FDA’s NSCLC Guidance Stresses Overall Survival Endpoint, Sets PFS Standards

FDA is using its new draft guidance on non-small cell lung cancer to once again stress the importance of using overall survival to prove efficacy and make clear that if progression-free survival measure is used, a “substantial and robust” effect should be shown.

It also notes that the actual magnitude of the benefit is important to demonstrate that the benefit is significant. “A statistically significant difference in PFS of short duration may not be clinically meaningful,” the draft guidance on "Clinical Trial Endpoints for the Approval of Non-small Cell Lung Cancer Drugs and Biologics" states.

That the magnitude of the benefit seen in PFS matters was the centerpiece of FDA's case that the metastatic breast cancer indication for Roche/Genentech's Avastin should be withdrawn since the confirmatory trials did not replicate the level of benefit seen in the pivotal trial (Also see "Avastin’s Breast Cancer Claim: Will FDA’s Hamburg Take A Middle Road?" - Pink Sheet, 4 Jul, 2011.)

The regulatory requirements for different cancer settings vary, but FDA has been consistent in its preference for overall survival (while allowing for alternate endpoints where appropriate) and that PFS values must be robust.

FDA's lung cancer guidance was informed by a public workshop held by the American Society of Clinical Oncology and FDA about trial endpoints in April 2003 (Also see "“Underpowered” Cancer Trials Are Becoming “Major Problem” – FDA’s Pazdur" - Pink Sheet, 21 Apr, 2003.). This was followed by a meeting in December of that year of the Oncologic Drugs Advisory Committee, which voted 11 to 8 to accept PFS as an endpoint for first-line metastatic disease, but not as a primary endpoint for earlier stage, inoperable, locally advanced NSCLC (Also see "Progression-Free Survival Is Good Endpoint For Metastatic NSCLC, Cmte. Says" - Pink Sheet, 22 Dec, 2003.).

According to the draft guidance, issued in mid-June, discussions during the workshop and committee meeting suggested that OS is the most appropriate, “definitive and easy to determine” endpoint, and a benefit shown in a well-run trial “could be directly attributed to the experimental therapy.”

Eight years after those meetings, the guidance concludes: “We consider OS to be the standard clinical benefit endpoint that should be used to establish efficacy of a treatment in patients with advanced or metastatic NSCLC.”

However, the guidance goes on, other endpoints can be considered for a regulatory decision, based on the population and risk/benefit profile of a drug. Sponsors are encouraged to meet with FDA to discuss endpoints prior to designing protocols for NSCLC trials.

PFS Effect Should Be Robust

FDA’s guidance places conditions on use of PFS as a primary endpoint. “Because of the subjectivity in the measurement of PFS assessments and the fact that the assessments depend on frequency, accuracy, reproducibility and completeness, the observed magnitude of effect should be substantial and robust.”

Criteria for disease progression and tumor response are often “poorly defined, not rigorously evaluated and potentially introduce bias.”

The guidance cautions that the criteria for disease progression and tumor response are often “poorly defined, not rigorously evaluated and potentially introduce bias,” particularly in open-label trials. In appendices, the NSCLC document offers detailed guidance on how to collect tumor measurement data and perform PFS analyses, including advice on handling of missing data.

OS The Norm In NSCLC

As noted in the guidance document, the three most commonly used endpoints in past NSCLC trials were OS, PFS and time-to-progression. But the majority of approvals have been based on a significant improvement in overall survival, which the document describes as “the optimal endpoint because the measurement is accurate and represents direct clinical benefit to the patient.” Improvement in patient-reported outcomes can also be used to demonstrate a drug’s benefit and to support approval.

The guidance concludes that interim analyses of OS may be appropriate, but discourages interim analyses of PFS before patient accrual is completed. If a trial is stopped early based on an interim PFS analysis, the results may be uninterpretable.

Use of patient-reported outcomes data has a relatively high profile in the guidance document. PRO measures are valuable for evaluating treatment response in a “predominantly symptomatic disease such as NSCLC,” but the guidance stresses that these data must be collected and assessed in very well-designed placebo-controlled trials.

FDA also nods to current research about subpopulations within NSCLC; the guidance notes that NSCLC is a heterogeneous disease with different efficacy and safety effects depending on the patient population (Also see "NSCLC Market Snapshot: Promising Biomarkers, Testing Challenges" - Pink Sheet, 30 May, 2011.). “We recommend that clinical trials be prospectively designed to evaluate such differences in treatment effect,” the guidance advises.

By Emily Hayes