Role of metabolic tumour burden derived from ¹⁸F-FDG PET/CT in ovarian and peritoneal cancers

Abstract

Achieving complete tumour debulking in patients with ovarian and peritoneal cancers is an independent prognostic factor of survival while the likelihood of achieving complete tumour debulking is affected by the presence of peritoneal carcinomatosis. This thesis aimed to evaluate the diagnostic characteristics of 18F-fluorodeoxyglucose Positron Emission Tomography/Computed Tomography (18F-FDG PET/CT) in ovarian and peritoneal cancers and the association between derived metabolic parameters and incomplete debulking surgery.

This retrospective study enrolled patients with primary and recurrent ovarian and peritoneal cancers who underwent 18F-FDG PET/CT before surgery. 18F-FDG PET/CT was qualitatively assessed by scrutinizing 15 anatomical sites of the peritoneum for the presence of peritoneal deposit using surgicopathological findings as standard of reference. The number of 18F-FDG avid peritoneal site was quantified. Lesions detected were semi-quantitatively measured [standard uptake value – SUV, metabolic tumour volume - MTV and total lesion glycolysis – TLG (both MTV and TLG were based on different threshold settings, from 30% to 70%)]. MTV and TLG of all peritoneal lesions were summated as PCMTV and PCTLG, respectively. Surgical outcome was dichotomized into complete debulking and incomplete debulking, and served as the main outcome measure. Univariate and multivariate analyses were performed to identify the independent risk factor for incomplete debulking surgery with subsequent receiver operating curve (ROC) to determine the optimal cut-off value. Based on different cut-off values, the rate of unnecessary surgical exploration and inappropriately unexplored rate were calculated.

Forty-nine patients (mean age 49 ± 15) had ovarian (n = 45) and peritoneal cancers (n = 4); 38 patients achieved complete debulking and 11 patients had incomplete debulking after surgery. Patients with 18F-FDG PET/CT detected peritoneal carcinomatosis had significantly higher rate of incomplete debulking (34.5% versus 5.0%) and lower rate of complete debulking (65.5% versus 95.0%) (p = 0.017). The site-base sensitivity and specificity for detection of peritoneal carcinomatosis (n = 735) were 75.9% and 99.4%, respectively while detection rate for peritoneal lesions > 1 cm was higher (sensitivity = 95.5%). PCMTV (for example PCMTV_50%: 63 ± 92 cm3 vs 22 ± 39 cm3 respectively, p = 0.012), PCTLG (for example PCTLG_50%: 137 ± 148 cm3 vs 45 ± 87 cm3 vs respectively, p = 0.004), SUVmax of the most avid peritoneal lesion (5.8 + 3.4 versus respectively 2.4 + 3.4, p = 0.006) and number of 18F-FDG avid peritoneal site (2.3 + 1.7 vs 0.6 + 0.8 sites respectively, p = 0.001) were significantly higher in incomplete debulking cohort than complete debulking cohort. Multivariate analysis showed the number of 18F-FDG avid peritoneal site was an independent risk factor for incomplete debulking (p = 0.042, standard error =0.525, odd ratio = 2.905 with 95% CI 1.039 to 8.126) with an optimal cut-off value of number of ≥ 4, resulting in 15.6% of unnecessary surgical exploration, 0% of inappropriately unexplored rate (AUC = 0.816).

To conclude, 18F-FDG PET/CT was moderately sensitivity and highly specific in detecting peritoneal carcinomatosis. The number of 18F-FDG avid peritoneal site was an independent risk factor for incomplete debulking in patients with ovarian and peritoneal cancers.