Analysis of Potential Safety Issues

With unsurpassed integrity and extensive experience in evaluating a wide range of potential and actual safety issues, you can depend on Wendy Stephenson & Associates (WSA) to provide expert advice from a medical, epidemiological and regulatory perspective. Analysis of ad hoc safety issues includes:

• Evaluation and assessment of safety data to determine the strength of the scientific evidence and provide advice on what, if any, actions are warranted (labeling changes, initiation of studies, etc.)
• Recommendations for potential sources of data, expertise and/or studies to further evaluate the significance of a safety signal
• Preparation of expert report with thorough analysis of the safety issue(s) and summary of recommendations.

Below are examples of safety issues and the approaches taken to better understand them:

Statins / lens opacities
Bisphosphonates / esophageal erosion
Drug / liver function
Varicella vaccine / secondary transmission
Antipsychotic / aplastic anemia
Rotavirus vaccine / intussusception
TNF antagonists / demyelination
Alzheimer’s vaccine / encephalitis

Statins / lens opacities
Before the approval and licensing of the first statin, there was some concern about the possibility of development of lens opacities associated with this new cholesterol-lowering therapy. The concern was based on the observation that administration of high doses to dogs resulted in the development of subcapsular lenticular opacities. Although the early clinical trials for the first statin included a series of ophthalmologic exams, the results were inconclusive, largely due to the lack of precision in the measurement of changes in the lens. Hence the initial product information included a warning about the possibility of lens opacity formation. A large post-marketing clinical trial of 8000 patients including highly specialized, objective measurements of lens changes was conducted. When the results showed conclusively that the drug did not cause lens opacities, the warning was removed from the label.

See also: Laties AM, Shear CL, Lippa EA, Gould AL, Taylor HR, Hurley DP, Stephenson WP, Keates EU, Tupy-Visich MA, Chremos AN: Expanded Clinical Evaluation of Lovastatin (EXCEL) Study Results II. Assessment of the Human Lens after 48 Weeks of Treatment with Lovastatin. Am J Card 67:447-453, 1991.

^ back to top

Bisphosphonates / esophageal erosion
Although the chemical nature of the compound would be expected to irritate the esophageal mucosa, the final tablet presentation and dosing instructions were designed to minimize such an effect. There was no indication of a serious problem in pre-approval trials of the first bisphophonate. Within the first year of marketing, reports of serious esophageal complications were received by the company. A thorough and detailed survey of patients revealed that problems tended to be associated with failure to follow dosing instructions. Education of health professionals and their patients was accomplished through letters to health care professionals, stronger warnings in the label and patient information leaflet and publication in a peer-reviewed journal.

See also: DeGroen PC, Lubbe DF, Hirsch LJ, Daifotis A, Stephenson WP, Freedholm D, Pryor-Tillotson S, Seleznick MJ, Pinkas H, Wang KK: Esophagitis Associated with the Use of Alendronate. NEJM 335(14): 1016-21, 1996.

^ back to top

Drug / liver function
A drug had been on the market for several years in several countries when one regulatory authority suggested that the product information include a recommendation for periodic liver function test monitoring. A thorough analysis of all available information was conducted. Although there had been sporadic post-marketing reports of elevated liver enzymes, there was no conclusive evidence of a causal association. The company’s very large clinical trial database of thousands of patients was reanalyzed with special attention to time-dependent patterns of liver function test results. The prevalence of enzyme elevations in the clinical trial database, using various cut-off points, was comparable to the general population based on data from the National Health and Nutrition Evaluation Survey (NHANES). An assessment of the impact of screening for liver enzyme elevations, when any elevations detected through such screening would more likely be due to random variation or another unrelated cause, successfully convinced the regulatory authority to withdraw its advice.

See also: Pratt DS, Kaplan MM. Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 342:1266-71, 2000.

^ back to top

Varicella vaccine / secondary transmission
Since varicella vaccine contains a live attenuated virus, when the vaccine first became available there was some concern about the possibility of transmission of the vaccine virus to other individuals; hence, the early recommendation to avoid contact with pregnant women and immunocompromised patients. When reports of possible secondary transmission were received, the company was interested in determining the actual extent of transmission and its significance. Since the vaccine strain and the wild strain were easily distinguished by polymerase chain reaction (PCR) analysis, a protocol was developed to obtain and test samples. The same protocol was used to test available samples when serious adverse reactions were reported. For example, a case of fatal pneumonia occurring in a recently vaccinated child was shown to be due to the wild form of the virus and not the vaccine. The laboratory studies added considerably to the company’s ability to monitor the safety of this new vaccine.

See also: Sharrar R, LaRussa P, Galea S, Steinberg S, Sweet A., Keatley RM, Wells M, Stephenson WP, Gershon A: The Postmarketing Safety Profile of Varicella Vaccine. Vaccine (19): 916-923, 2001.

^ back to top

Antipsychotic / aplastic anemia
Shortly after launch in Europe, there were several reports of aplastic anemia among patients taking a new antipsychotic. The 8 reports of aplastic anemia among the estimated 400,000 patients treated with the marketed drug were clearly higher than the background rate of 2 to 5 per million. On the other hand, clinical trials suggested few other side effects that tend to limit the usefulness of other antipsychotic drugs. The companies that shared licensing rights worked together to assemble an advisory panel of experts in psychiatry, hematology and epidemiology, to evaluate the benefit-risk of the drug and provide guidance. Taking all of the available information and outside assessments into account, the companies together agreed to withdraw the product from the market.

See also: Laidlaw ST, Snowden JA, Brown MF. Aplastic anemia and remoxipride. Lancet 342: 1244-45, 1993.

^ back to top

Rotavirus vaccine / intussusception
Cases of intussusception were reported in pre-approval clinical trials of the first rotavirus vaccine. These were the subject of discussion at the FDA Advisory Committee Meeting prior to approval, and at that time considered to be a possible signal that warranted mention in the label, but with no conclusive evidence of causal relationship. When the first 15 cases spontaneously reported to the company or VAERS were determined to equal the number of cases one would expect if every case were known, distribution of the vaccine was suspended. The company’s post-marketing surveillance study at Kaiser Northern was the largest single cohort of children receiving the vaccine, but even that was too small to detect a significant difference in a rare event. Thus the CDC launched a national case-control study to better understand the relationship between the vaccine and intussusception. As additional reports continued to be reported, the company identified several sources of background rates for comparison. These were remarkably consistent and confirmed the expected number of 15. In the end there were approximately 100 confirmed cases. The question was whether the risk outweighed the benefit. A carefully designed benefit-risk analysis suggested that, at least in the US where rotavirus may be responsible for a large number of hospitalizations but rarely results in death or permanent disability, the benefits no longer outweighed the risks. The same model, applied to other circumstances or other regions of the world, might result in different conclusions.

See also: Suspension of rotavirus vaccine after reports of intussusception—United States, 1999. [Erratum appears in MMWR 2004 Sep 24;53(37):879]. MMWR. Morbidity & Mortality Weekly Report. 53(34):782-9, 2004 Sep 3.

^ back to top

TNF antagonists / demyelination
Several reports of demyelination in patients taking a TNF-α antagonist, in and of themselves, would not have been cause for concern. However, in the process of conducting a complete evaluation of the issue a literature report of potential significance was discovered. The published paper reported on the results of a phase II clinical trial of another TNF-α antagonist designed to evaluate its efficacy as a treatment for multiple sclerosis. The study demonstrated that the drug was not effective and, to the contrary, appeared to worsen disease progression in patients with multiple sclerosis. The drug that was the subject of the study was not developed further. While there was no evidence that TNF-α antagonism can cause de novo multiple sclerosis, there was some evidence that it may exacerbate a pre-existing condition. Based on this information, it was decided to add a warning to the core safety information for a marketed TNF-α antagonist.

See also: Arnason BGW, et al. (Lenercept Multiple Sclerosis Study Group). TNF neutralization in MS: Results of a randomized, placebo-controlled multicenter study. Neurology 53:457, 1999.

^ back to top

Alzheimer’s vaccine / encephalitis
During clinical trials of a novel approach to treating Alzheimer’s disease, the company’s safety committee met to review the significance of two reports of encephalitis. Since encephalitis is extremely rare, occurring at an incidence of less than 1 in 100,000 in the general population, two reports were cause for concern but it was not possible to draw any firm conclusions about an association with the vaccine. The decision was made to notify all investigators, regulators and ethics committees and to update the informed consent, but to continue the trials. Within a week there were two additional reports of encephalitis, triggering another safety committee meeting. This time the decision was made to halt all trials and draft a protocol for systematic follow-up and assessment. In the end there were 18 confirmed cases of encephalitis out of 298 vaccinated subjects. This is an example of the importance of having a clear mechanism for addressing safety issues on an urgent basis and for making the right decisions without delay.

See also: 4. Orgogozo JM, Gilman S, Dartigues JF, et al. Subacute meningoencephalitis in a subset of patients with AD after A[beta] immunization. Neurology: 61: 46–54, 2003.

^ back to top