Highlights
– The Metformin Active Surveillance Trial (MAST) randomized 408 men on active surveillance (AS) for low‑risk prostate cancer to metformin 850 mg twice daily versus placebo for up to 36 months and found no reduction in progression (HR 1.09; 95% CI 0.79–1.52; P = .59).
– Negative biopsy rates at 36 months favored metformin numerically (41.0% vs 31.1%) but did not reach statistical significance (P = .181).
– A prespecified subgroup analysis identified an unexpected adverse association between metformin and pathologic progression among obese men (BMI ≥ 30; HR 2.36; 95% CI 1.21–4.59; P = .0092), warranting further study.
Background
Active surveillance (AS) is the preferred management strategy for many men with low‑risk localized prostate cancer because it balances the low short‑term mortality risk with the high morbidity of curative treatments. Despite this, a notable proportion of men on AS undergo therapeutic intervention or show pathologic progression over time. Identifying safe, well‑tolerated agents that might delay progression and prolong the period free from definitive treatment is an attractive clinical goal.
Metformin, a biguanide widely used for type 2 diabetes, has been of interest as a cancer prevention and adjuvant agent because of preclinical data demonstrating antiproliferative effects (AMPK activation, mTOR inhibition, decreased hepatic gluconeogenesis and insulin levels) and observational studies reporting associations between metformin use and reduced incidence or improved outcomes in several malignancies. However, observational data are vulnerable to confounding by indication and healthy‑user biases; randomized evidence in prostate cancer prevention or progression is sparse.
Study Design
The Metformin Active Surveillance Trial (MAST) was a multicenter, randomized, double‑blind, placebo‑controlled phase III trial assessing whether metformin could delay progression in men with low‑risk, localized prostate cancer managed with AS. Eligible participants were randomized 1:1 to metformin 850 mg twice daily or matching placebo and followed for up to 36 months.
The primary endpoint was time to progression, defined as a composite of therapeutic progression (initiation of definitive therapy) and/or pathologic progression on surveillance biopsy. Progression‑free survival (PFS) was evaluated using Kaplan‑Meier estimates and compared between arms with Cox proportional hazards models. Prespecified subgroup analyses included stratification by baseline body mass index (BMI).
Key Findings
Enrollment and follow‑up: 408 patients were randomized (205 to metformin, 203 to placebo) with a median follow‑up of 36 months. During follow‑up, 144 participants experienced progression events (70 in the metformin arm, 74 in the placebo arm).
Primary outcome
There was no statistically significant difference in progression‑free survival between metformin and placebo (hazard ratio [HR] 1.09; 95% confidence interval [CI], 0.79 to 1.52; P = .59). The point estimate was close to unity and the CI excludes a large protective effect, indicating that metformin at the studied dose did not delay progression in this population over three years.
Secondary outcome — biopsy results
At 36 months, the negative biopsy rate (i.e., proportion of men without pathologic evidence of progression on biopsy) was 41.0% in the metformin arm versus 31.1% in the placebo arm. Although numerically higher in the metformin group, this difference did not meet statistical significance (P = .181). The discordance between the neutral primary endpoint and a non‑significant trend toward more negative biopsies in the metformin arm may reflect variability in biopsy schedules, sampling differences, or chance.
Prespecified subgroup analyses
A prespecified analysis by baseline BMI revealed a concerning interaction: among obese participants (BMI ≥ 30), metformin was associated with increased pathologic progression (HR 2.36; 95% CI, 1.21 to 4.59; P = .0092). This effect was not anticipated based on prior observational signals and was not seen in the overall cohort. The report highlights this as a potentially clinically important and hypothesis‑generating finding.
Safety and tolerability
The published summary did not provide detailed breakdown of adverse events in the abstracted results here; the full manuscript should be consulted for comprehensive safety data. Metformin is generally well tolerated, with gastrointestinal side effects and a rare risk of lactic acidosis in patients with significant renal impairment being the principal concerns.
Expert Commentary and Interpretation
Clinical implications: The MAST trial provides the highest‑level evidence to date addressing whether metformin can be repurposed to slow progression in men with low‑risk prostate cancer on AS. The negative primary result argues against adopting metformin for this indication in unselected men. The trial’s sample size and 36‑month follow‑up were sufficient to detect moderate effects; the observed HR and confidence interval suggest an absence of clinically meaningful benefit within this timeframe.
Possible explanations for the findings
– Biological heterogeneity: Prostate cancer is biologically heterogeneous, and mechanisms by which metformin might exert antitumor effects (insulin/IGF axis modulation, AMPK activation) may not be relevant across the spectrum of low‑risk disease or may require longer exposure.
– Timing and dose: The chosen dose (850 mg twice daily) is a common therapeutic dose for diabetes, but optimal dosing, tissue concentrations in the prostate, or need for higher exposure remain uncertain.
– Composite endpoint complexity: The primary outcome combined therapeutic and pathologic progression. Physician or patient decisions to initiate treatment can be influenced by non‑biologic factors (anxiety, PSA kinetics, imaging), which may dilute a pure biologic signal.
– Subgroup signal in obesity: The increased pathologic progression in obese men is unexpected and biologically plausible yet unexplained. Obesity is associated with altered insulin signaling, chronic inflammation, and changes in drug pharmacokinetics and tissue distribution. It is conceivable that metformin interacts with these processes in ways that differ by adiposity, but this finding could also represent a chance interaction. Importantly, the effect was prespecified and statistically significant, which increases credibility, but independent replication and mechanistic work are needed before changing practice.
Limitations
– Follow‑up duration: Three years is meaningful for early progression signals but may be insufficient to capture longer‑term effects on disease trajectory or delayed benefits.
– Lack of granular safety reporting in the summary: Full adverse event tables and adherence data are necessary to interpret tolerability and exposure‑response relationships.
– Generalizability: Trial participants were men with low‑risk disease who consented to AS and a pharmacologic trial; results may not generalize to all men with low‑risk prostate cancer, men with higher‑risk disease, or those with diabetes already taking metformin.
Clinical Recommendations and Research Directions
For clinicians: Based on MAST, routine prescription of metformin to men on AS for the purpose of reducing progression cannot be recommended. Clinicians should not infer benefit from observational studies and should counsel patients accordingly. The unexpected signal of increased pathologic progression among obese men should prompt caution; clinicians should not initiate metformin for AS in obese patients outside of clinical trials.
Future research priorities
– Replication and pooled analyses: Independent confirmation of the obesity interaction in other randomized datasets or pooled individual participant data is an urgent priority.
– Mechanistic studies: Laboratory and translational work to explore how adiposity modifies metformin’s actions in prostate tissue, including pharmacokinetics, inflammation, insulin signaling, androgen metabolism, and tumor microenvironment effects.
– Longer follow‑up: Continued follow‑up of MAST participants (if planned) and other prospective studies will determine whether late effects emerge.
– Biomarker‑directed approaches: Future trials might target subgroups selected by metabolic biomarkers (insulin, HOMA‑IR), tumor genomics, or imaging features to identify patients more likely to benefit or be harmed.
Conclusion
The MAST phase III randomized trial found no evidence that metformin 850 mg twice daily reduces progression for men with low‑risk prostate cancer managed with active surveillance over 36 months. A prespecified but unexpected finding of increased pathologic progression among obese men raises concern and requires replication and mechanistic investigation. These results underscore the importance of randomized evaluation of repurposed drugs and caution against adopting metformin for cancer control based on observational data alone.
Funding and clinicaltrials.gov
Funding sources, trial registration, and detailed acknowledgments are reported in the original manuscript: Fleshner NE et al., Metformin Active Surveillance Trial in Low‑Risk Prostate Cancer. J Clin Oncol. 2025 Oct 30:JCO2501070. Readers should consult that publication for full disclosures and the ClinicalTrials.gov identifier.
References
1. Fleshner NE, Bernardino RM, Izawa J, et al. Metformin Active Surveillance Trial in Low‑Risk Prostate Cancer. J Clin Oncol. 2025 Oct 30:JCO2501070. doi: 10.1200/JCO-25-01070. Epub ahead of print. PMID: 41166665.
2. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Prostate Cancer. Latest version available at https://www.nccn.org. (Consult the guideline for recommendations on active surveillance and management of localized prostate cancer.)
Note: For additional background on metformin’s proposed anticancer mechanisms and prior observational data, readers may consult recent reviews on metformin and cancer biology as well as systematic reviews of epidemiologic studies.
