The Risk of Multiple Myeloma Is Reduced in Metformin Initiators: A Retrospective Cohort Study in Taiwanese Patients with Type 2 Diabetes Mellitus

SIMPLE SUMMARY: Metformin exerts anti-cancer effects but its effect on multiple myeloma requires investigation. This study used the nationwide database of Taiwan’s National Health Insurance to examine whether metformin use in patients with type 2 diabetes mellitus would have a reduced risk of multiple myeloma. Intention-to-treat analyses showed that patients who receive metformin treatment within the first 12 months of prescription of antidiabetic drugs have an approximately 30% lower risk than those who do not. In the per-protocol analyses, patients who adhere to metformin treatment will have an even lower risk reduction of approximately 65%. The findings of this study support an anti-cancer effect of metformin on multiple myeloma and provide a good reason for the recommendation of metformin as the first-line antidiabetic drug for patients with type 2 diabetes mellitus. In patients without contraindications, patients should be advised to maintain on metformin use because of its multiple pleiotropic benefits. ABSTRACT: Background: Whether metformin might reduce the risk of multiple myeloma (MM) has not been extensively researched in humans. Methods: The study subjects were enrolled from the reimbursement database of Taiwan’s National Health Insurance. A total of 739,553 patients who had a new diagnosis of type 2 diabetes mellitus during 1999–2009 were identified. They were categorized as metformin initiators (metformin (+)) and non-metformin initiators (metformin (−)) based on the prescriptions of antidiabetic drugs that included metformin and did not include metformin within the initial 12 months, respectively. MM incidence was calculated after the initial 12 months of treatment group assignment until 31 December 2011. Hazard ratios based on intention-to-treat (ITT) and per-protocol (PP) approaches were estimated by Cox regression weighted by propensity scores. Results: In the ITT analyses, the respective incidence rates for 497,248 metformin (+) and 242,305 metformin (−) were 9.97 and 14.33 per 100,000 person-years. The hazard ratio that compared metformin (+) to metformin (−) in the ITT analysis was 0.710 (95% confidence interval 0.593–0.850). In the PP analysis, the respective incidence rates were 5.14 and 13.98 per 100,000 person-years, and the hazard ratio was 0.355 (95% confidence interval, 0.270–0.466). The lower risk of MM among metformin (+) was supported by subgroup and sensitivity analyses. Conclusions: Type 2 diabetes patients who are initiated with metformin treatment have a significantly lower risk of MM, especially when they adhere to metformin treatment.