A prospective study of the insulin-like growth factor axis in relation with prostate cancer in the SU.VI.MAX trial.

Several epidemiologic studies have examined with diverging results the relationships between circulating levels of insulin-like growth factors (IGF) and of IGF-binding proteins (IGFBP) and prostate cancer risk. We assessed the association of prediagnostic plasma levels of IGF-I, IGF-II, IGFBP-2, and IGFBP-3 and subsequent occurrence of prostate cancer in a case-control study nested in the SU.VI.MAX trial. The SU.VI.MAX study was a primary prevention trial testing a daily supplementation with low-dose antioxidant vitamins and minerals in male and female middle-aged volunteers in France. One hundred prostate cancer cases were diagnosed among 4,855 SU.VI.MAX participants over a 9-year follow-up period. For each case, four age-matched controls were selected randomly. Frozen baseline plasma samples were used to measure IGF-I, IGF-II, IGFBP-2, and IGFBP-3. Conditional logistic regression was used to assess the association between these four biochemical markers and prostate cancer risk. After controlling for the intervention group in the trial and the other IGF axis variables, the odds ratios and 95% confidence interval (95% CI) comparing the upper quartile to the baseline quartile were 1.83 (95% CI, 0.85-3.95), 1.05 (95% CI, 0.35-3.18), 0.79 (95% CI, 0.39-1.58), and 0.42 (95% CI, 0.12-1.52) for IGF-I, IGF-II, IGFBP-2, and IGFBP-3, respectively. More suggestive associations for IGF-I and IGFBP-3 were observed with advanced and aggressive cancers. Our results are consistent with those of some previous prospective studies and suggest that IGF axis variables are not long-term predictors of the occurrence of prostate cancer.