Association between metformin medication, genetic variation and prostate cancer risk.

BACKGROUND

The relationship between metformin use and prostate cancer risk remains controversial. Genetic variation in metformin metabolism pathways appears to modify metformin glycemic control and the protective association with some cancers. However, no studies to date have examined this pharmacogenetic interaction and prostate cancer chemoprevention.

METHODS

Clinical data and germline DNA were collected from our prostate biopsy database between 1996 and 2014. In addition to a genome-wide association study (GWAS), 27 single nucleotide polymorphisms (SNPs) implicated in metformin metabolism were included on a custom SNP array. Associations between metformin use and risk of high-grade (Grade Group ≥ 2) and overall prostate cancer were explored using a case-control design. Interaction between the candidate/GWAS SNPs and the metformin-cancer association was explored using a case-only design.

RESULTS

Among 3481 men, 132 (4%) were taking metformin at diagnosis. Metformin users were older, more likely non-Caucasian, and had higher body mass index, Gleason score, and number of positive cores. Overall, 2061 (59%) were diagnosed with prostate cancer, of which 922 (45%) were high-grade. After adjusting for baseline characteristics, metformin use was associated with higher risk of high-grade prostate cancer (OR = 1.76, 95% CI 1.1-2.9, p = 0.02) and overall prostate cancer (OR = 1.77, 95% CI 1.1-2.9, p = 0.03). None of the 27 candidate SNPs in metformin metabolic pathways had significant interaction with the metformin-cancer association. Among the GWAS SNPs, one SNP (rs149137006) had genome-wide significant interaction with metformin for high-grade prostate cancer, and another, rs115071742, for overall prostate cancer. They were intronic and intergenic SNPs, respectively, with largely uncharacterized roles in prostate cancer chemoprevention.

CONCLUSIONS

In our cohort, metformin use was associated with increased risk of being diagnosed with prostate cancer. While SNPs involved in metformin metabolism did not have modifying effects on the association with disease risk, one intronic and one intergenic SNP from the GWAS study did, and these require further study.