Plasma IGF-1 and IGFBP-3 may be imprecise surrogates for breast concentrations: an analysis of healthy women

We investigated insulin-like growth factor (IGF)-1 and IGF binding protein (IGFBP)-3 concentrations in histologically normal breast tissues and assessed their association with plasma concentrations, and breast cancer risk factors. IGF-1 and IGFBP-3 were assessed in plasma and breast tissues of 90 women with no history of any cancer and undergoing reduction mammoplasty. Pearson correlations and ANOVAs were used to describe plasma-breast associations and biomarker differences by breast cancer risk factors, respectively. Multivariable regression models were used to determine associations between risk factors, and breast IGF-1 and IGFBP-3. The mean age of the study sample was 37.3 years, 58 % were white, and generally these women were obese (mean BMI = 30.8 kg/m(2)). We observed no plasma-breast correlation for IGF-1, IGFBP-3, or IGF-1/IGFBP-3 (r = -0.08, r = 0.14, and r = 0.03, respectively; p-values > 0.05). Through age- and BMI-adjusted analysis, BMI and years of oral contraceptive (OC) use were inversely associated with breast IGF-1 (p-values = 0.02 and 0.003, respectively) and age was associated with breast IGFBP-3 (p = 0.01), while breast IGF-1/IGFBP-3 was higher in blacks than whites (1.08 vs. 0.68, p = 0.04) and associated with age and BMI (p-values = 0.03 and 0.002, respectively). In multivariable-adjusted models, some breast cancer risk factors studied herein explained 24, 10, and 15 % of the variation in breast IGF-1, IGFBP-3, and IGF-1/IGFBP-3, respectively. While reasons for the lack of plasma-breast hormone correlations in these cancer-free women are unknown, several factors were shown to be associated with breast concentrations. The lack of correlation between blood and tissue IGF-1 and IGFBP-3 suggests that studies of breast cancer risk assessing blood IGF-1 and IGFBP-3 may have important limitations in understanding their role in breast carcinogenesis.