 |
 |
 |
FDA Home Page | Search
FDA Site | FDA A-Z Index | Contact
FDA
| Email
this Page To a Friend |
 |
By Michelle Meadows
A single report of a drug reaction in a 39-year-old woman ultimately contributed to the removal of the allergy drug Seldane (terfenadine) from the market in 1998.
Doctors at the National Naval Medical Center in Bethesda, Md., admitted the woman to the hospital because of fainting episodes. She had been prescribed Seldane 10 days before. She also started using the prescription drug Nizoral (ketoconazole) for a vaginal yeast infection. That combination caused potentially fatal changes in her heart rhythm.
The case, published in 1990 in the Journal of the American Medical Association, prompted further investigation of abnormal heart rhythms related to Seldane. The Food and Drug Administration issued warnings indicating that ketoconazole interfered with Seldane's metabolism, which resulted in increased levels of Seldane in the blood and slowed its elimination from the body. The FDA also warned that a similar effect could occur if Seldane was taken with the antibiotic erythromycin. People with liver disease, another barrier to properly metabolizing the drug, were also at increased risk.
Approved by the FDA in 1985, Seldane was the first prescription antihistamine that didn't cause drowsiness. This was considered a big benefit because it lowered the risk of falls and other accidents. Though warnings and labeling changes about Seldane's newly found risks worked to some extent, the FDA determined that the potential for heart rhythm problems due to dangerous interactions was too risky. The drug was taken off the market when Allegra (fexofenadine), a safer alternative, was approved.
According to FDA experts, discovering Seldane's interactions with other drugs marked a significant advance. Other drugs with interaction problems similar to Seldane's were also identified around the same time. The discoveries improved the ability of the FDA and drug manufacturers to test for drug interactions and to investigate risks of heart rhythm abnormalities before drugs could be marketed. This, in turn, helped the FDA in its mission to protect the public against dangerous medicines and ensure safety.
Early in a drug's development, companies conduct research to detect or predict potential interactions between drugs. FDA experts evaluate the drug-interaction studies as part of assessing a drug's safety. "We look at the issues that go into getting the best study design to identify drug interactions," says Larry Lesko, Ph.D., director of the FDA's Office of Clinical Pharmacology and Biopharmaceutics. "Once we have data in hand, we analyze the information and make decisions about whether a drug can be approved and how to communicate any risks of drug interactions to health professionals and consumers."
Mixing two drugs together could make one of the drugs ineffective. The combination also could increase a drug's effect, and be harmful. The result might be mild symptoms such as nausea, stomach upset, or headache, or more serious symptoms such as a dramatic drop in blood pressure, irregular heart beat, or damage to the liver-the primary way that drugs pass through the human body.
When a drug is taken orally, it usually travels from the stomach to the liver, where it can be metabolized-the process of breaking down and removing chemicals from the body. Enzymes are complex proteins that act as catalysts in starting or speeding up chemical reactions. They cause a specific chemical change in other substances without being changed themselves.
The most important enzymes in the liver that metabolize drugs are called the cytochrome P450 family of enzymes. These enzymes break down drugs when they pass through the liver or small intestine. Paul Watkins, M.D., professor of pharmacology at the University of North Carolina at Chapel Hill, says, "Just in the last 15 years, we've seen big advances in our ability to predict drug interactions because of the discovery of cytochromes P450 and what we've learned about the five major players in the family-CYP1A, CYP3A, CYP2C9, CYP2C19, and CYP2D6."
A drug may affect these enzymes by inhibiting them, which causes reduced activity of the enzyme and a buildup of the drug in the body. Or drugs may "induce" the enzymes, which causes increased activity of the enzyme and a reduction of the drug in the body. It turned out that ketoconazole inhibited the metabolism of Seldane through the CYP3A enzyme, causing a buildup of the drug in the blood. Found in the liver and the gastrointestinal tract, the CYP3A pathway metabolizes a majority of drugs on the market.
It used to be that the only way to test for drug interactions was in people, Watkins says. "Now drug companies can take five test tubes with the five major pathways for metabolism and put their drugs in to see whether it's metabolized by CYP450," he says. "This allows them to generate a list of possible interactions based on their findings."
This phase of research in test tubes, known as in vitro studies, allows researchers to perform drug-interaction studies in labs by testing a drug with other drugs that have the same route. "This has made the research faster and more accurate," says David Flockhart, M.D., Ph.D., chief of the division of clinical pharmacology at Indiana University's School of Medicine. "If two drugs go through the same enzyme, the presence of one drug can prevent the metabolism of the other. So this allows you to look at the worst-case scenarios and ask: 'What if we put this drug with that one, knowing that they have the same route?'"
Not everything that happens in a test tube will become meaningful in humans, though. "Results from these test-tube studies tell us whether we need to do further testing in people to find out if an interaction is clinically significant," says Steven Wrighton, Ph.D., a research fellow at Eli Lilly and Company in Indianapolis. "In vitro studies help us prioritize."
Three phases of clinical trials in humans must happen before a drug can be marketed. Phase 1 studies focus on a drug's side effects and how the drug is metabolized and eliminated from the body. Phase 2 studies focus on a drug's effectiveness. And Phase 3 studies gather more information on safety and effectiveness, which includes using the drug in combination with other drugs.
"Around Phase 1 of clinical development, we look for the ability of our drug to inhibit CYP450 and we determine whether our drug causes an increase in concentration of other drugs as a result of the inhibiting," Wrighton says. "By Phase 2, we know if our drug is an inducer or an inhibitor, and we know the enzyme that metabolizes it. This tells us what patient factors and other drugs influence the metabolism of our drug."
The FDA has published two guidance documents to help drug manufacturers conduct drug-drug interaction studies. One document focuses on laboratory (in vitro) studies, and the other deals with human (in vivo) studies. The FDA is revising this second guidance on evaluation of drug interactions in humans and its labeling implications. The revision will involve several significant changes, including a proposal to use a classification system for CYP3A inhibitors in drug labeling, according to Shiew-Mei Huang, Ph.D., who chairs the FDA working group responsible for revising the guidance. The revisions will highlight key drug interactions and improve the consistency of labeling.
Researchers say there are several important variables that affect individual differences in how drugs are metabolized, including race, gender, age, and health conditions. For example, people with kidney or liver disease don't eliminate drugs from their system as well as people who are healthy. Very young children and older people have slower drug metabolism than others, and women may metabolize drugs differently than men in some cases.
Wrighton says, "CYP2D6 is worth noting because roughly 6 percent to 8 percent of Caucasians don't have an active form of that enzyme." CYP2D6 metabolizes many cardiovascular and neurologic drugs. The CYP2C19 enzyme is absent in 20 percent to 30 percent of Asians. This enzyme metabolizes anticonvulsants, Valium (diazepam), and several tricyclic antidepressants.
"We are rapidly using genetics to predict drug toxicity and we want research to become even more targeted and specific," says Flockhart. "We want to identify the people who are vulnerable to the really bad things."
Watkins says there also needs to be more research on enzymes other than CYP450. "There is also increasing attention on drug interactions happening through "transporters," which control how drugs are taken up by the liver and pumped into the bile," he says. Bile is fluid that is secreted by the liver to aid in digestion.
Over the last several years, there has been a substantial increase in the number of drug-interaction studies the FDA sees in new drug applications. If drug interactions are significant enough, they can prevent a drug from being approved by the FDA. If the agency determines that known drug interactions can be managed and that a drug's benefits outweigh the risks for the intended population, a drug will be approved. Drug-interaction information then goes into the drug's labeling in the sections on "clinical pharmacology," "precautions," "warnings," "contraindications," and "dosage and administration."
Monitoring drug interactions after a drug is approved is a critical component of FDA oversight. The Seldane case, according to agency experts, is a perfect example of how reports of adverse events can uncover interactions that hadn't been anticipated earlier.
Because of the practical size of clinical trials, a drug's safety profile is not fully complete until a drug gets out in the wider market. Most drugs approved by the FDA are studied in a few thousand people, and a rare problem might not become apparent until 100,000 people are exposed to the drug. While the number of people in clinical trials is increasing in some areas of drug development, and researchers are getting better at predicting the rare events, it's not always possible to identify every risk of drug interactions.
Besides Seldane, three other drugs have been taken off the market because of significant interactions with other drugs--a heart drug called Posicor (mibefradil), the antihistamine Hismanal (astemizole), and the heartburn drug Propulsid (cisapride).
But it is rare that problems with drug interactions cause the FDA to reassess a drug's approval. The more usual case is that new information on drug interactions is added to the drug's labeling and doctors are informed through letters and other warning announcements. Then it takes time to evaluate additional reports and assess how well the labeling changes are working.
For example, in March 2001, the FDA advised women taking the prescription blood thinner warfarin to consult a doctor or pharmacy before using miconazole, an over-the-counter (OTC) antifungal drug for vaginal yeast infections. Manufacturers of vaginal creams and suppositories containing miconazole had to add a warning to the label that bleeding and bruising may occur in people taking the two drugs.
The FDA focus has been mainly on drug-drug interactions, but interactions between drugs and dietary supplements are gaining more and more attention. A recent Institute of Medicine report, sponsored by the FDA, estimated that Americans spend close to $16 billion each year on dietary supplements, which includes vitamins, minerals, herbs, and other plant-derived substances. Their widespread use poses health risks because, unlike new drugs, dietary supplements don't generally have to go through clinical testing and FDA approval before they can be marketed.
Under the 1994 Dietary Supplement Health and Education Act, once a dietary supplement is marketed, the FDA is responsible for showing it's unsafe before the agency can take action to restrict its use. The challenge is that some dietary supplements can interact with enzymes that metabolize drugs. The FDA and the American Society of Clinical Pharmacology and Therapeutics discussed the significance of the problem during a workshop on herb-drug interactions in 2002. The National Center for Complementary and Alternative Medicine, part of the National Institutes of Health, coordinates much of the federal research in this area.
The FDA released a public health advisory in 2000 on the risk of drug interactions when St. John's wort is taken with Crixivan (indinavir), a drug used to treat people infected with HIV, the virus that causes AIDS. St. John's wort is an herb commonly taken to ease the symptoms of depression. An NIH study showed the combination of the two lowered concentrations of Crixivan in the blood, thereby lowering the drug's effectiveness. Based on those results, the FDA determined that St. John's wort may decrease blood concentrations of other drugs that are metabolized in a way similar to Crixivan. St. John's wort increases activity of the CYP3A enzyme.
This can also make other drugs less effective. For example, there are reports of bleeding between menstrual periods and pregnancy in people taking St. John's wort with oral contraceptives. There are also reports of hypertension and a potentially fatal condition called serotonin syndrome when St. John's wort is used with selective serotonin reuptake inhibitors (SSRIs) for depression. Examples of drugs that have St. John's wort listed in the labeling as possibly lowering blood levels are Kaletra (lopinavir and ritonavir) for HIV/AIDS treatment, Mifeprex (mifepristone, RU486) for pregnancy termination, Nuvaring (etonogestrel/ethinyl estradiol) for contraception, Gleevec (imatinib) for cancer treatment, and drugs that prevent rejection of organ transplants such as Neoral (cyclosporine), Rapamune (sirolimus), and Prograf (tacrolimus).
In addition to having a good grasp of drugs and their effects, doctors take medication histories, and they consult with other members of their team to guide them in making decisions about drug interactions. They also use concise drug summaries and resources on pharmacological principles. Flockhart has created a Web site that allows health professionals to identify drug interactions by better understanding how the drugs are metabolized and which enzyme in the CYP450 family is involved.
Health professionals also use computer systems with drug-interaction screening software, electronic prescribing, and other technology. Mark Langdorf, M.D., chair of the department of emergency medicine at the University of California, Irvine, says, "In a busy emergency room, you have to quickly find out what a patient is taking and how those drugs could interact with other treatments."
He and his colleagues considered software programs available to emergency physicians a few years ago. "We found two types: one that checked for interactions but didn't write prescriptions, and one that wrote prescriptions but didn't check for interactions," Langdorf says. "None did both." In a chart review, he found that as many as 17 percent of emergency department patients were taking drug combinations that posed potential interactions. His findings prompted him to work with a software company to design a program.
"Rather than writing a prescription and having to wait until it's checked for drug interactions at the pharmacy," Langdorf says, "we wanted to see a standard computer system that identifies significant drug interactions before the prescription is written."
One challenge is that in an attempt to be comprehensive and avoid liability problems, software screening programs for drug interactions tend to flag every possible interaction, even those that aren't significant, says Daniel Malone, Ph.D., associate professor in the College of Pharmacy at the University of Arizona in Tucson. "Previous research has shown that health professionals are overburdened with alerts that aren't meaningful," Malone says. "We want to prevent useless alerts happening when physicians are trying to write a prescription so they don't get frustrated with the technology and abandon it."
In a CDC-sponsored study, Malone and his colleagues developed a list of 25 clinically important drug-drug interactions likely to occur in community and ambulatory settings. Their findings were published in the March-April 2004 issue of the Journal of the American Pharmacists Association.
"It's just not possible for health professionals to memorize or be aware of every drug interaction," Malone says. "The goal then becomes making the system of preventing drug interactions more efficient."
The large number of drugs on the market, combined with the common use of multiple medications, makes the risk for drug interactions significant. "Consumers need to tell doctors what they're taking and ask questions, and health professionals could do a better job at trying to get the information they want," says Timothy Tracy, Ph.D., a professor in the school of pharmacy at the University of Minnesota, Twin Cities campus. He says there is a pervasive attitude that the term "medications" only refers to prescription drugs. "So rather than asking patients what medications they take, doctors should make the questions specific: 'Are you taking any over-the-counter medication? Are you taking any herbal treatments or vitamins?'"
At the same time, Tracy suggests that consumers remind doctors of everything they take when they are prescribed a new medication. So a patient might say: "Now remember, I'm also taking birth control pills. Is there a risk of interaction with this new medicine?"
"Consumers also tend to think that just because something can be bought over the counter, it is without risk," Tracy says. In a study published in the April 2003 issue of the American Journal of Obstetrics and Gynecology, Tracy and his colleagues interviewed 578 pregnant women and found that some of them took medications that should not be used in pregnancy.
"We were surprised that close to 15 percent of the women took ibuprofen during pregnancy, including in the third trimester when ibuprofen is contraindicated," Tracy says. Ibuprofen should not be used in the third trimester because it could cause problems to the unborn child or complicate delivery.
Nearly 60 percent of the women in the study received a prescription medication other than prenatal vitamins, and many medications were prescribed by a physician other than the obstetrician. "We found that women taking long-term medication such as psychotropic agents don't always tell the obstetrician," Tracy says.
In a companion study of non-pregnant women, Tracy and his colleagues found that of 567 study participants, 92 percent took prescription medication, more than 96 percent self-medicated with OTC medications, and almost 60 percent used herbal medicine. The researchers found instances where the pain reliever codeine was taken with the antidepressant Paxil (paroxetine), a combination that can lower codeine's effectiveness. Some patients taking central nervous system (CNS) depressants also were taking the herb kava, which also acts as a CNS depressant.
"We discovered several women taking St. John's wort and SSRIs for depression," Tracy says. "They just had no idea of the interaction potential. Clearly, consumers and health professionals need to work together to prevent potential drug interactions."
Drugs with other drugs: This includes both prescription and over-the-counter medicines. Tricyclic antidepressants such as Elavil (amitriptyline) and Pamelor (nortriptyline) can interfere with blood pressure-lowering Catapres (clonidine). Taking the antibiotic Cipro (ciprofloxacin) with antacids lowers Cipro's effectiveness. Some antibiotics, such as rifampin, can lower the effectiveness of birth control pills. Sildenafil, the active ingredient in the erectile dysfunction drug Viagra, should not be taken with nitrates for heart treatment because of the potential for dangerously low blood pressure.
Drugs with dietary supplements: This includes herbs and vitamins, which can interact with drug-metabolizing enzymes. St. John's wort is an herb commonly used by people with cancer to improve mood, but research has shown it interferes with the metabolism of irinotecan, a standard chemotherapy treatment. Vitamin K (in dietary supplements or food) produces blood-clotting substances that may reduce the effectiveness of blood-thinning medicines like warfarin.
Drugs with food and beverages: Taking quinolone antibiotics such as ciprofloxacin with food and drinks such as colas, coffee, and chocolate that contain caffeine may cause excitability and nervousness. There can be a potentially fatal increase in blood pressure if food containing tyramine is eaten when taking monoamine oxidase inhibitors, drugs that treat mood disorders. Examples of food with tyramine are cheese and soy sauce. Grapefruit juice should not be taken with certain blood pressure-lowering drugs or cyclosporine for the prevention of organ transplant rejection. Alcohol should not be taken with pain relievers such as Tylenol (acetaminophen) or ibuprofen because of the increased risk of liver damage or stomach bleeding.