Work

Finite-Boundary Reduction and Exact Verification of Strong Familywise Error in Active-Count-Coupled Multi-Arm Efficacy--Toxicity Monitoring

Year: 2026

Type: preprint

Abstract: Initial public release accompanying the manuscript: "Finite-Boundary Reduction and Exact Verification of Strong Familywise Error in Active-Count-Coupled Multi-Arm Efficacy--Toxicity Monitoring" This release includes R code for: calibration and simulation of Strong-Independent and Proposed-AC monitoring procedures; K = 3, 4, and 5 scalability analyses; deterministic exact strong-FWER verification over the finite boundary set; and the ABROAD-informed prospective case study. The ABROAD case-study code uses phi_E = phi_T = 0.20 and does not impose or overwrite integer efficacy or toxicity boundaries.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2609.06886

Authors: Kojima, Masahiro, Sunami, Hisato, Takeda, Kentaro

Institutions: Kyowa Kirin (Japan), Chuo University, Astellas Pharma (United States)

Language: English

Topic: Statistical Methods in Clinical Trials

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: green

Funder: Japan Society for the Promotion of Science