Abstract 597: BLEACH&STAIN 15 marker multiplexed imaging in 3098 human carcinomas revealed six major PD-L1 driven immune phenotypes with distinct spatial orchestration

Abstract

Abstract Multiplex fluorescence immunohistochemistry (mfIHC) approaches were yet either limited to 6 markers or limited to a small (1.5cmx1.5cm) tissue size that hampers translational studies on large tissue microarray (TMA) cohorts. To be able to assess more marker in a large patient cohort, we have developed a BLEACH&STAIN multiplex fluorescence immunohistochemistry approach that enabled the simultaneous analysis of 15 biomarkers. To study the relationship of PD-L1 expression on multiple different cell types and the relationship with various lymphocyte subtypes, PD-L1, PD-1, CTLA-4, panCK, CD68, CD163, CD11c, iNOS, CD3, CD8, CD4, FOXP3, CD20, Ki67, and CD31 were analyzed in 3098 tumor samples from 44 different tumor types. An artificial intelligence-based framework - incorporating three different deep learning systems - for automated marker quantification on tumor as well as immune cells was further established to study the spatial interplay between PD-L1 expression and the composition of tumor infiltrating leucocytes (TILs). Comparing the automated deep learning-based PD-L1 quantification with conventional brightfield PD-L1 data revealed a high concordance in tumor cells (p<0.0001) as well as immune cells (p<0.0001) and an accuracy of the automated PD-L1 quantification ranging from 90% to 95.2%. Unsupervised clustering showed that a major proportion of the three PD-L1 phenotypes (i.e., PD-L1+ tumor and immune cells [G1], PD-L1+ immune cells [G2], PD-L1 negative [G3]) were either inflamed (G1.1, G2.1, G3.1) or non-inflamed (G1.2, G2.2, G3.2). In the inflamed PD-L1+ patients (G.1.1), spatial analysis revealed that an elevated intratumoral CD68+CD163+ M2 macrophages as well as CD11c+ dendritic cell infiltration (p<0.001 each) was associated with a high (CD3+CD4±CD8±FOXP3±) T-cell exclusion and a high PD-1 expression on T-cells (p<0.001 each). In breast cancer, a particular poor prognosis for the non-inflamed PD-L1+ breast cancer patients (G1.2, G2.2) was found and the PD-L1 fluorescence intensity on tumor cells showed a significantly higher predictive performance for overall survival with an area under receiver operating curves (AUC) of 0.72 (p<0.001) than the percentage of PD-L1+ tumor cells (AUC: 0.54). In conclusion, BLEACH&STAIN mfIHC in combination with a deep learning-based framework for automated PD-L1 assessment on tumor and immune cells enabled a rapid and comprehensive assessment of PD-L1 expression in different cell types and their interrelation with inflammatory cells. Our approach enabled the identification of six major PD-L1 phenotypes ranging from an PD-L1+ tumor cell inflamed phenotype (G1.1) with a spatial T-cell exclusion to a non-inflamed PD-L1+ immune cell phenotype showing a particular poor prognosis (G2.2) to a non-inflamed PD-L1 negative phenotype (G3.2). Citation Format: Elena Bady, Katharina Möller, Tim Mandelkow, Ronald Simon, Maximilian Lennartz, Claudia Hube-Magg, Guido Sauter, Niclas C. Blessin. BLEACH&STAIN 15 marker multiplexed imaging in 3098 human carcinomas revealed six major PD-L1 driven immune phenotypes with distinct spatial orchestration [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 597.