Objective Cancer of unknown primary (CUP) is a metastatic malignancy with an unidentifiable primary tumour origin. Previous studies suggest that type 2 diabetes mellitus (T2DM) and CUP risk are associated. This study examines the association in greater depth by investigating T2DM status, T2DM duration, T2DM age at diagnosis, and medication that was being used in relation to CUP. Methods Data were utilized from the Netherlands Cohort Study, a prospective cohort that includes 120 852 participants aged 55–69 years at baseline in 1986. All participants completed a self-administered questionnaire on cancer risk factors. CUP cases were identified through record linkage with the Netherlands Cancer Registry and Dutch Pathology Registry. After 20.3 years of follow-up, 963 incident CUP cases and 4288 subcohort members were available for case-cohort analyses. Proportional hazards models were employed to estimate multivariable-adjusted hazard ratios (HRs). Results Overall, we observed a nonsignificant positive association between T2DM status and CUP risk [HR, 1.35; 95% confidence interval (CI), 0.92–1.99], which increased in women after stratification for sex (HR, 1.55; 95% CI, 0.90–2.64). For participants who were aged less than 50 years at diagnosis of T2DM, a statistically significant positive association was found in relation to CUP (HR, 2.42; 95% CI, 1.26–4.65), compared with participants without T2DM. Conclusion Our findings indicate that there is a nonsignificant positive association between T2DM and CUP risk and that the association became stronger in women in stratified analyses.