Direct Comparison of Full Protocol Mri and Modified Non-Contrast Mri in Determination of Cancer Origin in Female Genital Tract Cancers: A Retrospective Study Publisher Pubmed



Zeinalkhani F ; Kamali Hakim P ; Mahdavi Sabet F ; Moradi Z ; Rahmani Firouzi S ; Zeinalkhani H ; Shakki Katouli F ; Azizinik F ; Delazar S ; Mohammadzadeh S
Source: Abdominal Radiology Published:2025

Abstract

Background and purpose: Cervical and endometrial cancers are among the most prevalent gynecologic malignancies, associated with high mortality and morbidity. Diagnostic challenges emerge when invasive gynecological cancers concurrently affect the lower uterine segment and cervix, complicating the identification of tumor origin. We aimed to evaluate and compare diagnostic performances of full protocol pelvic MRI and modified non-contrast pelvic MRI in determination of cancer origin in female genital tract cancers. We also compared performance of both protocol in characterization of genital tract cancers. Material and method: In this retrospective cross-sectional study, pelvic MRIs of 273 female patients with histopathologically confirmed genital tract cancer were evaluated. All patients were evaluated by two independent radiologists, with diagnoses initially established using the abbreviated non-contrast T2-weighted MRI protocol and subsequently by the dynamic contrast-enhanced full MRI protocol (T2-weighted and DWI). Primary site and origin of tumor, largest tumor diameter, and depth of myometrial invasion were assessed. Inter-protocol agreement was evaluated by Cohen's kappa. Results: Both abbreviated and full pelvic MRI protocols demonstrated high diagnostic accuracy for cervical (97.39%) and endometrial cancer (96.85–97.64%) when compared to histopathological evaluation as the gold standard. Similarly, sarcoma detection accuracies were identical (86.67%) between the two protocols. Near-perfect inter-protocol agreement for myometrial invasion (κ = 0.963), invasion degree (weighted κ = 0.964), cancer origin (κ = 0.994), and tumor site (κ = 0.912) further confirmed comparable diagnostic performance across protocols. Also, logistic regression analysis in both protocols revealed no significant predictors for accurate cancer origin classification. Conclusion: Based on our study, modified MRI had substantial agreement with full protocol MRI in determination of cancer origin in female genital tract cancers, suggesting its potential as a valuable contrast-free alternative for clinical practice. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.