8% in our control subjects. This frequency is also similar to the frequencies see more found in other studies that analyzed GSTP1 polymorphism [18–20]. Some studies have reported a relationship GS-9973 between GST variants and risk of prostate cancer [9, 10, 12, 13, 21]. Investigation of the GSTP1 gene did not reveal any significant association between heterozygous GSTP1 genotype (Ile/Val) and prostate cancer. However, our results suggest that Val/Val genotype of GSTP1

gene could modulate the risk of prostate cancer, even if this association did not reach statistical significance. It should be kept in mind that the inability to reject the null hypothesis could be due to low power of the test because of a relatively selleck products small sample size. Therefore, the lack of significance does not necessarily mean equality of the distributions. It is plausible that polymorphism at the GSTP1 locus can play an important role in the susceptibility to different types of cancer. Association of the GSTP1 Val allele with cancer could be expected since the conversion of the amino acid at codon 105 from isoleucine to valine substantially lowers activity of the altered enzyme. It has been predicted

from molecular modelling that the amino acid at this site lies in a hydrophobic binding site for electrophile substrates and thus affects the substrate binding [22]. On the other hand, there are also studies which did not prove any independent effect of this type of polymorphism on the susceptibility for prostate cancer [23–25]. In the present study, we did not observe significantly different crude rates of the GSTM1 and GSTT1 null genotypes in the men diagnosed with prostate cancer and those in the control group. Our

data and the data published by other research groups suggest that differences in the GST frequencies between prostate cancer patients and the control group are relatively small, which therefore makes it difficult to separate the groups from each other cAMP based on statistical data analysis. Once again, the high variability in the groups could mask statistical differences due to low power. The easiest way to improve precision is to increase the number of subjects and patients in the experimental design. However, this may not be applicable to all research conditions due to such factors as additional costs, poorer availability of resources, lower population, which compromises the number of subjects eligible for investigation. In order to achieve a power of at least 80%, we have to identify other explanatory variables and the control for them, and/or apply meta-analysis in order to increase sample size.