Selection of the initial design for the two-stage continual reassessment method

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmacometrics

Background:

• The two-stage continual reassessment method (CRM) is a model-based approach for dose escalation in clinical trials.
• Current methods for selecting the initial dose-escalating sequence in two-stage CRM are often arbitrary, relying on fixed cohorts or extensive simulations.
• Clinicians prefer designs that allow adequate patient allocation to lower dose levels before escalation.

Purpose of the Study:

• To propose a systematic algorithm for selecting the initial design in the two-stage CRM.
• To address the arbitrary nature of current initial design selection methods.
• To facilitate dose-finding studies where assigning patients to lower dose levels is a priority.

Main Methods:

• Development of a novel algorithm for systematic selection of the initial design for the two-stage CRM.
• Comparison of operating characteristics between the proposed initial design and the traditional cohort of three design.
• Provision of reference initial designs for planning two-stage CRM studies.

Main Results:

• The proposed systematic approach yields an initial design with superior operating characteristics compared to the standard cohort of three design.
• The algorithm simplifies the process of selecting an appropriate initial design for two-stage CRM.
• The study provides practical reference designs for implementing two-stage CRM.

Conclusions:

• The proposed systematic approach offers a more robust and efficient method for initial design selection in two-stage CRM.
• This advancement aids in the planning and execution of oncology dose-finding studies, aligning with clinical preferences.
• The developed algorithm and provided reference designs enhance the practical application of two-stage CRM.