[Relationship between tumor regression pattern and prognosis after neoadjuvant therapy for breast cancer]

Objective: To retrospectively analyze the histological regression patterns observed in residual tumor beds following neoadjuvant therapy (NAT) for breast cancer, with the aim of evaluating their prognostic significance. Methods: A total of 1 102 patients with pathologically confirmed invasive breast carcinoma of no special type (IBC-NST) who received surgical resection following NAT from 2015 to 2019 were enrolled at the Department of Pathology, the Fourth Hospital of Hebei Medical University. Histological evaluation was performed based on whole-slide images. Regression patterns were classified according to the spatial distribution of residual tumor. The associations between regression patterns and clinicopathological features were assessed. Independent prognostic factors for disease free survival (DFS) and overall survival (OS) were analyzed, and relevant results were visualized. Results: The age of 1 102 patients was 49 (42, 58) years, and the follow-up period was 81 (69, 93) months. Tumor regression patterns were classified into two categories based on histological characteristics: concentric regression and centrifugal regression. Concentric regression was further subdivided into four subtypes: isolated residual (Iso), concentric regression (Con), ring-like regression (Cir), and pathological complete response (pCR). Centrifugal regression comprised five subtypes: no/minimal response (No/Min), cord-like regression (Cor), clumpy regression (Clu), cribriform regression (Cri), and mixed regression (Mix). Significant differences in DFS and OS were observed among the nine subtypes. Overall, patients with concentric regression exhibited markedly better survival outcomes than those with centrifugal regression (P<0.05), and this difference remained statistically significant in the HR⁺HER2^- molecular subgroup. Conclusions: Histologic regression patterns in the residual tumor bed after neoadjuvant therapy were associated with prognosis in patients with breast cancer. Disease-free survival and overall survival were shorter in the centrifugal regression group than in the centripetal regression group. Histologic changes in the tumor bed may serve as an indicator for risk stratification in breast cancer.