The FDA’s Oncologic Drugs Advisory Committee recently voted to narrow the label for checkpoint inhibitors Keytruda and Opdivo in stomach and esophageal cancers based on PD-L1 expression levels—but the high unmet need in these patient populations should also be considered.
As the demand for precision medicine grows, the FDA has increasingly focused on biomarkers, reflecting a growing commitment to ensuring therapies are tailored to patient subgroups most likely to benefit while addressing safety concerns for those who may face unnecessary risks.
This regulatory shift was recently highlighted when the FDA’s Oncologic Drugs Advisory Committee (ODAC) voted in favor of narrowing the label for checkpoint inhibitors Keytruda and Opdivo in certain indications. Specifically, the ODAC voted 10-2, with one abstention, to recommend restricting the use of these therapies for patients whose HER2-negative gastric and esophageal cancers have low PD-L1 expression. It is unclear whether FDA Oncology will act on these votes to narrow currently labeled indications.
Data presented at the ODAC meeting showed that patients with high PD-L1 expression are more likely to benefit from immune checkpoint inhibitors, while those with low expression received limited therapeutic benefits and had increased risks of toxicity. Manufacturers have advocated for physician discretion in treating low PD-L1 patients. There are limited treatment options beyond these PD-L1 inhibitors and a high unmet need exists, and it is critical for pathologists and oncologists to assess patients individually. On the other hand, the committee voted in favor of FDA restricting the use of these treatments to ensure patient safety.
Now it’s up to the FDA to weigh the evidence.
As a former FDA regulator, I understand the challenging position the agency is now in. Safety is, of course, a priority, as the remit of the agency is to protect public health. However, it is important to balance safety standards with the potential benefit in diseases that have few treatment options. I may consider continued transparency about the risks associated with approved drugs, including how these risks vary by biomarker status. Ultimately, I support a balanced approach where the FDA provides insights but allows clinical discretion to ensure treatments remain available to those who need them, without compromising safety or ethical standards.
FDA’s Role in Aggregating and Sharing Biomarker Data
Historically, therapies were often approved with broad indications if they demonstrated overall efficacy, even in cases where biomarker-positive and -negative patient outcomes were mixed. However, the growing availability of biomarker data and a broader range of available treatments have prompted a closer analysis of patient subgroups.
Indeed, the FDA is the only regulatory body that routinely accesses comprehensive clinical trial data across multiple drugs and drug sponsors. By continually compiling these emerging data, the FDA can assess trends across therapeutic classes, like potential benefits or toxicities for certain biomarkers.
To date, the agency has used these data to guide providers on which patients may benefit most without restricting access for biomarker-negative patients. Data sharing preserves clinical discretion in patient care while enabling the FDA to provide guidance that ensures access to comprehensive efficacy and risk profiles for physicians and patients.
This is critical as, if a treatment is deemed safe and approvable, the final treatment decision lies in the hands of the physician and patient.
Balancing Data Transparency With Patient Access
While the FDA aims to improve patient safety through refined labels, it must also preserve access to therapies in high unmet need scenarios. In areas like gastric and esophageal cancers, where patients often have few or no alternative treatment options, even therapies with modest benefits might provide meaningful outcomes. For biomarker-negative populations, restricting access could inadvertently remove the last viable treatment option; instead of restricting a label, this information could be clearly laid out for physicians so they can evaluate their patients on a case-by-case basis.
The practical challenges of implementing and communicating label restrictions to providers add another layer of complexity. Part of the reason the FDA considers restricting labels is that many providers, specifically community oncologists, may lack the time or resources to stay up to date on nuanced biomarker data. While this is an important consideration, restrictive labels risk creating gaps in care by limiting informed oncologists’ ability to make case-specific decisions when setting forth a treatment plan.
The FDA’s role in ensuring access to comprehensive efficacy and risk profiles is essential in enabling providers to make the best decisions for their patients. By providing robust insights while allowing space for clinical judgment in high unmet need cases, the FDA can balance its responsibility to protect patient safety and maintain access to potentially beneficial therapies.
Trial Design Innovation for Biomarker-Driven Therapies
The regulatory shift we’re seeing highlights a pressing need for innovation in clinical trial design. Biomarker-driven trials allow for more precise patient selection and mitigate the need for restrictive post-approval label changes. Improving the quality of data can expedite regulatory approvals by demonstrating clear correlations between treatment efficacy and targeted patient outcomes.
Traditional approaches, which often test drugs on “all-comers” regardless of biomarker status, maximize inclusivity and broaden access. However, this approach can later complicate regulatory decisions, revealing discrepancies in outcomes, particularly for subgroups showing minimal benefit. To address these challenges, trial designs must evolve to balance inclusivity with the need for precise data on biomarker-driven outcomes.
Incorporating robust dose exploration studies and biomarker analyses earlier in the drug development process can also streamline regulatory approvals and mitigate challenges later. Integrating these critical activities earlier means developers can generate precise data and preserve both trial integrity and patient access.
Modernizing trial designs with a focus on biomarkers can also streamline drug development while improving patient outcomes. By selecting patients more precisely, trials can reduce sample sizes, shorten timelines and enhance the quality of data generated. These efforts minimize unnecessary exposure to potentially ineffective or harmful therapies while providing robust efficacy data for targeted subgroups. As the FDA implements a more robust biomarker-driven strategy, innovation in trial design can help balance inclusivity with precision, advancing patient care and more impactful oncology research.
