|| Checking for direct PDF access through Ovid
Prospective cohort studies on breast cancer risk among premenopausal women and insulin-like growth factor I (IGF-I) concentrations have so far included only few cases, and have shown inconsistent relative risk estimates. We pooled 220 cases of breast cancer diagnosed before age 50, and 434 control subjects, from three prospective studies in New York (USA), Umeå (Northern Sweden) and Milan (Italy), and we measured IGF-I and insulin-like growth factor binding protein 3 (IGFBP-3) with common enzyme-linked immunosorbent assays. Overall, IGF-I and IGFBP-3 measurements obtained by the common method showed a positive but not significant relationship with breast cancer risk (odds ratios (ORs) 0.90 [95% confidence intervals (95% CI) 0.50–1.62], 1.63 [0.89–2.97], 1.46 [0.78–2.73] and 1.41 [0.75–2.63] for quintiles of IGF-I, and ORs 0.98 [0.54–1.75], 1.06 [0.59–1.91], 1.04 [0.58–1.87] and 1.77 [0.97–3.24] for quintiles of IGFBP-3). Our results give only moderate support for an association of blood IGF-I with breast cancer risk in young women.