Delayed Separation of OS and PFS Kaplan Meier Curves Occurs in the Majority of IO Trials: A Meta-Analysis of Phase III Studies with Anti-PD-(L)1 Therapy

• Activity Number: 271 - Recent Statistical Advances of Handling Nonproportional Hazards in Study Design and Data Interpretation
• Type: Topic Contributed
• Date/Time: Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Magdalena Watras and Philip He and Nana Rokutanda* and Jennifer Stocks
• Companies: AstraZeneca and Daiichi Sankyo and AstraZeneca and AstraZeneca
• Keywords: immunotherapy; non-proportional hazards; meta-analysis

Abstract

Delayed separation of Kaplan-Meier (KM) curves in Overall Survival (OS) and Progression Free Survival (PFS) is commonly seen in oncology clinical trials with immunotherapy due to its mechanism of action. In this meta-analysis, a systematic literature search was performed, and immuno-oncology Phase 3 trials were reviewed with publications inclusive up to September 30 ,2021. Due to the limited trials in adjuvant/neo-adjuvant setting and with CTLA4 inhibitors to date, these trials are excluded in this meta-analysis. The KM plots are visually inspected whether delayed separation is evident. Three categories are considered in the assessment: no delayed separation, delayed separation and not applicable due to negligible treatment effect. Within the category of delayed separation, we also assess whether early curves crossover is observed. The meta-analysis may be helpful to support study design and conduct of trials with anti-PD-(L)1 therapy.