Work

MRI-derived tumor volume vs. pathological tumor diameter: a comparative study of prognostic value for disease-free survival in multifocal/multicentric breast cancer

Year: 2026

Type: article

Abstract: Background and objective Multifocal/multicentric breast cancer (MMBC) is characterized by multiple tumor foci in the same breast, but conventional T staging based on the largest lesion (Dmax) may underestimate true tumor burden. This study aimed to evaluate whether MRI-derived tumor volumes (Vmax and Vsum) provide superior prognostic value for disease-free survival (DFS) compared with pathological tumor diameters (Dmax and Dsum). Materials and methods We retrospectively reviewed 111 patients with MMBC treated between January 2015 and June 2020. Tumor volumes were quantified from preoperative 3T MRI, and pathological diameters and clinicopathological variables were collected from medical records. To normalize distributions, tumor size metrics were log 2 -transformed. Associations with DFS were assessed using Cox regression, and predictive performance was evaluated via Kaplan–Meier curves, receiver operating characteristic (ROC) analysis, and decision curve analysis (DCA). Results Log 2 -transformed tumor size metrics, positive lymph node count, and molecular subtype were significant prognostic factors for DFS (all P < 0.05). Volumetric metrics provided stronger risk stratification than diameter-based metrics (log-rank P < 0.001). A multivariate Cox model combining positive lymph node count with log 2 (Vsum) showed the best predictive performance (AIC = 138.42; C-index = 0.845) and remained robust after adjusting for molecular subtype and Ki-67. This model achieved higher sensitivity for 5-year DFS prediction (0.923 vs 0.769 for log 2 (Dmax)), and DCA demonstrated greater net clinical benefit at threshold probabilities below 15.5%. Conclusions MRI-derived sum of tumor volumes (Vsum) provides superior prognostic value for DFS in MMBC compared with diameter-based metrics. Integrating Vsum with lymph node burden may improve risk stratification and support personalized treatment planning.

Source: Frontiers in Oncology

DOI: 10.3389/fonc.2026.1749057