Lower risk of cancer in patients on metformin in comparison with those on sulfonylurea derivatives: results from a large population-based follow-up study

Abstract

Objective: Numerous studies have suggested a decreased risk of cancer in patients with diabetes on metformin. Because different comparison groups were used, the effect magnitude is difficult to estimate. Therefore, the objective of this study was to further analyze whether, and to what extent, use of metformin is associated with a decreased risk of cancer in a cohort of incident users of metformin compared with users of sulfonylurea derivatives.

Research design and methods: Data for this study were obtained from dispensing records from community pharmacies individually linked to hospital discharge records from 2.5 million individuals in the Netherlands. The association between the risk of cancer in those using metformin compared with those using sulfonylurea derivatives was analyzed using Cox proportional hazard models with cumulative duration of drug use as a time-varying determinant.

Results: Use of metformin was associated with a lower risk of cancer in general (hazard ratio 0.90 [95% CI 0.88-0.91]) compared with use of sulfonylurea derivatives. When specific cancers were used as end points, similar estimates were found. Dosage-response relations were identified for users of metformin but not for users of sulfonylurea derivatives.

Conclusions: In our study, cumulative exposure to metformin was associated with a lower risk of specific cancers and cancer in general, compared with cumulative exposure to sulfonylurea derivatives. However, whether this should indeed be seen as a decreased risk of cancer for the use of metformin or as an increased risk of cancer for the use sulfonylurea derivatives remains to be elucidated.

Figure 1

Risk of cancer in patients when comparing cumulative exposure to metformin with cumulative exposure to sulfonylurea derivatives. The full model included the covariables age at first OGLD prescription, sex, calendar time (calendar year in which the first prescription was dispensed), hospitalizations (number of hospitalizations in the year before the start of the OGLD), and unique drugs (number of unique drugs dispensed in the year before the start of the OGLD). Full model A additionally included the average DDD, which was the dosage calculated over all previous OGLD prescriptions. Full model B was stratified for dosage of the first OGLD prescription lower than the median dosage. Full model C was stratified for dosage of the first OGLD prescription higher than the median dosage. Subanalysis A included a 1-year latency period, in which exposure was cumulated until 1 year before the date of the cancer diagnosis. Subanalysis B included only those with more than 1 year of exposure since the start of the OGLD. In subanalysis C, additional censoring took place at the moment metformin users started sulfonylurea and at the moment sulfonylurea users started metformin. Subanalysis D included only those treated with monotherapy metformin or sulfonylurea during the study period. Subanalysis E included only those who were treated with metformin as well as sulfonylurea derivatives during the study period (♦, HR; —, 95% CI).