The FDA has approved cancer drugs at a pace and through a variety of regulatory mechanisms that would have been unrecognizable a generation ago. Between expedited pathways, surrogate endpoints, tissue-agnostic indications, and novel drug classes, oncology now accounts for more new approvals than any other therapeutic area. But the speed and volume raise real questions about how well the evidence holds up after approval, whether patients actually feel better on these drugs, and who gets left out of the trials that support them.
How Accelerated Approval Works in Oncology
The accelerated approval pathway, established in 1992, lets the FDA green-light drugs for serious diseases based on surrogate endpoints, such as tumor shrinkage or delayed progression, that are “reasonably likely” to predict a real clinical benefit like living longer or feeling better.1JNCI: Journal of the National Cancer Institute. Accelerated Approval of Oncology Products: The Food and Drug Administration Experience The deal is straightforward: patients get access to promising drugs sooner, and in return, the drug company commits to running confirmatory trials that prove the drug actually delivers the expected benefit. If those trials fail or never get done, the FDA can pull the approval.
Cancer drugs dominate this pathway. A cohort study covering 167 accelerated approval indications for 113 anticancer drugs found that by August 2024, about 61% had been converted to full regular approval, 19% had been withdrawn, and 20% were still sitting in accelerated-approval limbo.2PubMed Central. Predictors of withdrawal of anticancer drug indications granted accelerated approval: a retrospective cohort study That one-in-five withdrawal rate is not trivial. Among indications that had been around long enough to have a clear outcome, roughly 23% ultimately failed to confirm their clinical benefit and were pulled.
The timeline matters, too. For the indications that did convert to regular approval, the median time from accelerated approval to full approval was about three years.3JAMA Network Open. Factors in Time to Full Approval or Withdrawal for Anticancer Medicines Granted Accelerated Approval by the FDA But that median hides wide variation. When companies already had their confirmatory trial running at the time of accelerated approval, conversion happened much faster, around 2.8 years, compared with over 5.5 years when no trial had even started.4JAMA Network Open. Factors in Time to Full Approval or Withdrawal for Anticancer Medicines Granted Accelerated Approval by the FDA Among indications approved between 2013 and 2017 with more than five years of follow-up, about 63% had converted, 22% were withdrawn, and 15% remained in regulatory limbo after a median of over six years.5JAMA. Clinical Benefit and Regulatory Outcomes of Cancer Drugs Receiving Accelerated Approval
Legislative Reforms to Tighten the Process
The Food and Drug Omnibus Reform Act of 2022 (FDORA) gave the FDA sharper teeth. The law enables the agency to require that confirmatory studies be underway at the time of accelerated approval, or within a defined window afterward, rather than letting sponsors delay indefinitely.6Mayo Clinic Proceedings. Select Provisions of the Food and Drug Omnibus Reform Act of 2022 and Their Implications for Clinical and Pharmaceutical Sectors FDORA also expanded expedited withdrawal options, making it faster to pull an approval when a company drags its feet on confirmatory evidence. Separately, the FDA has floated a one-trial approach under its Project FrontRunner initiative, which would allow a single trial to serve as the basis for both accelerated and regular approval, potentially collapsing what used to be a two-study, multi-year process into a more efficient design.7PubMed. Success and Futility Criteria for Accelerated Approval of Oncology Drugs
Breakthrough Therapy Designation
Breakthrough therapy designation is often confused with accelerated approval, but it is a separate mechanism. It signals that a drug may offer a substantial improvement over existing treatments and triggers more intensive FDA guidance during development plus faster review times. Breakthrough-designated cancer drugs reached first FDA approval a median of about two years sooner than non-breakthrough drugs.8PubMed. Efficacy, Safety, and Regulatory Approval of Food and Drug Administration-Designated Breakthrough and Nonbreakthrough Cancer Medicines The catch: studies have found no statistically significant advantage in efficacy or safety for breakthrough-designated drugs compared with those that did not receive the designation. Rates of serious adverse events and deaths were similar between the two groups. The designation speeds things up, but speed alone does not guarantee a better drug.
When breakthrough designation was combined with accelerated approval, every single one of those approvals came with at least one postmarketing requirement focused on safety or efficacy. Among approvals without accelerated approval, about two-thirds still had postmarketing requirements.9JAMA. Clinical Trial Evidence Supporting FDA Approval of Drugs Granted Breakthrough Therapy Designation The overall message is that these designations create a faster on-ramp, but they shift more of the evidence burden to after the drug is already on the market.
The Surrogate Endpoint Problem
Most accelerated approvals in oncology lean on surrogate endpoints, especially progression-free survival (how long until the tumor grows or spreads) and overall response rate (the proportion of patients whose tumors shrink). These sound intuitively like they should track with living longer, but the relationship is not as strong as you might expect. A systematic review of studies evaluating whether progression-free survival predicts overall survival found that only about half concluded there was validated surrogacy. Among those, just 51% reported a trial-level correlation strong enough to be considered reliable.10British Journal of Cancer. Progression-free survival as a surrogate for overall survival in oncology trials: a methodological systematic review
The picture gets murkier with newer drug classes. For PD-1 inhibitors, which are among the most widely approved cancer immunotherapies, a meta-analysis found that the correlation between progression-free survival and overall survival was not statistically significant when measured by median values. The correlation between hazard ratios was significant but moderate, and the protective effect on overall survival actually exceeded what progression-free survival alone would have predicted.11JAMA Network Open. A Comparison of Response Patterns for Progression-Free Survival and Overall Survival Following Treatment for Cancer With PD-1 Inhibitors In other words, for these immunotherapies, progression-free survival may underestimate how much longer patients live, which is good news for patients but complicates the use of the surrogate as a reliable stand-in.
Single-Arm Trials and Clinical Benefit
Oncology is unusual in how often the FDA accepts single-arm trials, studies without a comparison group, as the basis for approval. These trials measure how many patients respond to the drug, but without a control arm, there is no way to use time-to-event endpoints like overall survival, and the safety dataset is smaller.12PubMed. Use of Single-Arm Trials for US Food and Drug Administration Drug Approval in Oncology, 2002-2021 The concern is that drugs approved this way may look promising but deliver less benefit than expected when tested more rigorously. One analysis scored drugs approved through single-arm trials using the European Society for Medical Oncology’s clinical benefit scale and found that only about 7% met the threshold for “substantial” clinical benefit. By contrast, drugs approved through randomized trials were significantly more likely to show substantial benefit.13JAMA Oncology. Magnitude of Clinical Benefit of Cancer Drugs Approved by the US Food and Drug Administration Based on Single-Arm Trials
Patient-Reported Outcomes Remain an Afterthought
One of the least-discussed gaps in FDA oncology approvals is how patients feel on these drugs. Patient-reported outcomes, which capture symptoms, side effects, and quality of life as the patient experiences them, are collected in many clinical trials but rarely make it into the drug’s label. An analysis of oncology drugs approved between 2017 and 2022 found that while patient-reported outcome data appeared in the clinical review documents of 45 products, only 2 of those had the information included in their FDA-approved labeling.14The Lancet Regional Health / eClinicalMedicine (The Lancet). Analysis of patient-reported outcomes in the approval of novel oncology drugs in the United States, 2017–2022 FDA reviewers commonly flagged methodological issues with the patient-reported outcome data, including problems with study design and analysis plans. The net result is that a doctor prescribing one of these drugs often has good data on tumor response and progression but very little on whether patients feel better, worse, or about the same during treatment.15BMJ. Health-related quality of life outcomes reporting associated with FDA approvals in haematology and oncology
Tissue-Agnostic Approvals
One of the most striking developments in recent years has been the FDA’s approval of drugs based on molecular markers rather than where the cancer originated in the body. If a tumor has a specific genetic alteration, the drug is approved regardless of whether the cancer started in the lung, the colon, or the thyroid. These tissue-agnostic approvals cover biomarkers like high tumor mutational burden, certain DNA repair deficiencies, specific gene fusions, and particular mutations.
A large real-world analysis of nearly 300,000 tumor samples across 57 cancer types found enormous variability in how common these targetable biomarkers are, ranging from 0% in pituitary carcinoma to 87% in basal cell skin cancer. Collectively, roughly one in five cancer patients carried at least one biomarker matching a tissue-agnostic drug indication. Even after excluding patients whose specific cancer type already had its own approved therapy for that marker, about 5% of cancer patients were candidates for a tissue-agnostic drug they might not otherwise have been offered.16Nature Communications. Real-world evidence provides clinical insights into tissue-agnostic therapeutic approvals These approvals are especially meaningful for rare cancers, including pediatric cancers, where the combination of an uncommon biomarker in an already rare disease makes traditional tumor-specific clinical trials nearly impossible to run.17PubMed Central. Ag(e)nostic precision oncology therapy approvals across the years
The Checkpoint Inhibitor Landscape and Limited Competition
Checkpoint inhibitors, the immunotherapy drugs that help the immune system recognize and attack cancer cells, have become the backbone of oncology treatment. Eleven FDA-approved checkpoint inhibitors now cover 43 distinct cancer indications. But the distribution is wildly lopsided. Pembrolizumab alone accounts for 35 of those indications, including 11 that no other checkpoint inhibitor is approved for. At the other end, three checkpoint inhibitors have just two indications each.18Journal of Clinical Oncology. Overlapping and non-overlapping indications for checkpoint inhibitors in the US.
The more concerning finding is that over a third of checkpoint inhibitor indications had only one approved drug. Among the overlapping indications where multiple drugs compete, most had approvals for just two or three drugs. Only one indication, non-small cell lung cancer, had seven approved checkpoint inhibitors. The researchers who mapped this landscape suggested that manufacturers may be avoiding seeking approval for indications where competitors are already established, which reduces the kind of competition that could push prices down.19Journal of Clinical Oncology. Overlapping and non-overlapping indications for checkpoint inhibitors in the US.
Who Gets Studied in Oncology Trials
The demographics of clinical trial participants have not kept up with the demographics of who gets cancer. An analysis of 372 pivotal trials supporting FDA oncology approvals between 2018 and 2026 found persistent underrepresentation of women, older adults, Black patients, and Hispanic patients relative to their share of U.S. cancer cases. Black patients were enrolled at about a quarter of what their cancer incidence would predict, and Hispanic patients at about half.20PubMed Central. Racial, ethnic, gender and age representation of clinical trials supporting FDA approval of cancer therapies An earlier analysis covering 2008 to 2018 found similarly low representation, with Black patients enrolled at 22% of expected levels and Hispanic patients at 44%, while Asian participants were substantially overrepresented at more than four times their expected proportion, likely reflecting the large number of trials conducted in Asian countries.21JAMA Oncology. Disparity of Race Reporting and Representation in Clinical Trials Leading to Cancer Drug Approvals From 2008 to 2018
The practical consequence is that safety signals, dosing needs, and efficacy differences that vary across populations may not be detected before a drug reaches the market. Older adults, who bear the greatest cancer burden, were enrolled at about two-thirds of their expected rate, which is particularly problematic since age-related differences in drug metabolism and tolerability can meaningfully affect outcomes.22PubMed Central. Racial, ethnic, gender and age representation of clinical trials supporting FDA approval of cancer therapers
Drug Pricing and Clinical Value
If you are hoping that the price of a cancer drug reflects how well it works, the evidence is not encouraging. A study examining U.S. cancer drug prices found only a weak correlation between price and improvements in overall survival or progression-free survival.23PubMed. Cancer Drug Prices in the United States: Efficacy, Innovation, Clinical Trial Evidence, and Epidemiology A separate international analysis found no significant association between monthly treatment costs and clinical benefit scores for solid tumor drugs in the U.S. or in England, Switzerland, and Germany.24The Lancet Oncology. Association of clinical benefit and drug prices with anticarcinoic drugs approved by the US Food and Drug Administration and the European Medicines Agency What a drug costs, in short, tells you almost nothing about how much it helps. Pricing appears driven more by market dynamics, intellectual property positioning, and the willingness of payers to cover treatments for serious illness than by the magnitude of therapeutic gain.
Project Orbis and Global Access Timelines
The FDA’s Oncology Center of Excellence launched Project Orbis in 2019 to let international regulators review cancer drug applications simultaneously rather than sequentially.25PubMed. Comparative Expedited Regulatory Programs of U.S Food & Drug Administration and Project Orbis Partners Between 2019 and late 2023, about a third of FDA cancer drug approvals were reviewed through this program. Drugs reviewed through Orbis were typically approved in the U.S. first, followed by Health Canada at a median lag of about 148 days and the UK’s MHRA at about 172 days.26The Lancet Oncology. Clinical benefit, reimbursement outcomes, and prices of FDA-approved cancer drugs reviewed through Project Orbis in the USA, Canada, England, and Scotland That is a meaningful compression of what used to be year-long or multi-year gaps between U.S. and international approvals.
Dose Optimization and Newer Drug Classes
Traditionally, oncology drug development followed a simple rule inherited from chemotherapy: find the highest dose patients can tolerate and use that. For targeted therapies and immunotherapies, where more drug does not always mean more benefit but does mean more side effects, that approach can be counterproductive. The FDA’s Project Optimus initiative, run by the Oncology Center of Excellence, aims to reform how dose selection studies are designed, pushing for evidence that the approved dose is genuinely the most appropriate one rather than just the maximum tolerable one.27PubMed Central. Project Optimus, an FDA initiative: Considerations for cancer drug development internationally, from an academic perspective The agency has increased its focus on dosing-related postmarketing requirements for oncology drugs, signaling that dose optimization is shifting from a suggestion to an expectation in the premarket setting.28PubMed Central. An Analysis of Dosing-Related Postmarketing Requirements for Novel Oncology Drugs Approved by the U.S. Food and Drug Administration, 2012-2022
Meanwhile, newer drug classes are arriving rapidly. Antibody-drug conjugates, bispecific antibodies, and CAR T-cell therapies have become standard treatments in blood cancers and are expanding into solid tumors.29American Society of Clinical Oncology Educational Book. Easy as ABC: Managing Toxicities of Antibody-Drug Conjugates, Bispecific Antibodies, and CAR T-Cell Therapies Each brings unique toxicity profiles that oncologists trained primarily on chemotherapy and targeted pills are still learning to manage. The expanding approval landscape means the gap between a drug reaching the market and clinicians fully understanding its real-world behavior is growing, not shrinking.
Real-World Evidence and Pediatric Development
As more evidence generation shifts to the post-approval setting, the FDA has begun incorporating real-world evidence, data from electronic health records, registries, and insurance claims, into its regulatory decisions. Cancer and blood-disease drugs account for the largest share of FDA decisions involving real-world evidence.30PubMed Central. Premarket and Postmarket Real-World Evidence Studies Supporting U.S. Food and Drug Administration Regulatory Decision-Making, 2016-2024 Among oncology drug approvals with postmarketing requirements, a small but growing number have specified real-world evidence, primarily requesting safety data from observational studies, registries, and expanded-access programs.31PubMed. The use of real-world evidence to support FDA post-approval study requirements for oncology drugs
Pediatric oncology has its own regulatory story. The RACE for Children Act, which took effect in 2020, closed loopholes in earlier legislation that allowed companies to avoid studying their cancer drugs in children. In its first year, three out of five eligible new oncology drugs were required to conduct pediatric trials, up from zero under the old rules. Ten of twelve drugs submitted just before the law took effect had molecular targets that would have qualified them for pediatric study requirements had the RACE Act been in place.32PubMed. The RACE for children act at one year: progress in pediatric development of molecularly targeted oncology drugs The law represents a shift from hoping that adult cancer drugs would eventually be studied in children to requiring it when the biology suggests they should be.