Seamless Phase I--II Cancer Clinical Trials Using Kernel-Based Covariate Similarity

2025-11-03

AI Summary

In oncology drug development, low reusability of Phase I data and small Phase II sample sizes lead to unreliable efficacy estimation. To address this, we propose a seamless Phase I/II clinical trial design based on kernel methods. Our method quantifies covariate distribution similarity between Phase I and Phase II patients using the kernel maximum mean discrepancy (MMD), enabling adaptive power prior construction for controlled borrowing of Phase I information within a Bayesian framework. We derive theoretical weight confidence intervals, permitting direct assessment of borrowing accuracy without resampling—particularly advantageous in small-sample settings. Simulation studies and real-data applications demonstrate that our approach significantly increases the probability that the objective response rate (ORR) credible interval exceeds the clinically meaningful threshold for effective doses, while reducing misclassification risk for weakly active doses. The design maintains statistical rigor and practical feasibility for early-phase oncology trials.

Technology Category

• Machine Learning: Kernel Methods
• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

In response to the U.S. Food and Drug Administration's (FDA) Project Optimus, a paradigm shift is underway in the design of early-phase oncology trials. To accelerate drug development, seamless Phase I/II designs have gained increasing attention, along with growing interest in the efficient reuse of Phase I data. We propose a nonparametric information-borrowing method that adaptively discounts Phase I observations according to the similarity of covariate distributions between Phase I and Phase II. Similarity is quantified using a kernel-based maximum mean discrepancy (MMD) and transformed into a dose-specific weight incorporated into a power-prior framework for Phase II efficacy evaluation, such as for the objective response rate (ORR). Considering the small sample sizes typical of early-phase oncology studies, we analytically derive a confidence interval for the weight, enabling assessment of borrowing precision without resampling procedures. Simulation studies under four toxicity scenarios and five baseline-covariate settings showed that the proposed method improved the probability that the lower bound of the 95% credible interval for ORR exceeded a prespecified threshold at efficacious doses, while avoiding false threshold crossings at weakly efficacious doses. A case study based on a metastatic pancreatic ductal adenocarcinoma trial illustrates the resulting borrowing weights and posterior estimates.

Problem

• Developing seamless Phase I-II oncology trials with covariate similarity
• Proposing nonparametric information-borrowing method using kernel-based MMD
• Improving efficacy evaluation accuracy while controlling false positives

Innovation

• Kernel-based MMD quantifies covariate distribution similarity
• Power-prior framework adaptively weights Phase I data
• Analytical confidence intervals assess borrowing precision