Supervisors info:
Σκορίλας Ανδρέας, Καθηγητής Τμήμα Βιολογίας ΕΚΠΑ (Επιβλέπων Καθηγητής),
Κοντός Χρήστος, Επίκουρος Καθηγητής Τμήμα Βιολογίας ΕΚΠΑ,
Αυγέρης Μαργαρίτης Αναπληρωτής Καθηγητής Τμήμα Βιολογίας ΕΚΠΑ
Summary:
Ovarian cancer (OC) represents the most fatal form of gynecological malignancy, primarily due to its insidious onset, non-specific clinical manifestations, and delayed diagnosis. Current screening methods are largely inadequate, exhibiting limited sensitivity and specificity in detecting early-stage disease. Approximately 70% of OC cases are diagnosed at advanced stages (III or IV), which are associated with poor five-year overall survival rates. Consequently, the development of a sensitive and specific biomarker for the early detection and screening of OC is of critical importance.
Diagnostic assessment relies on serum CA125 levels, menopausal status, and sonographic imaging features. However, accumulating evidence suggests that the combined use of CA125 and HE4 biomarkers, along with patient-specific clinical characteristics, may significantly enhance diagnostic accuracy in early-stage ovarian cancer. Emerging multivariate index assays such as OVA1, ROMA, and Overa incorporate multiple serum biomarkers into a single risk assessment tool, demonstrating promising diagnostic performance. Nevertheless, several barriers remain before their widespread implementation in clinical practice can be achieved.
This study evaluated the diagnostic accuracy and prognostic utility of selected biomarkers in a cohort of 40 women, including 21 individuals with ovarian cysts (only 4 of which were malignant) and 19 without cysts. The combination of HE4 and CA125 using the ROMA algorithm exhibited superior diagnostic performance compared to individual biomarkers, with an area under the ROC curve (AUC) of 0.971. This high AUC value indicates excellent overall sensitivity and specificity for the detection of malignancy. Nonetheless, three cases were falsely classified as malignant by the ROMA algorithm, underscoring the need for continued refinement of diagnostic strategies.
Comparative analysis of multiple biomarkers (HE4, CA125, CEA, CA153, CA19-9, CA72-4, SCC, β-hCG, FSH, LH, progesterone, estradiol, and AFP) revealed that certain markers—such as HE4, AFP, and β-hCG—were elevated in malignant cysts, suggesting potential utility in differentiating between benign and malignant lesions. Interestingly, CA125 levels were lower in some malignant masses, likely reflecting tumor-specific biological variability. These findings highlight the importance of integrated biomarker profiling and further underscore the necessity for ongoing research aimed at optimizing diagnostic algorithms—such as ROMA—for reliable malignancy prediction in ovarian cystic lesions.
Keywords:
molecular diagnostics, ovarian cystic lesions, ovarian cancer, biomarkers, comparative analysis, ROMA
