The influence of breast cancer subtypes on the response to anthracycline neoadjuvant chemotherapy in locally advanced breast cancer patients

Purpose: The objective of neoadjuvant chemotherapy (NACT) for locally advanced breast cancer (LABC) is down-staging to achieve resectability. According to the protocol for the treatment of LABC more than 10 years ago, the routine NACT for LABC in Serbia consisted of 4 cycles of FAC (fluorouracil, doxorubicin, cyclophosphamide). The aim of this analysis was to assess the influence of biologic subtypes of BC on the response to NACT and on the disease outcome in these patients. Methods: We analyzed 190 patients with median age of 52 years (range 26-74), diagnosed with LABC between Jun/2002 and Dec/2005 and treated with 4 cycles of FAC. Patients with clinical response to NACT (162/192;85.26%) were subjected to radical mastectomy after which the majority of them received 3 cycles of adjuvant FAC, adjuvant tamoxifen if HR-positive disease, and postoperative radiotherapy. We retrospectively determined by immunohistochemistry estrogen receptor (ER)/ progesterone receptor (PgR)/HER2 status from BC biopsies in all patients who were divided in 4 subgroups. Pathological complete remission (pCR) was defined as ypT0N0. The main end points were disease-free survival (DFS) and overall survival (OS). Statistics included Fisher's exact test, KaplanMeier product-limit method and Log-rank test. Results: After a median follow up of 76 months (range 3-128) 104/190 patients (54.74%) experienced disease relapse, while 78/190 (41.05%) died. Of 157 patients with known receptor status the numbers of 4 subtypes were as follows: 31/190 (16.32%) triple negative (TN) BC, 22/190 (11.58%) HR-/HER2+, 97/190 (51%) HR+/HER2- and 17/190 (8.95%) HR+/HER2+. Ten out of 190 patients (6.17%) achieved pCR and had significantly longer DFS (Log-rank test, p=0.042), and a trend to prolonged OS (Log-rank test, p=0.092). There was a significant difference (Fisher exact test, p=7.7x 10(-6)) between pCR rates among 4 BC subtypes: 3/31 (9.68%) in TNBC, 6/22 (27.27%) in HR-/HER2+, 0/97 in HR+/HER2- and 1/17 (5.88%) in HR-1-/HER2+ patients. This difference was achieved on the account of the difference between TNBC and HR-/HER2+ BC subtypes (Fisher's exact test, p=6.85x10(-6), Bonferroni correction: 0.05/6=0.0083). There were no differences in DFS and OS between the 4 BC subtypes. Conclusion: Although there was a significantly higher number of patients achieving pCR among HR-/HER2+ subtype compared to other BC subtypes, this did not translate into improvement in long-term disease outcome of these patients.