[18F]FDG-PET/CT Texture Analysis in “Hot” Thyroid Nodules: Results of the validation Cohort

Background: significance of thyroid nodule 2-deoxy-2-[18F]fluoro-D-glucose ([18F] FDG) uptake on positron emission tomography/computed tomography (PET/CT) scans remains controversial. We aimed to validate our previous results on the ability of texture analysis to predict final diagnosis in [18F]FDG-avid thyroid nodules. Methods: We retrospectively evaluated medical records of all patients who performed an [18F]FDG-PET/CT from January 2017 to December 2018. Those patients who presented a thyroid nodule described in the medical records and performed a fine needle aspiration in our institution were considered for the analysis. Cytological and/ or histological results were used as reference standard to define the final diagnosis. In case of negative cytology, the nodule was considered benign. In case of non-diagnostic or inconclusive results ultrasound, follow-up and/or further cytology/histology was used as final diagnosis. LifeX software was used for image segmentation (threshold of 40% SUVmax) and for textural features extraction. The 7 features previously identified as possible diagnostic biomarkers were tested by logistic regression. Receiver operating characteristic (ROC) curves were plotted and for each curve, the area under the curve (AUC) was calculated. Statistical analysis was performed with STATA package. Results: We identified 498 medical reports describing [18F]FDG uptake in a thyroid nodule. Fifty-two patients performed further diagnostic steps in our institution. Twelve patients/52, referred to ultrasonographers, did not perform cytology. Eight/52 patients were excluded from the analysis because a final diagnosis was not available. Accordingly, 32 patients were analyzed. Nineteen out of 32 patients had benign nodules while in 13/32 cases a malignancy (primary thyroid cancer = 12, metastases = 1) was diagnosed. Conventional PET parameters and histogram-based features were calculated for all patients, while second and higher-order features were available for only 15/32 patients. Skewness and kurtosis resulted significantly different in benign and malignant nodules (p = 0.03 and = 0.02, respectively) and showed the best AUC (=0.75). Conclusions: texture analysis may be useful to characterize [18F]FDG-avid thyroid nodules and to refine the selection of patients to be referred for cytology.