Amide proton transfer-weighted magnetic resonance imaging for predicting histopathology and biomarkers in rectal adenocarcinoma.

Background: Preoperative evaluation of histopathological features and biomarkers is critical for personalized treatment of rectal adenocarcinoma. Amide proton transfer-weighted (APTw) magnetic resonance imaging (MRI) is a promising noninvasive functional imaging technique, but its predictive value in this tumor remains unclear. This study aimed to assess preoperative APTw MRI as a noninvasive imaging biomarker for predicting histopathological characteristics and biomarkers in rectal adenocarcinoma. Methods: This single-center prospective study enrolled 57 consecutive patients with biopsy-proven rectal adenocarcinoma who were treatment-naive and scheduled for radical surgery, between July 2020 and December 2021. Patients who had received neoadjuvant chemoradiotherapy were excluded to avoid confounding effects of treatment on tumor biology and imaging signal, thereby enabling a clear assessment of baseline APTw signal in relation to pathological features. All participants underwent preoperative APTw MRI and the mean intratumoral APTw signal intensity (APTw SImean) was quantitatively measured and correlated with key pathological parameters including tumor grade, lymph node (N) stage, extramural vascular invasion (EMVI), circumferential resection margin (CRM) and lateral pelvic lymph node metastasis (LPLNM), as well as immunohistochemical markers Ki-67 and p53. Diagnostic efficacy was quantitatively assessed via receiver operating characteristic (ROC) curve analysis, with corresponding area under the curve (AUC) values calculated. Results: Quantitative analysis revealed that APTw SImean was significantly higher in moderately differentiated (G2) versus well-differentiated (G1) tumors [mean difference: 0.523%, 95% confidence interval (CI): 0.248-0.798; P<0.001] in node-positive (pN1-2) versus node-negative (pN0) diseases (mean difference: 0.404%, 95% CI: 0.093-0.715%; P=0.034). Significant differences were also observed between EMVI-positive and negative status (mean difference: 0.310%, 95% CI: 0.012-0.608%; P=0.027), CRM-positive and negative involvement (mean difference: 0.384%, 95% CI: 0.068-0.700%; P=0.021), and LPLNM presence versus absence (mean difference: 0.574%, 95% CI: 0.246-0.902%; P=0.001). Tumors with high Ki-67 expression showed higher APTw SImean than those with low expression (mean difference: 0.520%, 95% CI: 0.207-0.833%; P=0.004), while no significant difference was found for p53 status (P=0.469). Quantitative ROC result showed favorable diagnostic performance of APTw SImean for discriminating tumor grade (AUC =0.811), Ki-67 expression level (AUC =0.813), LPLNM status (AUC =0.792), EMVI status (AUC =0.667), and CRM involvement (AUC =0.727). Notably, the highest AUCs (>0.8) were observed for Ki-67 prediction and tumor grade. Conclusions: Preoperative APTw MRI is a noninvasive biomarker for predicting aggressive histopathological features in rectal adenocarcinoma. Quantitative APTw SImean is significantly associated with tumor grade, Ki-67 expression, and LPLNM presence.