Trebananib (T), an angiopoietin 1/2 neutralizing peptibody that inhibits interaction with TIE2 receptors, was available to all HR/HER2 subtypes in the I-SPY2 TRIAL. The agent did not achieve the prescribed graduation threshold for any eligible signatures prior to accrual of maximum sample size. We postulated that response to a drug that blocks TIE2 receptor-ligand interaction could be predicted by the measurement of basal TIE2 phosphorylation and downstream signaling in the pre-treatment biopsies.
Of 267 patients in the T and control arms, 203 patients (T: 128, controls: 73) had reverse phase protein microarray (RPPA) and pCR data available. RPPA data for 33 (phospho- and total) proteins involved in TIE2 signaling were evaluated for association between biomarker and response in the T and control arms alone (likelihood ratio test), and relative performance between arms (biomarker x treatment interaction) using a logistic model (LM). Analysis was also performed adjusting for HR/HER2 status. Markers were analyzed individually; p-values are descriptive and were not corrected for multiple comparisons.
In the TN subpopulation, TIE2 receptor levels (p = 0.037), ERBB3 (p = 0.048), total ERα (p = 0.05) and ERα S118 (p = 0.016) were negatively associated with response to T. In HER2+ patients, phospho-TIE2 Y1119 (p = 0.001) and Y992 (p = 0.0007) were positively associated with T response, as were downstream AKT-mTOR signaling activation proteins such as eIF4G S1108 (p = 0.005), p70S6K T389 (p = 0.011) and T412 (p = 0.038) and FOXO3a S253 (p = 0.041). ERBB2 Y877 (p = 0.028) was negatively associated with response in these patients. TIE2 Y1119, TIE2 Y992, eIF4G S1108, ERBB2 Y877, and FOXO3a S253 all demonstrated a significant treatment interaction by LM.
While small sample sizes preclude drawing definitive conclusions, our results suggest that activation levels of the TIE2 receptor may be predictive of T efficacy in HER2+ patients and signaling activation downstream of TIE2 such as AKT-mTOR signaling may correlate with response in the HER2+ and TN populations. These results need to be independently validated to determine the significance of these findings. Clinical trial information: NCT01042379