Proceedings: AACR Annual Meeting 2020; April 27-28, 2020 and June 22-24, 2020; Philadelphia, PA
Introduction: The tumor microenvironment can be classified into three subtypes: the immune inflamed, excluded and desert, by pathologic evaluation of tumor tissue. The three subtypes demonstrate different sensitivity to immune checkpoint inhibitors (ICIs). However, the transcriptomics and biologic translation of the subtypes in non-small cell lung cancer (NSCLC) are not fully investigated. We previously developed a deep learning-based algorithm, Lunit-SCOPE, which describes tissue phenomics and predicts ICI response. Here, we utilize quantitative parameters provided by Lunit-SCOPE to classify NSCLC into the three immune subtypes and validate the biological relevance of this classification by describing their transcriptomic and immunologic landscape.
Methods: For the current analysis, tissue H&E images and sequencing data from The Cancer Genome Atlas (TCGA) were utilized. Lunit-SCOPE was trained with 1,824 H&E images of advanced NSCLC from Samsung Medical Center which were annotated by pathologists. It identifies lymphocytes and quantifies the density in the area of cancer epithelium (CE-Lym), cancer stroma (CS-Lym) and the combined area (C-Lym). Tumors in the lowest tertile of C-Lym were classified as Immune desert. Then, the remaining tumors were classified with the ratio of CE-Lym / CS-Lym into Immune inflamed and Immune excluded by the median. The xCell was used to estimate the proportion of cell subsets from the bulk sequencing data. We then identified differentially expressed genes between the three subtypes and further analyzed the differentially enriched gene sets between Immune inflamed and Immune excluded.
Results: Among 965 samples of NSCLC, the frequency of inflamed, excluded and desert subtypes were 201 (40.7%), 138 (27.9%), and 155 (31.4%) in LUAD, and 122 (25.9%), 185 (39.3%), and 164 (34.8%) in LUSC. Tumors of inflamed subtype had higher proportion of CD8-positive T cells (Inflamed: 2.4 [0.5-5.4] % vs others: 0.9 [0-2.5]%, P<.001) and M1 macrophages (Inflamed: 9.2 [5.6-13.0] % vs others: 5.8 [3.3-8.7]%, P<.001) in tumor microenvironment. The cytolytic activity score was higher in the inflamed subtype (194 [107-371] (inflamed) vs 167 [92-336] (others); P=0.03). Interactive genomic analysis showed that inflamed subtype significantly enriched interferon-gamma pathway (NES, 2.1; FDR P-value=0.008). On the other hand, excluded subtype enriched glycolysis (NES, -2.05; FDR P-value=0.02), fatty acid (NES, -1.49; FDR P-value = 0.02) and cholesterol (NES, -1.87; FDR P-value=0.02) metabolism pathways.
Conclusion: Tissue H&E based tissue phenomics is applicable and objectively classifies tumors from patients with NSCLC into three immune subtypes. The distinct transcriptomic and immunologic landscape demonstrates the biological relevance of the classification. The transcriptomic enrichment of glycolysis and cholesterol metabolism pathway in the immune excluded subtype suggests a potential mechanism of immune evasion.
Citation Format: Jonghanne Park, Chan-Young Ock, Kyunghyun Paeng, Sehhoon Park, Horyun Choi, Dongyup Lee, Gahyun Gim, Sukjoo Cho, Yoon La Choi, Se-Hoon Lee, Young Kwang Chae. Comprehensive deep learning analysis of H&E tissue phenomics reveals distinct immune landscape and transcriptomic enrichment profile among immune inflamed, excluded and desert subtypes in non-small cell lung cancer [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 5076.
Jonghanne Park, Chan-Young Ock, Kyunghyun Paeng, Sehhoon Park, Horyun Choi, Dongyup Lee, Gahyun Gim, Sukjoo Cho, Yoon La Choi, Se-Hoon Lee, Young Kwang Chae
Abstract: Deep learning-based predictive biomarker for adjuvant chemotherapy in early-stage hormone receptor-positive breast cancer
Abstract: Pan-cancer analysis of tumor microenvironment using deep learning-based cancer stroma and immune profiling in H&E images
Abstract: Deep learning-based predictive biomarker for immune checkpoint inhibitor response in metastatic non-small cell lung cancer
Abstract: Deep-learning analysis of H&E images to define three immune phenotypes to reveal loss-of-target in excluded immune cells as a novel resistance mechanism of immune checkpoint inhibitor...
Abstract: Deep learning-based immune phenotype analysis reveals distinct resistance pattern of immune checkpoint inhibitor in non-small cell lung cancer
Reducing Domain Gap by Reducing Style Bias
Predicting breast tumor proliferation from whole-slide images: The TUPAC16 challenge
Accuracy and efficiency of an artificial intelligence tool when counting breast mitoses
Abstract: Artificial intelligence-powered spatial analysis of tumor infiltrating lymphocytes (TIL) to reflect target gene expressions of novel immuno-oncology agents.
Abstract: Distinct subset of immune cells assessed by multiplex immunohistochemistry correlates with immune phenotype classified by ...
Abstract : Pathologic validation of artificial intelligence-powered prediction of combined positive score of PD-L1 immunohistochemistry in urothelial carcinoma.
Abstract : Clinical performance of artificial intelligence-powered annotation of tumor cell PD-L1 expression for treatment of immune-checkpoint inhibitor (ICI) ...
Abstract : Artificial intelligence-powered spatial analysis of tumor-infiltrating lymphocytes predicts survival after immune checkpoint inhibitor therapy across multiple cancer types.
Abstract : Deep learning based radiomic biomarker for predicting the presence of high-grade histologic patterns in lung adenocarcinoma
Abstract : Artificial intelligence-powered tissue analysis reveals distinct tumor-infiltrating lymphocyte profile as a potential biomarker of molecular subtypes in endometrial cancer
Abstract : Artificial intelligence-powered spatial analysis of tumor-infiltrating lymphocytes reveals distinct genomic profile of immune excluded phenotype in pan-carcinoma
Deep learning from HE slides predicts the clinical benefit from adjuvant chemotherapy in hormone receptor-positive breast cancer patients
Using artificial intelligence to read chest radiographs for tuberculosis detection: A multi-site evaluation of the diagnostic accuracy of three deep learning systems
Abstract : Interim results of phase I dose escalation study of YBL-006: A novel anti-PD-1 monoclonal antibody in advanced solid tumors
Abstract : Assistance with an artificial intelligence-powered PD-L1 analyzer reduces interobserver variation in pathologic reading of tumor proportion score in non-small cell lung cancer
Abstract : AI-powered whole-slide image analysis of tumor-infiltrating lymphocytes for prediction of prognosis in colorectal cancer
Deep Learning Analysis of CT Images Reveals High-Grade Pathological Features to Predict Survival in Lung Adenocarcinoma