Anti-diabetic treatment and cancer occurrence among patients with type II diabetes mellitus

This is a retrospective cohort study, with nested case-control analyses, of type II diabetes patients, to evaluate the association between anti-diabetic treatment and cancer incidence. Methods: Patients who were initially stabilized on an oral hypoglycemic agent (OHA) mono-therapy, including metformin, sulfonylurea, TZD, and meglitinides, in the GPRD were included in the cohort. New diagnoses of cancer during cohort follow up were compared among exposure groups. In addition to the cohort analyses, solid tumors, as well as breast cancer and colorectal cancer, identified during the cohort follow-up were matched to controls and case-control analyses were performed within the cohort. Results: Compared to metformin, there was no difference in risk of malignant solid tumors or hematological malignancy with other major classes of OHAs in the cohort. The case-control analyses further supported the findings that these OHA classes or their combinations did not alter the risk of malignant solid tumors. iii In the case-control analyses, exposure to insulin was associated with a 64% (95% CI: 1.24, 2.16) higher risk of malignant solid tumors and this risk was modified when combined with different OHAs. When sulfonylurea was added to insulin, the risk of cancer was increased to almost 3 times (OR = 2.75, 95% CI: 1.51, 5.03), while adding metformin to insulin changed the odds ratio to 1.35 (95% CI: 0.94, 1.94). Similar results were found for breast cancer in which insulin was associated with a significant increase of more than 2 times in risk and other OHAs without insulin were not related to altered risk. For colorectal cancer (CRC), both sulfonylurea and TZD were associated with around a 2-fold significant risk increase compared to metformin. The odds of developing CRC were also elevated after starting insulin (OR = 1.41), although this did not reach statistical significance in these data. Conclusion: This study provides evidence that neither sulfonylurea nor TZD substantially alter the risk of solid tumor compared to metformin among type II diabetes patients. Insulin is associated with higher risk of cancer and this risk may be diminished by concomitant use of metformin, but appears to be magnified by concomitant use of sulfonylurea.Ph. D.Includes bibliographical referencesby Hong Qi