PET/CT com 18F-FDG no planejamento radioterápico dos pacientescom carcinoma pulmonar não pequenas células: proposta de Threshold individualizado

Abstract

Introduction: Lung carcinoma has significant relevance in the global and national epidemiological profile due to its high incidence and high mortality rate. The most common histological type is non-small cell lung carcinoma (NSCLC) which accounts for approximately 85% of all lung carcinoma cases. Most of these patients have advanced stage disease at diagnosis due to late onset of symptoms, leading to a significant reduction in survival. The NSCLC treatment is based on an accurate staging and advanced cases usually are treated with radiation therapy. Conformationals radiationg treatment planning need precise tumor delineation and fluorine 18 labeled fluordeoxyglucose positron emission tomography computed tomography (18F-FDG PET/CT) has became an important tool, however there is no consensus on how to use this technology. Objectives: To evaluate an individualized segmentation algorithm using 18F-FDG PET/CT images to generate gross tumor volume (GTV) in NSCLC patients and asses the clinical impact. Methods: Twentyfive patients with NSCLC whom underwent 18F-FDG PET/CT staging examinations were included in this study. All lesions were segmented using an semi-automated algorithm for segmentation of GTV 2SD. Fixed threshold (40%, 50% and 60%) were also applied for comparisons (GTV 40%, 50%, 60%). Clinical impact was verified in an small subsample (n=4). Results: The threshold for tumor segmenation (and the GTV delineation) was smaller, as long as the SUVmax increase, following an quadratic regression model. Regardless the positive linear correlation between fixed threshold and the GTV delineated using CT images, significant subestimation were observed. At clinical practice, GTVs generated with an individualized semi-automated segmentation algorithm were smaller than those from CT-based. Conclusions: 18FFDG PET / CT is an important tool in the radiation therapy, being more effective with the use of an individualized algorithm for more precise GTVs. However, the clinical impact of this methodology need to be evaluated in a larger sample for confirmation.