Utility of a multigene prognostic algorithm in combination with TP53 expression for prediction of benefit from adjuvant taxane-containing chemotherapy in node-positive breast cancer patients

Abstract

593 Background: Recently, we have shown that the Siemens Prognostic Score (SPS) based on mRNA expression of nine informative genes predicted outcome in node-positive (N+) patients with breast cancer (SABCS 2008, abstract 6044). The aim of this retrospective biomarker study was to examine the utility of the SPS in combination with TP53 expression to predict benefit from adjuvant taxane therapy. Methods: The 211 N+ patients included in this study were treated in the context of a randomized two-arm phase III study (E-T-CMF vs. E-CMF) investigating adjuvant dose-dense sequential chemotherapy with epirubicin (E) followed by CMF with or without paclitaxel (T). RNA was isolated from formalin-fixed, paraffin-embedded tissue samples, using a Siemens proprietary method, followed by kinetic one-step RT-PCR for assessment of mRNA expression of the nine SPS genes, TP53 and two normalization genes. The continuous SPS was calculated using a linear combination of expression values of the SPS genes. Patients were separated into a high- and low-risk group using a cutoff at the median of the SPS. Optimal cutoff for low or high TP53 expression was defined on the basis of a ROC curve in SPS high-risk patients. Distant metastasis-free survival (MFS) and overall survival (OS) were estimated by the Kaplan-Meier method and compared using the log-rank test. Results: For patients with high SPS or high TP53 expression, we observed a trend for a better MFS in the E-T-CMF arm (SPS: p = 0.18; HR = 0.66; TP53: p = 0.23; HR = 0.67; n = 211). Combining both parameters, patients with high SPS and high TP53 expression (n = 44) had a significantly better MFS following E-T-CMF compared to E-CMF (5-year MFS 80% vs. 40%, p = 0.003, HR = 0.21; OS: p = 0.09; HR = 0.34). On the other hand, patients with high SPS and low TP53 expression (n = 32) showed a trend for a worse outcome with E-T-CMF (MFS: p = 0.09; HR = 3.54; OS: p = 0.35, HR = 1.90). Conclusions: Our prognostic algorithm combined with TP53 mRNA expression predicts the benefit from the addition of paclitaxel to E-CMF and might be used for identification of patients who should be considered for adjuvant taxane therapy. This hypothesis needs to be confirmed in an independent clinical study. [Table: see text]