FDA's Accelerated Approval Pathway Is Broken

When We Reward Failure, Why Should We Expect Success?

Bishal Gyawali, MD, PhD


August 04, 2021

During a 3-day meeting in April, the FDA's Oncologic Drugs Advisory Committee (ODAC) debated six so-called "dangling" accelerated approvals — drugs that initially received accelerated approval but have since failed to improve clinical outcomes in confirmatory trials.

Ultimately, the ODAC committee voted to keep four of the six "dangling" approvals on the market.

I will be honest: I was upset with the outcome, for two reasons. First, the ODAC meeting seemed like a way to legitimize not taking any action against these indications. If I am supposed to pass an exam to get my license to practice medicine but I fail, the licensing body does not hold a meeting to vote on whether I should be allowed to practice anyway. It would not be fair for those who actually passed the test, and it would also undermine the integrity of the test.

In a similar manner, the ODAC's decision to maintain marketing approval for cancer drugs that failed to improve outcomes in confirmatory trials undermines the accelerated approval pathway. It is also, in a sense, "unfair" to the drugs that have actually demonstrated that they improve outcomes in clinical trials. The withdrawal of accelerated approval when the confirmatory trials fail should be automatic, as we have recently argued.

I find solace in the fact that the FDA is not bound to abide by the ODAC vote and could still revoke the approvals. But I am aware that I'm being naively optimistic. Take, for example, the recent approval of aducanumab for Alzheimer's disease, in which the FDA approved a drug despite the drug advisory committee's recommendation against it.

But my real concern is the discussion that happened during the ODAC meeting. I realized that some of the core concepts we talk about in cancer drug development are either not understood or are being ignored.

Both scenarios should make all of us as oncologists concerned.

ODAC Talking Points

These are some of the talking points that ODAC committee members put forward to argue why the approval status for these drugs should be maintained despite their failure to improve outcomes in confirmatory trials.

The most common was "unmet need," a term to describe a lack of other treatment options in a particular disease setting. But this argument doesn't hold water for two reasons.

First, this is simply false. For the four indications in which the committee voted to maintain approval, there was no unmet need. Take atezolizumab in triple-negative breast cancer. There is another checkpoint inhibitor, pembrolizumab, approved in the same setting with more robust data. Atezolizumab is not filling any void here.

Next, pembrolizumab for liver cancer. Atezolizumab (plus bevacizumab) has solid overall survival (OS) data in hepatocellular carcinoma, so there is no need for another immunotherapy that has failed to improve OS. I'd even argue that giving pembrolizumab instead of atezolizumab in this context is potentially harmful for patients.

The other two cases involved both atezolizumab and pembrolizumab for urothelial cancer; the mere existence of one negates the "unmet need" argument for the other.

But even if there is a genuine "unmet need" in a particular setting, these drugs did not meet the standard of improving survival. An unmet need doesn't imply that the treatment void should be filled with a drug that provides nothing of value to patients. When we talk about an unmet need, we are speaking of drugs that provide a clinical benefit; any true unmet needs will continue to exist despite maintaining these approvals.

The other argument is even more bizarre — the "no harm" argument. During the ODAC meeting, committee members argued for maintaining a drug's approval because there was no signal of harm nor a suggestion of detriment in survival.

To be clear: We prescribe drugs in oncology to improve survival. Prescribing drugs because they do not worsen survival is against all principles of oncology and medicine. How do you even take an informed consent from the patient? I will never be able to counsel a patient to take a medication with a litany of adverse events, potentially serious and even fatal, because the medication won't worsen their survival.

Here's another strange argument I heard during the meeting: Approval should be maintained because the National Comprehensive Cancer Network and other guidelines recommend the treatment. This is strange because it is supposed to be the other way around: These drugs are put in the guidelines because the FDA approved them. The FDA doesn't wait and see what drugs are put into guidelines and then subsequently approve them on that basis. This is without even addressing the conflicts of interest and other sources of bias that affect guidelines and have been extensively documented in the literature.

Others suggested that we need to run more trials for the same drug and simply wait until one is positive. Put another way, this is arguing that the results of the confirmatory trials presented were false negatives.

There are multiple problems with this argument. For one, you can just as easily argue that a positive trial is probably a false positive. But more importantly, if you run a number of trials, one of them will eventually be positive simply by chance alone. Our goal as cancer researchers shouldn't be to get a positive trial by any means necessary; we should be trying to understand whether these drugs provide any meaningful clinical benefit. And in the absence of positive data and the presence of negative data, it is not ethical to continue using a drug hoping that one day a future trial will turn out to be positive. If this is how we practice oncology, the industry will be incentivized not to run additional trials in the future or to delay them as long as possible.

Bear in mind that these confirmatory trials were run with the consultation of the FDA. We cannot fault the trial's design simply because the results did not go the way they were expected.

I hope that policymakers and all stakeholders — including physicians, patients, and caregivers — recognize the faulty reasoning that led to favorable ODAC votes for these cancer drugs. This is not simply about the four indications maintained last April; it is about the principles underlying our accelerated approval pathway for cancer drugs.

Bishal Gyawali, MD, PhD, is an associate professor in the Departments of Oncology and Public Health Sciences and a scientist in the Division of Cancer Care and Epidemiology at Queen's University in Kingston, Ontario, Canada, and is also affiliated faculty at the Program on Regulation, Therapeutics, and Law in the Department of Medicine at Brigham and Women's Hospital in Boston. His clinical and research interests revolve around cancer policy, global oncology, evidence-based oncology, financial toxicities of cancer treatment, clinical trial methods, and supportive care. He tweets at @oncology_bg.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.