Association of MRI biomarkers on survival in TNBC patients: A retrospective, multi-center cohort study.

Abstract

e13155 Background: Triple-negative breast cancer (TNBC) is an aggressive subtype associated with heterogenous radiological and histopathological findings. Emerging MRI-based biomarkers—such as peritumoral edema and rim enhancement—may better predict disease progression and recurrence. However, multi-center evidence remains limited, warranting further investigation. Methods: In this retrospective study, women diagnosed with triple-negative breast cancer (TNBC) in the Canadian province of Saskatchewan from 2017 to 2021 were identified. Radiologist MRI reports were independently evaluated and coded to document the presence of peritumoral edema and rim enhancement. Multifactorial data on patient demographics, treatment, and staging were also collected. The relationship between these MRI features and overall survival was analyzed by a non-parametric survival analysis and Cox regression analyses. Chi-squared tests were conducted to evaluate differences in key clinical and pathological outcomes among patients with or without peritumoral edema and rim enhancement. Results: 380 patients with TNBC were included in the study, of which 174 received a breast MRI within 6 months of their diagnosis. Of these patients, 38 (10.0%) had peritumoral edema pre-systemic therapy, and 44 (11.5%) had rim enhancement pre-systemic therapy. The median overall survival for those with peritumoral edema was 1.88 years (95% CI, 1.29 - 6.15) and 2.47 years (95% CI, 1.35 – 6.15) for those with rim enhancement, as compared to 5.10 years (95% CI, 4.30-6.79) for the entire study population. Multivariate analysis demonstrated that peritumoral edema (p = 0.36, HR [95% CI] = 1.53 [0.61 –3.81] and rim enhancement (p = 0.21, HR [95% CI] = 1.81 [0.71 – 4.63] were not significant predictors of overall survival. Those with peritumoral edema were more likely to undergo first-line dose reduction in systemic therapy (p = 0.01, OR [95% CI] = 2.96 [1.25 –7.04]), though this finding was not significant for those with rim enhancement (p = 0.12, OR [95% CI] = 2.08 [0.83 –5.23]). Amongst patients with post-resection pathology post-neoadjuvant systemic therapy, those with rim enhancement and/or peritumoral edema exhibited a greater rate of residual disease versus complete pathological response, though this result was statistically non-significant (p = 0.12, OR [95% CI] = 2.13 [0.83 –5.43]). Similar results were observed with tumor necrosis on post-resection pathology post-neoadjuvant chemotherapy (p = 0.23, OR [95% CI] = 2.03 [0.65 –6.38]). Conclusions: The presence of peritumoral edema and rim enhancement were associated with poor survival outcomes in TNBC. However, these findings are not definitively extended to treatment response, treatment course and pathological outcomes. Larger datasets are required to better elucidate a possible connection between MRI-biomarkers and post-treatment pathology and prognosis.