Abstract No. 
2013 San Antonio Breast Cancer Symposium
December 10-14

Protein pathway activation mapping of LCM tumor epithelium from I-SPY 1 TRIAL surgical specimens reveals activation of receptor tyrosine kinase drug target signaling networks in the non-responding patient cohort

Wulfkuhle JD, Gallagher RI, Wolf D, Yau C, I-SPYTrial1Investigators, Liotta LA, van'tVeer L, Esserman L, PetricoinIII EF

Background: Identification of alternative therapeutic strategies is critically needed for patients who do not achieve complete clinical response in the neoadjuvant setting, and in particular, from that cohort with the most aggressive disease (shortest relapse-free survival (RFS)). The I-SPY 1 TRIAL (CALGB 150007/150012, ACRIN 6657) is a trial of neoadjuvant anthracycline- and taxane-based chemotherapy that provides longitudinal biopsy specimens and extensive molecular and clinical/pathological characterization of non-responding patients. 

Methods: Tumor epithelium was procured by Laser Capture Microdissection (LCM) from frozen surgical specimens of 27 patients who did not achieve complete pathological response in the I-SPY1 TRIAL. Reverse Phase Protein Microarray (RPPA) technology was used to measure the levels of 120 key signaling proteins including drug targets for Phase I-III and FDA-approved therapeutics. Cox proportional hazard univariate and multivariate analyses were used to evaluate RFS correlates. Significance threshold was p = <0.05. 

Results: Univariate analysis of pathway activation revealed a number of actionable/druggable targets, particularly receptor tyrosine kinase (RTK) activation, where increased phosphorylation of ALK (Y1586), EGFR (Y1173), ERBB3 (Y1289), cABL (T735), and FAK (Y576/Y577) was significantly associated with aggressive disease (shorter RFS). Higher levels of SRC (Y527) inhibitory phosphorylation were also significantly associated with early recurrence. Coordinate activation of downstream targets of these RTKs, such as RAS-GRF S916, SHC (Y317), MEK (S217/221) and the Aurora kinases were also associated with early relapse in this study set. In multivariate analyses controlling for receptor subtype (HR+/HER2-, triple-negative (TN), HER2+), the associations between shorter RFS and RTK pathway activation including EGFR, ERBB3, cABL and FAK activation and SRC inhibition, as well as RAS-GRF and SHC phosphorylation remained significant. 

Conclusions: This unique functional protein pathway/proteomic analysis of LCM-isolated tumor epithelium by RPPA in surgical specimens taken from patients who did not achieve complete pathological response provides evidence for a significant number of druggable RTK pathway targets in patients with residual disease who are most at risk to die from breast cancer. Activation of ALK, EGFR, ERBB3, ABL, FAK, along with downstream RAS-ERK and Aurora kinase signaling activation in this cohort provide new insights into tumor progression along with distinct, actionable information on new therapeutic modalities to test in patients with progressing disease. Identification of new therapeutic strategies for patients who progress or recur early after neoadjuvant therapy is critical and, if validated in larger independent cohorts, could lead to rationalized personalized therapy-based trials. This strategy is being utilized in the I SPY 2 TRIAL as we introduce pathway-directed therapeutics in the neoadjuvant setting and are employing the RPPA platform to determine whether these agents are impacting their intended pathway. This work was partially supported by NIH grants CA31946 and CA33601.

View original