Making cancer clinical trials more relevant

When a new cancer drug improves survival in a clinical trial, it too often remains unclear whether the drug will be effective in clinical practice, asserts a provocative commentary (co-authored by an OHSU physician) that’s making headlines.

Figure: Composition of National Cancer Institute cooperative group clinical trials in 2002 (OHSU/Joe Rojas-Burke)

The problem stems from the fact that sponsors of clinical trials largely enroll subjects who represent only a part of the cancer population. Clinical trial participants are disproportionately white and they are younger and healthier than the majority of people with cancer. Only about 20 percent of  subjects in pivotal trials from 1992 to 2002 were older than 70, compared with 46 percent of real-world cancer patients. (After adjusting for age, cancer type, and sex, enrollees were 24 percent less likely to be black than white).

It’s clear that older patients experience more severe adverse effects of treatment and are more likely to stop therapy or undergo dose reductions that compromise efficacy. And “at a certain point the balance between benefit and harm tips toward more harm than benefit,” write the authors, Sham Mailankody of Memorial Sloan Kettering Cancer Center  and Vinay Prasad, M.D., M.P.H., an OHSU Knight Cancer Institute hematologist-oncologist and assistant professor of medicine.

Prasad and Mailankody offer two solutions:  The FDA could demand that data supporting the efficacy of novel cancer drugs come from trials including a more realistic range of patients for age, comorbidities, social status, and sex.

Or alternatively, the FDA could deem that overall survival in idealized trials is a surrogate end point, comparable to other surrogates, such as response rate and progression free survival:

“Thus, instead of providing traditional (full) approval to drugs such as sorafenib—a status that seldom entails further postmarketing efficacy studies—they may instead grant accelerated approval to these agents. Accelerated approval would empower the FDA to demand and enforce postmarketing studies that may confirm or refute the drugs’ benefit on survival in real-world populations. This policy would benefit patients with cancer as well as physicians, as considerable data relevant for clinical practice would be generated, and ineffective drugs could later be removed from the market.”

You can read the full commentary online at JAMA Oncology.


Overall Survival in Cancer Drug Trials as a New Surrogate End Point for Overall Survival in the Real World by Sham Mailankody and Vinay Prasad. JAMA Oncology (2016)

Participation in Cancer Clinical Trials: Race-, Sex-, and Age-Based Disparities by Vivek H. Murthy, Harlan M. Krumholz, and Cary P. Gross. JAMA (2004)