Pharmacologic Therapy of Diabetes and Overall Cancer Risk and Mortality: A Meta-Analysis of 265 Studies.
Level 1 - systematic review of randomized trials
Systematic review and meta-analysis incorporating 182 randomized controlled trials alongside observational studies.
PubMed 26076034 · doi:10.1038/srep10147
What was done
A systematic review and meta-analysis was conducted across 265 studies (182 randomized controlled trials, 44 cohort studies, and 39 case-control studies) to evaluate the association between various anti-diabetic medications (ADMs) and cancer incidence. The dataset comprised approximately 7.6 million patients from observational studies and 137,540 patients from randomized controlled trials.
What was found
Overall risk of bias was moderate across included studies. A lower risk of incident cancer was observed with metformin (RR = 0.86, 95% CI 0.83–0.90; I² = 88.61%) and thiazolidinediones (RR = 0.93, 95% CI 0.91–0.96; I² = 0.00%). Conversely, an increased risk was found with insulin (RR = 1.21, 95% CI 1.08–1.36; I² = 96.31%), sulfonylureas (RR = 1.20, 95% CI 1.13–1.27; I² = 95.02%), and alpha-glucosidase inhibitors (RR = 1.10, 95% CI 1.05–1.15; I² = 0.00%). Other classes of ADMs showed no significant association with cancer risk.
Why it matters
This large meta-analysis provides comparative estimates across diabetes drug classes, offering quantitative data to inform medication choices in patients with diabetes who have heightened baseline cancer risk.
Limits
The overall risk of bias was moderate, and analyses for metformin, insulin, and sulfonylureas demonstrated extreme statistical heterogeneity (I² > 88%). The abstract does not break down results by specific cancer sites or separate RCT outcomes from observational data.
Cited by
- context Metformin has been shown in short-term studies to possibly prevent cancer prophylactically.