Signal enhancement ratio of multi-phase contrast-enhanced MRI: an imaging biomarker for survival in pancreatic adenocarcinoma

• DOI: 10.1007/s00330-024-10746-z
• Publication Date: 2024-05-15T16:01:45Z
• Authors: Cong Xia; Jin-rong Qu; Yi-ping Jiao; Chun-qiang Lu; Ben Zhao; Rong-jun Ge; Yue Qiu; Bu-yue Cao; Qian Yu; Tian-yi Xia; Xiang-pan Meng; Yang Song; Li-hua Zhang; Xue-ying Long; Jing Ye; Zhi-min Ding; Wu Cai; Sheng-hong Ju
• Keywords: Imaging biomarker

Abstract Objectives To evaluate signal enhancement ratio (SER) for tissue characterization and prognosis stratification in pancreatic adenocarcinoma (PDAC), with quantitative histopathological analysis (QHA) as the reference standard. Methods This retrospective study included 277 PDAC patients who underwent multi-phase contrast-enhanced (CE) MRI whole-slide imaging (WSI) from three centers (2015–2021). SER is defined (SI lt − SI pre )/(SI ea ), where , ea, represent intensity of tumor pre-contrast, early-, late post-contrast images, respectively. Deep-learning algorithms were implemented to quantify stroma, epithelium, lumen on WSIs. Correlation, regression, Bland-Altman analyses utilized investigate associations between QHA. The prognostic significance overall survival (OS) was evaluated using Cox regression Kaplan–Meier curves. Results internal dataset comprised 159 patients, which further divided into training, validation, test datasets ( n = 60, 41, 58, respectively). Sixty-five 53 two external datasets. Excluding lumen, demonstrated significant correlations stroma r 0.29–0.74, all p < 0.001) epithelium −0.23 −0.71, across a wide post-injection time window (range, 25–300 s). revealed small bias QHA quantifying stroma/epithelium individual validation (all within ± 2%), 4%). Moreover, SER-predicted low stromal proportion independently associated worse OS (HR 1.84 (1.17–2.91), 0.009) training datasets, remained combined 1.73 (1.25–2.41), 0.001). Conclusion CE-MRI allows PDAC. Clinical relevance statement can serve novel biomarker characterizing composition holds potential improving patient therapy Key Points Imaging biomarkers are needed better characterize adenocarcinoma. Signal ) -predicted stromal/epithelial showed good agreement histopathology measurements distinct centers. be an independent factor Graphical