This study is aimed to illustrate whether Radiomics combining multiparametric MRI before and after neoadjuvant chemotherapy (NACT) with clinical data is a good way to predict axillary lymph node metastasis and prognosis in invasive-breast-cancer.

This study proposes to build a clinical predictive model to predict axillary lymph node metastasis and prognosis in invasive-breast-cancer patients who received neoadjuvant chemotherapy before surgery. The model is built based on breast MRI signatures extracted and analyzed via deep machine-learning algorithm methods. Invasive breast cancer patients undergo multiparametric MRI at baseline, then undergo multiparametric MRI after received neoadjuvant chemotherapy for at least 4 cycles as planned. After the surgery, responses to neoadjuvant chemotherapy are determined according to the histopathologically examination of the surgically resected specimens. After completion of treatment procedure, patients are followed up for 5 years.

Inclusion Criteria:

1. Primary lesion diagnosed as invasive breast cancer;
2. Imaging examination confirmed no distant organ metastasis;
3. Received neoadjuvant chemotherapy for drugs such as taxanes, anthracyclines, and platinum as planned;
4. Completed breast MRI examination before or after neoadjuvant chemotherapy;
5. Accepted breast cancer surgery and axillary lymph node dissection;
6. Eastern Cooperative Oncology Group performance status 0-2.

Exclusion Criteria:

1. History of ipsilateral axillary or breast surgery;
2. Inflammatory breast cancer;
3. Bilateral breast cancer;
4. Malignant tumor history in 5 years;
5. Patients with cervical or contralateral axillary lymph node metastasis;
6. Incomplete imaging or medical history data.

Invasive breast cancer patients with no distant organ metastasis who had neoadjuvant chemotherapy before the surgery

Requests for the individual data or study documents will be considered where the proposed use aligns with public good purposes, does not conflict with other requests, and the requestor is willing to sign a data access agreement. Contact is though the corresponding author.