Neoadjuvant chemotherapy (NCT) in 710 patients for operable breast cancer: Comparison of 2 pathological classifications

Abstract

10502 Background: In a database harbouring the patients of several prospective phase II neoadjuvant trials (avcf, avcfm, fec 50–100, nem, net, tncf, taxotère/tncf, taxotere alone), clinical and pathological responses rates were studied in 710 women with stage II-III operable breast cancer treated 1982–2004. Pathological review was conducted to assess the residual disease in breast and nodes according to the 2 most used classifications in Europe: CHEVALLIER’s and SATALOFF’s (Am J Clin Oncol 1993; J Am Coll Surg 1995). Methods: Median age of the patients was 49.5 years [26–81]. Median diameter of the invasive tumour was 40 mm [10–130]. 555 (78%) patients had a canalar, 90 (13%) a lobular, 18 (2.5%) a mixed or invasive carcinoma, 36 (5%) neoplasic cells only and 11 (1.5%) another carcinoma. 25.6% of the tumours were grade III SBR, 28.3% grade 4 or 5 MSBR. The median number of NCT courses was 6 [1–9] followed by a surgery for 92%, a radiotherapy for 95%, an adjuvant chemotherapy (17%) and/or a hormonotherapy (52%). A breast evaluation was realized before, during and after NCT (Amat et al, Breast Cancer Res Treat 2005). Results: Intent to treat (n = 710), overall response rate was 68% (16% complete). The complete pathological response (pCR) rate was 14.18% according to Chevallier’s and 22.50% according to Sataloff’s classification. On 656 patients operated, 470 (72%) had a conservative surgery. On 520 patients with an axillary dissection, 272 (52.3%) had involved nodes (median number:1 [0–20]). After a median follow-up of 93 months, DFS and actuarial survival at 120 months were 54.9% and 66.5%, respectively.Chevallier’s classification level 1 (pCR in breast and axilla) was the most predictive of a good DFS, with a plateau appearance near 80%. Chevallier’s classification level 2 (in situ only) and Sataloff grade A (pCR and isolated tumor cells) yielded a lesser DFS (p < 0.01). With a long follow-up, a complete pCR is the most favourable prognostic factor, followed by in situ only or isolated tumor cells, then residual tumor. Conclusions: Sataloff’s gives higher pCR figures than Chevallier’s without evidence of superior predictive value. The classification used is important to predict outcome after NCT. No significant financial relationships to disclose.