Cancer immunotherapy trial design with cure rate and delayed treatment effect

Area of Science:

• Oncology
• Biostatistics
• Clinical Trials

Background:

• Cancer immunotherapy trials exhibit unique characteristics, including delayed treatment effects and potential cure rates.
• These features violate standard proportional hazards model assumptions, potentially compromising trial design and statistical power.

Purpose of the Study:

• To propose a novel statistical model that accurately incorporates both delayed treatment effects and cure rates in cancer immunotherapy trial design.
• To develop a precise sample size formula for weighted log-rank tests, enhancing the statistical power of these trials.

Main Methods:

• A piecewise proportional hazards cure rate model was developed to address the complexities of immunotherapy trial data.
• A sample size formula was derived for a weighted log-rank test under a fixed alternative hypothesis.
• Simulation studies were conducted to assess the accuracy of the new sample size formula and compare it with existing methods.

Main Results:

• The proposed piecewise proportional hazards cure rate model effectively incorporates delayed treatment effects and cure rates.
• The derived sample size formula provides accurate calculations, outperforming existing methods in simulation studies.
• Illustrative analysis of a real immunotherapy trial highlights the practical application and balance between sample size and follow-up time.

Conclusions:

• The developed statistical framework and sample size formula are crucial for optimizing the design of cancer immunotherapy trials.
• Accurate consideration of delayed effects and cure rates is essential to avoid substantial loss of statistical power.
• This approach offers a more robust methodology for planning and conducting future immunotherapy studies.