Abstract PURPOSE Epidemiological studies suggest that biological sex influences overall survival (OS) in Glioblastoma (GBM), possibly due to immunological variances between the sexes. Further, while prognostic in other cancers, the association of tumor-infiltrating lymphocytes (TILs) to OS in GBM requires additional exploration. We hypothesize that histological features corresponding to spatial organization of TILs from within the GBM microenvironment, when evaluated in a sex-specific manner, will be associated with OS. METHODS Digitized Hematoxylin-and-Eosin (H&E) slides of resected GBM tumors were collected to establish a multi-site cohort of Nf435 patients. The training set included Nf201 patients from eleven contributing sites of the Cancer Genome Atlas (TCGA). The validation set included Nf234 patients consisting of cases from two hold-out TCGA sites and the Clinical Proteomic Tumor Analysis Consortium. Each digitized H&E slide was assessed for quality, followed by employing a pre-trained classifier to label TIL and non-TIL cells. Spatial features related to the connectivity of nuclei centroids and corresponding cell cluster graphs were extracted for TILs, non-TILs, and between-group interactions. These TIL-centric histological features were fed into a Cox Hazard regression model with least absolute shrinkage and selection operator feature selection, to stratify patients as low or high-risk. Survival models were evaluated using Kaplan-Meier estimates, across male, female, and combined cohorts. RESULTS Selected features (i.e., measures of area and proximity across TIL clusters) were significantly associated with OS in the validation cohort (hazard ratio (HR)=1.95, confidence interval (CI)=1.19–3.2, p=0.00278), for the female cohort. For the male cohort, the selected spatial TIL features’ association with OS for the validation cohort yielded HR=1.43, CI=0.986–2.07, p=0.0667. The spatial TIL survival validation model on the combined cohort did not yield significant survival risk stratification. CONCLUSION Our findings suggest that sex-specific TIL organization histological features may be prognostic of OS in GBM tumors, specifically for the female cohort.