Abstract We hypothesized that the associations between coffee intake and colorectal cancer (CRC) incidence might differ by immune cell densities in CRC tissue. Utilizing the Nurses’ Health Study and the Health Professionals Follow-up Study, we examined the association of coffee intake with incidence of CRC classified by intraepithelial or stromal T-cell subset densities by multiplex immunofluorescence assay for CD3, CD4, CD8, CD45RO (PTPRC), and FOXP3. We applied inverse probability weighted Cox proportional hazards regression model to control for selection bias and potential confounders. During follow-up of 133,924 participants (3,585,019 person-years), we documented 3,161 incident CRC cases, including 908 CRC cases with available data on T-cell densities in tumor tissue. The association between coffee intake and CRC was not statistically significantly different by intraepithelial or stroma T-cell subset (Pheterogeneity > .38). Hence, there is no sufficient evidence for differential effect of coffee intake on incidence of CRC subtypes classified by T-cell infiltrates.