Metabolic & Cardiometabolic
metformin decreases cancer incidence
In plain terms: Does metformin reduce cancer risk?
Contested Metabolic & Cardiometabolic 🔬 Includes disconfirming
Part of: 💊 metformin
RefutedContestedStrong support
consensus score -0.03
Contested — observational cohorts and their meta-analyses suggest lower cancer incidence, but those associations are heavily confounded by time-related (immortal-time) bias, and bias-corrected analyses, target-trial emulations, and the one cancer-endpoint RCT show little or no true effect.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
6 support 4 contradict 2 tested null 8 mixed · 20 sources, 10 independent groups
What the evidence shows
Observational cohorts and their meta-analyses associate metformin with lower cancer incidence, and two small randomised trials in 405 non-diabetic adults after polypectomy found fewer colorectal adenomas. The cohort studies are prone to time-related biases that inflate the effect, and genetic instruments give mixed answers, so this stays contested rather than supported.
The evidence (22)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Ansar F et al 2026 · Qeios | observational | supports | moderate | Meta-analysis of 2 RCTs, n=405, non-diabetic post-polypectomy adults. Metformin reduced 1y colorectal adenoma recurrence: RR 0.61 (0.48-0.78), low risk of bias. |
| Nie Z, Zhu H, Gu M 2016 · Pharm Biol | meta-analysis | supports | low | Meta of observational studies: metformin associated with reduced colorectal cancer incidence in T2D; observational-only, susceptible to bias. |
| Lee HS, Kang SH, Eun HS, Lee ES, Moon HS, Kim SH, Sung JK, Lee BS 2026 · The Korean journal of helicobacter and upper gastrointestinal research | observational | supports | low | Retrospective cohort, T2D endoscopy patients, Korea. Metformin (no insulin) assoc w/ lower gastric adenocarcinoma incidence; no effect on dysplasia. |
| Feng Y, Qiao Y, Li H, Shen B, Ding J, Hua D 2026 · BMC cancer | observational | supports | low | Two-sample Mendelian randomization (IVW), GWAS-derived SNP instruments. Genetically-proxied metformin exposure assoc w/ lower lung cancer risk: OR 0.249 (0.065-0.950, P=0.041). |
| Col NF, Ochs L, Springmann V, Aragaki AK, Chlebowski RT 2012 · Breast Cancer Res Treat | meta-analysis | mixed | moderate | Breast cancer meta + critical review: apparent risk reduction but authors explicitly flag methodological bias undermining the association. |
| Getu MA, Zhang X, Wang J, Xu K, Ying Y, Ma L, Gong P 2026 · Current cancer drug targets | meta-analysis | mixed | moderate | Meta-analysis, 18 studies (10 cohort+8 RCT), 1.3M participants. OS HR 1.02 (0.80-1.30, ns); PFS HR 1.00 (0.76-1.31, ns); but cancer incidence HR 0.90 (0.86-0.94, reduced). |
| Yu OHY, Suissa S 2023 · Diabetes Care | observational | contradicts | high | Narrative methodological review, no new analysis: earlier studies were 'affected by time-related biases, such as immortal time bias'; 'The observational studies that avoided these biases did not find an association'; the largest adjuvant RCT (3,649 women) 'found no benefit'. The old extract's 're-analysis' is not this paper's. Review — zero weight. |
| Huang Y, Santoni G, Sparrelid E, Xie SH 2026 · JNCI cancer spectrum | observational | contradicts | high | Swedish nationwide cohort, n=744,173, Cox regression. Metformin assoc w/ INCREASED HCC risk (HR 1.15, 95%CI 1.02-1.30); decreased in obesity subgroup (HR 0.56); null for cholangiocarcinoma. |
| Campbell JM, Bellman SM, Stephenson MD, Lisy K 2017 · Ageing Res Rev | meta-analysis | mixed | low | Meta reports lower all-cause mortality and some cancers in metformin users vs non-diabetics, but design cannot separate drug effect from healthy-user/survivor bias. |
| Mehta V, Fraile JF, Mathur A, Bhadania M, Galletti C 2026 · BMC cancer | meta-analysis | mixed | low | Umbrella review, 5 studies, head & neck cancer. Metformin assoc w/ 0.74x survival (OS) vs non-users (95%CI 0.67-0.81); authors flag publication bias (Egger's test) and low confidence. |
| Jackson JW, García-Albéniz X 2018 · Cancer Epidemiol Biomarkers Prev | observational | contradicts | high | Methods commentary (PubMed: Comment), no own data: 'In this commentary, using the study of metformin use on colorectal cancer risk by Bradley and colleagues (2018) as motivation, we compare the etiologic versus action-focused analysis of epidemiologic data.' No metformin–cancer estimate in the abstract. Review — zero weight. |
| Hiroki S, Fukasawa T, Honda M, Kawakami K 2026 · Pharmacoepidemiology and drug safety | observational | contradicts | high | Target trial emulation, metformin vs DPP-4i, T2D, Japan claims data. 5y CRC risk 1.55% vs 1.26%; RR 1.23 (0.91-1.66). No CRC benefit. |
| Shen X, Luo S, Zheng J, Chui CSL, Wong ICK, Wan EYF, Schooling CM, Au Yeung SL 2026 · Diabetes, obesity & metabolism | observational | mixed | moderate | Drug-target Mendelian randomization, 11 metformin-target genes vs 5 cancers. PRKAG1 (AMPK-gamma1) assoc w/ reduced CRC risk: OR 0.74 per mmol/mol HbA1c reduction (0.63-0.87, P=0.001), HbA1c-independent. |
| Xu S, Zheng B, Su B, Finkelstein SN, Welsch R, Ng K, Shahn Z 2025 · Am J Epidemiol | observational | contradicts | high | Target-trial emulation vs sulfonylureas accounting for biases: metformin did NOT reduce cancer incidence — well-designed null undercutting the protective claim. |
| Lei D, Wu JH, Zeng XH, Xu R, Zheng SS, Zhu XJ, Yang JA 2026 · Urol Oncol | meta-analysis | tested-null | moderate | SR/MA (PROSPERO CRD420251149266) of 85 cohort, case-control and RCT studies across all antidiabetic drug classes in T2D. For metformin the abstract reports only 'No significant association' with bladder cancer; the metformin study count, pooled OR and CI are not given. Single cancer site; mostly observational data. |
| Zhang S, Chen B, Hu X, Hu S, Li Y, Han J 2026 · Frontiers in pharmacology | meta-analysis | tested-null | moderate | Meta-analysis of 5 RCTs, 705 patients, gynecologic cancers, metformin+standard therapy. PFS HR 0.76 (0.55-1.03, ns); OS HR 1.20 (0.88-1.62, ns). No survival benefit. |
| Szymczak-Pajor 2026 · Int J Mol Sci | meta-analysis | supports | low | Meta-analysis of 16 observational studies found metformin use associated with reduced pancreatic cancer risk versus other antidiabetic drugs. |
| O'Connor L, Bailey-Whyte M, Bhattacharya M, Butera G, Hardell KNL, Seidenberg AB, Castle PE, Loomans-Kropp HA 2024 · J Natl Cancer Inst | meta-analysis | mixed | moderate | Updated systematic review/meta: pooled observational data show lower cancer incidence with metformin, but authors stress residual confounding and bias limit causal inference. |
| Peerless Y, Strauss G, Margalit O, Shacham-Shmueli E, Yang YX, Boursi B 2026 · Cancer epidemiology | observational | mixed | moderate | VA nested case-control, n=31,078 CRC cases/310,621 controls, diabetics. No effect right-sided CRC (OR~1.0); reduced left-sided CRC (OR 0.90, 0.82-0.98 at 1-3y; 0.87 at 3-5y). |
| Suissa S, Azoulay L 2012 · Diabetes Care | observational | contradicts | high | Demonstrates time-related (immortal-time, time-window) biases inflate apparent metformin anticancer benefit; correcting them largely abolishes the association. |
| Sacco M, Ribaldone DG, Saracco GM 2023 · Viruses | observational | supports | low | Cohort in diabetic chronic-HCV patients: metformin use associated with reduced hepatocellular carcinoma; observational, confounding-prone. |
| Higurashi T, Hosono K, Takahashi H, Komiya Y, Umezawa S, Sakai E, Uchiyama T, Taniguchi L, Hata Y, Uchiyama S, Hattori A, Nagase H, Kessoku T, Arimoto J, Matsuhashi N, Inayama Y, Yamanaka S, Taguri M, Nakajima A 2016 · Lancet Oncol | RCT | mixed | moderate | Only cancer-endpoint RCT: low-dose metformin modestly reduced metachronous colorectal adenoma recurrence in non-diabetics — small, single-center, surrogate (adenoma not cancer) endpoint. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.