Accuracy of Preoperative Magnet Resonance Imaging to Predict Pathologic T-Stage in Patients With Cervical Cancer

Background/aim: The trend in today's surgical gynecological oncology is to provide equal oncological safety with less radical surgery. The SHAPE trial demonstrated the non-inferiority of a simple hysterectomy compared to a radical hysterectomy in low-risk cervical cancer. As a result, the accuracy of preoperative diagnostics has become increasingly important to avoid both under- and overtreatment. The aim of the study was to investigate the accuracy of MRI-based T-stage.

Patients and methods: Forty-five patients who were surgically treated for an initial diagnosis of a primary cervical carcinoma at the University Hospital Cologne in the Department of Gynecology and Obstetrics between 2015 and 2021 were included in the study. All patients underwent MRI prior to their surgical treatment.

Results: In 44.4% of cases, the pathological tumor size in the surgical specimen was consistent with the preoperative tumor size determined by MRI. In 28.9% of the cases, MRI overestimated the final pathologic T-stage while in 26.7% of cases, MRI underestimated it. Furthermore, we were able to show that overall survival was significantly poorer (p<0.05) in patients whose preoperative MRI had underestimated the final T-stage in our study cohort.

Conclusion: Preoperative MRI diagnostics alone are not reliable enough for accurate T-stage estimation. Multimodal diagnostic approaches are essential for accurate preoperative staging. Prospective trials are needed to evaluate preoperative staging strategies to optimize sizing accuracy.