Abstract
A challenge arising in cancer immunotherapy trial design is the presence of a delayed treatment effect wherein the proportional hazard assumption no longer holds true. As a result, a traditional survival trial design based on the standard log-rank test, which ignores the delayed treatment effect, will lead to substantial loss of statistical power. Recently, a piecewise weighted log-rank test is proposed to incorporate the delayed treatment effect into consideration of the trial design. However, because the sample size formula was derived under a sequence of local alternative hypotheses, it results in an underestimated sample size when the hazard ratio is relatively small for a balanced trial design and an inaccurate sample size estimation for an unbalanced design. In this article, we derived a new sample size formula under a fixed alternative hypothesis for the delayed treatment effect model. Simulation results show that the new formula provides accurate sample size estimation for both balanced and unbalanced designs.
Original language | English |
---|---|
Pages (from-to) | 202-213 |
Number of pages | 12 |
Journal | Pharmaceutical Statistics |
Volume | 19 |
Issue number | 3 |
DOIs | |
State | Published - May 1 2020 |
Bibliographical note
Publisher Copyright:© 2019 John Wiley & Sons, Ltd.
Keywords
- cancer clinical trial
- delayed treatment effect
- piecewise weighted log-rank test
- sample size
ASJC Scopus subject areas
- Statistics and Probability
- Pharmacology
- Pharmacology (medical)