Ricerc@Sapienza

PURPOSE: The aim of the study is to identify CT texture features able to distinguish malignant from benign pulmonary nodules in cases of PET-negative diagnosis (no 18-FDG uptake at the PET scan).
MATERIAL AND METHODS: For this single-center retrospective study we are going to enroll cohort of patients with diagnosis of suspected lung nodule (maximum diameter > 8mm), admitted in our institution, Sant¿Andrea Hospital, Rome, Italy, for a FDG-PET/CT.
Patients with diagnosis of not 18-FDG avid nodule (Standardized Uptake Value, SUV CT-Texture Analysis (CTTA) was performed on the pretreatment unenhanced CT images by using TexRAD (version 3.2.0). TexRAD will be used to segment target nodules drawing a region of interest (ROI) for each CT slice which displayed the lesion, thus the entire nodule¿s volume will be analyzed. CT texture parameters, including Kurtosis, Standard Deviation (SD), Mean and Skewness will be extracted for each spatial scale image filtration (SSF) of texture parameters. Correlation between CTTA features values and pulmonary nodules¿ histological report (malignant/ benign) will be tested. The P-value will be obtained confronting the two groups (malignant/benign lesions).
EXPECTED RESULTS: CT texture analysis could be useful in primary care settings in making quick diagnoses of lung cancer in patient with PET-negative pulmonary nodule.
In preliminary data on 10 patients Kurtosis showed a significant difference for each SSF (all P¿0.0015); SD, skewness and mean were significant relatively to the SSF.
ROC analysis showed significant AUC for Kurtosis at SSF3 (P EXPECTED CONCLUSIONS: CTTA could be a promising radiological tool which can distinguish a benign from a malignant nodule even in those cases without an altered glucose metabolism at the FDG-PET/CT scan.