Approved But Not Proven: What's Up With FDA, Cancer Drugs?

Nick Mulcahy

October 22, 2015

In approving new cancer drugs, the US Food and Drug Administration (FDA) is now heavily relying on surrogate markers of effectiveness, such as tumor shrinkage, instead of proof that an agent improves survival, according to a new analysis.

The investigators conclude that this might be having a deleterious effect on patients, public health, and healthcare costs.

"Our results suggest that the FDA may be approving many costly, toxic drugs that do not improve overall survival," write Chul Kim, MD, MPH, from the National Cancer Institute (NCI), and Vinay Prasad, MD, MPH, from the Knight Cancer Center at the Oregon Health and Sciences University in Portland.

Their analysis showed that 36 of 54 (67%) cancer drug approvals from 2008 to 2012 were made on the basis of surrogate markers — either tumor response rate or progression-free survival, in about equal frequency.

But over a median follow-up period of 4.4 years, only five of those 36 drugs were shown, in randomized studies, to improve overall survival.

The team's research letter was published online October 19 in JAMA Internal Medicine.

Only two things matter in patient care: living longer and living better.

"Doctors must remember that only two things matter in patient care: living longer and living better. Surrogate end points address neither issue head on, and always require follow-up," Dr. Prasad told Medscape Medical News.

Lack of follow-up is part of the problem here.

Postmarketing studies with results are missing for about one-third of the drugs approved with surrogates. This means that the survival effect of 13 of the 36 drugs approved on the basis of surrogate markers is still untested or without reported results.

Enforcement of postmarketing studies is of "critical importance," Drs Kim and Prasad write.

Their study is timely, given Congressional debate on how the FDA approves drugs and medical devices. In July, the House passed the 21st Century Cures Act, which encourages the use of surrogate markers as one way to speed new drugs to market. Opponents of the legislation, however, say that its lax standards would permit unsafe and ineffective products to reach the marketplace. The Senate is deliberating its own version of the House bill.

Surrogate markers, especially tumor response rate, should not be used to approve cancer drugs, according to a critic of the approval process for cancer drugs in the United States.

It's outrageous that tumor shrinkage is a basis of approval.

"It's outrageous that tumor shrinkage is a basis of approval," said Diana Zuckerman, PhD, from the National Center for Health Research in Washington, DC, who was asked for comment.

"Cancer drugs are good at killing cancer cells," Dr. Zuckerman told Medscape Medical News. Thus, tumors respond to the toxic agents and die or shrink, which is typically captured by radiographic imaging and is reported as an overall rate in clinical trials. But the problem is that they also kill healthy cells, and overall survival might not improve.

"It's important to know if the patient will live longer," she said.

In their study, half the drugs (18 of 36) approved on the basis of surrogate markers were eventually found to not improve survival, Drs Kim and Prasad report.

Pharmaceutical companies have "no incentives" to speed up the survival discovery process, unless early signs of survival improvement are obvious, said Dr. Zuckerman, whose nonprofit think tank receives funding from individuals and foundations and focuses on the safety, efficacy, and quality of healthcare in the United States.

But the FDA has incentives to approve drugs on standards less stringent than improved survival, she noted.

"The FDA is trying so hard to please Congress and industry that they are approving drugs on the basis of flimsy evidence, and patients are being harmed," she said.

The FDA is trying so hard to please Congress and industry.

Agency budget is one of the motivators for the FDA to loosen its standards, Dr. Zuckerman explained.

The use of surrogates in cancer drug approval seems to be here to stay, she said, citing a personal conversation she had with an ex-FDA official. There is no real debate about using progression-free instead of overall survival. "That ship has sailed," she said, repeating her source's comments.

The NCI has also transitioned to emphasizing progression-free survival as a primary outcome in its funded studies, she observed.

However, some of the drugs approved using surrogates during the study period have been hailed as practice-changing and have led to dramatic improvements. For example, vismodegib (Erivedge), which was approved in 2012 for locally advanced or metastatic basal cell carcinoma on the basis of treatment response rate, was called "the greatest advance in therapy yet" in an editorial published that year in the New England Journal of Medicine (2012;366:2225-2226).

Dr. Zuckerman was not swayed. "Some patients have been helped by approvals using surrogates, but the vast majority have not," she opined. As Drs Kim and Prasad report, half of the drugs did not improve survival despite approval, and only five of 36 have been proven to improve survival, she added.

Crossover occurred in 11 of 36 trials (31%), which happens in a trial when an experimental treatment is found early on to be significantly more effective than standard treatment, the investigators explain. However, crossover did not differ between trials that found a survival advantage and those that did not (1 of 5 [20%] vs 10 of 18 [55%]; = .16).

Dr. Prasad said he believes that patients need to intervene on their own behalf to learn whether they are getting a truly valuable therapy.

"Patients with cancer should ask their doctor if the drugs they are offered have been shown to improve survival or quality of life or merely a surrogate. This information may be helpful as they try to balance the side effects and benefits of a treatment and make healthcare choices that are right for them," he said.

Dr Kim, Dr Prasad, and Dr. Zuckerman have disclosed no relevant financial relationships.

JAMA Intern Med. Published online October 19, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.