FDA officials said the questions about accelerated approval of PD-1 and PD-L1 drugs have been largely resolved, thanks in part to guidance from the Oncologic Drugs Advisory Committee.
Last year, the agency focused on 10 indications for PD-1 and PD-L1 drugs that received accelerated approvals, but hadn’t demonstrated patient benefit in confirmatory trials. Nine of the 10 original accelerated approvals—dubbed “dangling indications” by the agency—were based on single-arm studies that produced low response rates.
If a clinical trial does not meet its endpoints, the drug should be evaluated in the context of the available treatments for the indication.
Richard Pazdur
An accelerated approval allows FDA to extend the benefit of doubt on an indication based on an intermediate endpoint judged as “reasonably likely to predict” patient benefit, and now—following review of PD-1 and PD-L1 drugs—the agency appears to have soured on accelerated approvals based on single-arm studies producing low response.
“Our evaluation of the PD-1 space and accelerated approvals showed that for this class of drugs, single arm trials with low response rates do not always reflect eventual long-term benefit and as such this type of data is unlikely to support future accelerated approvals for this class of drugs,” said Richard Pazdur, director of the Oncology Center of Excellence, and acting director of the FDA Office of Oncologic Diseases.
“As accelerated approval is not an incentive program for drug companies, when a confirmatory trial does not meet its endpoints, we need to look at the reasons for this and the treatment landscape to see if an unmet need still exists and another trial may be used as alternative confirmatory evidence,” Pazdur said to The Cancer Letter.
The next two meetings of ODAC—on Oct. 28 and Dec. 2—will review accelerated approvals of cancer drugs beyond the PD-1 and PD-L1 antagonists.
A table summarizing the outcome of the agency’s examination of accelerated approvals of PD-1 and PD-L1 drugs is available here.
“There are a small number of other dangling approvals for which we are in discussions with companies, and which will be addressed in the near future,” Julia Beaver, chief of medical oncology, Oncology Center of Excellence, and acting deputy director of the Office of Oncologic Diseases of the FDA Center for Drug Evaluation and Research, said to The Cancer Letter. “As recently announced, there will be ODAC meetings discussing melphalan flufenamide, panobinostat, and vincristine sulfate liposome injection this fall.
Pazdur and Beaver said the agency plans to post web pages that will list the accelerated approvals, the status of the confirmatory trials, and converted or withdrawn approvals.
FDA has been granting accelerated approvals since 1992, and the agency has conducted three evaluations of progress in confirmatory trials.
One highlight of that exercise 18 years ago was a single two-day meeting, where seven sponsors regaled ODAC with dispatches from their ongoing confirmatory trials (The Cancer Letter, March 10, 2003; Feb. 11, 2011).
Over the past six years, the agency has approved 86 indications for checkpoint inhibitors, drugs that target the PD-1/PD-L1 proteins. As of August 2021, the agency has rolled the dice on 38 of these indications, giving them accelerated approvals.
Then, last year, as the results of confirmatory trials required for maintaining accelerated approvals rolled in, no benefit was shown in 10 of these indications.
Starting late last year, FDA engaged in what might be called quiet diplomacy, convincing four drug companies in as many months to voluntarily yank four dangling indications (The Cancer Letter, March 12, March 5, 2021).
Then, the remaining indications were thrown to ODAC, which held a three-day meeting April 27, 28, and 29, which reviewed six dangling indications (The Cancer Letter, April 30, 2021).
PD-1/PD-L1 drugs often get accelerated approvals based on tumor shrinkage and are allowed to keep the indication while conducting confirmatory trials, which usually focus on progression-free or overall survival. Since a massive number of trials are focused on PD-1/PD-L1 drugs, unmet medical need can evaporate while the trials are being conducted.
After a June meeting, ODAC demonstrated its reluctance to accept low responses in single-arm trials as a basis for accelerated approvals for PD-1 and PD-L1 drugs. On June 24, the committee voted 13-4 against approval of retifanlimab, a PD-1 inhibitor, for the treatment of squamous carcinoma of the anal canal (The Cancer Letter, June 25, 2021).
Beaver and Pazdur spoke with Paul Goldberg, editor and publisher of The Cancer Letter.
Paul Goldberg: Could we talk about accelerated approval in general—how this mechanism is being used by FDA, particularly in oncology? How has this regulatory mechanism been used in evaluation of anti-PD-1 antibodies approved under accelerated approval?
Julia Beaver: Oncology has used the accelerated approval pathway extensively, and we have learned a great deal. The majority of accelerated approvals have been in oncology, compared to other disease areas, with over 85% in oncology in the last 10 years.
When a confirmatory trial does not meet its endpoints, we need to look at the reasons for this and the treatment landscape to see if an unmet need still exists and another trial may be used as alternative confirmatory evidence.
Richard Pazdur
Overall, we have more than 165 indications approved in oncology using the accelerated approval pathway, with almost half converted to regular approval in a median of three years. Less than 10% of the indications have been withdrawn, and the majority of the remaining indications were approved in the last three years, with ongoing clinical trials.
Richard Pazdur: With the large number of anti-PD-1 antibody approvals in the last seven years and now more complete picture of this class of drug, we conducted an evaluation of the accelerated approvals for these agents and identified 10 “dangling” accelerated approval indications.
These were situations where the confirmatory trial or trials had not verified clinical benefit and the accelerated approval remained on the market. We then examined the evidence in support of the original accelerated approval in the context of the treatment landscape, paying particular attention to other approvals for the same class of drug that may now be accepted standards of care.
I have said before that a failed clinical trial does not mean a failed drug. However, if a clinical trial does not meet its endpoints, the drug should be evaluated in the context of the available treatments for the indication.
You coined the phrase “dangling indications.” Some of these dangling approvals were withdrawn, some were taken to an advisory committee meeting. Can you go over the thought process behind convening the ODAC meetings in April 2021?
RP: First, I want to note that four indications were withdrawn prior to the ODAC meetings through communication with the companies and an understanding that for those, the treatment landscape had moved on.
These were for two indications in refractory small cell lung cancer and two indications in refractory bladder cancer. We brought the remaining six forward to discussion, bringing up particular points, including the reevaluation of treatment landscape as well as the low response rate seen for many of the drugs.
Can you describe the advice you received from the ODAC, lessons learned?
JB: In general, the ODAC votes were consistent with unfavorable votes for the indications with changes in treatment—for instance, in situations where another checkpoint inhibitor was approved in an earlier-line setting based on an overall survival benefit.
The votes in favor of retaining indications were not for an indefinite retention, but rather until other clarifying trials could be completed—taking into consideration the evidence of activity supporting the original accelerated approvals and the continued unmet need.
The committee also took the status of alternative confirmatory trials into account, voting more favorably when trials were near completion.
So, since you initiated this evaluation, seven of the 10 PDL-1 dangling indications have been either removed from marketing or planned for removal? What about the remaining three? Is anything still dangling out there?
RP: Yes, after the advisory committee meeting and the four prior withdrawals, three additional indications have been removed or announced for withdrawal.
The ODAC voted unfavorably for two of these. While one of these indications was given a favorable vote, the treatment landscape changed after the ODAC meeting, with another checkpoint inhibitor converting to regular approval and reporting survival benefits in the same indication.
JB: One of the three remaining indications, pembrolizumab for the treatment of patients with bladder cancer, was converted to regular approval after revising the indication to a narrower patient population with residual unmet need—those ineligible for any platinum-based chemotherapy—with confirmatory evidence from further response rate and duration data in that narrowed indication, and supportive overall survival improvement in a later-line setting.
The other two indications remain under discussion, one with a confirmatory trial that does not yet have final results, and one that recently reported favorable results.
How are you evaluating other therapies that may be dangling?
JB: There are a small number of other dangling approvals for which we are in discussions with companies, and which will be addressed in the near future.
As recently announced, there will be ODAC meetings discussing melphalan flufenamide, panobinostat, and vincristine sulfate liposome injection this fall.
RP: Ultimately, these evaluations and our follow-up will preserve the integrity of the Accelerated Approval Program. In addition, we will be posting web pages that will make information on our accelerated approvals more transparent. This will list the approvals, the status of the confirmatory trials, and converted or withdrawn approvals.
This will be fun reading. I, for one, will enjoy it. When will you make this content available?
JB: We hope the web pages will be posted in a matter of weeks. In addition, our clinical reviewers write manuscripts on many of our oncology approvals, including accelerated approvals, so that the cancer field can learn more about our risk-benefit analysis for each product and indication.
These are published in Clinical Cancer Research, The Oncologist, and other journals.
What are the lessons learned for future accelerated approvals? What should sponsors expect?
RP: Our evaluation of the PD-1 space and accelerated approvals showed that for this class of drugs, single arm trials with low response rates do not always reflect eventual long-term benefit and as such this type of data is unlikely to support future accelerated approvals for this class of drugs.
In general, the ODAC votes were consistent with unfavorable votes for the indications with changes in treatment.
Julia Beaver
As accelerated approval is not an incentive program for drug companies, when a confirmatory trial does not meet its endpoints, we need to look at the reasons for this and the treatment landscape to see if an unmet need still exists and another trial may be used as alternative confirmatory evidence.
JB: In addition, our evaluation has shown that the rapid conduct of confirmatory trial or trials is critical, and we are going to be working with companies from the initial drug development planning stages to come up with more comprehensive development plans if accelerated approval seems appropriate to encourage proactively initiated confirmatory trials.
