Table 2.

The rs1447295 variant in 8q24 and prostate cancer risk by tumor stage and grade, and age at onset in the Australian Risk Factors for Prostate Cancer case-control study

A allele carriers, no. of cases (%)OR* (95% CI)PP heterogeneity
Tumor stage0.8
    Stage I-II158 (28.0)1.56 (1.20-2.02)0.001
    Stage III53 (26.9)1.48 (1.03-2.12)0.04
    Stage IV13 (23.6)1.24 (0.65-2.37)0.5
Tumor grade0.7
    Gleason score 5-6125 (27.0)1.48 (1.13-1.95)0.005
    Gleason score 7-10101 (28.2)1.58 (1.18-2.11)0.002
Age at diagnosis (y)0.9
    <5530 (27.0)1.49 (0.94-2.35)0.09
    55-64120 (27.8)1.54 (1.17-2.04)0.002
    65-6976 (27.3)1.51 (1.10-2.08)0.01
  • NOTE: Estimates from a dominant model.

  • * ORs and 95% CIs from unadjusted polytomous logistic regression analysis (mlogit function in STATA 8.2). The dependent variables included four categories for stage (0 for controls, 1 for stage I-II, 2 for stage III, and 3 for stage IV tumors) and age at diagnosis (0 for controls, 1 for <55-year-olds, 2 for 55- to 64-year-olds, and 3 for 65- to 69-year-olds), and three categories for Gleason score (0 for controls, 1 for Gleason score 5-6 tumors, and 2 for Gleason 7-10 tumors). Adjustment for family history of prostate cancer, age, country of birth, body mass index, and smoking history did not materially change the OR estimates.

  • Test for association between genotype and prostate cancer risk (likelihood ratio test).

  • Test for homogeneity of ORs across tumor stage, grade, and age at onset (likelihood ratio test).