Abstract Background: Studies support the hypothesis that high circulating levels of the hormone prolactin increase breast cancer risk. It remains unclear if genetic factors are associated with interindividual variation in circulating prolactin and subsequent breast cancer risk. We examined the relationship between single nucleotide polymorphisms (SNPs) in prolactin (PRL) and prolactin receptor (PRLR), serum prolactin levels and breast cancer risk in a population based case-control study. Methods: We genotyped 8 PRL and 20 PRLR tag SNPs in 1943 cases and 2229 controls (82% and 89% of enrolled, respectively). Cases were women 20-74 years old, diagnosed with in situ or invasive breast cancer between 2000 and 2003, and living in Warsaw or Lodz, Poland. Controls were cancer-free women selected from a population registry, matched to cases on age and city. Blood samples were provided at the study interview. Serum prolactin levels were measured in a subset of 773 controls using the Bayer Advia Centaur immunoassay. Odds ratios and 95% CIs for genotype-breast cancer associations were estimated using unconditional logistic regression, adjusting for age and city. Geometric mean prolactin levels and 95% confidence intervals (CI) were estimated using linear regression, adjusting for age, city, blood collection time, and day of menstrual cycle (premenopausal). Log-additive model Wald P-values were used to test for linear trends. Analyses were stratified by menopausal status, and all statistical tests were 2-sided (alpha=0.05). To account for multiple comparisons, P-values were adjusted using the Benjamini and Hochberg method. Results: Three PRLR SNPs were nominally associated with breast cancer: in premenopausal women, rs249537 (per-allele OR 1.39, 95% CI 1.07 - 1.82, P=0.0140); and in postmenopausal women, rs7718468 (per-allele OR 1.16, 95% CI 1.03 — 1.30, P=0.0137) and rs13436213 (per-allele OR 1.13 95% CI 1.01 — 1.26, P=0.0397). We examined serum prolactin levels by genotype for these three SNPs to determine whether the potential risk alleles were associated with increased circulating prolactin, but mean serum prolactin levels were similar by genotype in controls: rs249537 TT=3.83 ng/ml, CT=5.51 ng/ml, CC=5.50 ng/ml, P-trend=0.8381; rs7718468 CC=6.41 ng/ml, CT=6.86 ng/ml, TT=6.47 ng/ml, P-trend=0.5643; and rs13436213 AA=6.48 ng/ml, AG=6.52 ng/ml, GG=6.29 ng/ml, P-trend=0.4766. None of the nominal associations between PRLR genotypes and breast cancer risk were significant after adjusting for multiple comparisons (multiple comparison adjusted P-values: rs249537 P=0.4340; rs7718468 P=0.4340; rs13436213 P=0.8205). PRL SNPs were not associated with breast cancer risk in premenopausal or postmenopausal women. Discussion: Our data do not support an association between PRL or PRLR tag SNPs and breast cancer. Though some associations were nominally significant, these results may be due to random error and require further study. Furthermore, data do not support a relationship between riskassociated SNPs and differences in prolactin levels. Future research should focus on non-genetic factors that affect circulating prolactin levels. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-12-05.
Read full abstract