A randomized phase II neoadjuvant trial in patients with stages II-III and inflammatory breast cancer.

3515 Background: Complete pathologic response (CR) following neoadjuvant chemotherapy (NCT) is achieved in only a third of patients, although the CR rate doubles in those with HER2 over-expressing (+) tumors. Therefore, more efficient regimens as well as better predictors of outcome are needed. Objectives: To assess the feasibility and efficacy (defined by CR rate) of a novel regimen in comparison to standard therapy in the neoadjuvant setting; to assess the feasibility of sequential tissue procurement and to identify molecular predictors of CR with emphasis on markers of receptor/signal transduction pathways and their feedback loops. Methods: Patients with stages II-III HER- breast cancer are randomized on a prospective phase II trial to receive 6 cycles of docetaxel 75 mg/m 2 , doxorubicin 50 mg/m 2 , cyclophosphamide 500 mg/m 2 (TAC, arm A) versus doxorubicin 60 mg/m 2 , cyclophosphamide 600 mg/m 2 given every 2 weeks x 4, followed by 3 weekly doses of carboplatin (AUC 2) and paclitaxel albumin-bound 100 mg/m 2 given every 4 weeks x 3 cycles (ACAC; arm B). A separate stratum of patients with HER2 + tumors are included in this trial and they receive ACAC and 12 weekly doses of trastuzumab given during the administration of carboplatin and paclitaxel albumin-bound (ACAC-T, arm C). Core biopsies of the untreated primary breast tumor are performed prior to treatment and mid-therapy, and tumor is also procured at completion from the definitive surgical specimens. Fresh frozen, formalin-fixed paraffin-embedded, and ethanol preserved tumor and adjacent stromal tissues are procured for extraction of RNA and protein analysis, allowing assessment of protein and gene expression patterns. The immunohistochemically targeted proteins include ER, PR, HER2, EGFR, p-Stat3, p-AKT, p-Src, Survivin, p21, p27, Ki67. Results: Twenty seven patients received at least one cycle of therapy (41 were screened, but due to upstaging or other eigilibility problems only 30 have or are about to receive treatment). For the 27 patients median age is 50 years (range-31-69); 13 patients presented with stage II and 14 with stage III disease (6 patients were enrolled with inflammatory breast cancer); 29% of tumors are HER2 +. Pre-treatment biopsies were successfully procured from all, and 70% of study-required mid-treatment biopsies were performed. So far, no study-limiting toxicities have been observed on either arm: grade 4 hematologic toxicities were seen on arm A (n:5); B (n:4); and C:(n:4 cases). There were no grade 2 or greater neurotoxicities. All 5 evaluable patients treated on Arm C were found to have either CR (n:3), or minimal microscopic residual disease. Conclusion: The study design is feasible, and the novel regimen is active. Scheduled comparative analysis of toxicities, CR rate, and assessment of biological markers in the first 35 patients who will have undergone definitive surgery will be presented.