In preclinical work and retrospective population studies, the anti-diabetic drug metformin has been associated with antineoplastic activity and decreased burden of many cancers, including pancreatic cancer. There is therefore interest in the hypothesis that this drug might be repurposed for indications in oncology. We aimed to assess the efficacy of the addition of metformin to a standard systemic therapy in patients with advanced pancreatic cancer, and provide the first report of a clinical trial with a survival endpoint of metformin for an oncological indication.

We did this double-blind, randomised, placebo-controlled phase 2 trial at four centres in the Netherlands. Patients aged 18 years or older with advanced pancreatic cancer were randomly assigned (1:1), via a permutated computer-generated block allocation scheme (block size of six) to receive intravenous gemcitabine (1000 mg/m²) on days 1, 8, and 15 every 4 weeks and oral erlotinib (100mg) once daily in combination with either oral metformin or placebo twice daily. Metformin dose was escalated from 500 mg (in the first week) to 1000 mg twice daily in the second week. Randomisation was stratified by hospital, diabetes status, and tumour stage. The primary endpoint was overall survival at 6 months in the intention-to-treat population. This trial is complete and is registered with ClinicalTrials.gov, number NCT01210911.

Between May 31, 2010, and Jan 3, 2014, we randomly assigned 121 patients to receive gemcitabine and erlotinib with either placebo (n=61) or metformin (n=60). Overall survival at 6 months was 63·9% (95% CI 51·9–75·9) in the placebo group and 56·7% (44·1–69·2) in the metformin group (p=0·41). There was no difference in overall survival between groups (median 7·6 months [95% CI 6·1–9·1] vs 6·8 months [95% CI 5·1–8·5] in the metformin group; hazard ratio [HR] 1·056 [95% CI 0·72–1·55]; log-rank p=0·78). The most frequent grade 3–4 toxic effects were neutropenia (15 [25%] patients in placebo group vs 15 [25%] in metformin group), skin rash (six [10%] vs four [7%]), diarrhoea (three [5%] vs six [10%]), and fatigue (two [3%] vs six [10%]).

Addition of a conventional anti-diabetic dose of metformin does not improve outcome in patients with advanced pancreatic cancer treated with gemcitabine and erlotinib. Future research should include studies of more potent biguanides, and should focus on patients with hyperinsulinaemia and patients with tumours showing markers of sensitivity to energetic stress, such as loss of function of AMP kinase, a key regulator of cellular energy homoeostasis.

Academic Medical Centre, Amsterdam, and The Terry Fox Foundation, Vancouver, Canada.