Overall response rates and progression-free survival in patients with advanced non–small-cell lung cancer are strongly associated at the trial level, and responders have better progression-free survival and overall survival than do nonresponders, according to findings from Food and Drug Administration trial– and patient-level analyses.
Based on data from 14 trials of treatments for advanced NSCLC, which involved 12,567 patients and which were submitted to the FDA between 2003 and 2013, the association between overall response rates (ORR) and progression-free survival (PFS) was strong using scatterplots of the treatment effects on the log-scale (R2 = 0.89), but no association was seen between ORR and overall survival (R2 = 0.09) or between PFS and overall survival (R2 = 0.08) – possibly because of factors confounding overall survival analysis, such as cross-over, subsequent therapies, and long post-progression survival, reported Dr. Gideon M. Blumenthal and his colleagues at the FDA, White Oak, Md.
Patient-level analyses showed that those who achieved a response had improved PFS and overall survival compared with nonresponders (hazard ratio, 0.40 for both), the investigators noted (J. Clin. Oncol. 2015 Feb. 9 [doi:10.1200/JCO.2014.59.0489]).
The findings, which suggest that a therapy for advanced NSCLC that has a large magnitude of effect on ORR may also have a large effect on PFS, have implications for drug approvals, as “accelerated approval can be granted based on improvement in a surrogate endpoint reasonably likely to predict clinical benefit, such as ORR of large magnitude and long duration,” the researchers wrote, noting that “ORR may not be the optimal endpoint for hypothesis generation or expedited approval pathways for cytostatic therapies and immunotherapies, in which alternate endpoints may be needed to estimate activity and PFS or OS may be required to confirm clinical benefit,” and that novel methods for assessing drug activity warrant further investigation.
The authors reported having no disclosures.
