Cancer, Autoimmunity and Diabetes: A Critical Narrative Review of Shared Immunometabolic Pathways, Bidirectional Risks and Therapeutic Intersections
Domina Petric *
Department of Emergency Medicine, University Hospital Centre Zagreb, Zagreb, Croatia.
*Author to whom correspondence should be addressed.
Abstract
Cancer, autoimmune disease and diabetes mellitus are usually studied within separate clinical and scientific traditions, yet they share immune, metabolic and pharmacological pathways that link their occurrence, course and treatment. Epidemiological studies associate type 2 diabetes with an excess risk of several cancers, several autoimmune diseases with site-specific malignancy, and particular cancers with the onset of both autoimmunity and diabetes. The introduction of immune checkpoint inhibitors has converted these links into a routine clinical concern, because therapeutic release of anti-tumour immunity can precipitate insulin-deficient autoimmune diabetes. This critical narrative review examines the strength, consistency and causal credibility of the evidence connecting the three disease groups. Literature published between January 2000 and July 2026 was identified through searches of biomedical and multidisciplinary scholarly indexes, supplemented by citation tracking, and was appraised with attention to study design, confounding, time-related bias and replication. The synthesis indicates that the association between type 2 diabetes and cancers of the liver, pancreas and endometrium is robust to plausible unmeasured confounding, whereas many other site-specific associations are weaker and heterogeneous. Genetic studies implicate hyperinsulinaemia more consistently than hyperglycaemia. Type 1 diabetes shows a modest and site-specific pattern of cancer risk that is inconsistent across cohorts. Among systemic autoimmune diseases, the most consistent excess risk concerns lymphoma, and evidence from rheumatoid arthritis suggests that cumulative inflammatory activity contributes more than conventional treatment. Cancer can also initiate autoimmunity, as shown by tumour-associated autoantibodies in dermatomyositis and systemic sclerosis, and can cause diabetes, most clearly in pancreatic ductal adenocarcinoma. Checkpoint inhibitor-associated diabetes is rare, arises predominantly with programmed cell death protein 1 pathway blockade and differs from classical type 1 diabetes in its autoantibody profile and tempo. Observational claims that metformin prevents cancer have not been confirmed by randomised trials. Major unresolved questions include the causal role of insulin, the prediction of immune-related diabetes, and the oncological safety of immunomodulation in people with autoimmune disease. The evidence supports an integrated but differentiated model in which some links are established, others are plausible, and several remain speculative.
Keywords: Diabetes mellitus, neoplasms, autoimmunity, immune checkpoint inhibitors, hyperinsulinaemia, mendelian randomisation, paraneoplastic syndromes, metformin