Short Course: Drug development of non-cytotoxics

Subtype Identification and Strategies for Trial Design

Half-day Shortcourse

Shortcourse: Drug development of non-cytotoxics: Subtype Identification and Strategies for Trial Design

Presenters:

Brian P. Hobbs, (hobbsb@ccf.org)

Associate Staff; Section Head of Cancer Biostatistics

Taussig Cancer Institute and Lerner Research Institute

Cleveland Clinic

Michael J. Kane, (michael.kane@yale.edu)

Assistant Professor of Biostatistics

Yale School of Public Health

Yale University

Description:

Advances in biology and immunology continue to refine our understanding of cancer pathogenesis, elucidating potential mechanisms of tumor-cell growth, survival, angiogenesis and the systematic suppression of cancer immunity progressing toward precision medicine. Implicit to the concept of precision medicine is heterogeneity of treatment benefit among patients and patient sub-populations. Yet, precision medicine presents challenges to traditional paradigms of clinical translational for which estimates of population-averaged effects are used as the basis for selecting dose-scheduling strategies as well as demonstrating comparative benefit with randomized study.

Aspects of the traditional clinical research paradigm may not ideally suit the development of non-cytotoxics that challenge its foundational assumptions pertaining to dose-response and inter-patient exchangeability. Tumor biology and/or host immunity may better delineate target treatment populations than histology. Several emerging molecularly targeted and immunotherapeutic agents have produced durable responses in first-in-human trials. The U.S. regulatory landscape has also changed, with a growing number of accelerated approvals on the basis of single-arm trials. Collectively, these phenomena have prompted innovations with drug development strategies devised to consolidate phases in the traditional paradigm and rapidly expand accrual with “seamless” trial designs. This short-course is intended to elucidate issues that limit the effectiveness of traditional clinical research and offer solutions to elucidate subpopulation heterogeneity from clinical datasets and subsequent trial designs devised to characterize and adapt to evidence of benefit heterogeneity.

Specifically, analysis and design methodology will consider characterization of localized treatment benefit, basket trial design based on multi-source exchangeability modeling (MEM), and how to leverage existing databases to design single-arm trials that incorporate baseline prognostic classifiers and thereby facilitate counterfactual comparisons in early-phase trials. The methodology will be illustrated by actual clinical cancer trials, including analysis and permutation studies of actual data reported from a recent basket trial designed to estimate the effectiveness of vemurafenib in BRAF mutant non-melanoma among six clinical sites.