Exploring the relationship between Overall Survival (OS), Progression Free Survival (PFS) and Objective Response Rate...

From 2011 to 2016, 13 randomized clinical trials with active controls were submitted to U.S. Food and Drug Administration (FDA) for the treatment of advanced melanoma. While regular approval is generally granted due to a substantial improvement in overall survival (OS), or a large, clinically meaningful improvement in progression-free survival (PFS), accelerated approval can be granted based on incremental change in a surrogate end point reasonably likely to predict clinical benefit, such as objective response (ORR) of large magnitude and long duration. However, the relationship between objective response rate and progression free survival or objective response rate and overall survival in advanced melanoma has not been established.

We conducted analyses to assess the correlation of objective response rate with progression free survival as well as overall survival by examining all advanced melanoma trials submitted to the FDA between 2011 and 2016. In order to examine these relationships, associations between trial-level hazard ratio (HR) of progression free survival, hazard ration of survival and odds ratio of objective response rate were analyzed using a weighted linear regression model. Patient-level responder analyses comparing progression free survival and overall survival between patients with and without an objective response were performed using pooled data from all studies.

In the trial-level analysis, the linear relationships between progression free survival and objective response rate, and overall survival and objective response rate were weak (R²_adj = 0.019 and 0.093, respectively). The linear relationship between overall survival and progression free survival (R²_adj = 0.075) was also weak. In the patient-level responder analyses, patients who achieved a response had better progression free survival and overall survival compared with non-responders in both the control drug treatment and the experimental drug treatment.

Based on this analysis, use of objective response rate as a surrogate endpoint for overall survival or progression free survival in this population appears not appropriate. However, due to the nature of heterogeneity, interpretation needs to be cautious.