Anterior uterocervical angle and cervical tumor topography: a hypothesis-generating model for future colposcopic research

To investigate the association between the anterior uterocervical angle (aUCA) and the predominant topographic localization of cervical tumors, and to evaluate the agreement between transvaginal ultrasonography (TVUS) and magnetic resonance imaging (MRI) measurements of the aUCA. In this retrospective study, 62 women with surgically treated cervical cancer were included. The aUCA was measured on preoperative MRI and routinely acquired TVUS images. Tumor localization was determined according to postoperative histopathological findings and categorized as anterior, right lateral, left lateral, posterior or circumferential. Associations between aUCA and tumor topography were evaluated using group comparisons, exploratory receiver operating characteristic (ROC) analysis, multivariable linear regression and sensitivity analyses. Agreement between TVUS- and MRI-derived measurements, and of MRI-based localization with histopathology, was assessed. Mean MRI- and TVUS-measured aUCAs were 143.3° ± 20.7° and 142.5° ± 21.2°, respectively. Posterior tumors were associated with significantly wider aUCAs than anterior tumors on both MRI (157.4° ± 23.4° vs. 134.2° ± 14.8°; Dunn–Bonferroni p = 0.001) and TVUS (156.5° ± 24.1° vs. 133.5° ± 15.1°; p = 0.001). In an exploratory ROC analysis of anterior versus posterior tumors ( n = 51), MRI-measured aUCA showed moderate discrimination (area under the curve [AUC] = 0.78, 95% CI 0.64–0.92); a sample-derived cut-off of ≥ 159° yielded 57.9% sensitivity and 93.8% specificity against anterior tumors. Posterior tumor localization remained associated with a wider aUCA in multivariable linear regression (B = 22.5°, 95% CI 11.0–34.0) and in all sensitivity analyses. TVUS and MRI measurements were highly concordant (intraclass correlation coefficient [ICC] = 0.96), with a mean bias of − 0.8° and 95% limits of agreement of − 12.5° to 10.9°, and MRI-based tumor localization showed near-perfect agreement with postoperative histopathology (κ = 0.95). In this selected surgical cohort, a wider anterior uterocervical angle was associated with posterior cervical tumor localization. These hypothesis-generating findings require prospective validation in women with high-risk HPV infection or cervical intraepithelial neoplasia before any role in colposcopic evaluation can be considered.