Dr. Janet Arrowsmith-Lowe, M.D., F.A.C.P., offered the following expert report on Wyeth’s behalf in the case of Wright v. American Home Products.
C. FDA Approvals of Pondimin and Redux
I agree with the FDA that the NDA data for Pondimin and Redux supported a conclusion that the drugs were safe and effective when used in accordance with their labeling. In other words, the potential benefits outweighed the known risks. My opinions on the adequacy of the data and labeling are based on my reviews of NDA data, FDA's reviews of the data, and my experience with the FDA review processes.
The FDA determined that the data submitted for Pondimin established the safety and efficacy of the drug for short-term use as adjunctive therapy for the treatment of obesity. Pondimin was found to have a lower abuse potential than amphetamine-like drugs, which for many years had been used widely as appetite suppressants.
The Redux NDA contained extensive information on safety and effectiveness. This included years of widespread use in Europe where it had been marketed since the mid-1980s. Included in the NDA were controlled clinical trials of up to one year duration demonstrating safety and effectiveness, as well as detailed post-marketing safety reports. The studies demonstrated that dexfenfluramine was more effective than placebo in promoting weight loss, when administered to patients with instructions regarding diet and activity.
Also included in the Redux NDA were results from the International Primary Pulmonary Hypertension Study (IPPHS), which addressed the risk associated with anorexigen drug exposure in the development of primary pulmonary hypertension (PPH). The FDA also convened an Advisory Committee to review the safety and effectiveness of Redux, including data relating to PPH. The FDA was well apprised of the risk of PPH in patients who took Redux and Pondimin before the NDA was approved. After reviewing the data on PPH and the IPPHS, as well as the post-marketing safety data from Europe, the Advisory Committee recommended the approval of Redux based on a favorable risk-benefit ratio. FDA approved the drug for marketing in the United States in April 1996. In my opinion, the initial Redux labeling adequately and appropriately conveyed the safety data learned in both the pre-approval studies and the European post-marketing experience with dexfenfluramine.
I. Qualifications
I have had over 11 years of experience at the United States Food and Drug Administration (“FDA”). As described in my CV, attached as Exhibit A, this included a two-year assignment to FDA by the Centers for Disease Control concerning epidemiological matters. I was also an FDA Medical Review Officer in premarket New Drug Application (“NDA”) review divisions in FDA and was a staff epidemiologist in the Center for Drug Evaluation and Research. In addition, I was acting director of the newly formed Office of Surveillance in Biometrics in the Center for Devices and Radiological Health. I am a physician board certified in Internal Medicine and a fellow of the American College of Physicians. I also have received formal training in epidemiology with a focus on pharmacoepidemiology, which studies the effects of drugs after they become available for general use in medical treatment. I am an elected member of the American College of Epidemiology. As also reflected in my CV, I have published in epidemiology and pharmacoepidemiology.
I have been asked to serve as an expert by Wyeth in the area of FDA practice and requirements involving drugs, including specifically issues relating to Wyeth's compliance with applicable laws and regulations with regard to Pondimin and Redux, the purpose and goals of those laws and regulations, as well as industry practices generally. In addition, as noted above, I have expertise in epidemiology and have opinions regarding the conclusions that can be drawn from epidemiological studies relating to the relationship between Pondimin (fenflurarnine) and Redux (dexfenfluramine) use and heart valve regurgitation. The matters on which I express opinions in this report as well as on which I expect to testify are within my expertise. To the extent that my opinions in this report deal with medical and scientific issues, they are made to a reasonable degree of medical and scientific certainty. The opinions expressed in this report are based on the materials I have reviewed to date and I reserve the right to amend and supplement them as I obtain and review additional documents or to respond to any points that plaintiffs are permitted to present at trial.
II. Materials Reviewed
I have reviewed numerous documents relating to Pondimin and Redux. I selected these documents from a group of documents that were available for my examination. A listing of materials reviewed is attached as Exhibit B. I also reviewed various medical articles and texts, as well as FDA regulations in 21 C.F.R. parts 200 and 300. I have also reviewed a number of company employee, expert witness, and other depositions and trial testimony, all of which are included within Exhibit B.
III. FDA PRACTICE AND REQUIREMENTS
A. FDA's Regulation of Drugs
The FDA is a federal regulatory agency whose broad mission is to protect the public health in the area of drugs, as well as medical devices, devices that utilize and emit radiation, biologics, veterinary products, foods and cosmetics. Several thousand scientists and other professionals and support staff are engaged in this mission.
The drug section of the FDA, CDER, is the largest drug regulatory agency in the world. The FDA regulates virtually every aspect of the manufacturing, distribution, study, labeling and post-marketing surveillance of drugs in the United States. The FDA has been in the forefront of setting standards for drug and other product regulation worldwide. The FDA's standards for drug approval and post-marketing surveillance of drugs have been, and are being adopted, by regulatory agencies in Europe and throughout the world.
B. The New Drug Process
Investigational New Drugs: Under the provisions of Section 505 of the Federal Food Drug and Cosmetic Act, the FDA requires that an IND be submitted to the agency prior to administration of any new pharmaceutical compound to any human being in the United States. The IND application must contain extensive information on the manufacture of the drug, animal studies demonstrating safety for humans, pharmacologic studies, and protocols for the first studies in humans. Based on safety information from the initial studies in healthy volunteers, the drug product can be studied in more extensive clinical trials to establish effectiveness as well as safety in the intended population. The IND stage of drug development represents the gateway to further development. FDA can and sometimes does place a hold on further development at the IND stage when the safety of the product for use in humans is in question. The foremost consideration for FDA in clinical investigations is drug safety.
New Drug Application Review: When a sponsor feels that sufficient information to demonstrate that a drug is safe and effective has been complied, it submits to FDA an NDA containing this information together with proposed labeling. In general, under FDA regulations the NDA must contain: (i) reports of all investigations (pre-clinical and clinical); (ii) a summary of the safety data; (iii) composition of the drug; (iv) methods and facilities used in the manufacture of the drug; (v) samples; and (vi) proposed labeling.
The NDA is then rigorously reviewed by FDA officials to determine whether the information is sufficient to establish that the drug meets FDA's safety and efficacy requirements. FDA is well acquainted with a drug when the NDA is submitted because it has followed the drug in the prior IND process. A team of scientists and physicians conducts a careful review of the NDA and prepares detailed reports on their findings. The FDA review teams are experts in their fields. The FDA also may convene advisory committees of outside experts when it wishes to obtain additional opinions on safety and efficacy. The recommendations of these experts are carefully considered by the FDA but are only recommendations and are not binding on the agency. After intensive review by the FDA team and their supervisors, the FDA then issues a letter to the sponsor regarding whether the NDA is approvable and if not, what the deficiencies are and what additional information is required.
Labeling Review: The final product label is a result of discussion between the company and FDA. However, FDA makes the final determination of the content and language of the label based on its judgment and in accordance with the regulations. The labeling is reviewed on an ongoing basis after the drug is approved for marketing for possible revisions based on post-marketing safety data and other information by both the company and the FDA.
Risk-Benefit Assessment: All drugs, and particularly prescription drugs, by their very nature involve risks. When the FDA approves an NDA, it has reached a decision that the drug has been shown to be safe and effective as labeled. This means that the FDA has found that the potential benefits outweigh its known risks when used in accordance with the approved labeling. The risk-benefit ratio is an essential part of the drug approval process and involves judgment on the part of the reviewers.
C. FDA Approvals of Pondimin and Redux
I agree with the FDA that the NDA data for Pondimin and Redux supported a conclusion that the drugs were safe and effective when used in accordance with their labeling. In other words, the potential benefits outweighed the known risks. My opinions on the adequacy of the data and labeling are based on my reviews of NDA data, FDA's reviews of the data, and my experience with the FDA review processes.
The FDA determined that the data submitted for Pondimin established the safety and efficacy of the drug for short-term use as adjunctive therapy for the treatment of obesity. Pondimin was found to have a lower abuse potential than amphetamine-like drugs, which for many years had been used widely as appetite suppressants.
The Redux NDA contained extensive information on safety and effectiveness. This included years of widespread use in Europe where it had been marketed since the mid-1980s. Included in the NDA were controlled clinical trials of up to one year duration demonstrating safety and effectiveness, as well as detailed post-marketing safety reports. The studies demonstrated that dexfenfluramine was more effective than placebo in promoting weight loss, when administered to patients with instructions regarding diet and activity.
Also included in the Redux NDA were results from the International Primary Pulmonary Hypertension Study (IPPHS), which addressed the risk associated with anorexigen drug exposure in the development of primary pulmonary hypertension (PPH). The FDA also convened an Advisory Committee to review the safety and effectiveness of Redux, including data relating to PPH. The FDA was well apprised of the risk of PPH in patients who took Redux and Pondimin before the NDA was approved. After reviewing the data on PPH and the IPPHS, as well as the post-marketing safety data from Europe, the Advisory Committee recommended the approval of Redux based on a favorable risk-benefit ratio. FDA approved the drug for marketing in the United States in April 1996. In my opinion, the initial Redux labeling adequately and appropriately conveyed the safety data learned in both the pre-approval studies and the European post-marketing experience with dexfenfluramine.
D. Post Marketing Safety Surveillance for Approved Drug Products
When a drug has been approved for marketing, the FDA conducts post-marketing safety surveillance in accordance with the requirements of its regulations. Safety surveillance is based on review and analysis of adverse event reports submitted by drug manufacturers, by health professionals such as physicians, pharmacists and nurses, and by individuals as well as reports from in the published scientific literature. The FDA often receives information from foreign drug regulatory authorities, from other federal agencies, and from other sources. The FDA uses outside databases to conduct epidemiological studies testing hypotheses concerning drug exposure and adverse events.
Virtually all serious, unexpected adverse event reports and all adverse event reports from health care providers are individually reviewed by FDA for new safety information. Any significant new safety issues are discussed with the review division responsible for approval of the suspect drug(s). FDA follows adverse event reports closely and determines whether labeling changes are needed. Each time the FDA approves a labeling change, the Agency reaffirms that the drug is safe and effective when used in accordance with the labeling. FDA makes the final determination of the location and language of any labeling changes.
E. FDA Post-Marketing Reporting Requirements
Under 21 C.F.R. § 314.80, FDA has established a systematic approach to the reporting of adverse drug events associated with the use of approved drugs. These regulations are based on federal law and were enacted in 1985. Adverse event reports submitted to FDA are generally evaluated by two groups in the agency. One group responsible for the review, coding and processing of such reports, formerly known as the Office of Epidemiology and Biostatistics, includes physicians, epidemiologists, pharmacists and other scientists. The other group monitoring postmarket safety reports is the review team responsible for approval and labeling of the drug. Adverse event reports are evaluated and independently coded by the FDA and entered into a computerized, automated database. Detailed reports of the review of ADEs reported with certain products or product classes can be prepared using ADE data alone or in combination with other information. The FDA records I have reviewed indicate that the agency carefully monitored and evaluated ADEs reported in association with use of Pondimin and Redux, looked at other databases estimating adverse event outcomes, and made independent judgments concerning Pondimin and Redux ADEs.
Under 21 C.F.R. § 314.80, FDA has established a triage system for various types of adverse event reports. Serious, unlabeled events from any source are submitted to FDA by the manufacturer within 15 days after the manufacturer determines that the event is a serious, unlabeled event. Serious in this context means that the event meets certain FDA criteria, as spelled out in the regulations. Serious labeled reactions from foreign or domestic sources do not fall under the 15 day requirement. Other reporting requirements include submission of periodic reports (either annually or quarterly) listing those reports identified in 21 C.F.R. § 314.80(c)(iv)(2) and submission of annual reports covering information specified in 21 C.F.R. § 314.81(b)(2). Prior to April 1998, companies were required to evaluate rates of serious, labeled events and report an increased frequency of such reports. It is my opinion that FDA's reporting requirements and regulations are based on good science and sound judgement. I believe that Wyeth complied with the reporting requirements in effect during the marketing of Pondimin and Redux. In fact, FDA thoroughly reviewed Wyeth's ADE reporting system during an inspection in the fall of 1997 and did not fault the company for failing to submit any reports.
F. Labeling Requirements
FDA's labeling requirements for prescription drugs are set out in 21 C.F.R. § 201.57. Significant safety information is included under the Contraindications, Warnings and Precautions sections under 21 C.F.R. § 201.57(d)-(f). These regulations spell out what must be included in each section of the labeling. “Adverse Reactions” are included in a separate part of the labeling. Adverse events which occur during the use of a drug are not necessarily caused by the drug nor are they necessarily related to its use. Only those events which are considered to be reasonably associated with the drug are to be included in the “Adverse Reaction” section of the labeling. In my view these requirements are appropriate and Wyeth was in compliance with the labeling requirements for both Redux and Pondimin.
The FDA has complete authority over the use of so-called “black box” warnings. FDA is careful to limit their use to clearly appropriate situations so as to avoid a proliferation of black box warnings that will dilute their impact. The FDA specifically considered whether black box warnings were necessary for the Redux labeling. The FDA concluded that a black box warning for the risk of PPH was not warranted. The FDA agreed that a bold type warning at the beginning of the “Warnings” section of the labeling was appropriate. There are many drugs that have serious and even fatal side effects for which the FDA has concluded black box warnings to be unnecessary.
G. Pondimin and Redux
It is my opinion that labeling for Pondimin and Redux complied with FDA's labeling requirements throughout the relevant periods. The initial Redux labeling accurately conveyed safety data from the clinical trials and European post-marketing experience with dexfenfluramine. From the beginning, it contained information about the IPPHS in prominent, bold-faced text and accurately described other adverse events that had been reported in the European post-marketing experience and in the clinical trials. In my opinion, the Redux labeling continued to adequately and appropriately convey the known risks of Redux during the time it was on the market.
With respect to Pondimin, information on pulmonary hypertension was added to the Precautions section of the labeling in 1983 and it was revised in 1987. The 1987 revision expanded on the earlier information and stated that a fatality had occurred. Beginning in 1983, the labeling included the statement that patients taking the drug should be advised to report immediately any deterioration in exercise tolerance. It is my understanding that this is the most common presenting symptom of pulmonary hypertension. Until June 1996, the information on pulmonary hypertension was appropriately placed in the Precautions section as required by the FDA's labeling regulations.
In my opinion, the inclusion of pulmonary hypertension in the Precautions section of the Pondimin labeling adequately conveyed to physicians the possibility that the condition might be associated with use of the drug. Prior to the Pondimin labeling revisions in 1996, the labeling contained an outdated reference to the number of pulmonary hypertension reports that had been received by Wyeth in persons taking Pondimin. In my opinion, the fact that the number of cases was not updated did not render the labeling inadequate. The purpose of providing safety information in the labeling is to apprise physicians of the known risks involved in prescribing a drug. The number of cases stated in the label did not quantify the risk. In July 1996, Wyeth revised the Pondimin labeling to add information derived from the IPPHS. The IPPHS provided important new information concerning the relative risk of PPH in persons taking appetite suppressants. This new information from a well-designed case-control study was appropriately placed in the “Warnings” section of the labeling.
Internal FDA documents indicate that the Agency did not find it necessary to add any additional information concerning pulmonary hypertension to the Pondimin labeling until the IPPHS was concluded. The FDA specifically reviewed the adequacy of the Pondimin labeling in 1994 and determined that the labeling did not need to be changed. Notably, after the IPPHS results became available, the FDA and Wyeth reached agreement concerning labeling for both Pondimin and Redux that did not include a specific number of case reports but instead provided a relative risk calculation based on the IPPHS. The FDA decided to address changes to the Pondimin labeling after it had determined the final language for the Redux labeling.
It is my opinion that Wyeth acted properly in reporting Pondimin and Redux ADEs to the FDA. Wyeth did not conceal adverse event information with respect to Pondimin and Redux from the FDA.
With respect to claims concerning cardiac valvular abnormalities, I have reviewed computer printouts of adverse drug events, copies of Adverse Drug Event Reports, other safety data relating to cardiac valvular abnormalities, and other relevant documents.
It is my opinion that the 1997 Mayo/Fargo case series provided the first signal that heart valve abnormalities might be related to use of Pondimin or Redux. This opinion is supported by (1) my review of NDA data for both Pondimin and Redux; (2) the post-marketing safety information from decades of safe use of Pondimin and over a decade of safe use for Redux; (3) information published prior to the 1997 Mayo/Fargo case reports; (4) the 1997 report of the Belgian Centre for Pharmacovigalance; and (5) other internal FDA memoranda concerning this issue.
The Mayo/Fargo case series included histopathologic examination of a few heart valves following surgery. The Mayo/Fargo physicians developed a hypothesis to link the histopathologic findings to the anorexigenic agents taken by the patients. By contrast, although Wyeth received some ADEs prior to 1997 that contained references to valvular regurgitation, usually as a lab test finding and not as the reported adverse event, such ADEs were confounded by the use of multiple drugs, the presence of other diseases, or had insufficient information. Many of the ADEs involved patients who had been treated with a slimming cocktail that included Chinese herbs, which are known to cause renal failure as well as other problems. It is also important to note that cardiac valve abnormalities are not uncommon in the general population and that cardiac valve abnormalities are almost never seen as a side effect of prescription drug use. Thus, there was no basis to conclude that there was a reasonable association between use of fenfluraminc or dexfenfluramine and the development of valvular regurgitation prior to the summer of 1997. Indeed, other authorities who reviewed ADEs received prior to 1997 that mention heart valve regurgitation also did not conclude that they showed a connection between Pondimin and Redux use and heart valve abnormalities. For example, during its extensive review of the Redux NDA and post-marketing ADEs the FDA never concluded that Redux and Pondimin were associated with cardiac valve abnormalities. The Pondimin and Redux labeling adequately reflected the knowledge about Pondimin and Redux in the time period before the products were voluntarily withdrawn from the market in September of 1997.
I am aware that Pondimin was being used by physicians for periods of time beyond the few weeks recommended in the product label and that it was being used in conjunction with phentermine. This did not place the company in violation of FDA regulations and it did not obligate the company to specifically discourage such use. The FDA has often stated that it does not dictate the practice of medicine. Physicians are free to use drugs in a manner not stated in the labeling when they feel that it is appropriate and safe to do so. In fact, studies sponsored by the National Institutes of Health and conducted by a respected clinical investigator, Dr. Weintraub, supported the safety and efficacy of the combination of Pondimin and phentermine when used together for more than a year. Moreover, drugs are commonly used in combination, even though such use is not described in the labeling.
It is my opinion that Wyeth neither violated FDA regulations nor improperly withheld safety and medical information from the FDA or medical practitioners. In addition, it is my opinion that Wyeth acted prudently, responsibly, and in a manner consistent with industry standards with respect to the issues discussed in this report.
IV. Relationship Between Diet Drug Use and Heart Valve Regurgitation[FN1]
FN1. I have used the FDA case definitions in which aortic regurgitation is defined as mild or greater and mitral regurgitation is defined as moderate or greater. The definition is appropriate as it filters out levels of heart valve regurgitation that have no clinical significance and occur frequently in the general population of people who never used diet drugs. Using these levels of regurgitation makes the comparisons between those exposed and controls more meaningful.
In addition to serving as an expert in the area of FDA practice and requirements and drugs, I have been asked by Wyeth to review and analyze the various scientific studies investigating the potential relationship between Pondimin and Redux and heart valve regurgitation and offer opinions concerning their findings and conclusions. This portion of the report summarizes my opinions in this area.
The published epidemiological studies concerning Pondimin and Redux do not show a statistically significant increased risk of FDA case definition aortic regurgitation from taking fenfluramine or dexfenfluramine for less than three to six months. The studies have shown that people who used the drugs for longer periods of time are at a statistically significant increased risk of developing aortic regurgitation, which is usually mild. None of these case-controlled or nested studies have shown a statistically significant increased risk of the development of FDA case definition mitral regurgitation. A recently published meta-analysis has demonstrated a statistically significant increased risk of mitral regurgitation for persons who used the drugs for more than ninety days. The study methodology, which is somewhat controversial among epidemiologists and statisticians, required that data from several of the published studies, each with different populations and different protocols, be pooled for the purpose of increasing the population then subjected to this new methodology. It is important to note that none of the studies from which the data were drawn for the meta-analysis demonstrated an increased risk of mitral regurgitation at the FDA defined levels in their original analyses despite the fact that FDA-level mitral regurgitation was included as an endpoint in the original study designs.
The controlled, clinical epidemiological studies have consistently shown that there is no association between Pondimin or Redux use and the development of tricuspid regurgitation, pulmonic regurgitation or other cardiac valvular abnormalities. Also, the studies consistently have shown that patients who took Pondimin or Redux have not experienced more significant cardiovascular events, including surgeries, when compared to controls.
I have reviewed a number of studies that have evaluated the natural history of valvular regurgitation observed in former users of Pondimin and Redux. These studies consistently have shown that the regurgitation does not worsen and is much more likely to remain stable or improve over time. I am aware that studies that have evaluated the natural history of valvular regurgitation in people who never used Pondimin or Redux have reported some progression. However, these studies are not applicable to the question of whether heart valve regurgitation in former users of Pondimin and Redux progresses or regresses because such studies do not involve valvular regurgitation that may be attributable to diet drug use. As in the case of many other drug effects, it is understood that cessation of use of diet drugs removes the agent that may be causing the valvular regurgitation and also stops any progression or may result in regression.
In addition, the controlled epidemiological studies have also consistently shown no evidence that Pondimin and Redux use led to the latent development of heart valve regurgitation.
V. Compensation
When I serve as an expert in litigation I receive $400 an hour or $4,000 a day for testimony.
VI. Prior Testimony
A listing of cases in which I have testified as an expert at trial or by deposition within the proceeding four years is attached as Exhibit C.
Comments