Pharmacoepidemiology: Scrutinizing drug safety in dermatology

According to Joel M. Gelfand, M.D., M.S.C.E., pharmacoepidemiology has timely significance for dermatologists, given recent national publicity focused on the associations of isotretinoin with suicide and lymphoma with psoriasis.

Dr. Gelfand, assistant professor of dermatology and director of the Clinical Studies Unit at the University of Pennsylvania Health System's department of dermatology, addressed such issues in a seminar on pharmacoepidemiology during the 63rd Annual Meeting of the American Academy of Dermatology (AAD) here.

"In dermatology specifically, there are multiple examples of the need for ongoing study of medications to monitor safety. It took several years to demonstrate that PUVA is associated with the development of non-melanoma skin cancers and over a decade to establish an association with melanoma, for example. And several commonly used antihistamines, such as terfenadine, have been removed from the market due to fatal interactions with other medications."

Dr. Gelfand says many adverse drug reactions are not detected prior to drug approval because they are statistically rare.

"For example, a study of 1,000 patients can only detect adverse effects that occur in more then one in 333 patients," he says.

"Many serious medical events, such as cancer, thrombosis and suicide, are actually 'rare,' and therefore premarketing studies are unlikely to determine the safety of therapeutics with respect to these outcomes. Some adverse effects, such as carcinogenesis, are even more difficult to detect because of their delayed nature."

Dr. Gelfand notes that rare side effects can be a public health problem because drugs are often used in millions of patients, resulting in potentially thousands of affected patients.

Epidemiologic studies Epidemiologic studies - that is, those that focus on the distribution and determinants of health and disease in large populations - are critical in evaluating drug safety, Dr. Gelfand says, adding that the most commonly employed design is that of case reports and series based on spontaneous reports to the FDA. There is, however, a caveat.

"It has been well documented that there can be significant under- and over-reporting of adverse events," he says.

Limitations "A serious limitation of this system is that only rarely can a causal relationship be established by case reports," he says.

"Furthermore, this system does not allow one to determine the incidence of a suspected adverse event, which is critical in determining the potential public health impact. More scientifically sound study designs, such as case-control studies, cohort studies and controlled clinical trials, are therefore often necessary to study drugs when used in large populations."

As an example of this process, Dr. Gelfand cites the reported association of isotretinoin with suicide. He says isotretinoin has been used by more than 12 million people worldwide, and due to the widespread use of this medication, one would expect by pure coincidence to observe suicides, given that suicide is the third most common cause of death in patients in the age group commonly prescribed the drug.

"Some case reports have established a temporal nature, or symptoms have improved and recurred with drug removal and rechallenge," he says.

"Though large observational cohort studies have not shown an association between isotretinoin and suicide, a modest increase in risk cannot be excluded. To date, there is limited scientific evidence demonstrating a causal relationship between isotretinoin and suicide."