Abstract P3-10-13: Prognostic Value of Genomic Analysis after Neoadjuvant Chemotherapy for Breast Cancer

Background Oncotype DX® Recurrence Score® (RS) provides both predictive and prognostic information for estrogen receptor positive (ER+) breast cancer. Whether RS evolves after exposure to chemotherapy (C), or if RS post-neoadjuvant C has prognostic capacity for outcomes after bevacizumab, is not known. Methods DFCI 05-055 treated patients (pts) with residual invasive disease following neoadjuvant C with sequential bevacizumab-containing regimens. Archival FFPE tissue was obtained from participants at 3 potential timepoints: pre-C core biopsy, post-C surgical residual disease, and core biopsy at time of systemic recurrence. Standard RS testing was performed on all samples. Results A total of 162 pts were enrolled on the parent study; of that, 118 samples from 80 pts were available for testing, including 48 pre-C core biopsies, 68 post-C surgical samples, and 34 paired samples before and after C. Comparison of ER/PR/HER2 testing by local IHC vs RT-PCR for the pre-C core biopsies showed excellent concordance. For the entire cohort of 80, 20 pts (25%) experienced distant recurrence (9 ER+, 11 ER-by IHC). RS risk distribution in this group included 14 high, 3 intermediate (int), and 3 low. High RS was positively associated with distant recurrence for the entire cohort (t-test p=0.04); this association remained consistent whether RS was tested pre-C (t-test p=0.07) or post-C (t-test p=0.02). Specimen characteristics of the 34 paired pre-C/post-C samples are described below. Of the 34 paired samples, RS was highly correlated before and after exposure to neoadjuvant C, (r=0.85, 95% CI 0.72-0.92), suggesting little, if any, change in RS from C exposure. Specific changes over time included 24 without change in RS group, 3 high to int, 1 high to low, 1 int to low, and 5 low to int. Testing by RT-PCR remained consistent for ER (r=0.91, 95% CI 0.82-0.95), PR (r=0.79, 95% CI 0.62-0.89) and HER2 (r=0.88, 95% CI 0.76-0.94) before and after neoadjuvant C. Two pts had biopsy samples from systemic recurrence; testing showed RS, ER, PR, HER remained in similar range from core biopsy to surgery to disease recurrence. Conclusions In this evaluation of samples pre-and post-neoadjuvant C, high RS determined either before or after neoadjuvant C predicted for disease recurrence after adjuvant bevacizumab treatment. RT-PCR and IHC values for ER/PR/HER2 were concordant. Paired RS and RT-PCR ER/PR/HER2 remained correlated despite interval exposure to neoadjuvant C. Post neoadjuvant C RS may provide valid prognostic information for breast cancer disease recurrence and warrants further study. Specimen Characteristics for Paired Samples (n=34) Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-10-13.