P1-06-04: PAM50 Proliferation Index Predicts Response to Weekly Adjuvant Paclitaxel in Node-Positive Operable Breast Cancer.

Abstract

Abstract BACKGROUND: Predictive factors of survival benefit from taxane over non-taxane based adjuvant chemotherapy regimens are needed. Recent studies have suggested that “intrinsic” breast cancer subtypes may differ in their responsiveness to specific chemotherapeutics. Tumor samples from GEICAM 9906 study (Randomized Phase 3 Trial of Fluorouracil, Epirubicin, and Cyclophosphamide Alone [FEC] or FEC Followed by Paclitaxel [FEC-P] for Early Breast Cancer, JNCI 100:805, 2008) were gene expression profiled using the RT-qPCR PAM50 clinical test in order to identify potential predictive markers of taxane clinical benefit. METHODS: 793 formalin-fixed paraffin-embedded tumors were studied and classified into intrinsic subtypes (Luminal A & B, HER2−enriched, Basal-like, and Normal) using the PAM50 test. The assay also provided gene expression scores for the standard protein biomarkers usually measured by immunohistochemistry (ESR1/ER, PGR/PR, and ERBB2/HER2) and a meta-gene score for proliferation (proliferation signature) and for a luminal gene signature. Intrinsic subtypes, individual genes, or meta-genes were correlated with disease-free survival (DFS). Multivariable Cox regression analyses were performed to determine the significance of the interaction between treatment and intrinsic subtypes, single genes and meta-genes, adjusting for standard clinicopathological factors. RESULTS: A 8.7 years follow-up update of GEICAM 9906 trial confirmed a statistically significant advantage of FEC-P over FEC in terms of DFS (p=0.016) and OS (p=0.013). Exploratory analyses for prognostic factors showed that treatment arm (p=0.016), tumor size (p<0.0001), type of surgery (p=0.003), tumor grade (p=0.001), nodal status (p<0.0001), intrinsic subtypes (p<0.0001), ERBB2 (p=0.031), PGR (p<0.0001), Luminal meta-gene (p=0.006) and Proliferation meta-gene (p<0.0001) were associated with DFS. A Cox multivariate analysis using the backward and forward method showed that only the treatment arm (P=0.052), tumor size (p=0.0003), nodal status (p=0.001), intrinsic subtypes (p=0.045) and Proliferation (p=0.008) were prognostic for DFS. Concerning predictive factors, exploratory analyses were performed and interaction tests were calculated. Theses analyses showed that FEC-P was superior to FEC in low PGR expression (HR= 0.68, p=0.034) and not in high PGR group (HR=0.83, p=0.245); interaction test p=0.358. Similarly, FEC-P was superior in low ERBB2 expression (HR=0.67; p=0.005) and not in the high ERBB2 group (HR=0.92, P=0.707); interaction test p=0.256. Interestingly, superiority of FEC-P was observed for the low Proliferation Signature group (HR=0.58, p= 0.006) in contrast to the high proliferation group (HR=0.93, p=0.633); the interaction test showed an almost statistical significant difference (p=0.069). The FEC-P arm showed improved outcome in all genomic intrinsic subtypes, although no subtype alone reached statistical significance. CONCLUSION: Our study suggests that the PAM50 proliferation signature could be predictive of benefit for adding weekly paclitaxel to the adjuvant chemotherapy FEC regimen. These results need further validation in an independent study. The last two authors of this abstract have contributed equally to this study. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-06-04.