AI–Discovery of Cellular Morphometric Biomarkers Reveals Unique Tumor Immune Microenvironments in High-Grade Serous Ovarian Cancer Patients of African Ancestry

High-Grade Serous Ovarian Carcinoma (HGSOC) is a highly heterogeneous disease. Machine learning-based cellular morphometric biomarkers (CMB-ML) have been identified across multiple tumor types, capturing tissue heterogeneity, predicting tumor microenvironments (TME) and clinical outcomes. We aimed to identify ethnicity-specific CMBs in African American and White HGSOC patients using whole-slide images (WSIs) and assess their associations with overall survival (OS). Analysis of 109 patients from The Cancer Genome Atlas Ovarian Cancer (TCGA-OV) cohort was validated in WSIs from 22 patients in the Loma Linda University (LLU) cohort. We linked immune checkpoint markers (ICM) to CMB scores and confirmed via immunohistochemistry. We identified and validated three ethnicity-specific CMBs — 73, 80, and 215 with significant frequency differences. Higher CMB 73 and 80 frequencies correlated with shorter OS in African Americans (p=0.022, p=0.023), while higher CMB 215 frequency was linked to improved OS in White patients (p=0.051). Molecular analysis of TCGA-OV cohort revealed lower immune infiltration in African Americans and higher ICM expression in Whites (PDCD1: p=0.033, PDCD1LG2: p=0.014, CD8A: p=0.014). Immunohistochemistry in the LLU-OV cohort showed expression of predicted markers CD3, CD8, and PDCD1. Although there were no significant differences between the two ethnic groups, CMB-ML explores a new avenue for understanding health disparities.