The purpose of this trial is to compare the 3 years DFS of dose-dense epirubicin and cyclophosphamide followed by paclitaxel with paclitaxel plus carboplatin as adjuvant therapy for triple-negative breast cancer.

The other purpose of this trial is to observe the patient's tolerance.

Breast cancer are heterogeneous group of tumors with diverse behavior, outcome, and sensitivity to therapy.In recent years, the term triple negative (TN) breast cancer has emerged to describe those cancers which do not express oestrogen (ER) , progesterone (PR) receptors, or Her2. Many studies had estimated that TN cases represents between 12%-20% of all breast cancers. Those TN case constitute one of the most challenging breast cancer groups, with only systemic chemotherapy is currently available for their treatment.

BRCA1 protein normally functions as a negative regulator of the cell cycle, also, BRCA1-positive tumors encompass a heterogeneous group of tumors that show distinctive pathological and clinical features. BRCA1-associated cancers are typically high-grade invasive duct carcinoma and are mostly triple negative.The phenotypic and molecular similarity of the TNBCs to BRCA1-associated BCs might be of use in designing their treatment protocol. There is increasing evidence that the DNA repair defects that are characteristic of BRCA-1 related cancers may provide sensitivity to certain systemic agents to treat TNBC patients such as the bifunctional alkylating agents and platinum drugs.

Dose density refers to the administration of drugs with a shortened intertreatment interval. It is based on the observation that in experimental models, a given dose always kills a certain fraction, rather than a certain number, of exponentially growing cancer cells. Because human cancers in general, and breast cancers in particular, usually grow by nonexponential Gompertzian kinetics, this model has been extended to those situations. Regrowth of cancer cells between cycles of cytoreduction is more rapid in volume-reduced Gompertzian cancer models than in exponential models. Hence it has been hypothesized that the more frequent administration of cytotoxic therapy would be a more effective way of minimizing residual tumor burden than dose escalation. In the INT C9741 trial, the dose-dense schedule is accomplished by using granulocyte colony-stimulating factor (filgrastim) to permit every-2-week recycling of the drugs A, T and C at their optimal dose levels rather than at the conventional 3-week intervals.Sequential therapy refers to the application of treatments one at a time rather than concurrently. It does not challenge the concept that multiple drugs are needed to maximally perturb cancers that are composed of cells heterogeneous in drug sensitivity. Rather, it hypothesizes that for slow-growing cancers like most breast cancers, it is more important to preserve dose density than to force a combination, especially if that combination would be more toxic and requires dose-reductions or delays in drug administration. If dose density is the same in a sequential combination chemotherapy regimen and a concurrent combination regimen, theoretical considerations indicate that the therapeutic result should be the same, even if the sequential pattern happens to be less toxic.

In our trial, we want to compare the 3 years DFS of dose-dense epirubicin and cyclophosphamide followed by paclitaxel with paclitaxel plus carboplatin as adjuvant therapy for triple-negative breast cancer.The other purpose of this trial is to observe the patient's tolerance.

Inclusion Criteria:

1. Patient must accept the modified radical mastectomy
2. Patients with histologically confirmed ER(-) PR(-) and HER-2(-)
3. Positive axillary lymph nodes;negative axillary lymph node with age< 35 years or Ⅲ grade or intravascular cancer embolus.
4. Age between 18 years to 65 years
5. Able to give informed consent
6. Patients with an Eastern Cooperative Oncology Group (ECOG) performance score of 0 or 1.
7. Not pregnant, and on appropriate birth control if of child-bearing potential.
8. Adequate bone marrow reserve with ANC > 1000 and platelets > 100,000.
9. Adequate renal function with serum creatinine < 2.0.
10. Adequate hepatic reserve with serum bilirubin < 2.0, AST/ALT < 2X the upper limit of normal, and alkaline phosphatase < 5X the upper limit of normal. Serum bilirubin > 2.0 is acceptable in the setting of known Gilbert's syndrome.
11. No active major medical or psychosocial problems that could be complicated by study participation.

Exclusion Criteria:

1. received neo-adjuvant therapy
2. Cardiac dysfunction documented by an ejection fraction less than the lower limit of the facility normal by multi-gated acquisition (MUGA) scan, or 45% by echocardiogram. -The rate of Disease recurrence
3. Uncontrolled medical problems.
4. Evidence of active acute or chronic infection.
5. Pregnant or breast feeding.
6. Hepatic, renal, or bone marrow dysfunction as detailed above.