Checkpoint Inhibitor Changes Take Hold: Approval Standards Stir Debate

After 7 years of explosive growth, the development of PD-1/PD-L1 immune checkpoint inhibitor (ICI) therapy underwent a course correction last year, with the withdrawal of a range of indications due to study results that failed to reach thresholds for confirming clinical benefit. These withdrawals have had a muted impact on clinical practice so far, experts say, but that may change amid increasing scrutiny of the accelerated approval (AA) pathway that brought many of these drugs to patients.

Since 2014, the FDA has approved 7 ICIs, with more than 85 indications directed at the PD-1/ PD-L1 immune checkpoint, and development is continuing with over 2000 clinical trials registered for testing existing and emerging agents in this drug class. The race to provide these therapies across a spectrum of cancer stages and settings has resulted in a “Wild West of drug development,” according to 2 high-ranking FDA officials.¹

In a recent article in the New England Journal of Medicine, Julia A. Beaver, MD; and Richard Pazdur, MD, warned drug sponsors the FDA would no longer look favorably upon ICI approval applications based on single-arm clinical trials in patients with refractory disease.¹ Beaver is the chief of medical oncology in the FDA’s Oncology Center of Excellence (OCE) and acting deputy director in the Office of Oncologic Diseases in the Center for Drug Evaluation and Research. Pazdur, a 2019 Giants of Cancer Care® award winner for Community Outreach, is the director of the OCE.

They noted that approximately 45% of the indications for PD-1/PD-L1–directed ICIs have been approved through the AA pathway based on data from nonrandomized, single-arm trials, many of which have used overall response rate (ORR) and duration of response to predict clinical benefit.¹ However, clinical trial findings have shown that initial ORRs for this class of ICIs do not consistently predict long-term outcomes, Beaver and Pazdur said.¹ In March 2021, the FDA announced the agency was evaluating a group of approvals for ICIs that gained indications through the AA pathway but had not demonstrated clinical benefit in confirmatory trials.²

Of 10 dangling approvals that were evaluated, 9 used ORR as a primary end point for AA.³ The manufacturers withdrew 6 of these indications, including 4 before the Oncologic Drugs Advisory Committee (ODAC) met in April 2021 to consider their future status. Another indication that received a negative ODAC vote is scheduled to be withdrawn, whereas 2 indications that had positive ODAC votes remain under discussion. Meanwhile, an indication for pembrolizumab (Keytruda) in urothelial carcinoma, which ODAC backed, subsequently was converted to a regular approval for a revised patient population based on overall survival data from a similar clinical trial.¹

Moving forward, regulators may be less likely to approve applications based on ORRs. “For the anti–PD-1/PD-L1 antibody class of drugs, where we were seeing low response rates and variable correlation to outcome, we will not be accepting low response rate single-arm trials to support [AA] for this class of drug,” Beaver wrote in an email interview with OncologyLive^®. “For higher response rate trials with supportive duration, there is still a possibility of [AA].”

The agency is taking additional steps to improve the use of the AA pathway and other facets of cancer drug development, Beaver said. These efforts include Project Confirm, an OCE initiative that features searchable databases for oncology drugs approved through the AA pathway, including indications with verifiable clinical benefit and those that have been withdrawn.¹⁰

The project is intended to promote transparency about the AA pathway, Beaver said. The agency remains confident in the need for the AA process, which was introduced in 1992 to expedite drug development for patients with life-threatening unmet medical needs.³ “[AA] has been and will continue to be a very effective pathway for drug approval to bring promising therapies to patients years earlier,” Beaver said. “Even with the recent advisory committee meetings and withdrawals, there are still less than 10% of drugs approved under [AA] that have not verified benefit.”

Call for Reforms

The agency’s efforts come at a time when the AA pathway has drawn increasing scrutiny in the oncology community.¹¹ In April 2021, the Institute for Clinical and Economic Review (ICER) proposed several reforms, including stronger selection of surrogate end points, greater use of random-ized controlled trials, and increased enforcement of requirements to complete confirmatory trials. Although FDA officials have cited the relatively small number of withdrawn indications as a sign of the program’s success, the ICER report ques-tioned whether the standards for moving drugs to full approval require enough evidence and noted delays in conducting confirmatory studies.¹²

More than 60% of the 278 applications granted under the AA program since its inception have been for oncology indications.^13,14 Of 173 oncology AA indications, 89 (51%) have not been converted and 84 (49%) have been converted to full approv-als, according to a Friends of Cancer Research dashboard. The median time elapsed since the AA was granted is 1.8 years for indications not yet converted, whereas the median time for the converted indications to reach full approval is 3.1 years. The site lists 17 withdrawn AA indications as of January 31, 2022, including the dangling approvals for ICIs and other oncology drugs with diverse mechanisms. The median time from AA to withdrawal for these therapies is 3.8 years.¹⁴

The Friends group decided to create the dashboard, which is publicly available on the organization’s website, to enable a fuller understanding of the AA pathway, accord-ing to Jeff Allen, PhD, president and CEO of Friends of Cancer Research. “These programs have shown they have fulfilled the intention for which they were designed in terms of expediting new drug access to address serious life-threatening illnesses when there’s an unmet medical need,” Allen said in an interview with OncologyLive^®.

The data show, among other things, that it takes approximately 3 years to either confirm the bene-fit of a therapy or withdraw the drug, Allen noted. Studies that have not yet been completed are “not the same as a demonstrated absence of benefit,” he said. “The studies may be ongoing, or perhaps they have faced some difficulties in terms of being able to be completed or sufficiently accrued. That could be a very good focus point in terms of how to enhance these programs—what strategies can be deployed by the FDA and sponsors to expedite the development of the postmarket commitments to get important information about the long-term outcomes associated with these products.”

Single-arm studies play a vital role in oncol-ogy, helping to evaluate initial signals of activity and to determine initial dosing, Allen noted. At the same time, a randomized approach is needed in many cases to not only streamline drug devel-opment but also evaluate a therapy in an early disease state, he said. Once a drug receives an approval, enrolling participants to a subsequent randomized trial becomes particularly chal-lenging because patients want access to the new therapy, not the former standard.

The Friends group also has recommended several changes in the AA process, including requiring confirmatory trials be initiated in a premarket setting and making the withdrawal process “more nimble.”²⁰ Overall, Allen sees the increased focus on the AA pathway as a positive development. “As science changes, it’s good to continually evaluate the policy that helps guide it, and that’s the source of some of these [ongo-ing] discussions,” he said. “Hopefully these discussions will help optimize these programs so they can best support the science coming in the future.”

Implications for Clinical Practice

Although ICIs have become an important facet of the oncology treatment landscape, the with-drawals of indications so far do not appear to be hampering clinical practice, particularly in tumor types with multiple approved ICIs. In urothelial carcinoma, ICI therapy has been “a bit of a roller coaster” since the novel agents were first approved in the malignancy, according to Matthew Galsky, MD, codirector of the Center of Excellence for Bladder Cancer and associ-ate director for Translational Research, both at The Tisch Cancer Institute at Mount Sinai in New York, New York.

Both atezolizumab (Tecentriq), a PD-L1 inhibitor, and pembrolizumab were approved under the AA pathway in 2017 for the treatment of patients with locally advanced or metastatic urothelial carcinoma who are not eligible for cisplatin-containing chemotherapy.^21,22 Less than a year later, based on interim study data, the FDA limited both drugs in the frontline setting to cisplatin-ineligible patients whose tumors expressed PD-L1 or in those who are not eligible for any platinum-containing therapy regardless of PD-L1 status.⁶ After the ODAC hearing in April 2021, the FDA converted the pembrolizumab indication to a regular approval for patients who are not eligible for any platinum-containing therapy. The first-line indication for atezolizumab remains under discussion.^1,9

Meanwhile, Genentech and AstraZeneca voluntarily withdrew AA indications for atezoli-zumab and durvalumab (Imfinzi), respectively, for patients with urothelial carcinoma that progressed during or following platinum-containing chemotherapy or within 12 months of neoadjuvant or adjuvant treatment with platinum-containing chemotherapy.¹ Despite these changes, 4 ICIs retain indications in urothelial carcinoma (pembrolizumab, atezoli-zumab, nivolumab [Opdivo], and avelumab [Bavencio]).^23-26 “There were so many of these PD-1 and PD-L1 inhibitors approved, at least in patients who had progressed despite prior chemotherapy, that the removal of 1 or 2 [indications] hasn’t had a clinical impact because there [are] still other drugs in that same class avail-able,” Galsky said. At least part of 1 approval is retained in the frontline platinum-ineligible setting, he added.

“If there weren’t other drugs in the same class available, and there was the inabil-ity to prescribe these drugs at all, then that would take away a very important part of our treatment armamentarium,” Galsky said. “These drugs are incredibly effective for a subset of patients and would have lost their approvals based on the consequences of the way the [AA] pathway is set up.”

Galsky, who was a leading investigator on the pivotal phase 2 IMvigor210 study (NCT02108652) into frontline atezolizumab monotherapy,²⁷ said confirmatory trials are challenging to conduct. “They’re difficult to do in the exact same patient population because the drugs are already available. That certainly impacts whether patients would be willing to enroll on a study and investigator enthusiasm for enrolling patients on a study. Even if patients do [enroll], there’s the potential for contamination of the results as patients come off study and can get that treatment as standard care,” he said. “That created this complexity once you have the [AA]. How do you definitively prove the drugs work, because you can’t do it in the same patient population?”

Looking at AA indications from a broader scope, Galsky said the potential impact on clinical practice looms larger than immediate circumstances. “There is potentially an impact from the clinical science standpoint in that it makes us think about the consequences of the [AA] pathway as currently devised and the potential implications if we hadn’t had all those other drugs in the same class available—if this would have led to the inability to prescribe these drugs at all, to certain patient populations, which, based on the combination of data, are an important part of our treatment armamentarium,” he said.

In triple-negative breast cancer (TNBC), the changing ICI landscape also provides patients with other options. In August 2021, Genentech decided to voluntarily withdraw the AA indication for atezolizumab, even though the ODAC voted by a 7 to 2 margin to recommend the indication be retained pending additional trial data. Atezolizumab was approved through the AA pathway in March 2019 in combination with nab-paclitaxel (Abraxane) in patients with metastatic TNBC whose tumors express PD-L1 based on an improvement in progression-free survival in the phase 3 IMpassion130 study (NCT02425891). It was the first immunotherapy approved for breast cancer.^1,28

In announcing the decision, Genentech noted that the treatment landscape for TNBC had changed, and “the FDA no longer considers it appropriate to maintain the [AA].”²⁸ In July 2021, a month before the announcement, the FDA granted regular approval to pembrolizumab in combination with chemotherapy for patients with locally recurrent unresectable or metastatic TNBC whose tumors express PD-L1, a decision that converted an AA indication from November 2020. Pembrolizumab also received a regular approval for high-risk, early-stage TNBC in combination with chemotherapy as neoadjuvant treatment, then continued as a single agent as adjuvant treatment after surgery.²⁹

Both approvals are based on findings from the phase 3 KEYNOTE-522 (NCT03036488) in which 1174 patients with newly diagnosed, previously untreated, high-risk, early-stage TNBC were randomized 2:1 to receive chemotherapy with pembrolizumab or placebo. The main efficacy outcome measures were pathological complete response (pCR) rate and event-free survival (EFS). The pCR rate was 63% (95% CI, 59.5%-66.4%) for patients who received the pembrolizumab combination compared with 56% (95% CI, 50.6%-60.6%) for those in the placebo arm. The number of patients who experienced an EFS event was 123 (16%) and 93 (24%), respectively, in the pembrolizumab and placebo groups (HR, 0.63; 95% CI, 0.48-0.82; P = .00031).²⁹

The availability of pembrolizumab for this patient population provides an ICI option despite the withdrawal of the atezolizumab indication, noted Kevin Kalinsky, MD, MS, the Louisa and Rand Glenn Family Chair in Breast Cancer Research and director of Glenn Family Breast Center and Breast Medical Oncology at the Winship Cancer Institute of Emory University in Atlanta, Georgia.

Some patients would still be candidates for atezolizumab therapy because different assays are used to assess PD-L1 status for each drug. Atezolizumab was approved for patients with PD-L1–stained tumor-infiltrating immune cells covering greater than or equal to 1% of the tumor area. For pembrolizumab, PD-L1 positivity is defined as a combined positive score greater than or equal to 10. The Society for Immunotherapy of Cancer (SITC) guidelines recommend the VENTANA PD-L1 (SP142) assay be used for atezolizumab and the PD-L1 IHC 22C3 assay be used for pembrolizumab.³⁰

Although many patients meet the threshold for PD-L1 expression on both tests, atezolizumab remains an option for those who are positive on SP142 but not on 22C3, said Kalinsky, who was a member of the SITC guidelines panel. “For those patients, there are communications that can happen directly with the company, as well as directly with insurance companies, to see [whether] there can be coverage to patients who were previously on atezolizumab and responding,” he said. “We’ve been able to continue that therapy. If there [are] any barriers, we speak to the insurance company and/or Genentech directly.”

For practicing oncologists sorting out revised rules for using ICIs, keeping up with the changes can be challenging. However, Galsky said there are indications that community oncologists can navigate the shifting landscape, at least in urothelial carcinoma. He noted research f indings suggest practicing oncologists were able to adapt to changes in ICI therapy in 2018 after the FDA restricted atezolizumab and pembrolizumab use to patients with PD-L1–positive tumors.³¹

“That analysis showed there was a dramatic change in the prescribing of frontline immune checkpoint blockade to patients who were eligible before and after the FDA label change, suggesting that yes, indeed, the community can keep up with these changes in the context of [AA], that the data do disseminate, and prescribing on label is pursued,” Galsky said, who participated in the study.

References

1. Beaver JA, Pazdur R. The Wild West of checkpoint inhibitor development. N Engl J Med. 2021;10.1056/NEJMp2116863. doi:10.1056/NEJMp2116863
2. FDA in brief: FDA Oncologic Drugs Advisory Committee to review status of six indications granted accelerated approval. FDA. March 11, 2021. Accessed February 1, 2022. https://bit.ly/3cgG1bm
3. Beaver JA, Pazdur R. “Dangling” accelerated approvals in oncology. N Engl J Med. 2021;384(18):e68. doi:10.1056/NEJMp2104846
4. Withdrawn | cancer accelerated approvals. FDA. Updated January 18, 2022. Accessed January 31, 2022. https://www.fda.gov/drugs/resources-information-approved-drugs/withdrawn-cancer-accelerated-approvals 
5. Merck provides update on Keytruda (pembrolizumab) indication in third-line gastric cancer in the US. News release. Merck; July 1, 2021. Accessed January 31, 2022. https://bit.ly/32PaHA2
6. FDA limits the use of Tecentriq and Keytruda for some urothelial cancer patients. FDA. July 5, 2018. Accessed January 31, 2022. https://bit.ly/3Hj99x1 
7. Fuchs C. Sponsor’s position on maintaining the accelerated approval for atezolizumab(Tecentriq) in first line metastatic urothelial cancer. Presented at: 2021 Meeting of the Oncologic Drugs Advisory Committee; April 27-29, 2021; virtual. Accessed January 31, 2022. https://www.fda.gov/media/147926/download
8. Merck announces Keytruda (pembrolizumab) met primary endpoint of overall survival (OS) in patients with advanced hepatocellular carcinoma previously treated with sorafenib. News release. Merck; September 27, 2021. Accessed January 31, 2022. https://bit.ly/3zLvpLq 
9. FDA approves updated indication for Merck’s Keytruda (pembrolizumab) for treatment of certain patients with urothelial carcinoma (bladder cancer). News release. Merck; August 31, 2021. Accessed January 31, 2022. https://bit.ly/3IICBNo
10. Project Confirm. FDA. Updated January 4, 2022. Accessed February 1, 2022. https://bit.ly/3ITFJ9c
11. Mullard A. Accelerated approvals under the microscope. Nat Rev Drug Discov. 2021;20(6):409-411. doi:10.1038/d41573-021-00089-5
12. Kaltenboeck A, Mehlman A, Pearson SD. Strengthening the accelerated approval pathway:an analysis of potential policy reforms and their impact on uncertainty, access, innovation, and costs. Institute for Clinical and Economic Review. April 26, 2021. Accessed February 1, 2022. https://bit.ly/3IVegUz
13. CDER drug and biologic accelerated approvals based on a surrogate endpoint. FDA. December 31, 2021. Accessed February 1, 2022. https://bit.ly/3geM7f1
14. Accelerated approvals in oncology (1992-present). Friends of Cancer Research. Accessed February 1, 2022. https://bit.ly/3rjJQ8z
15. Idelalisib. FDA. Updated September 15, 2015. Accessed January 31, 2022. https://bit.ly/3ukzhUR
16. Gilead statement on Zydelig U.S. indication for follicular lymphoma and small lymphocytic leukemia. News release. Gilead; January 14, 2022. Accessed January 31, 2022. https://bit.ly/3Ggp263
17. Oncopeptides withdraws Pepaxto in US, scale down organization and focus on R&D. News release. Oncopeptides AB; October 22, 2021. Accessed February 1, 2022. https://bit.ly/3AWesjF
18. Gefitinib (Iressa). FDA. Updated July 13, 2015. Accessed January 31, 2022. https://bit.ly/3rdQj4O
19. FDA approves gemtuzumabozogamicin for CD33-positive AML. FDA. September 1. 2017. Accessed January 31, 2022. https://bit.ly/3IT6db2
20. Optimizing the use of accelerated approval. Friends of Cancer Research. Presented at: Friends of Cancer Research Annual Meeting 2020; November 19, 2020; virtual. Accessed February 1, 2022. https://bit.ly/3IXftdY
21. FDA grants Roche’s Tecentriq (atezolizumab) accelerated approval as initial treatment for certain people with advanced bladder cancer. News release. Roche; April 18, 2017. Accessed January 31, 2022. https://bit.ly/3IQuFcT
22. Pembrolizumab (Keytruda): advanced or metastatic urothelial carcinoma. FDA. Updated May 19, 2017. Accessed February 1, 2022. https://bit.ly/3eTrkOq
23. Keytruda. Prescribing information. Merck Sharp & Dohme Corp; 2021. Accessed February 2, 2022. https://bit.ly/3AQHlxH
24. Tecentriq. Prescribing information. Genentech; 2022. Accessed February 2, 2022. https://bit.ly/3IWT14H
25. Opdivo. Prescribing information. Bristol Myers Squibb; 2021. Accessed February 2, 2022. https://bit.ly/3Hh4ADv
26. Bavencio. Prescribing information. EMD Serono; 2020. Accessed February 2, 2022. https://bit.ly/3sdS1m2
27. Balar AV, Galsky MD, Rosenberg JE, et al; IMvigor210 Study Group. Atezolizumab as first-line treatment in cisplatin-ineligible patients with locally advanced and metastatic urothelial carcinoma: a single-arm, multicentre, phase 2 trial. Lancet. 2017;389(10064):67-76. doi:10.1016/S0140-6736(16)32455-2
28. Genentech provides update on Tecentriq U.S. indication for PD-L1-positive, metastatic triple-negative breast cancer. News release. Genentech; August 27, 2021. Accessed February 2, 2022. https://bit.ly/3L4uHA0
29. FDA approves pembrolizumab for high-risk early-stage triple-negative breast cancer. FDA. July 27, 2021. Accessed February 2, 2022. https://bit.ly/3okq9f0
30. Emens LA, Adams S, Cimino-Mathews A, et al. Society for Immunotherapy of Cancer (SITC) clinical practice guideline onimmunotherapy for the treatment ofbreast cancer. J Immunother Cancer. 2021;9(8):e002597. doi:10.1136/jitc-2021-002597
31. Parikh RB, Adamson BJS, Khozin S, et al. Association between FDA label restriction and immunotherapy and chemotherapy use in bladder cancer. JAMA. 2019;322(12):1209-1211. doi:10.1001/jama.2019.10650