Immune-Related Pathological Features and Predictive Score in Intrahepatic Cholangiocarcinoma Treated With Combined Conversion Immunotherapy.

Immune checkpoint inhibitor (ICI)-based combined neoadjuvant/conversion therapy is increasingly used in the management of initially unresectable intrahepatic cholangiocarcinoma (iCCA). Reliable biomarkers in predicting treatment and survival outcome remain underdeveloped. Here, we assessed immune-related pathological features in pretreatment biopsy specimens of initially unresectable iCCA treated with ICI-based conversion therapy. A predictive scoring system (immune therapy predictive score [iTPS]) was developed to predict the patients' oncological outcome. We evaluated 13 immune-related pathological features (tumor-infiltrating lymphocytes, mature tertiary lymphoid structure, lymphoid aggregate, dense plasma cell infiltration, granuloma, neutrophil, foamy macrophage, necrosis, cholesterol cleft, hemosiderin deposition, giant cell formation, neovascularization, and mature fibrosis) in hematoxylin-eosin-stained biopsy specimen slides from 95 iCCA patients. Cox regression analysis was used to evaluate the correlation between these features and recurrence-free survival (RFS) and overall survival (OS). Features showing correlation with survival were selected for iTPS to develop a binary iTPS scheme. Four features (immature fibrosis, tumor-infiltrating lymphocytes, lymphoid aggregate, and hemosiderin) were included in iTPS. A binary iTPS scheme showed patients with low iTPS displaying significantly better RFS (hazard ratio, 1.83; 95% CI, 1.02-3.27; P = .04) and OS (hazard ratio, 12.9; 95% CI, 3.07-54.18; P < .001). In conclusion, we developed an iTPS scheme based on routine hematoxylin-eosin-stained pretreatment biopsy slides. The iTPS can predict the RFS and OS of iCCA patients treated with combined ICI-based conversion therapy. iTPS is a good candidate predictive biomarker in iCCA and provides a new perspective in understanding the tumor immune microenvironment in immunotherapy.