Microsatellite profile in hormonal receptor genes associated with breast cancer

Given that breast cancer is depending on multiple hormonal influences, the nuclear receptors, estrogen receptor alpha, estrogen receptor beta and androgen receptor, are candidates for cancer susceptibility markers. We conducted an association study in a case-control population (139 cases and 145 controls) by genotyping three potentially functional microsatellites (TA)n, (CA)n and (CAG)n in the ERa, ERb and AR genes respectively. For (CAG)n polymorphism, a significant difference was observed using a cut-off 15 repeats CAG between genotypes short-short/short-long/long-long in cases and control subjects (p = 0.009) and also between the distribution of short/long allele in the two groups of individuals (p = 0.001). Genotypes comprising one or two short (CAG)n sequences had higher risk of breast cancer compared to genotypes with two long allele (odds ratio = 1,93; confidence interval = 1.05-3.55; p = 0.03). No significant difference was observed in allele frequency or in short/long allele percentage for (CA)n or (TA)n polymorphism (cut-off 22 CA and 19 TA repeats), neither in genotype frequencies (short-short, short-long or long-long). When the three microsatellite genotype were taken in analysis, the profile short CA-long TA-short CAG could clearly discriminate between cases and controls (p = 0.006). Also, this combined genotype profile has greater predictive values for breast cancer than (CAG)n genotype alone (predictive positive value 57,1% versus 53,7% and predictive negative value 53% versus 23% respectively). Our results sustain a polygenic model of breast cancer with gene-gene interactions; combined effects of three low-risk polymorphisms conferred significant genetic predisposition. Genotyping hormonal receptor genes ERa, ERb and AR could be a useful genetic marker for defining disease risk.