e12118 Background: HR-/HER2+ breast cancer is a subtype with aggressive characteristic and poor survival. More clinical evidence are needed for choice of therapeutic strategies. Methods: Patients with T1-3N0-3M0 received preoperative chemotherapy (PTX 175 mg/m2, CBP AUC 4, q2w*6) combined with trastuzumab (2mg/kg qw) or standard postoperative chemotherapy such as ddAC-PH, AC-PH, TCH. The primary endpoint was RFS. Results: 86 patients were enrolled, 43 received preoperative chemotherapy (pre arm) and the other 43 received postoperative chemotherapy (post arm). There was no significant difference in baseline between the two arms. 22.1% of patients were stage IIA, 25.6% IIB, 34.9% IIIA, and 18.6% IIIC. At a median follow-up of 33.4 months, 16 patients had relapsed (pre arm 8, post arm 8). The median time from diagnosis to relapse was 22.8 months (7.1-49.2) and 23.8 months (11.4-37.4) in pre and post arm. Kaplan-Meier survival analysis estimated that the 3-year RFS were similar (pre vs post: 73.4% vs 75.4%, p= 0.631). Only 1 death occurred in post arm. Table showed that in subgroups, there was no statistical difference in risk of recurrence between pre and post arms. In pre arm, ORR was 97.7% clinically, and pCR (ypT0/TisN0) was 39.0%. No patients achieved pCR relapsed, and the residual invasive lesions indicated poor prognosis. Table showed that Neo-Bioscore 4-5 was related to recurrence event significantly ( p= 0.021). The rate of breast-conserving in pre arm was higher (19.5% vs 9.3%), and PCb regiments every 2 weeks had similar adverse effects with standard chemotherapy, and less patients had dose reductions (18.6% vs 25.6%). Conclusions: Preoperative chemotherapy versus standard postoperative chemotherapy results in similar RFS among HR-/HER2+ patients. Preoperative chemotherapy can identify prognosis of patients early by Neo-Bioscore and adjuvant therapy should be strengthened for high-risk patients. PCb every 2 weeks combined with trastuzumab can be an option of preoperative therapy for HER2+ breast cancer. Clinical trial information: NCT02934828. [Table: see text]
Read full abstract