Pearson-Stuttard · Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 2021 · umbrella review of observational meta-analyses and Mendelian randomization studies · n=?

Type 2 Diabetes and Cancer: An Umbrella Review of Observational and Mendelian Randomization Studies.

Cited 266 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic umbrella review of observational meta-analyses and Mendelian randomization studies

PubMed 33737302 · doi:10.1158/1055-9965.EPI-20-1245 · record verified 2026-08-26

What was done

The authors conducted an umbrella review synthesizing evidence from published meta-analyses of observational studies evaluating associations between type 2 diabetes mellitus (T2DM) and site-specific cancer incidence or mortality. They also systematically reviewed Mendelian randomization (MR) studies examining genetically predicted T2DM, fasting insulin, and fasting glucose concentrations in relation to cancer risk.

What was found

Observational meta-analyses evaluated T2DM and cancer incidence across 18 sites, cancer mortality across 7 sites, and combined incidence or mortality across 4 sites. Strong or highly suggestive positive observational evidence was identified for six cancers: colorectal, hepatocellular, gallbladder, breast, endometrial, and pancreatic. Across MR studies (8 on genetically predicted T2DM, 7 on fasting insulin, and 8 on fasting glucose), genetically predicted T2DM and fasting insulin were positively associated with risk of six cancers: endometrial, pancreatic, kidney, breast, lung, and cervical. Genetically predicted fasting glucose showed no association with cancer risk except for squamous cell lung carcinoma. The abstract reported no numerical effect sizes or confidence intervals.

Why it matters

By combining observational meta-analyses with Mendelian randomization studies, this review highlights fasting insulin and diabetes-related genetic traits as potential causal drivers for specific cancers, whereas fasting glucose alone showed minimal association.

Limits

The abstract reports no numerical risk estimates, confidence intervals, or total participant sample sizes. Observational meta-analyses remain susceptible to residual confounding and exposure misclassification, and Mendelian randomization studies depend on genetic instrumental variable assumptions that cannot be verified from the abstract alone.

Cited by