TITE-gBOIN-ET: Time-to-event generalized Bayesian optimal interval design to accelerate dose-finding accounting for...

Area of Science:

• Oncology
• Clinical Trial Design
• Biostatistics

Background:

• Identifying optimal doses (OD) for novel molecular-targeted and immune-oncology agents is crucial for cancer drug development.
• New therapies often cause low-to-moderate toxicities, complicating traditional dose-finding methods.
• Accelerating early-phase trials is essential, but challenges like late-onset outcomes and differing evaluation periods hinder adaptive decision-making.

Purpose of the Study:

• To propose a novel time-to-event generalized Bayesian optimal interval (TITE-gBOIN-ET) design for oncology dose-finding trials.
• To accelerate dose-finding by accounting for both efficacy and toxicity grades.
• To provide a model-assisted, straightforward design for practical implementation.

Main Methods:

• Developed a time-to-event generalized Bayesian optimal interval design (TITE-gBOIN-ET).
• Incorporated efficacy and toxicity grades into the dose-finding algorithm.
• Utilized simulation studies to evaluate design performance.

Main Results:

• The TITE-gBOIN-ET design significantly shortens trial duration compared to designs without sequential enrollment.
• Demonstrated comparable or higher performance in selecting the correct optimal dose.
• Showcased efficient patient allocation to optimal doses across various settings.

Conclusions:

• The TITE-gBOIN-ET design offers an accelerated and effective approach to oncology dose-finding.
• This model-assisted design is practical for real-world clinical trials.
• The design improves efficiency and accuracy in identifying optimal doses for novel cancer therapies.