Prospective Individual Patient Data Meta-Analysis: Evaluating Convalescent Plasma for COVID-19

• Activity Number: 406 - Novel Approaches for Handling Complex Data in Treatment Diagnosis and Evaluation
• Type: Topic-Contributed
• Date/Time: Thursday, August 12, 2021 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Danni Wu* and Eva Petkova and Keith Goldfeld and Thaddeus Tarpey
• Companies: NYU Grossman School of Medicine and New York University School of Medicine and NYU Grossman School of Medicine and NYU
• Keywords: international consortium for data sharing from ongoing RCTs; statistical analysis plan; Bayesian data and safety monitoring; stopping rules

Abstract

Numerous clinical trials were initiated to find treatment for COVID-19. However, clinical trials were frequently launched in a region after the peak of pandemic locally. It might take more COVID-19 surges at the same location over years to achieve full enrollment and to find answers about the efficacy of treatments. We propose a statistical plan for pooling patient-level data from ongoing randomized clinical trials (RCTs) of convalescent plasma (CP) that are not originally configured as a network of sites. We employ an adaptive Bayesian approach for continuously monitoring the accumulating pooled data for safety, efficacy, and harm. We describe the statistical challenges: complex hierarchical modeling and the choice of prior distributions. We have done extensive simulations using high-performance computing to assess and calibrate the operating characteristics of the monitoring rules as well as to understand the behavior of the models for estimation of the effect of CP in a variety of realistic situations. We expect the proposed framework can also be applied to pooling data from RCTs for other therapies and disease settings to find answers in weeks or months, rather than years.