Surrogate outcomes in clinical trials are deceptive and harmful says York U professor

Regulators often approve drugs without any hard evidence, relying on the belief that surrogate markers are accurate predictors of the health benefits of the drug being tested. However, researchers including York University Professor Joel Lexchin, say the contrary can be the case.

“Sometimes faith is misplaced and these drugs end up harming people more than helping. For example, clofibrate, a drug that lowers cholesterol, was approved because it was believed that lowering cholesterol would reduce the risk of heart disease. But in fact it ended up causing more deaths than it prevented,” says Lexchin of York’s School of Health Policy and Management, Faculty of Health.

JoelLexchinJoel Lexchin

“Surrogate markers have dangerously failed as guides to appropriate treatment because favourable effects on them do not automatically translate into benefits to health,” says Lexchin, who coauthored “Surrogate Outcomes in Clinical Trials: A Cautionary Tale”, published online in the Journal of the American Medical Association (JAMA)-Internal Medicine.

According to the authors, surrogate measurements are a routine part of clinical practice. Clinicians feel that these measurements provide them with a sense of understanding and control, and use them as shortcuts in making decisions.

“For example, using hemoglobin A1c, a surrogate marker, to monitor diabetes treatment means that its short-term effects are visible, but the true outcomes may be years later. This situation confuses the actual disease and its surrogate, so the surrogate evidence from clinical trials is easily accepted,” explains Lexchin.

The researchers have cited several drugs approved on the basis of how they affect surrogate markers, such as blood pressure, cholesterol and diabetes, that ended up causing harm. Their list of prescription drugs includes one for controlling blood pressure that causes congestive heart failure, another for lowering cholesterol levels that increases the risk for cardiovascular disease and breast cancer, and a third for diabetes that caused heart attacks. The harmful effects of these drugs were proven through clinical trials or meta-analysis that looked at real health outcomes.

Lexchin, who is also attached to the Department of Emergency Medicine at the University Health Network, coauthored the research article with Staffan Svensson in the Department of Pharmacology at the Sahlgrenska University Hospital in Sweden and David Menkes of Waikato Clinical School, University of Auckland in New Zealand.