HNF1B and JAZF1 genes, diabetes, and prostate cancer risk

Abstract

Background: Epidemiologic studies have shown that men with type II diabetes have a lower risk of prostate cancer than non-diabetic men. Recently, common variants in two genes, HNF1B and JAZF1, were found to be associated with both of these diseases.

Methods: We examined whether the relationship between HNF1B and JAZF1 variants and decreased prostate cancer risk may potentially be mediated through diabetes in two large prospective studies, the Cancer Prevention Study II Nutrition Cohort and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial.

Results: Three HNF1B SNPS, rs11649743, rs4430796, and rs7501939, were associated with decreased risk of prostate cancer and were also associated, with marginal statistical significance, with increased risk of diabetes. The JAZF1 SNPs rs6968704 and rs10486567 were associated with decreased risk of prostate cancer but were not associated with diabetes. All five SNP-prostate cancer relationships did not substantially differ when the analyses were stratified by diabetic status or when diabetic status was controlled for in the model. Furthermore, the association of diabetes with prostate cancer was not altered when the SNPs were included in the logistic model.

Conclusions: These findings indicate that the HNF1B variants are directly associated with both diabetes and prostate cancer, that diabetes does not mediate these gene variant-prostate cancer relationships, and the relationship between these diseases is not mediated through these gene variants.

Figure 1

Per allele odds ratios (OR) for associations of the minor alleles for 3 HNF1B SNPs and 2 JAZF1 SNPs with either prostate cancer (A) or diabetes (B). Summary estimates for each SNP are indicated by the diamonds. The minor allele nucleotides are: HNF1B rs11649743; A, HNF1B rs4430796; G, HNF1B rs7501939: T, JAZF1 rs6968704; C, and JAZF1 rs10486567; G.

Figure 2

Linkage disequilibrium (LD) matrix for the HNF1B gene. The relative positions of the four SNPs investigated in this study are shown. The three SNPs in blue were found to be significantly associated prostate cancer risk. The general area in which the diabetes-associated SNPs are located is under the diabetes heading. The area in which the prostate cancer-associated SNPs reside is larger and encompasses the diabetes-associated SNPs.

Figure 3

LD matrix for the JAZF1 gene. The relative positions of the seven SNPs investigated in this study are shown under the prostate cancer heading towards the 3” end of the gene. The two SNPs in blue were found to be significantly associated prostate cancer risk. The diabetes-associated SNP rs864745 is near the 5’ end of this gene.

Figure 4

Odds ratios for associations of the minor alleles for 3 HNF1B SNPs and 2 JAZF1 SNPs with prostate cancer. The results from the analysis in (A) excluded cases with diabetes while those in (B) are limited only to cases with diabetes. Summary estimates for each SNP are indicated by the diamonds.