recommend a surrogate marker such as serum ferritin which is as has been mentioned likely to be significantly altered by the underlying inflammation or steatosis or what else have you that you could run into problems.

DR. ALLEN: I think given the potential for changes in medical practice and the difficulty of changing a drug label after approval you know my recommendation would be something that you know before putting the patient on it adequate baseline studies you know, whatever, but without getting into specific methodologies you know to have on the label a requirement for or a strong recommendation even for liver biopsy given where practice is likely to be at some point in the future my personal feeling is I wouldn't be uncomfortable although I 100 percent agree with the general statement that you made.

Other questions or comments on that point or any other point on this?

DR. KLEIN: I would just like to say I agree with you entirely but I think there should be the medical indication, the clinician's decision and not a requirement to start the drug on the label.

DR. ALLEN: Yes, I would agree with that and that was said much better. It was the essential point I was trying to make, too.

All right, anything else on question 4?

We will move on to question 5. The preamble for question 5 is safety review indicates that treatment with Exjade was associated with adverse events in the kidney, liver and other organs. The adverse events were most often considered non-serious and did not usually lead to drug discontinuation. Treatment with Exjade will likely be lifelong. Long-term safety data are limited.

Question, does the safety database provide an adequate characterization of the safety of Exjade to allow adequate benefit/risk assessment and adequate labeling?

DR. BRITTENHAM: I really think this is a determination that can be made only after the additional analyses that we have discussed.

DR. QUIROLO: The sickle cell population I think needs to be looked at carefully because those individuals can have pre-existing renal disease and the exclusionary criteria for the study really eliminated anybody with sickle cell disease particularly who might be at risk for problems and those tests maybe won't be done in the areas outside of centers. So, we have to be really careful we don't hurt people with this.

DR. KLEIN: I, also, think if the question you are asking is whether a prolonged Phase IV ought to be done I have to say absolutely without any question because despite the fact that this is the largest randomized trial for a chelator the numbers really are very small and this is only a 48-week follow-up. So, yes, you clearly for relatively serious but unusual complications you definitely have to follow up.

DR. GHANY: I would also like to say that I think the company should get data on other subsets of patients who have underlying liver disease especially since there is a concern about hepatotoxicity and we know from other studies that patients with hepatitis C for instance have high incidence of drug reactions and hepatotoxicity and a significant proportion of multi-transfused patients have underlying hepatitis virus infections.

DR. ALLEN: I think that is a very important point to follow up also.

DR. SCHREIBER: This is somewhat related to the database but I am a little bit dismayed at in fact we are drawing conclusions on parts of the study that were not statistically powered to address to start. It was powered and it failed the test that it was designed for and the sponsor agreed to. There were some obvious differences in a couple of subgroup analyses but now we are trying to draw conclusions from subgroup analyses because I believe we actually want to get this product out on the market but it was never powered to answer most of these questions that we are asking and I guess it brings back in my mind what is the role of the negotiations of the FDA in clinical trial design because you did negotiate end points and you did negotiate what you are willing to accept in terms of analysis of efficacy and safety and I can see where we are going because there are a lot of things that are unanswered but I think they are unanswered because it was never powered to do that and it was never powered or designed to answer the question of ferritin and that is what we are addressing now but if that was the predominant role that you wanted it should have been addressed front on in the design of the study. So, I think that the subsequent reanalysis of the data which I think we all agree you probably have and we haven't seen would probably give us a lot more leeway in looking and actually being able to see what the outcomes were but without that I find it very frustrating from a statistical standpoint.

DR. PAZDUR: Usually, well, in most cases studies are powered for efficacy endpoints, superiority trials to look at treatment A, B, non-inferiority trials, not so much for the description or not for detecting safety differences. Usually that is a descriptive type of analysis that is made and not a power situation where you are powering to detect a difference between creatinine elevations between two arms.

The primary analysis of the study and the randomized study obviously was to take a look at efficacy. We frequently see this in a field that I am more familiar with, the oncology field of primary analysis usually and efficacy analysis and then there is a description of toxicity and this is a similar situation that we are at and we have to make a decision based on the amount of information that is carried in the randomized trial.

I think here we hear your concerns and we do have them also that there is a need for a long-term follow-up and that will be done. We really want to take a look at long-term toxicities of the drug as well as better delineation of how this drug would be used in both renally and hepatically impaired patients, but we don't live in a perfect world and we are given the database that we have at this time and a need as you heard from the open public hearing that there is a need for this drug and frequently people are not getting deferoxamine. So, we have to balance sometimes these needs of getting a drug out for a life-threatening condition such as this.

This is not uncommon in life-threatening disease that we have this situation.

DR. ALLEN: Dr. Portman, do you have any comments with regard to the question 5?

DR. PORTMAN: Only to reiterate what I said before. I think this is a situation where significant renal toxicity is possible. We just don't know and I think it is a good drug that we want to get out there, too, but I think that it needs to be studied in much more detail. What they have now should be examined and postmarketing studies should definitely be done including what they have under way.

DR. ALLEN: Thank you.