Objective:: This retrospective study examined whether the primary tumour 18F-FDG uptake features could predict the high-risk of recurrence in differentiated thyroid cancer (DTC) patients.
Methods:: The enrolled 114 DTC patients underwent preoperative 18F-FDG-PET/CT. SUVmax, SUVmean, metabolic tumour volume (MTV), total lesion glycolysis (TLG) and 6 texture parameters were obtained. Because the texture features can be confounded by the tumour volume effects, 18F-FDG-avid tumour patients were divided into two groups (tumours with MTV ≤ 10.0 cm3 and >10.0 cm3). Diagnostic performance for predicting the high-risk was evaluated by the area under the curve (AUC) by the ROC curve analysis.
Results:: Eighty eight 18F-FDG-avid tumours revealed more advanced-risk classification (p = 0.015 → 0.02) than 26 18F-FDG-nonavid tumours, which yielded no high-risk patients. In the 44 MTV > 10.0 cm3 18F-FDG-avid tumour patients, 8 high-risk patients revealed significantly higher SUVmax, SUVmean, MTV, TLG, intensity variability and size-zone variability, and lower zone percentage than 36 non-high-risk patients (p < 0.001-0.016). Their AUC (diagnostic accuracy) ranged between 0.77 (66%) and 0.92 (91%). When each parameter was scored as 0 (negative for high-risk) or 1 (positive for high-risk) according to each threshold criterion, and the 7 parameter summed score ≥5 was defined as high-risk, the accuracy was 93.2% (AUC: 0.98) in the MTV > 10.0 cm3 18F-FDG-avid tumour patients.
Conclusion:: For primary MTV > 10.0 cm3 18F-FDG-avid DTCs, the combined use of SUV-related, volumetric, and texture parameters may be more useful to identify high-risk patients than the individual parameters.
Advances in knowledge:: Combined use of SUV-related, volumetric, and texture parameters may be useful to identify high-risk DTC patients.