Neoadjuvant chemotherapy (NAC) has become the standard therapy for both locally advanced and early-stage breast cancer in recent years for the improvement breast conserving surgery rate and the evaluation of treatment response in vivo. Pathological complete response (pCR) is an independent prognostic factor irrespective of breast cancer intrinsic subtypes after NAC. The trial is designed to compare effectiveness between anthracycline and/or taxane as neoadjuvant chemotherapy for operable advanced breast cancer in different molecular typing. In this trial the investigators will randomly assign 200 primary breast cancer patients to receive six cycles of fluorourcil, epirubicin,and cyclophosphamide(FEC), or four cycles of epirubicin and cyclophosphamide (EC) followed by four cycles of docetaxel(T), or six cycles of docetaxel and cyclophosphamide (TC). Trasuzumab was recommended combining docetaxel to patients if HER-2 positive.The effectiveness of therapy will be estimated after every two cycles of neoadjuvant chemotherapy. Surgery will be performed after completing designated full cycles of neoadjuvant chemotherapy. The primary endpoint is to assess pathologic complete response (pCR, ypT0/is ypN0) rate in different regiments. The secondary endpoint is to assess the relationship between pCR rate with molecular typing in different regiments, so that the investigators could optimize neoadjuvant chemotherapy regiment according to molecular typing.

The trial is designed to compare effectiveness between anthracycline and/or taxane as neoadjuvant chemotherapy for operable advanced breast cancer in different molecular typing. In this trial the investigators will randomly assign 200 primary breast cancer patients to receive six cycles of fluorourcil, epirubicin,and cyclophosphamide(FEC), or four cycles of epirubicin and cyclophosphamide (EC) followed by four cycles of docetaxel(T), or six cycles of docetaxel and cyclophosphamide (TC). Trasuzumab was recommended combining docetaxel to patients if HER-2 positive.The effectiveness of therapy will be estimated after every two cycles of neoadjuvant chemotherapy. Surgery will be performed after completing designated full cycles of neoadjuvant chemotherapy. The primary endpoint is to assess pathologic complete response (pCR, ypT0/is ypN0) rate in different regiments. The secondary endpoint is to assess the relationship between pCR rate with molecular typing in different regiments, so that the investigators could optimize neoadjuvant chemotherapy regiment according to molecular typing.

Inclusion Criteria:

• All patients were required to give written informed consent.
• Patients present with operable breast cancers that were diagnosed by histopathology and have no distant metastasis.
• Have no history of anti-cancer therapies including chemotherapy, radiation therapy, hormone therapy and surgical therapy.
• Have normal cardiac functions by echocardiography.
• ECOG scores are ≤ 0-1.
• Patients are disposed to practice contraception during the whole trial.
• The results of patients' blood tests are as follows:

Hb ≥ 90 g/L WBC ≥ 3.0×109/L Plt ≥ 100×109/L Neutrophils ≥ 1.5×109/L ALT and AST ≤ 2.5 times of normal upper limit. TBIL ≤ 1.5 times of normal upper limit. Creatinine ≤ 1.5 times of normal upper limit.

Exclusion Criteria:

• Have other cancers at the same time or have the history of other cancers in recent five years, excluding the controlled skin basal cell carcinoma or skin squamous cell carcinoma or carcinoma in situ of cervix.
• Active infections
• Severe non-cancerous diseases.
• The patients are undergoing current administration of anti-cancer therapies, or are attending some other clinical trails.
• Inflammatory breast cancer.
• Pregnant or lactational, or patients refuse to practice contraception during the whole trial.
• The patients are in some special conditions that they can't understand the written informed consent, such as they are demented or hawkish.
• Have allergic history of the chemotherapeutic agents.
• Bilateral breast cancers.