Towards Efficient Time-to-Event Dose-Escalation Guidance of Multi-Cycle Cancer Therapies

Area of Science:

• Oncology
• Clinical Trial Design
• Biostatistics

Background:

• Cancer treatment has evolved significantly, yet early phase I trials often use basic dose-escalation designs focused only on the first cycle.
• This limited focus is insufficient for modern multi-cycle therapies where toxicities can occur later.
• Increased trial complexity necessitates advanced designs that accommodate drug combinations and varied dosing schedules.

Purpose of the Study:

• To introduce a novel multi-cycle time-to-event model (TITE-CLRM) for guiding dose-escalation trials in oncology.
• To address the limitations of single-cycle focused designs in evaluating multi-cycle cancer therapies.
• To balance the need for monitoring safety over longer treatment periods with continuous patient enrollment.

Main Methods:

• Development of a multi-cycle time-to-event model, Time-Interval-To-Event Complementary-Loglog Regression Model (TITE-CLRM).
• Formulation of the model as an extension of established principles like escalation with overdose control.
• Evaluation of the model's performance through a simulation study, motivated by a current drug development project.

Main Results:

• The proposed TITE-CLRM model provides a framework for dose-escalation trials involving multi-cycle therapies.
• The model allows for monitoring safety beyond the first treatment cycle, accommodating later-occurring toxicities.
• Simulations demonstrated the model's potential to guide trials while ensuring patient safety and continuous enrollment.

Conclusions:

• The TITE-CLRM model represents a significant advancement for early-phase oncology trial design.
• It offers a more robust approach to dose escalation for complex, multi-cycle cancer treatments.
• This model facilitates safer and more efficient drug development for novel cancer therapies.