Genetic variation in AKT1, PTEN and the 8q24 locus, and the risk of testicular germ cell tumor

STUDY QUESTION Is there an association between testicular germ cell tumor (TGCT) and genetic polymorphisms in AKT1, PTEN and the 8q24 locus? SUMMARY ANSWER Our findings suggest that genetic variation in PTEN may influence the risk of TGCT. WHAT IS KNOWN ALREADY There is strong evidence that genetic variation influences the risk of TGCT. The oncogene, AKT1, the tumor suppressor gene, PTEN and the chromosome 8q24 locus play important roles in cancer development in general. STUDY DESIGN, SIZE, DURATION We have conducted a population-based Norwegian-Swedish case–parent study, based on cases diagnosed in 1990–2008, including 831 triads (TGCT case and both parents), 474 dyads (TGCT case and one parent) and 712 singletons (only the TGCT case). In addition we expanded the study to include 3922 unrelated male controls from the TwinGene project. PARTICIPANTS/MATERIALS, SETTING, METHODS We genotyped 26 single nucleotide polymorphisms (SNPs) in AKT1, PTEN and the 8q24 locus. First, triads and dyads were included in a likelihood-based association test. To increase the statistical power, case singletons and controls from the TwinGene project were included in a single test for association. We examined if the allelic effect on TGCT risk differed by histological subgroup, country of origin or parent of origin. Odds ratios (ORs) and 95% confidence intervals (CI) were calculated with Bonferroni correction (Pbonf) for multiple testing. MAIN RESULTS AND THE ROLE OF CHANCE In the case–parent analyses, none of the 26 SNPs were significantly associated with TGCT. Of the 23 SNPs investigated in the combined study, one SNP in PTEN (rs11202586) remained associated with TGCT risk after adjusting for multiple testing (OR = 1.16, 95% CI = 1.06–1.28, Pbonf = 0.040). We found no difference in risk according to histological subgroup, parent of origin or between countries. LIMITATIONS, REASONS FOR CAUTION Our study is strengthened by the population-based design and large sample size, which gives high power to detect risk alleles. The reported association was not highly significant, and although it was based on an a priori hypothesis of this tumor suppressor gene being implicated in the etiology of TGCT, replication studies, as well as functional studies of this polymorphism, are warranted. WIDER IMPLICATIONS OF THE FINDINGS We report, to our knowledge, a novel association between TGCT and a marker in the tumor suppressor gene PTEN. Previous studies have linked PTEN to TGCT etiology, and th