Magnetic resonance imaging and ultrasound examination in preoperative pelvic staging of early-stage cervical cancer: post-hoc analysis of SENTIX study.

objectivesSENTIX was a prospective, single-arm, international multicenter study that evaluated sentinel lymph node biopsy without pelvic lymph node dissection in patients with early-stage cervical cancer. We aimed to evaluate the concordance between preoperative imaging modalities (magnetic resonance imaging (MRI) and ultrasound) and final pathology in the clinical staging of early-stage cervical cancer by post-hoc analysis of the SENTIX study data.

methodsIn total, 47 sites across 18 countries participated in the SENTIX study. Patients with Stage IA1/lymphovascular space invasion-positive to IB2 (International Federation of Gynecology and Obstetrics (FIGO) classification (2018)) cervical cancer, with usual histological types and no suspicious lymph nodes on imaging, were prospectively enrolled between May 2016 and October 2020. Preoperative pelvic clinical staging on either pelvic MRI or ultrasound examination was mandatory. Tumor size discrepancy (< 10 mm vs ≥ 10 mm) between imaging and pathology, as well as the negative predictive value (NPV) of MRI and ultrasound for parametrial involvement and lymph node macrometastasis, were analyzed.

resultsAmong 690 eligible prospectively enrolled patients, MRI and ultrasound were used as the staging imaging modality in 322 (46.7%) and 298 (43.2%) patients, respectively. A discrepancy of tumor size ≥ 10 mm was reported between ultrasound and final pathology in 39/298 (13.1%) patients and between MRI and pathology in 53/322 (16.5%), with no significant difference in the accuracy of tumor measurement between the two imaging modalities. The NPV of ultrasound in assessing parametrial infiltration and lymph node involvement was 97.0% (95% CI, 0.95-0.99%) and 94.0% (95% CI, 0.91-0.97%), respectively, and that of MRI was 95.3% (95% CI, 0.93-0.98%) and 94.1% (95% CI, 0.92-0.97%), respectively, with no significant differences between the parameters. Ultrasound and MRI were comparable regarding the tumor size measurement (P = 0.452), failure to detect parametrial involvement (P = 0.624) and failure to detect macrometastases in sentinel lymph node (P = 0.876).

conclusionsPelvic ultrasound examination and MRI had similar concordance with histology in the assessment of tumor size and of parametrial and lymph node invasion in early-stage cervical cancer. Ultrasound examination should be considered part of preoperative pelvic clinical staging in early-stage cervical cancer, especially in limited-resource regions where MRI is unavailable. © 2025 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.