A common 8q24 variant and prostate cancer in a large nested case-control study.

A195

Cancer of the prostate (CaP) is the most common cancer, besides skin related malignancies, among U.S. men. Although an estimated 234,460 new cases and 27,350 deaths are expected in 2006, very few unequivocal risk factors are known. Aging, family history and ethnicity are known risk factors for CaP. For example, the incidence rate for African Americans (AA) is nearly 1.6-fold higher than Caucasians. Family history and ethnicity suggest a genetic cause. Recently, two reports independently identified the 8q24 region as being associated with CaP risk in Caucasians and AA. The findings were reported in moderately sized studies. Additionally, one study reported a greater effect in AA diagnosed at an earlier age (<72), while the other reported a slightly greater risk for aggressive prostate cancers defined by Gleason score at diagnosis (<6 vs. ≥6). Here, we replicate the overall association within a large nested case-control study from the NCI Breast and Prostate Cancer Cohort Consortium (BPC3). The BPC3 is comprised of seven individual cohorts, which collectively include over 248,000 men with a blood sample. We genotyped over 7,000 cases and 8,000 matched controls for the 8q24 marker (rs1447205). This SNP was strongly associated (p=4.00x10^-19) with CaP in the overall population. Compared with wild-type homozygotes, carriers with one copy of the minor allele had an OR_AC=1.33 (99%CI: 1.20, 1.46) and carriers with two copies of the minor allele had an OR_AA=1.87 (99%CI: 1.44, 2.42). As previously reported, the association between CaP and the marker was only detected in AA diagnosed at an early age (≤65). However, the increased risk was present in Caucasians regardless of age at diagnosis. The main effect for the marker was slightly greater in aggressive CaP, defined by Whitmore-Jewett stage or mortality, although the difference was statistically non-significant. The genotype risk estimates for aggressive cancers defined by stage (C, D or metastases/death due to CaP) at diagnosis were OR_AC=1.46 (99%CI: 1.20, 1.78) and OR_AA=2.60 (99%CI: 1.68, 4.04). The risk estimates for non-aggressive cancers (stage A or B and no metastases/deaths due to CaP) were OR_AC=1.35 (99%CI: 1.20, 1.52) and OR_AA=1.85 (99%CI: 1.36, 2.50). The validation of this marker within a large sample size leaves little room for the possibility of a false-positive result. However, as indicated in the previous reports, a risk gene for prostate cancer has yet to be identified in the 8q24 region.