Assessment of the efficacy of dynamic contrast-enhanced magnetic resonance imaging quantitative parameters combined with serological biomarkers in diagnosing lymph node metastasis of cervical cancer

This work aimed to study the diagnostic value of quantitative parameters and serological biomarkers (SBs) in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for lymph node metastasis (LNM) in cervical cancer (CC), and construct a predictive model, this paper was conducted. Data of patients with pathologically confirmed CC were retrospectively analyzed. DCE-MRI quantitative parameters and serological indicators (SCC-Ag, VEGF, MMP-9) were collected. Build a nomogram prediction model and assess the quality of this model. Univariate regression analysis showed that Ktrans, Kep, Ve, SCC-Ag, VEGF, and MMP-9 were all significantly associated with LNM in CC (all P < 0.001). Multiple linear regression analysis showed that Ktrans, Kep, Ve, SCC-Ag and MMP-9 were independent predictors (all P < 0.001), but VEGF was not. The nomogram model in the development cohort had an AUC of 0.86 (95% CI 0.78–0.95), a diagnostic sensitivity of 85.7% and a specificity of 83.2%. After internal validation using 1,000 bootstrap resamples, the bias-corrected AUC was 0.81 (95% CI 0.72–0.90). The AUC value for the external validation cohort is 0.87 (95% CI 0.79–0.94), with a sensitivity of 83.3% and a specificity of 81.2%. The prediction and actual probabilities of the calibration curve were in good agreement (Hosmer-Lemeshow test P = 0.646), and according to decision curve analysis, notable net benefit was confirmed across a wide range of risk thresholds. A nomogram model combining quantitative DCE-MRI parameters and serum biomarkers (SCC-Ag, MMP-9) has achieved good preoperative predictive performance for lymph node metastasis in cervical cancer in the study cohort of this paper.