Biomarker threshold adaptive designs for survival endpoints

• Published in: Journal of biopharmaceutical statistics Vol. 28; no. 6; pp. 1038 - 1054
• Main Authors: Diao, Guoqing, Dong, Jun, Zeng, Donglin, Ke, Chunlei, Rong, Alan, Ibrahim, Joseph G
• Format: Journal Article
• Language: English
• Published: England Taylor & Francis 02.11.2018; Taylor & Francis Ltd
• Subjects: Adaptive enrichment design; Algorithms; Antineoplastic Agents - therapeutic use; Antineoplastic Agents, Immunological - therapeutic use; Biomarkers, Tumor - genetics; Biomarkers, Tumor - metabolism; Biostatistics - methods; Clinical Decision-Making; Clinical Trials, Phase III as Topic - methods; Clinical Trials, Phase III as Topic - statistics & numerical data; Computer Simulation; Data Interpretation, Statistical; ErbB Receptors - antagonists & inhibitors; ErbB Receptors - genetics; ErbB Receptors - metabolism; Head and Neck Neoplasms - drug therapy; Head and Neck Neoplasms - genetics; Head and Neck Neoplasms - metabolism; Head and Neck Neoplasms - mortality; Humans; Models, Statistical; Neoplasms - drug therapy; Neoplasms - genetics; Neoplasms - metabolism; Neoplasms - mortality; Panitumumab - therapeutic use; Patient Selection; Precision Medicine - methods; Precision Medicine - statistics & numerical data; predictive biomarker; Predictive Value of Tests; PTEN Phosphohydrolase - genetics; PTEN Phosphohydrolase - metabolism; Randomized Controlled Trials as Topic - methods; Randomized Controlled Trials as Topic - statistics & numerical data; Research Design - statistics & numerical data; Squamous Cell Carcinoma of Head and Neck - drug therapy; Squamous Cell Carcinoma of Head and Neck - genetics; Squamous Cell Carcinoma of Head and Neck - metabolism; Squamous Cell Carcinoma of Head and Neck - mortality; Survival Analysis; survival endpoint; Time Factors; Treatment Outcome; two-stage design; Adaptive enrichment design; predictive biomarker; two-stage design; survival endpoint
• DOI: 10.1080/10543406.2018.1434191
• ISSN: 1054-3406; 1520-5711

Due to the importance of precision medicine, it is essential to identify the right patients for the right treatment. Biomarkers, which have been commonly used in clinical research as well as in clinical practice, can facilitate selection of patients with a good response to the treatment. In this paper, we describe a biomarker threshold adaptive design with survival endpoints. In the first stage, we determine subgroups for one or more biomarkers such that patients in these subgroups benefit the most from the new treatment. The analysis in this stage can be based on historical or pilot studies. In the second stage, we sample subjects from the subgroups determined in the first stage and randomly allocate them to the treatment or control group. Extensive simulation studies are conducted to examine the performance of the proposed design. Application to a real data example is provided for implementation of the first-stage algorithms.