Abstract 802: Implementation of optimized research biopsy analyses for clinical pharmacodynamic (PD) studies

Abstract

Abstract Research biopsies collected for evaluation of pharmacodynamic (PD) biomarkers present requirements beyond those obtained for diagnostic purposes. Despite establishing stringent post-collection handling and processing procedures, we have found that many biopsy pairs collected from the early phase trial patient population demonstrate significant tissue heterogeneity and viable tumor content too low to adequately assess biomarker response in the tumor cell population. A retrospective analysis of samples collected from NCI's Developmental Therapeutics Clinic (DTC) from 1/2010-9/2016 revealed an initial success rate for analysis of paired biopsies of only 53%; this was primarily due to a high incidence of biopsies with no or very low tumor content. Based on this initial high failure rate, several clinical and laboratory initiatives were undertaken to identify and mitigate factors which contribute to biopsy unsuitability. After consultation with other translational oncology teams to review best practices, we have published and implemented critical recommendations to improve research biopsy quality at the clinical project team level (PMID: 30285529). In addition to important updates to the clinical trial strategy, several actionable improvements have been identified and implemented at the laboratory level to significantly increase the success rate of research biopsy analyses. Improvements in the biopsy processing approach include use of multiple 18-gauge biopsy cores per timepoint as a standard approach for all PD trials, implementation of methods to prevent biopsy curling during fixation and processing to ensure full longitudinal facing of the tissue during sectioning, and deeper sectioning of the blocks with H&E evaluation of flanking slides by an anatomic pathologist with subsequent release of nonadjacent slides from the most optimal slide range for the primary PD analyses. Development of advanced image analysis methods including the use of β-catenin as a tissue segmentation marker for multiplex immunofluorescence assays has allowed for successful analysis of biopsies with significant heterogeneity and low tumor content. Additionally, defining the minimum cell number required to reach significance for PD marker measurements and capturing additional image fields in low tumor content biopsies to reach the required threshold has led to the implementation of highly robust methods across multiple clinical trials and laboratories. An analysis of biopsies collected from 10/2016-9/2019, after which several clinical and laboratory improvements had been implemented, revealed a significantly improved success rate for paired biopsies of 83%. Funded by NCI Contract No HHSN261200800001E. Citation Format: Katherine V. Ferry-Galow, Deborah Wilsker, Robert J. Kinders, Hala R. Makhlouf, Angie Dull, Tony Navas, Rachel E. Andrews, Geraldine O'Sullivan Coyne, Alice P. Chen, James H. Doroshow, Ralph E. Parchment. Implementation of optimized research biopsy analyses for clinical pharmacodynamic (PD) studies [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 802.