P5-18-13: Adjuvant Chemotherapy with Vinorelbine+5FU or Capecitabine in Poor Responders to Neoadjuvant EC- Docetaxel Chemotherapy (NAC) for Locally Advanced Breast Cancers

Background Post-operative chemotherapy in poor responders after NAC is questionable. We evaluated the use of adjuvant Vinorelbin (NVB) + continuous infusion 5FU or Capecitabine (Cap) in women with poor pathological response to preoperative sequential Epirubicin Cyclophosphamide- Docetaxel.(ECT) Materials and methods: From May 2004 till October 2007, 340 patients (pts) with stage II or III breast adenocarcinoma received neoadjuvant ECT +/− Traztuzumab (HER2+), +/− Celecoxib (HER2 negative) in a randomized phase II trial (Pierga, 2010). 53 patients (15.5 %) had pathological complete response (no infiltrative or in situ carcinoma in the breast and lymph nodes). Eleven pts were not operated. 276 pts had grade 3–4 Chevallier residual disease. These poor responders could be optionally offered post surgery chemotherapy (NVB+5-FU or Cap)+/− Traztuzumab.This regimen started concomitantly with post-operative RT. All positive hormonal receptors received hormonotherapy. We attempted to evaluate DFS according to post operative CT in the 276 pts. To take into account the differences between populations, treated and untreated patients were matched (1:1) according to variables with a significant difference between the two groups, analyses were adjusted with all other prognostic factors. Results: Out of 276 pts with 3–4 chevallier's grade after NAC, 198 pts did not receive any CT (group A), 78 received adjuvant CT (group B): NVB + 5-FU (FUN: N=41) or Cap (N=17) + Traztuzumab (21pts, 11 FUN and 10 CapNVB), 28 pts HER2+ in the group A receive Traztuzumab alone after surgery. The main characteristics of these patients differed for the initial lymph nodes status (N0: 15 (19 %) in group A vs 88 (44 %) in group B, p = With a median follow up of 38.9 months, 63/276 pts (22.8 %) recurred, 44 (22 %) in the group A and 19 (24.3 %) in the group B. DFS for the 276 pts is 95% at 12 months (m) and 86% at 24 m, 95% and 87% in group A and 96% and 83% in group B. (logrank test: p= 0.63). We matched the 2 groups (124 pts; 62/group) according to initial lymph nodes status, local response (Chevallier grade 3 or 4) and pN status after NAC. Characteristics of this matched population were similar except for more ER negative in chemo group, 19 pts (30.6 %) and 9 pts (14.5 %) in non chemo arm. In this population, 27 events occurred, with 13 in the chemo group and 14 in the non chemo'one. Treatment effect was non significant in univariate analysis (p = 0.93). In multivariate analysis adjusted for initial tumor size (TNM), age, molecular status (hormonal receptor and HER2 status), histological grade, initial lymph node status, histological response after NAC, treatment effect was non significant (Hazard Ratio = 1 IC95% [0.38; 2.67], p = 0.99). There was no difference in treatment effect according to molecular status (test for interaction p = 0.30).This multivariate analysis was also completed in the 276 pts with similar results. Conclusion: In poor responders patients after Anthracyclins-Taxanes NAC, adjuvant chemotherapy with Vinorelbine- 5-FU or Capecitabine- did not appear to prolong DFS. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-18-13.