Surrogate Endpoints and Hidden Uncertainty in Oncology Accelerated Approval
Voices in Bioethics · 29.08.2026 · 12
DOI: 10.52214/vib.v12i.14892 ↗Kurz gefasst
Fachartikel in der Zeitschrift Voices in Bioethics, online erschienen am 29.08.2026. Verfasst von Nathan Grubach. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Photo ID 39534183 © Luchschen | Dreamstime.com Introduction When a cancer drug is described as approved by the Food and Drug Administration (FDA), patients may reasonably assume that approval means the drug has been shown to provide clinically meaningful benefit. Accelerated approval complicates that assumption because it can permit a drug to enter the market before clinical benefit has been directly confirmed. FDA treats accelerated approval as different from traditional approval: the pathway permits earlier access for serious conditions based on surrogate or intermediate endpoints that are reasonably likely to predict clinical benefit, with confirmatory trials required to verify that benefit after approval.[1] This regulatory category may not be obvious to patients deciding whether to begin treatment. Within oncology, endpoints such as tumor response and progression-free survival can support accelerated approval even when improvements in overall survival (OS) or quality of life (QoL) have not yet been demonstrated.[2] The challenge is how to weigh that expedited approval against the burden it creates for patients. For patients whose cancer is progressing, waiting years for traditional endpoints such as OS may itself cause harm. This is the core rationale for accelerated approval: shortening the time for a drug to become available may itself produce benefit. One analysis estimates that accelerated approval has allowed market access approximately three years earlier than traditional approval for cancer drugs.[3] Earlier access can therefore have real moral force, especially when disease progression is time sensitive. Because accelerated approval is used most heavily in oncology, (80 percent of all accelerated approvals granted are for cancer therapies), this risk-benefit tradeoff is especially relevant to oncology.[4] Because it can cloud the patient’s treatment decision, this uncertainty is not merely regulatory. The concern is sharper when approval rests on early or noncomparative evidence, including single-arm trials with small sample sizes. These studies can identify early antitumor activity, but without a randomized comparison group, they make it harder to know whether tumor shrinkage reflects meaningful benefit. A response rate may be encouraging, but when it comes from early, noncomparative trials, it should not be treated as equivalent to patient-centered benefit. Fewer than half of all cancer drugs granted accelerated approval from 2013 to 2017 and followed for more than five years demonstrated clinical benefit in confirmatory trials.[5] This article argues that accelerated approval in oncology is not problematic simply because it involves provisional evidence. The harder question is when that provisional evidence remains acceptable in patient care. At the patient level, the central issue is informed decision making: accelerated approval is defensible only when patients understand what evidence supported approval and what outcomes remain unproven. Its ethical justification weakens when approval language and patient-facing materials obscure the limits of the supporting evidence, allowing surrogate endpoints to appear equivalent to proven patient benefit. Case Study: Belantamab Mafodotin Belantamab mafodotin, marketed as Blenrep, is a biologic drug developed for relapsed or refractory multiple myeloma (MM). Its original accelerated approval applied to adults who had received at least four prior therapies.[6] The evidence supporting the original approval came largely from the 97-patient 2.5 mg/kg cohort in DREAMM-2, an open-label, two-arm, noncomparative phase II trial. DREAMM-2 showed antimyeloma activity, with an overall response rate of roughly one-third in heavily pretreated patients, but that response rate did not establish improvement in OS, QoL, or symptoms.[7] Accelerated approval carries particular weight in oncology because cancer treatment is often burdensome even when it is justified. Blenrep was not a low-burden treatment. DREAMM-2 reported substantial side effects, including a high rate of eye problems that could affect vision, low platelet counts, and anemia.[8] Because the original approval rested on a limited, noncomparative trial, that burden was harder to justify. At the same time, Blenrep did not enter an empty treatment landscape. Although available treatments were limited, other disease-specific options had already been approved for overlapping relapsed or refractory multiple myeloma.[9] Accelerated approval in cancer often adds another option to an existing treatment pathway, rather than offering the first disease-specific treatment available to patients. This makes accelerated approval in oncology more questionable than in non-oncologic rare-disease settings, where patients may lack any FDA-approved treatment.[10] DREAMM-3 served as the confirmatory trial for the original approval. It did not show enough benefit to support continued approval of single-agent Blenrep for that use, and the original accelerated approval indication was withdrawn.[11] Blenrep later returned to the US market in October 2025 through regular approval, rather than accelerated approval, but that approval involved a different regimen and treatment setting than the original single-agent accelerated approval.[12] Testing often continues on oncology drugs in other combinations, lines of therapy, or patient populations after an accelerated approval is granted. A later approval regimen may show that the drug has clinical value, but it does not necessarily validate the earlier use that exposed patients to treatment before benefit had been confirmed. Therefore, the later approval does not eliminate the ethical concern raised by the original indication: accelerated approval can create a window in which patients bear the burdens of an unproven treatment before its benefit has been confirmed. The exposure window seen with Blenrep is part of a larger issue in oncology accelerated approval. Some accelerated approval indications are later confirmed, but others remain unresolved for years or are withdrawn after clinical benefit is not verified. The Ethical Case for Accelerated Approval and Its Limits In clinical research, exposing participants to a possibly ineffective intervention can be ethically justified in part because the study is designed to produce reliable knowledge that may improve scientific progress. Accelerated approval changes that context. Once a drug is approved, patients may receive it in ordinary clinical care, not primarily as participants in a study designed to answer whether the drug works. In that setting, patients may bear risks similar to those in research, but without the same formal protections or knowledge-generating purpose. Accelerated approval can be defended on consequentialist grounds: patients with cancer may gain earlier access to promising therapies, and drug development may move faster in areas of unmet need. However, that defense should not rest only on aggregate gains. Current research on the matter is mixed. One model estimated substantial population-level gains from oncology accelerated approval, while another analysis found that survival gains were unevenly distributed and concentrated in a small number of drugs.[13] Patients and physicians must therefore consider the possibility that accelerated approval can make a drug appear more established than the evidence shows. Even if accelerated approval produces life years overall, individual patients should not be treated simply as the cost of population benefit. The Belmont Report allows future benefit to justify risk, but it does not treat social value as enough on its own. The individual patient’s burden still matters. The risks must be reasonable, the burdens must be fairly imposed, and the patient must understand the uncertainty before accepting treatment.[14] When surrogate endpoints obscure the true patient benefit and survival gains are unevenly distributed, accelerated approval is defensible only if patients clearly understand what benefit has been shown and what remains unproven. Proposed Reforms: Making Provisional Approval Visible Once a drug is in routine care, clinicians should inform patients that the clinical benefit has not been verified. If patients understand what approval has established and what remains uncertain, then for serious conditions the accelerated approval path may be justifiable. In clinical trials, IRBs oversee a formal research consent process including oversight and federal human-subject protections.[15] Once a drug moves from a clinical trial into routine care, patients still must give informed consent before treatment. However, that consent is no longer governed by the same research-consent framework. Disclosure instead occurs through ordinary treatment counseling, institutional consent forms, FDA-approved patient labeling, and any required Risk Evaluation and Mitigation Strategy (REMS) materials. For Blenrep, these materials focused on adverse events and did little to address the issue of transparency.[16] While informed consent requires clinicians to disclose the benefits of a drug, this standard requirement does not always highlight the surrogate endpoints used for accelerated approval. The remaining gap is a lack of standardized disclosure that brings this information together when a patient is deciding whether to accept treatment. A short treatment consent addendum could build on existing FDA transparency efforts without requiring patients to navigate regulatory databases or technical labeling on their own. At minimum, that disclosure should identify the endpoint supporting approval, the limits of that endpoint, any major study limitations, the outcomes still unproven, the major risks of treatment, the confirmatory evidence still pending, and brief context about how often oncology accelerated approvals have ultimately demonstrated clinical benefit for the approved indication. This would ensure consent is genuinely informed. Because most oncology accelerated approvals are later confirmed either with surrogate endpoints or without evidence of an overall survival benefit, confirmatory trials should be designed with more than the initial surrogate endpoint in mind.[17] Although the FDA has strengthened the pathway by adding requirements for confirmatory trials within a specified period after approval, it still does not clearly specify what kind of evidence should be used in follow-up trials.[18] Surrogate endpoints may still be appropriate when direct outcomes would take too long to measure. But when confirmatory evidence relies on those endpoints, regulators and clinicians should explain why the endpoint is adequate in that disease setting and what uncertainty remains for patients. Reform should also include more independent research on the patient-level effects of accelerated approval, including toxicity, clinic time, financial burden, anxiety, and whether standardized disclosures improve patient understanding in routine care. These reforms would not remove uncertainty from oncology accelerated approvals. That would be unrealistic, and it would risk weakening a pathway that can provide meaningful access for patients with serious disease. The more practical goal is to prevent uncertainty from becoming invisible once a drug is called “FDA approved”. That requires patient-facing disclosure that makes the limits of approval understandable and confirmatory evidence that can support continued use or justify withdrawal. Conclusion Accelerated approval is a useful tool because it offers patients earlier access to promising cancer therapies, but it is ethically risky because it asks them to consider treatment before the full clinical value of that treatment is known. Clinicians, regulators, and scientists must therefore weigh the harms of adding another potentially toxic intervention against the harms of waiting for trials that use more definitive clinical endpoints. The danger comes when that uncertainty disappears into the ordinary language of FDA approval. A drug may be legally available, clinically promising, and still not yet proven to improve OS or QoL. Patients weighing that treatment are therefore weighing more than side effects; they are also weighing the type of benefit that has been shown and whether it is meaningful. When accelerated approval indications are later withdrawn, as with Blenrep, patients have been exposed to treatment that may not have benefited them. Earlier treatment may help cancer patients when options are limited, and accelerated approval may produce real gains across a population when confirmatory research proves efficacy for some patients. Regardless of whether later confirmatory trials validate the treatment or lead to withdrawal, patients assume the risk while the clinical benefit remains uncertain. The balance depends on whether patients understand that tradeoff. The practical response is to keep uncertainty visible. That requires clearer patient-facing disclosure at the time of treatment and confirmatory trials with endpoints strong enough to support continued use or justify withdrawal. - [1] U.S. Food and Drug Administration. (2026, February 27). Accelerated approval program. https://www.fda.gov/drugs/nda-and-bla-approvals/accelerated-approval-program [2] Chen, E. Y., Raghunathan, V., & Prasad, V. (2019). An overview of cancer drugs approved by the U.S. Food and Drug Administration based on the surrogate end point of response rate. JAMA Internal Medicine, 179(7), 915–921. https://doi.org/10.1001/jamainternmed.2019.0583; Gyawali, B., Hey, S. P., & Kesselheim, A. S. (2019). Assessment of the clinical benefit of cancer drugs receiving accelerated approval. JAMA Internal Medicine, 179(7), 906–913. https://doi.org/10.1001/jamainternmed.2019.0462 [3] Tibau, A., Hwang, T. J., Romano, A., Borrell, M., Gich, I., Molto, C., & Kesselheim, A. S. (2025). Factors in time to full approval or withdrawal for anticancer medicines granted accelerated approval by the FDA. JAMA Network Open, 8(3), e252026. https://doi.org/10.1001/jamanetworkopen.2025.2026 [4] Liu, I. T. T., Kesselheim, A. S., & Cliff, E. R. (2024). Clinical benefit and regulatory outcomes of cancer drugs receiving accelerated approval. JAMA, 331(17), 1471–1479. https://doi.org/10.1001/jama.2024.2396 [5] Liu, I. T. T., Kesselheim, A. S., & Cliff, E. R. (2024). Clinical benefit and regulatory outcomes of cancer drugs receiving accelerated approval. JAMA, 331(17), 1471–1479. https://doi.org/10.1001/jama.2024.2396 [6] U.S. Food and Drug Administration. (2024, March 7). FDA granted accelerated approval to belantamab mafodotin-blmf for multiple myeloma. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-granted-accelerated-approval-belantamab-mafodotin-blmf-multiple-myeloma [7] Lonial, S., Lee, H. C., Badros, A., et al. (2020). Belantamab mafodotin for relapsed or refractory multiple myeloma (DREAMM-2): A two-arm, randomised, open-label, phase 2 study. The Lancet Oncology, 21(2), 207–221. https://doi.org/10.1016/S1470-2045(19)30788-0 [8] Lonial, S., Lee, H. C., Badros, A., et al. (2020). Belantamab mafodotin for relapsed or refractory multiple myeloma (DREAMM-2): A two-arm, randomised, open-label, phase 2 study. The Lancet Oncology, 21(2), 207–221. https://doi.org/10.1016/S1470-2045(19)30788-0 [9] Bhatnagar, V., Gormley, N. J., Luo, L., et al. (2017). FDA approval summary: Daratumumab for treatment of multiple myeloma after one prior therapy. The Oncologist, 22(11), 1347–1353. https://doi.org/10.1634/theoncologist.2017-0229; Gormley, N. J., Ko, C.-W., Deisseroth, A., et al. (2017). FDA drug approval: Elotuzumab in combination with lenalidomide and dexamethasone for the treatment of relapsed or refractory multiple myeloma. Clinical Cancer Research, 23(3), 675–679. https://doi.org/10.1158/1078-0432.CCR-16-1930; U.S. Food and Drug Administration. (2020). SARCLISA (isatuximab-irfc) prescribing information. https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/761113s000lbl.pdf [10] Fermaglich, L. J., & Miller, K. L. (2023). A comprehensive study of the rare diseases and conditions targeted by orphan drug designations and approvals over the forty years of the Orphan Drug Act. Orphanet Journal of Rare Diseases, 18, 163. https://doi.org/10.1186/s13023-023-02790-7 [11] Dimopoulos, M. A., Hungria, V. T. M., Leleu, X., Sborov, D. W., Moreau, P., Mateos, M. V., Baz, R., Zamagni, E., Bahlis, N. J., Weisel, K., Oriol, A., Facon, T., Voorhees, P. M., Usmani, S. Z., Nooka, A. K., Chari, A., Rodriguez-Otero, P., et al. (2023). Efficacy and safety of single-agent belantamab mafodotin versus pomalidomide plus low-dose dexamethasone in patients with relapsed or refractory multiple myeloma (DREAMM-3): A phase 3, open-label, randomised study. The Lancet Haematology, 10(10), e801–e812. https://doi.org/10.1016/S2352-3026(23)00243-0 [12] U.S. Food and Drug Administration. (2025, October 23). FDA approves belantamab mafodotin-blmf for relapsed or refractory multiple myeloma. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-belantamab-mafodotin-blmf-relapsed-or-refractory-multiple-myeloma [13] Benedict, Á., Marczell, K., Szabo, G., Odom, D., Sylvester, S., & Doherty, B. (2024). Life years gained from the FDA accelerated approval program in oncology: A portfolio model. Journal of the National Comprehensive Cancer Network, 22(6), 382–389. https://doi.org/10.6004/jnccn.2024.7023; Naci, H., Shahzad, M., Murphy, P., Zhang, Y., Costa, R., Ross, J. S., & Wagner, A. K. (2025). Costs and benefits of early access to new cancer drugs through the US Food and Drug Administration’s accelerated approval pathway: Retrospective observational study and economic evaluation. BMJ Medicine, 4(1), e001934. https://doi.org/10.1136/bmjmed-2025-001934 [14] National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (1979). The Belmont Report: Ethical principles and guidelines for the protection of human subjects of research. U.S. Department of Health, Education, and Welfare. https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/read-the-belmont-report/index.html [15] 21 C.F.R. §§ 50.20–50.27. https://www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-50 [16] U.S. Food and Drug Administration. (2020, August). BLENREP (belantamab mafodotin-blmf) for injection, for intravenous use: Prescribing information and Medication Guide. https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/761158s000lbl.pdf; U.S. Food and Drug Administration. (2020, August 5). BLENREP Risk Evaluation and Mitigation Strategy (REMS). https://www.accessdata.fda.gov/drugsatfda_docs/nda/2020/761158Orig1s000REMS1.pdf [17] Chen, E. Y., Raghunathan, V., & Prasad, V. (2019). An overview of cancer drugs approved by the U.S. Food and Drug Administration based on the surrogate end point of response rate. JAMA Internal Medicine, 179(7), 915–921. https://doi.org/10.1001/jamainternmed.2019.0583; Gyawali, B., Hey, S. P., & Kesselheim, A. S. (2019). Assessment of the clinical benefit of cancer drugs receiving accelerated approval. JAMA Internal Medicine, 179(7), 906–913. https://doi.org/10.1001/jamainternmed.2019.0462; Liu, I. T. T., Kesselheim, A. S., & Cliff, E. R. (2024). Clinical benefit and regulatory outcomes of cancer drugs receiving accelerated approval. JAMA, 331(17), 1471–1479. https://doi.org/10.1001/jama.2024.2396 [18] U.S. Food and Drug Administration. (2025, January). Accelerated approval and considerations for determining whether a confirmatory trial is underway: Draft guidance for industry. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/accelerated-approval-and-considerations-determining-whether-confirmatory-trial-underway
Zitieren
APA
Grubach, N. (2026). Surrogate Endpoints and Hidden Uncertainty in Oncology Accelerated Approval. Voices in Bioethics, 12. https://doi.org/10.52214/vib.v12i.14892
Vancouver
Grubach N. Surrogate Endpoints and Hidden Uncertainty in Oncology Accelerated Approval. Voices in Bioethics. 2026;12. doi:10.52214/vib.v12i.14892
BibTeX
@article{Grubach2026,
author = {Grubach, Nathan},
title = {Surrogate Endpoints and Hidden Uncertainty in Oncology Accelerated Approval},
journal = {Voices in Bioethics},
year = {2026},
volume = {12},
doi = {10.52214/vib.v12i.14892}
}
EUVIMED