[Bayesian Regression Modeling of the Correlation between Post-Progression Survival and Overall Survival in Immune...

Area of Science:

• Oncology
• Immunotherapy
• Biostatistics

Context:

• Advanced or recurrent cancer treatment efficacy is typically measured by progression-free survival (PFS) and overall survival (OS).
• Immune checkpoint inhibitor (ICI) therapy effectiveness, particularly in non-small cell lung cancer (NSCLC), presents challenges where PFS improvements do not always correlate with extended OS.
• Delayed responses and pseudoprogression are known phenomena in ICI therapy that can complicate PFS assessments.

Purpose:

• To analyze the relationship between PFS, OS, and post-progression survival (PPS) in advanced NSCLC patients treated with atezolizumab and pembrolizumab.
• To evaluate the predictive accuracy of PPS for OS in ICI therapy.
• To highlight the importance of PPS in capturing the full survival benefit of ICI treatments.

Summary:

• Bayesian statistical analysis of advanced NSCLC patients treated with atezolizumab and pembrolizumab revealed that extended PPS significantly contributed to prolonged OS for both agents.
• The study found that PPS demonstrated high predictive accuracy for OS in the pembrolizumab group, suggesting its utility as a reliable metric.
• Findings underscore the need to consider PPS alongside PFS for a comprehensive evaluation of ICI therapy, accounting for delayed responses and pseudoprogression.

Impact:

• This research emphasizes the critical role of post-progression survival in understanding the long-term benefits of immune checkpoint inhibitors.
• The findings suggest that PPS may serve as a more stable and accurate predictor of overall survival than progression-free survival in certain ICI treatment scenarios.
• Future research directions include exploring nonlinear models to further refine the predictive power of PPS for overall survival in cancer immunotherapy.