Integrated MRI-Derived, Pathologic, and Molecular Biomarkers for Prediction of Neoadjuvant Chemotherapy Response in Breast Cancer

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Aim: To investigate the predictive value of pretreatment breast magnetic resonance imaging (MRI) findings, clinicopathological characteristics, and molecular tumor features for pathological response to neoadjuvant chemotherapy (NACT).Method: This retrospective study included 194 patients with invasive breast cancer who underwent pretreatment breast MRI, received NACT, and subsequently underwent surgery between 2015 and 2020. Pathological response was assessed using the Miller–Payne grading system and categorized as complete/near-complete response (grades 4–5), partial response (grade 3), or limited response (grades 1–2). Quantitative and morphological MRI parameters, together with clinicopathological and molecular characteristics, were analyzed. Independent predictors of pathological response were identified using multinomial logistic regression analysis.Results: According to the Miller–Payne classification, 80 patients (41.2%) achieved complete/near-complete response, 98 (50.5%) partial response, and 16 (8.2%) limited response. Histologic subtype, tumor grade, HER2 status, Ki-67 index, molecular subtype, ADC, BPE, tumor margins, enhancement pattern, and MSI were significantly associated with pathological response. In multivariable analysis, non-ductal histology and minimal BPE independently predicted limited versus complete/near-complete response, whereas non-luminal molecular subtype, minimal BPE, and MSI ≥429 independently predicted limited versus partial response. Lower ADC independently distinguished partial from complete/near-complete response. However, substantial overlap in biological and imaging features remained across response groups.Conclusion: Pretreatment MRI biomarkers may provide complementary information for predicting NACT response when interpreted with clinicopathologic and molecular features. Lower ADC was not an independent predictor of limited response. The occurrence of similar biological and imaging features in both highly responsive and limited-response tumors highlights substantial treatment-response heterogeneity and supports an integrated multiparametric assessment rather than reliance on any single biomarker.