TSNP: A two-stage nonparametric phase I/II clinical trial design for immunotherapy

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacology

Background:

• Identifying the optimal biological dose (OBD) for immunotherapies is crucial for maximizing therapeutic effects while minimizing toxicity.
• Existing clinical trial designs may lack efficiency or flexibility in balancing complex toxicity-efficacy profiles.

Purpose of the Study:

• To develop a transparent and efficient two-stage nonparametric (TSNP) clinical trial design for identifying the OBD of immunotherapy.
• To propose a nonparametric approach for estimating joint toxicity-efficacy probabilities under monotonicity constraints.

Main Methods:

• A two-stage design: Stage 1 explores toxicity, Stage 2 identifies OBD using a utility function balancing toxicity and efficacy.
• Nonparametric estimation of joint toxicity-efficacy response probabilities with closed-form solutions.
• Development of a concise dose-finding algorithm.

Main Results:

• The TSNP design provides closed-form estimates for toxicity-efficacy profiles.
• Simulations demonstrate superior operating characteristics compared to existing Bayesian parametric designs.
• User-friendly computational software is available for practical application.

Conclusions:

• The TSNP design offers an efficient and practical method for immunotherapy dose-finding.
• The nonparametric approach effectively handles toxicity-efficacy profiles under monotonicity.
• The availability of software facilitates the implementation of this advanced clinical trial design.