Transrectal Biplane Ultrasound Features and Clinical Characteristics Associated With T3-T4 Staging of Rectal Cancer:...

To evaluate the predictive value of transrectal biplane ultrasound (TRBUS) imaging features and clinical characteristics for distinguishing T3-T4 stage rectal cancer (RC) and to improve preoperative staging accuracy and guide treatment planning.

A retrospective analysis of 170 RC patients who underwent preoperative TRBUS was conducted. We assessed 10 clinical variables (eg, age, body mass index, carcinoembryonic antigen [CEA] levels) and 7 ultrasound features (eg, lesion thickness, outer membrane spiculation [defined as spiky edges indicating tumor breakthrough of the serosal layer or outer membrane], vessel penetration [defined as color Doppler flow signal extending beyond the outer membrane]). Univariate and multivariate logistic regression analyses were performed to identify independent predictors of T3-T4 stage, and receiver operating characteristic curve analysis was subsequently used to assess their diagnostic performance.

Among 170 RC patients, 87 were diagnosed with T3-T4 stage and 83 with T1-T2 stage. Univariate analysis linked T3-T4 stage with clinical symptoms, higher histologic grade (P ≤ .031), elevated CEA levels (24.1% versus 12.0%, P = .041), the outer membrane spiculation sign (89.7% versus 19.3%, P < .001), thicker lesions (18.8 ± 10.9 versus 14.0 ± 4.6 mm, P = .012), the presence of ulcers (14.9% versus 4.8%, P = .028), and the vessel penetration sign (86.2% versus 2.4%, P < .001). Stepwise regression analysis confirmed CEA (B = 0.111, P = .022), the outer membrane spiculation sign (B = .183, P < .001), and the vessel penetration sign (B = .698, P < .001) as independent predictors. Combining spiculation and vessel penetration achieved AUC = 0.985 (sensitivity 94.3%, specificity 80.7%). TRBUS staging showed good agreement with pathology (kappa = 0.702, P < .001).

TRBUS features, the outer membrane spiculation sign and the vessel penetration sign, are critical predictive markers for T3-T4 stage RC. These imaging biomarkers significantly improve the accuracy of preoperative T staging, providing clinicians with robust evidence to guide therapeutic decision-making.