Embargoed Until:
August 5, 2008
12:00 a.m. Eastern Time



Christopher Fleming

Enlarging Pre-Approval Clinical Trials Could Be
A Cost-Effective Way Of Improving Drug Safety

Bethesda, MD -- Requiring larger pre-approval clinical trials could be a cost-effective method of reducing post-approval injuries and deaths caused by new medications, according to a study published today on the Health Affairs Web site: http://content.healthaffairs.org/cgi/content/abstract/hlthaff.27.5.w360

Drug safety has been a much-discussed topic recently, but most of the attention has focused on strengthening post-marketing surveillance of prescription drugs. In their paper, researchers from Duke University point out the potential for strategies at the pre-approval stage -- before the Food and Drug Administration has cleared the drug -- to detect adverse events caused by new medications, known as "adverse drug events"(ADEs).

"Most industry-sponsored pre-approval clinical trials are designed to detect differences in efficacy -- in other words, to get the drugs approved. Detection of ADEs, if it happens at all, occurs as a byproduct," said lead author Shelby Reed, an associate professor in medicine at the Duke Clinical Research Institute. "We believe the ability to detect adverse events should be an explicit factor in designing clinical trials, and in our paper we provide a model for evaluating the trade-offs between higher costs and improved detection of ADEs. Pre-approval trials can't be the whole story when it comes to drug safety, but they can be a crucial part of an overall strategy."

How The Model Works: The COX-2 Inhibitor Example

The model put forward by Reed and coauthors examines how increasing the size of a clinical trial affects the odds of detecting an ADE. In general, the researchers say, the effect of increasing the number of patients in the trial depends on the "background" frequency of the ADE in the general population and the increased odds of encountering the ADE among those taking the drug (the "odds ratio").

A recent drug safety controversy focused on pain relievers known as COX-2 inhibitors, which caused cardiovascular and cerebrovascular symptoms that pre-approval clinical trials had failed to detect. Using their model, the Duke researchers found that a clinical trail with 2,000 patients in each treatment arm would be expected to detect the ADEs associated with COX-2 inhibitors 76 percent of the time. Doubling the size of the trial to 4,000 patients in each treatment arm would increase the detection rate to 96 percent, but a further increase to 8,000 patients per arm would provide little additional detection power, the researchers found.

When they plugged in the cost of running a larger clinical trial, Reed and colleagues found that increasing the size of a pre-approval trial from 2,000 to 4,000 patients per treatment arm could be a cost-effective method of avoiding the ADEs. The incremental cost was approximately $27,100 per life-year saved.

"The assumptions used in our model were based on risks associated with COX-2 inhibitors, but this is just one example. For any given drug, research sponsors and regulatory agencies should identify potential ADEs of interest, as well as the odds ratios at which the drug's side effects might be seen as outweighing its benefits. Pre-approval trials can then be designed with enough patients to be able to reliably detect the ADEs at those odds ratios," Reed said.

Reed and colleagues note that the point at which ADEs might outweigh a given drug's benefits will depend on factors such as the severity of disease being treated, the number of people suffering from that condition, and the availability of alternative therapies. The authors also note that the ability of clinical trials to detect ADEs can be strengthened by including patients at high risk for those ADEs, as well as increasing the size of the trials.

Garber: The Best Strategy For Evaluating New Therapies Will Vary From Drug To Drug

In a Perspective on the article by Reed and coauthors, Alan Garber calls the work of the Duke researchers "a step toward the comprehensive, formal analysis needed to make rational decisions about drug safety." He also stresses the need for a broad focus that considers the cost of delaying a drug's availability and recognizes the improvements that have occurred in the rigor of post-marketing drug-safety monitoring. http://content.healthaffairs.org/cgi/content/abstract/hlthaff.27.5.w371

Garber, a staff physician at the Veterans Affairs Palo Alto Health Care System and the director of the Center for Health Policy and Center for Primary Care and Outcomes Research at Stanford, observes that the appropriate strategy for evaluating a medication's risks and benefits will vary from drug to drug. "The amount and type of data that should be collected to determine safety and efficacy of a COX-2 inhibitor that will often be used off-label -- when the drug is less likely to have a favorable risk-benefit profile -- will almost certainly be different than for a second-line therapy for a lethal cancer."

The article by Reed and coauthors will be available when the embargo lifts, at http://content.healthaffairs.org/cgi/content/abstract/hlthaff.27.5.w360.

Garber's Perspective will be available at http://content.healthaffairs.org/cgi/content/abstract/hlthaff.27.5.w371.


Health Affairs, published by Project HOPE, is the leading journal of health policy. The peer-reviewed journal appears bimonthly in print with additional online-only papers published weekly as Health Affairs Web Exclusives at www.healthaffairs.org. The full text of each Health Affairs Web Exclusive is available free of charge to all Web site visitors for a two-week period following posting, after which it will switch to pay-per-view for nonsubscribers. Web Exclusives are supported in part by a grant from the Commonwealth Fund.


©2008 Project HOPE–The People-to-People Health Foundation, Inc.