Inflammatory Indices vs. CA 125 for the Diagnosis of Early Ovarian Cancer: Evidence from a Multicenter Prospective Italian Cohort

Abstract

Objective: Ovarian cancer (OC) remains a diagnostic challenge, often leading to late-stage detection and poor prognosis. CA 125 is widely used in clinical practice, but its limited specificity necessitates complementary biomarkers. This study evaluates the diagnostic performance of systemic inflammatory indices, specifically the Systemic Inflammation Response Index (SIRI) and the Systemic Inflammatory Response (SIR), in differentiating benign ovarian masses, borderline ovarian tumors (BOTs), and OC. Methods: A prospective multicenter observational study was conducted on 94 patients with adnexal masses undergoing primary surgical staging with adnexal masses. The cohort included 31 patients with benign tumors, 42 with BOTs, and 21 with OC. Inclusion criteria encompassed patients with a unilateral adnexal mass identified via transvaginal ultrasound, no suspected extra-ovarian disease, and complete preoperative blood work including CA 125 and systemic inflammatory indices. Exclusion criteria included chronic inflammatory diseases, history of malignancies within the past three years, endometriosis, recent corticosteroid therapy, or incomplete clinical data. Preoperative levels of CA 125, SIRI, and SIR were measured, and their diagnostic performance was assessed using ROC curve analysis. Linear regression models were applied to determine associations between inflammatory indices and ovarian tumor types. Results: SIRI and SIR were significantly higher in OC and BOTs compared to benign tumors (p < 0.001). ROC curve analysis demonstrated that SIRI had a superior diagnostic performance (AUC = 0.71) compared to CA 125 (AUC = 0.59). Regression analysis confirmed that SIRI was independently associated with non-benign ovarian tumors (p = 0.01), while CA 125 did not reach statistical significance. Conclusions: SIRI and SIR show promising diagnostic potential in differentiating ovarian tumor types, outperforming CA 125 in specificity and predictive value. Given their accessibility and cost-effectiveness, these indices could be integrated into preoperative evaluation strategies to enhance risk stratification and early detection of malignancy.