The focus on diabetic mortality in cancer patients remains superficial. The objective of our study is to identify cancer patients with the highest risk of diabetic mortality compared with other cancer patients and the general US population. The representative data of cancer patients dying from diabetes between 1975 and 2016 was retrieved from the Surveillance, Epidemiology, and End Results program. Standardized mortality ratios (SMRs) and excess risks for multiple cancer sites were calculated. Cox regression analysis was performed to identify potential risk factors of death from diabetes. Among 9,043,788 cancer patients diagnosed between 1975 and 2016, 51,611 patients died from diabetes. the SMR of diabetic death was 2.15 compared with the general population. Malignancies of pancreas, liver, and brain had the highest SMR (>15) compared with the general population. The risk of death from diabetes was increasing in more recent years. The majority of deaths from diabetes occurred in those >45 years of age diagnosed with prostate, breast, colorectum, lung, or bladder cancer. Patients with older age, male sex, black race, higher histologic grade, unmarried status, and not undergoing surgery are at higher risk compared with other cancer survivors. Compared to the general population, cancer patients are at elevated risk of death from diabetes throughout the follow-up period. Elderly, black, unmarried males with distant metastases and without receiving surgery are recommended with earlier detection and more efficient diabetic care, especially for those with prostate, breast, colorectum, lung, or bladder cancer.