Nomifensine (Merital®), an antidepressant that had been available in Germany since 1976, had been prescribed to an estimated ten million patients prior to its marketing in the U.S. in July, 1985^(1,2). Initial labeling for the product reflected a variety of long-recognized hypersensitivity reactions, including fever, liver injury, hemolytic anemia and eosinophilia, that were apparently all readily reversible.⁽³⁾

At the time of U.S. approval, FDA was aware of reports of less than twenty hemolytic anemia cases, all non-fatal; however, in 1985, when foreign adverse reaction reports showed the hemolytic anemia might be fatal, labeling was revised to reflect the potential seriousness of the reaction⁽³⁾. Due to an increase in serious hemolytic anemia cases seen in Europe, marketing of nomifensine was reconsidered by the manufacturer, who announced a worldwide withdrawal of the drug on January 21, 1986 ^(3,4).

The case of nomifensine illustrates that the safety profile of a drug evolves over its lifetime on the market. Even after almost ten years experience, or longer, new information that will impact the clinical use of a medical product can be detected. Consequently, all medical products need to be continually assessed for safety within the context of their perceived benefit.

Medical product safety monitoring is an ongoing process accomplished through Postmarketing Surveillance, the collection of data about drugs [or any other medical product] once they are marketed and thus available to the general population.⁽⁵⁾ This process encompasses adverse event reports evaluation, generation of safety-related hypotheses and use of techniques to evaluate these hypotheses.

Return to the Table of Contents

While the U.S. has one of the most rigorous approval processes in the world, it is not possible to detect all potential problems during premarketing clinical trials. Medical product studies, ranging from preclinical animal testing and medical device bench testing to final tests in humans, have inherent limitations no matter how well they are designed or conducted. The need for postmarketing surveillance is a direct result of these limitations.

Premarketing Animal Studies

Most medical products are first tested in animals prior to introduction into humans. Animal studies have limitations in their ability to predict human toxicity; this is demonstrated by the case of practolol, a ß₁-adrenoreceptor blocking agent withdrawn from the U.K. market in 1976 after several years of widespread use^(6,7), and never marketed in the U.S.

The U.K. action was prompted by the serious adverse reactions of dermatitis, keratoconjunctivitis and sclerosing peritonitis, collectively termed the oculomucocutaneous syndrome^(6,7). This syndrome had not been seen during extensive preclinical animal testing conducted within required guidelines⁽⁷⁾.

Subsequent toxicity studies in several small animal species (both those that metabolize practolol similarly to humans and those whose practolol metabolism is more extensive than humans) found no animal model for the observed human adverse reactions⁽⁸⁾. The lack of reproduction of these particular adverse reactions in any laboratory animal species⁽⁹⁾ demonstrates that animal studies, no matter how appropriate or well-performed, are not necessarily predictive of human pathology.

Premarketing Human Clinical Studies

There are intrinsic limitations to premarketing human clinical trials with respect to their ability to detect adverse events. Short duration, narrow population, narrow set of indications and small size are major factors in this regard ⁽¹⁰⁾, irrespective of the type of medical product being studied.

The capability of premarketing clinical trials to discover rare adverse events is particularly affected by their size. In order to have a 95 percent chance of detecting an adverse event with an incidence of 1 per 1,000, 3,000 patients at risk are required⁽¹¹⁾; with no more than 3,000 to 4,000 individuals usually exposed to a medical product prior to marketing, only those adverse events with approximately 1/1,000 or greater incidence can be expected to be found.

While medical products are usually studied for several years before they are marketed, an individual patient in a clinical trial is generally exposed to the product for less than a year. Even long-duration premarketing clinical trials, which can last several years, do not provide the degree of patient exposure that will occur postmarketing with a chronically used medical product. In addition, the relatively short durations of clinical trials mitigate against the detection of adverse events with long latency.

Because of these limitations, premarketing clinical trials seldom detect or define the frequency of all important adverse events. As a result, the official labeling/product information at the time of approval of a medical product reflects what is known about that product's risk at that point in time. The controlled environment under which clinical trials are conducted means that the safety data presented in the original labeling of a product usually represents actual occurrence rates in the defined population that has been studied.

Postmarketing Experience

Health professionals should be aware that this is not the case with postmarketing data. Once a product leaves the controlled study environment and enters general clinical use, the ability to detect the actual incidence of an adverse event can essentially be lost. On the other hand, once a new product is marketed, there are great increases in the number and variety of patients exposed, including those with multiple medical problems and undergoing treatment with numerous concomitant medical products.

As a result, the population experience with the product will be much broader than that derived from the clinical trials. One particular safety-related advantage this offers is a generally greater capability to detect adverse events possibly related to interactions with other medical products than is available in the premarketing phase.

The major changes in the size and nature of the exposed patient population that occur once a medical product is available for widespread use emphasize the great importance of adverse event detection and reporting by health professionals.

Return to the Table of Contents

It is with these considerations in mind that MedWatch, the FDA Medical Products Reporting Program, was established⁽¹²⁾. While FDA's long-standing postmarketing surveillance programs predate MedWatch, this educational/promotional initiative was designed to emphasize the responsibility of healthcare providers to identify and report adverse events related to the use of medical products. Through the MedWatch program health professionals can report serious adverse events and product problems that occur with such medical products as drugs, biologics, medical and radiation-emitting devices, and special nutritional products (e.g., medical foods, dietary supplements and infant formulas).

Causality is not a prerequisite for MedWatch reporting; suspicion that a medical product may be related to a serious event is sufficient reason for a health professional to submit a MedWatch report. However, a report on every adverse event is not sought - what is desired is an increase in the reporting of serious events. In that regard, TABLE 1 offers a guideline for adverse event reporting. However, health professionals are welcome to report any adverse event that they judge to be clinically significant.

Return to the Table of Contents