Conditional power in vaccine trials with seasonal variations

Area of Science:

• Biostatistics
• Clinical Trials
• Vaccinology

Background:

• Conditional power (CP) and predictive power (PP) are vital for interim analyses in clinical trials, particularly for futility stopping and sample size adaptation.
• Seasonal variations in disease incidence and vaccine efficacy (VE) can impact the accuracy of CP/PP calculations in vaccine trials, potentially leading to flawed decisions.

Purpose of the Study:

• To investigate the performance of different choices for the drift parameter in computing CP or PP for end-of-season interim analyses in vaccine trials with seasonal variations.
• To compare the effectiveness of various CP and PP methods in informing futility stopping and sample size adaptation decisions.

Main Methods:

• Conducted simulation studies to evaluate seven different drift parameter choices for calculating CP and PP.
• Assessed performance based on type II error rates, statistical power, and expected sample sizes.

Main Results:

• CP under the hypothesized effect and a weighted PP under a normal prior distribution showed better performance for futility stopping regarding type II error rates.
• All CP and PP methods yielded comparable results for power and sample size adaptation.
• Weighted CP/PP methods demonstrated greater robustness to discrepancies between observed and future data due to seasonal variations.

Conclusions:

• Weighted CP/PP strategies are recommended for guiding trial operations in vaccine trials with seasonal variations due to their enhanced robustness.
• Decisions regarding trial operations should integrate plausible CP/PP calculations that consider data, prior hypotheses, and new evidence on clinical relevance.