FDA Panel: Tolerating Approval Outliers Would Set ‘Dangerous Precedent’

An unprecedented wait for final FDA approval will drag on for two orphan lymphoma drugs, at least a little while longer, as the agency decides how to end the delays and prevent them from happening again.

The dihydrofolate reductase inhibitor pralatrexate (Folotyn) received accelerated approval in 2009 and the histone deacetylase (HDAC) inhibitor belinostat (Beleodaq) in 2014, both for adults with relapsed/refractory peripheral T-cell lymphoma (PTCL). As is often the practice for accelerated approval, both drugs received conditional approval on the basis of small single-arm trials with a primary endpoint of objective response rate (ORR), which was 26-27% in both studies. Final approval was contingent on one or more confirmatory clinical trials.

After multiple delays, Acrotech Biopharma now anticipates finishing a phase III multi-arm trial involving both drugs by February 2030.

Tasked with providing the FDA with guidance on the situation, members of the Oncologic Drugs Advisory Committee (ODAC) expressed mixed feelings about what they heard during a half day of presentations and questioning. On the one hand, they almost universally agreed that the delay since accelerated approval is unacceptable, regardless of the explanations. However, they also were uneasy about the possibility of withdrawing one drug, much less two, for a disease that already has limited treatment options.

“In terms of the delays in the post-approval confirmatory trials, and whether the current plan to verify the clinical benefit is reasonable … I think the consensus of the advisory committee is that we have significant concerns about the very prolonged delay in getting these confirmatory studies underway,” said ODAC acting chair Andy Chen, MD, of the Knight Cancer Institute at Oregon Health & Science University in Portland. “I think we have concerns about the dosing and whether or not these are the appropriate studies to be doing, whether there should be an additional study in a subset of T-cell lymphoma.”

“We would like the FDA and sponsor to strategize about other possible ways to have a shorter study readout than waiting another 7 years from now, which would essentially be 20 years from the initial approval of pralatrexate,” he added.

The approval landscape has changed since pralatrexate and belinostat received accelerated approval, Chen continued. In particular, the FDA now requires that a sponsor have confirmatory studies underway prior to accelerated approval. That change alone has been shown to reduce the timeline to final approval by 4 years (from 7.3 to 3.1).

Few Answers

The committee had few recommendations about how to move the process along faster. One member suggested conducting smaller studies with a higher threshold for clinical benefit, which could help with accrual for a trial involving an uncommon disease.

“It’s a bit of a mixed bag. We do not have a direct answer for this,” said Chen. “I would like to note that the question for this committee from the FDA is different from the typical ODAC. We are not being asked to approve or revoke approval. That is not on the agenda.”

Issues related to slow patient accrual, dosing, toxicity, and dose optimization, among others, contributed to delays in meeting several deadlines. Along the way the FDA requested additional studies, and both drugs changed hands more than once as a result of sales and acquisitions.

Current sponsor Acrotech Biopharma plans to conduct a two-step phase III trial involving both drugs. The first step will address dose optimization, and the second step will be a randomized comparison of each drug plus chemotherapy versus chemotherapy alone. The company anticipates completing the first step by October 2025. The clinical benefit study will begin 2 months later and has an anticipated completion date of February 2030, including 2 years of follow-up.

Pralatrexate and belinostat have the longest pending approvals since the accelerated approval program began in 1990. In that time, the number of accelerated approvals increased dramatically from 38 in the 1990s to 121 in the 2010s. Thus far, 89 accelerated approvals have been granted in the 2020s, according to data presented by the FDA. Oncology has accounted for 60% of all accelerated approvals.

Among 187 accelerated approvals, 65 are ongoing and 96 have progressed to traditional approval, a median of 3.1 years after accelerated approval. Indications have been withdrawn 26 times, a median of 4.1 years after accelerated approval. More than 80% of ongoing oncology accelerated approvals have been ongoing for less than 5 years.

The U.S. might learn from other countries’ approach to accelerated, or expedited, approval. England, the European Union, Australia, and Switzerland have a renewal process for expedited approval. Switzerland has a 2-year maximum on expedited approval and Australia has a 6-year limit.

‘Dangerous Precedent’

Late last year, Congress passed the Food and Drug Omnibus Reform Act, giving the FDA additional tools to keep accelerated approvals moving toward final approval. The FDA can now require sponsors to have confirmatory trials underway prior to granting accelerated approval. Sponsors must submit biannual progress reports, and a streamlined withdrawal process has been created for drugs that do not verify clinical benefit.

Speaking for the sponsor, lymphoma specialist Owen A. O’Connor, MD, PhD, of the University of Virginia Comprehensive Cancer Center in Charlottesville, emphasized the logistical issues for evaluating potential therapies for PTCL.

“There are about 10,000 to 15,000 cases per year in the United States,” he said. “Even the most common subtype might have only 2,000 to 3,000 cases per year. There are about 15,000 medical oncologists in the United States, so this means that each oncologist could expect to see a case of PTCL about once a year, if the cases were spread out evenly.”

PTCL is remarkably heterogeneous, O’Connor continued. The 2022 World Health Organization classification system recognized 36 distinct subtypes of PTCL, including many indolent subtypes.

“Most forms of PTCL are considered highly aggressive diseases,” he said. “In essence, the PTCLs are 36 orphan diseases lumped under an orphan disease. These features all conspire to make the conduct of clinical trials, let alone randomized clinical trials, exceedingly difficult.”

“Importantly, frontline conventional chemotherapy programs are not highly effective, and there is no unified standard of care,” O’Connor said. “CHOP-based chemotherapy is often regarded as the standard of care. CHOP was developed in patients with aggressive B-cell malignancy, so a radically different disease.”

Unlike B-cell lymphomas, PTCL outcomes remain poor, he added.

FDA and ODAC participants at the meeting acknowledged the unmet therapeutic need in PTCL, but several participants expressed skepticism that Acrotech can complete the confirmatory process in the current timeline, given the history of delays.

“The development of this drug has been, for lack of a better word and not to be overly critical, let’s just say suboptimal, and I’m being kind by using suboptimal rather than other words here,” said Richard Pazdur, MD, director of FDA’s Center for Oncology Excellence. “I don’t want be back here years later with a negative trial and be in the same situation.”

“Because this drug has not been developed, could the company have a more robust program for the determination of clinical benefit?” Pazdur continued. “Do two trials, one in the relapsed and refractory setting. It could be a very simple trial…. We’ve seen that multiple times in solid tumors where there are very few effective therapies.”

ODAC member Toni Choueiri, MD, of Dana-Farber Cancer Institute in Boston, was less charitable, expressed concern about setting a “dangerous precedent to have such outliers.”

“I think Dr. Pazdur was quite kind in mentioning the word ‘suboptimal’ in the development,” said Choueiri. “I will be maybe one level less kind. I would say the development of this drug has been sloppy, and that is being somewhat kind. There have been perhaps many justifications why, [but] no matter what there could have been at least one randomized study.”

Please enable JavaScript to view the

comments powered by Disqus.