Abstract 2680: Exploration of tumor environment classification using mIHC in pan-cancer tissue samples

Abstract

Abstract Background: Besides three major biomarkers (PD-L1, TMB and MSI), tumor immune microenvironment is also elemental to predict how patients respond to immunotherapy and their prognosis. According to PD-L1 expression and tumor infiltrating lymphocytes, the tumor environment was classified into four types by Chen L. The multiplex immunohistochemistry (mIHC) provides the possibility to characterize the infiltrating immune cells in a single tissue section and classified patients more precisely with multi-biomarkers. Methods: We collected FFPE tissues from 170 cancer patients and performed Tumor Immune Microenvironment tests using a validated commercial assay, which included PD-L1 test (22C3, DAKO) and 7-plex mIHC to quantify and locate CD8+ T cells, M1 macrophages, M2 macrophages, CD56dim NK cells and CD56bright NK cells. CD8+ T cell positive was defined as ≥median CD8+ T cells density (cells/mm2) in tumor. Positive PD-L1 expression was defined as ≥1% of tumor cells with membranous staining(TPS). Results:In our study, the tissues were collected from colorectum cancer(n=46), biliary tract cancer(n=40), liver cancer(n=38), lung cancer(n=18)and others (n=28). Of the 170 patients, 44.1% were identified as tumor environment Type I (PDL1- /CD8-),11.1% Type II (PDL1+ /CD8+),38.8% Type III (PDL1- /CD8+) and 5.9% Type IV (PDL1+ /CD8-). Type II patients were the most likely subgroup benefit from anti-PD1/PDL1 therapy, who make up 22.2%, 12.5%, 7.8% and 4.3% in Lung, biliary tract cancer, liver cancer, colorectum cancer, respectively. We used unsupervised clustering to cluster patients into 3 groups: immune hot, immune cold and intermediate immunosuppressed, according to the above five kinds of immune cells. The immune hot environment was rich in immune cells and the immune cold environment was lack of immune cells, between which was the intermediate immunosuppressed environment. The proportion of immune hot was 23.5% in 170 patients, 44.8% in PDL1 positive patients and 19.1% in PDL1 negative patients. In addition, no correlation between Immune cells density in tumor /stroma and PD-L1 expression was observed. In our cohort, there was one lung adenocarcinoma patient, female, 37 years old, had positive PDL1 expression (TPS=95%), low TMB, MSS, EGFR/ALK- and carried KRAS G12D mutation, but was identified as immune cold tumor environment, who also had lower immune cell density in tumor than in stroma. During three months of receiving pembrolizumab plus pemetrexed and carboplatin, she quickly had multiple sites metastases. Conclusions: In our study, PDL1 expression and Tumor immune microenvironment are independent. And tumor immune microenvironment varies between Cancer types. Taking multi-biomarkers into consideration might help to evaluate immunotherapy response and resistance comprehensively. Citation Format: Tianhang Luo, Bei Zhang, Yifan Zhou, Fei Wang, Hui Chen, Xiaochen Zhao, Yuezong Bai, Li Li. Exploration of tumor environment classification using mIHC in pan-cancer tissue samples [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 2680.