Abstract LB-154: Pathologic features of response to neoadjuvant anti-PD-1 in resected non-small cell lung carcinoma (NSCLC): A proposal for quantitative immune-related pathologic response criteria (irPRC)

Abstract

Abstract There is great interest in using PD-(L)1 blockading drugs as neoadjuvant therapy for patients with resectable NSCLC. Early results demonstrated a 45% (9/20) major pathologic response (MPR) rate in patients with Stage I-III NSCLC after receiving nivolumab (NCT02259621). Major pathologic response (MPR) criteria were developed in the context of cytotoxic chemotherapy, defined as ≤10% residual viable tumor cells (RVT). The features of immune-mediated tumor regression following anti-PD-1 have yet to be described. We reviewed H&E-stained slides from resection specimens in 19 patients treated with neoadjuvant nivolumab [n=9 MPR, n=3 partial responders, n=7 non-responders (>70% RVT)] to identify histopathologic features of immune-mediated tumor regression. Specimens were assessed for immune characteristics (tumor infiltrating lymphocyte (TIL) and macrophage density, and presence/absence of, lymphoid aggregates, tertiary lymphoid structures (TLS), dense plasma cell infiltrates, neutrophils, giant cells, etc.) and non-immune features (necrosis, hemosiderin, hyalinized and proliferative fibrosis). We found that immune-mediated tumor regression is characterized by a fibroinflammatory stroma with features of (1) immune activation, including dense TIL and macrophages, TLS, and granulomas; (2) massive [tumor] cell death, including cholesterol clefts and giant cells; and (3) tissue repair, including neovascularization and proliferative fibrosis (each enriched in MPR vs. non-responders, Fisher's exact test p<0.05). An “outside-in” pattern of regression was noted, which has important implications for defining total tumor bed area. As such, we propose “Immune-Related Pathologic Response Criteria” (irPRC), with tumor bed defined by RVT + necrosis + surrounding fibroinflammatory stroma. The areas of each are summed across all slides to calculate %RVT (RVT area/tumor bed area). This differs from chemotherapy MPR criteria, where %RVT is determined for each slide and then averaged, and the distinct fibroinflammatory regression stroma and peripheral regression bed are not acknowledged. The surgical resection specimens were then evaluated by four independent pathologists blinded to response to assess inter-observer variability. Compared to %RVT using chemotherapy criteria, irPRC had improved inter-observer variability [median per-case %RVT variability 5% (0-29%) vs. 10% (0-58%), paired t test p=0.007] and a two-fold decrease in median standard deviation across pathologists within a sample (4.6 vs 2.2, F-test p=0.002). We propose irPRC to standardize pathologic assessment of immune-mediated tumor regression and immunotherapeutic efficacy. Long-term follow up is needed to determine the reliability of irPRC as a surrogate for clinical outcomes such as recurrence-free and overall survival. Citation Format: Tricia R. Cottrell, Julie E. Stein, Jamie E. Chaft, Elizabeth D. Thompson, Natasha Rekhtman, Valsamo Anagnostou, Kellie N. Smith, Amy S. Duffield, Robert A. Anders, James M. Isbell, David R. Jones, Jonathan D. Cuda, Richard Battafarano, Stephen C. Yang, Peter B. Illei, Edward Gabrielson, Frederic Askin, Moises Velez, Matthew D. Hellmann, Jennifer L. Sauter, Ludmila Danilova, Victor E. Velculescu, Jedd D. Wolchok, Suzanne L. Topalian, Julie R. Brahmer, Drew M. Pardoll, Ashley Cimino-Mathews, Patrick M. Forde, Janis M. Taube. Pathologic features of response to neoadjuvant anti-PD-1 in resected non-small cell lung carcinoma (NSCLC): A proposal for quantitative immune-related pathologic response criteria (irPRC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-154.