Background: In the I-SPY 2 trial, the pan-ERBB inhibitor, neratinib (N) arm was open to all HR/HER2 subtypes but graduated in the HR-/HER2+ signature. Exploratory analysis of protein signaling was performed to identify biomarker candidates correlated with pCR in the TN population. We evaluated 110 key signaling proteins using reverse phase protein microarray (RPPA) data from pre-treatment LCM purified tumor epithelium. Methods: Of 59 TN patients, 49 (N: 30, concurrent controls: 19) had RPPA and pCR data. RPPA data was correlated to pCR in both the treated and untreated patients using parametric (t-test) or non-parametric (Wilcoxon) statistical analysis, depending on data distribution. Only analytes whose levels were associated with response in the N but not the control arm were selected for further analysis. Markers are analyzed individually; p-values are descriptive and were not corrected for multiple comparisons. ROC analysis identified an optimal cut point and pCR rates of biomarker positive patients were assessed using that cut point. Results: Out of 110 analytes analyzed, only activation of HER2 Y1248, p = 0.03; EGFR Y1173, p = 0.009; were found to be positive predictors of pCR, and 3 proteins, TIE2 Y992, p = 0.02; LC3B, p = 0.02, and A-RAF S299, p = 0.0008, were found to be negative predictors of pCR. pCR rates in the biomarker positive group of 62.5% (10/16), 66% (11/16), 55%(10,/18), 67% (12/17). 61% (11/18) were determined for phospho- HER2, EGFR, TIE2, A-RAF and total LC3B, respectively, compared to a pCR rate of 40% ffor the IHC/FISH-based TN subgroup where RPPA data were available. Conclusions: Our sample size is too small to draw definitive conclusions. However, activation of HER2-EGFR in HER2- tumors may identify patients who respond to N. Low levels of activated A-RAF and LC3B also correlated with response. The results imply that there is a subset TN patients that paradoxically exhibit HER family signaling activation and may achieve clinical benefit with N. These findings merit future consideration as we develop trials for patients with suboptimal response to neoadjuvant therapy where biomarkers could be used as the basis for treatment reassignment. Clinical trial information: NCT01042379