Sequential doxorubicin/paclitaxel (AT) followed by CMF treatment was shown to be an active neoadjuvant chemotherapy regimen in the first European Cooperative Trial in Operable Breast Cancer (ECTO I trial). The aim of the current study (ECTO II) is to assess the complete pathological response (pCR) rate following three different anthracycline and taxane-containing neoadjuvant chemotherapy regimens, with or without capecitabine (X). Patients with operable, invasive breast cancer > 2.0 cm in diameter, were randomized to AT→CMF, AT→CMX or AC→TX regimens in two parallel, randomized, open-label, phase II trials (within a single study) in patients with estrogen receptor negative (ER-) and estrogen receptor positive (ER+) diseases, respectively. Exemestane was delivered concomitantly with neoadjuvant chemotherapy in ER+ tumors. Achievement of pCR was more common in ER- than ER+ women (45.3 vs. 10.4%). Capecitabine was only associated with a higher frequency of pCR in ER+ patients receiving AT→CMX. Overall response rates (ORR) ranged from 88 to 97%, and this translated into high rates of breast-conserving surgery (67% of ER- patients and 72% of ER+ patients). All three regimens were well tolerated. Febrile neutropenia and gastrointestinal effects were the most common grade ≥ 3 adverse events. As expected, the ECTO II study showed higher pCR rates in patients with ER- disease. Substituting capecitabine for fluorouracil (± methotrexate) in anthracycline/taxane-containing regimens appeared to be beneficial only in ER+ tumors. Translational studies investigating interactions between therapeutic agents and tumor biology are warranted to refine patient selection and improve the results of neoadjuvant chemotherapy.