Predictive Effect of IGFBP-3 on Low-Dose Tamoxifen Efficacy in Noninvasive Breast Cancer in the Phase III Tam-01 Trial

Abstract Purpose: Low-dose tamoxifen 5 mg/day (babytam) for 3 years can decrease the incidence of new breast cancer events in women with breast intraepithelial neoplasia by 42% with limited toxicity, which provides a new treatment option for these disorders. However, predictive biomarkers of babytam efficacy are lacking. We studied whether baseline levels of insulin-like growth factor-1 (IGF-I), IGF-binding protein-3 (IGFBP-3), estradiol, and sex hormone–binding globulin (SHBG) and their ratios predict babytam efficacy on breast cancer events in a preplanned secondary analysis. Patients and Methods: Within a 1:1 placebo-controlled, multicenter randomized trial of babytam or placebo administered for 3 years after surgery in women with hormone-sensitive or unknown breast intraepithelial neoplasia, including atypical ductal hyperplasia and lobular or ductal carcinoma in situ, 406 of 500 participants consented to blood sampling at baseline and at 1 and 3 years. Serum IGF-I, IGFBP-3, estradiol, and SHBG levels and their ratios were measured using chemiluminescent immunoassays. Biomarker changes were estimated using mixed-effects models, and incidence rate ratios were calculated after 10 years of follow-up with Poisson regression. Subgroup analyses were performed using an interaction test and subpopulation treatment effect pattern plot. Results: Baseline levels of IGFBP-3 in the three top quartiles (≥3.44 µg/mL), but not in the lower quartile, predicted greater babytam efficacy compared with placebo (Pinteraction = 0.006). Baseline IGF-I, estradiol, or SHBG levels were not predictive of babytam efficacy, whereas the IGF-I/IGFBP-3 ratio was borderline significant (Pinteraction = 0.067). Conclusions: High baseline levels of IGFBP-3 (≥3.44 µg/mL) predicted babytam efficacy and may help differentiate which women benefit most from this treatment.