Progression-Free Survival As a Surrogate for Overall Survival in Unresectable, Locally Advanced, or Metastatic Non-Small Cell Lung Cancer: A Targeted Literature Review

OBJECTIVES: Overall survival (OS) has historically been the ‘gold standard’ endpoint in oncology trials, but necessitates prolonged follow-up and may be confounded by subsequent therapies. To avoid delays in patient access, other endpoints are therefore sometimes considered by regulatory authorities and reimbursement agencies to inform decision-making. This study aimed to summarize the available synthesized literature on progression-free survival (PFS) as a surrogate for OS in unresectable, locally advanced or metastatic non-small cell lung cancer (NSCLC).

METHODS: A de novo targeted literature review (TLR) was conducted by searching MEDLINE, Embase, the Cochrane Library, the Database of Abstracts of Reviews of Effects and preselected regulatory and HTA body websites through May 2023. Two independent reviewers screened records. Eligible records reported on a statistical assessment of the relationship of PFS to OS in patients with unresectable, locally advanced or metastatic NSCLC.

RESULTS: Of 1,482 records identified, 27 unique studies were included in the TLR. Included studies reported on absolute outcomes (e.g. median PFS/OS; n=12), relative treatment effects (e.g. hazard ratios; n=9) or both (n=6). Studies varied in patient eligibility criteria, disease stage, and included treatments.

Where strength of relationship between PFS and OS was categorized and reported (absolute outcomes: 39 analyses, 13 studies; relative treatment effects: 51 analyses, 12 studies), the majority of analyses identified a moderate-to-strong relationship: 82% (32 analyses, 13 studies) for absolute outcomes, and 67% (34 analyses, 10 studies) for relative treatment effects. Where reported, R² values were assessed against published strength of relationship thresholds, and although less clear for relative effects, these were generally supportive of results based on author-selected thresholds.

CONCLUSIONS: Overall, the evidence is supportive of using PFS as a surrogate for OS. Notably, when focusing on associations of relative treatment effects, a key requirement for surrogacy assessment, there is a greater degree of heterogeneity between results than for absolute outcomes.