Early-phase trial efficacy and safety effect estimates regress upon testing in phase 3 trials

Area of Science:

• Oncology
• Clinical Trials
• Biostatistics

Background:

• Decisions in drug development and patient care frequently rely on phase Ib/II trial data.
• Early-phase trial results may overestimate treatment efficacy due to regression to the mean and bias.

Purpose of the Study:

• To evaluate the agreement between efficacy and safety estimates from phase Ib/II (P1b/2) and phase III (P3) trials for solid tumor therapies.
• To determine if P3 effect sizes fall within the confidence intervals of P1b/2 results and analyze factors influencing regression.

Main Methods:

• Identified P3 solid tumor trials (2014-2020) and matched them with corresponding P1b/2 trials for the same drug-indication.
• Assessed correlations for objective response rate (ORR), median overall survival (mOS), and serious adverse events (SAEs).
• Examined regression effects in relation to trial design features, including risk of bias (RoB).

Main Results:

• Significant positive correlations found between P1b/2 and P3 trials for ORR (β=0.78), mOS (β=0.67), and SAE rates (β=0.54).
• Regression was more pronounced in P3 trials with low RoB compared to high RoB.
• P3 estimates were more likely to be lower than P1b/2 estimates (ORR: 21% vs 10%; mOS: 25% vs 6%; SAE: 22% vs 4%).

Conclusions:

• Efficacy and toxicity estimates from early-phase (P1b/2) solid tumor drug trials frequently exhibit regression in later-phase (P3) studies.
• The observed variability highlights the need for caution when using early-phase trial data for clinical decision-making.