Randomized phase II trial of capecitabine and lapatinib with or without cixutumumab in patients with HER2+ breast cancer previously treated with trastuzumab and an anthracycline and/or a taxane: NCCTG N0733 (Alliance)

Background: Crosstalk between the insulin-like growth factor (IGF) and HER2 pathways in multiple preclinical models suggest a mechanism of resistance to HER2-targeted therapy. Cixutumumab (CIX, IMC-A12), an anti-IGF-1 Receptor monoclonal antibody, was investigated in combination with capecitabine and lapatinib (cape/lap) to determine if the addition of an IGF target agent would improve PFS compared to cape/lap in unselected HER2+ metastatic breast cancer (mbc) patients (pts). IGF binding protein (IGFBP) levels may be important in predicting tumor dependence on IGF signaling and are potential biomarkers. Methods: Following an initial safety cohort, this open label phase II study randomized (2:1) pts to cape/lap with CIX (Arm B) or cape/lap (Arm A). Primary endpoint was progression-free survival (PFS) with 144 pts planned for a target hazard ratio of 0.56. PFS was estimated using Kaplan-Meier and compared using Cox regression. Secondary endpoints included assessing the overall survival and treatment tolerability. Correlative studies included assessments of IGFBPs 1-6 using a Luminex multiplex assay on pt serum. Results: 64 evaluable pts were accrued, including 8 to the safety cohort. The most common grade 3 events were diarrhea, fatigue and mucositis. Dose reductions in the safety cohort for 6 of 8 pts prompted a dose reduction of cape/lap in Arm B of the randomized portion. The randomized portion of the study was closed early at 56 pts due to slow accrual. Median PFS by in Arm A was 6.0 mo vs. 4.9 mo in Arm B (p= 0.89). Median overall survival (16.8 vs 14.7 mo, p=0.93), time to treatment failure (TTF) (4.6 vs 4.4 mo, p=0.66), and duration of response (4.8 vs 6.9 mo, p=0.26) were not different between arms. In Arm B, increase in IGFBP5 correlated with poor PFS (1.4 vs 9.9 mo, p= 0.02) and high IGFBP5 correlated with improved TTF (9.9 vs 3.9 mo, p= 0.03). Conclusions: CIX is reasonably tolerated in combination with cape/lap following a dose reduction, but does not improve PFS in unselected pts with HER2+ mbc. IGFBP5 may be an important determinant of benefit from IGF targeted therapy and warrants further investigation.