The Need for Post-Marketing Surveillance of Drugs

When a new drug is approved and enters the marketplace, often the only safety and efficacy information available is based on a few thousand people who took the drug during strictly controlled clinical trials. Not surprisingly, these trials are designed to focus on the drugs’ benefits, and may not include a large enough sample size to elicit serious adverse effects. Once the drug is available for widespread use, we are able to better evaluate the real safety profile of the drug.

A recent study published in JAMA evaluates the issue of post-marketing surveillance related to biologic agents. Biologic agents are a relatively new class of drugs, and include recombinant products, monoclonal-antibody-based products, and recombinant vaccines. Recombinant insulin was the first biologic agent approved in the US in 1982, and now there are more than 250 agents that have received approval. Many of these agents are invaluable in cancer therapy today.

However, since 1995, nearly 25% of biologic agents approved in the US and Europe have had safety-related regulatory actions issued against them. More than 1in 10 of the agents had black-box warnings placed on the prescribing information. Most safety problems were related to immunomodulatory effects, and more than half of the safety issues involved system organ classes, including infections, immune system disorders, and benign and malignant neoplasms.

From January 1995 to June 2007, 174 biologic agents were approved for use in the US and Europe. A total of 41 agents received safety-related regulatory actions, including 46 “dear healthcare professional” letters, 17 direct healthcare professional communications, and 19 black box warnings. The first-in-class agents that obtained approval were nearly 4 times as likely to receive safety-related regulatory actions versus agents receiving later approval. The agents most likely to receive a black-box warning in the US included monoclonal antibodies, cytokines, enzymes, growth factors, hormones, and interferons.

Adverse events that may trigger regulatory action from the FDA may occur as infrequently as 1 in 50,000 or more. Also, many adverse events are not seen initially, but appear with a drug’s long-term use. Last, once a drug enters the market, it can be used for off-label purposes, which were never studied in clinical trials to evaluate safety and efficacy. For these reasons, post-marketing surveillance is paramount to making drug safety a top priority.

The data for post-marketing surveillance comes from voluntary reporting systems. Healthcare providers, consumers, and pharmaceutical companies submit nearly 500,000 reports of adverse events to the FDA each year. Most of the reports come from healthcare providers, but 90% of the reports are routed through the drug manufacturer before being sent to the FDA. The FDA estimates that the submitted reports represent only 10% of the actual adverse events experienced. There is no system in place for the FDA to track how many people are taking a particular drug or how it is being used.

In addition to its insufficient voluntary reporting mechanisms, the FDA lacks federal authority to require pharmaceutical companies to conduct any post-marketing surveillance. They can request studies, but there is no consequence for noncompliance. If the FDA calls for additional warnings, labeling changes, patient education, patient registries, or restricted advertising, there is no procedure for enforcement.

The current JAMA article underscores the need for more transparency in the oversight of drug approval and drug safety. Many politicians and independent organizations have suggested the formation of autonomous safety oversight boards, or the development of a pharmacist-led adverse event reporting system, but none of these proposals have gained traction to date. Currently, the best scenario involves increased evaluation of new drugs during the approval process to better predict possible adverse events that may occur, as well as increased monitoring in the post-marketing period. Even with the safety of a handful of agents possibly in question, however, people are still better off with most of these drugs than without them.

References

T. J. Giezen, A. K. Mantel-Teeuwisse, S. M. J. M. Straus, H. Schellekens, H. G. M. Leufkens, A. C. G. Egberts (2008). Safety-Related Regulatory Actions for Biologicals Approved in the United States and the European Union JAMA: The Journal of the American Medical Association, 300 (16), 1887-1896 DOI: 10.1001/jama.300.16.1887

S. Okie (2005). Safety in Numbers — Monitoring Risk in Approved Drugs New England Journal of Medicine, 352 (12), 1173-1176 DOI: 10.1056/NEJMp058029