Diagnostic and prognostic evaluation of fluorodeoxyglucose positron emission tomography/computed tomography and its correlation with serum cancer antigen-125 (CA125) in a large cohort of ovarian cancer patients

Objective: We evaluated the efficacy of 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) in recurrent disease, response to therapy, and long-term follow-up of ovarian cancer (OC) patients in relation to cancer antigen-125 (CA125) levels and the prognostic meaning of this modality in this subset of subjects.

Material and methods: Between 2005 and 2015, we retrospectively evaluated 125 patients affected by OC who underwent FDG PET/CT imaging at our institution. The indications for PET/CT were recurrence of disease in 78 patients, therapy response assessment in 29, and follow-up in 18. The results of FDG PET/CT were compared with those of histopathology and clinical and radiological progression during follow-up for at least 6 months. The median long-term follow-up was 33 months. The diagnostic accuracies for the different clinical settings were evaluated. The relationships among global survival (GS), FDG PET/CT results, and CA125 levels were evaluated by both Kaplan-Meier and Cox regression analysis.

Results: CA125 results were positive (>35 UI/mL) in 62 patients and negative in 63 (49% vs. 51%). The sensitivity and specificity of CA125 were 72% and 91%, respectively. PET/CT imaging showed a sensitivity of 98.6% and a specificity of 77.8% for the assessment of recurrent disease, and a sensitivity of 72.7% and a specificity of 88.9% for therapy evaluation. Meanwhile, in 18 patients evaluated during follow-up, the specificity was 82.3%. GS was significantly higher in case of negative CA125 values at the time of FDG PET/CT, of a negative PET/CT scan and when no evidence of peritoneum recurrence and distant metastases was determined by PET. Multivariate regression analysis showed that only age and peritoneum recurrence as determined by PET were identified as independent predictors of poor prognosis.

Conclusion: Metabolic imaging with FDG PET/CT proved useful in patients where OC recurrence was suspected, even when the value of tumor marker CA125 was in a normal range. A positive PET/CT scan and the presence of peritoneum recurrence at PET were associated with a poor prognosis after approximately 30 months.