Spatial profiling reveals distinct immune microenvironments in early vs late onset colorectal cancer

Colorectal cancer (CRC) is the fourth most common cancer and the second leading cause of cancer related deaths in the United States. While CRC incidence increases with age, rates are rising among individuals younger than 50 years old (early-onset CRC; EOCRC). Most sporadic EOCRC tumors are microsatellite stable (MSS), and MSS tumors generally do not respond favorably to immunotherapy. The age-associated differences within the tumor microenvironment (TME) remain poorly understood. We hypothesize that EOCRC tumors have a more immunosuppressive microenvironment, which may influence treatment response. Methods: Tumor tissues were collected under an approved IRB protocol from Waterbury Hospital and UConn John Dempsey Hospital. Patients ≤50 years were classified as EOCRC (n = 8) and ≥60 years as late-onset (LOCRC; (n = 10). A total of eighteen regions of interest (ROIs) from tumor sections were selected by a pathologist. Imaging mass cytometry (IMC; 27-marker panel) was performed to spatially characterize the epithelial and immune cell populations. Images were processed to identify single cells, the data were adjusted to remove technical differences, and cells were grouped based on markers and visualized using MCD Viewer software. We also examined how different cell types were spatially arranged and interacted within the tissue. Statistical comparisons were performed using Welch's t-test and permutation-based tests. Results: Using IMC, a total of 142,992 single cells across 18 ROIs were analyzed. Unsupervised clustering identified 19 cell populations, including epithelial (40%), stromal (18%), CD45⁺ lymphocytes (11%), myeloid subsets (11%), and rare intra-epithelial lymphocytes (<1%). Cluster identities were validated by marker expression and spatial localization. EOCRC tumors showed a significantly higher abundance of FoxP3⁺ CD4⁺ Tregs (P = 0.047), indicating an immunosuppressive TME. LOCRC samples trended toward higher intra-epithelial lymphocytes. Spatial analysis suggested a direct Treg-M2 macrophage interaction due to their colocalization, suggesting the possibility of suppressive niches forming within EOCRC. Immune-enriched stromal regions were slightly more abundant in EOCRC (20.3% of cells) than in LOCRC (16.5%). These regions were enriched with Tregs, CTLs, and M2 macrophages, with Treg infiltration being significantly higher in EOCRC (P = 0.032). These findings support a localized immunosuppressive phenotype. Conclusion: EOCRC tumors display a more immunosuppressive microenvironment, characterized by increased Treg infiltration and a higher M2 macrophage presence in immune cell-enriched regions. These findings suggest that age-related immune differences may exist within tumors of CRC patients. Ongoing validation by multiplexed immunohistology will strengthen these results and inform the development of age-tailored therapeutic strategies.