Automatic assessment of body composition in children with lymphoma: results of a [18F]FDG-PET/MR study

Abstract

Abstract Objectives To use Dixon-MR images extracted from [18F]FDG-PET/MR scans to perform an automatic, volumetric segmentation and quantification of body composition in pediatric patients with lymphoma. Materials and methods Pediatric patients with lymphoma examined by [18F]FDG-PET/MR at diagnosis and restaging were included. At each time point, axial fat and water Dixon T1w images of the thighs were automatically segmented and muscle volume, subcutaneous, intramuscular, and intermuscular fat volume were quantified. The metabolic activity of the largest nodal lesion and of muscles and subcutaneous fat was recorded. The paired samples t-test and Spearman’s correlation coefficient were applied to evaluate potential differences between the two time points and the relationship between metabolic and body composition metrics, respectively. By logistic regression analysis, the prognostic role of the investigated variables was assessed. The applied significance level was p < 0.05 for all analyses. Results Thirty-seven patients (mean age ± SD 14 ± 3-years-old; 20 females) matched the inclusion criteria. After chemotherapy (interval between the two PET/MR scans, 56–80 days; median 65 days), muscle volume significantly decreased (629 ± 259 cm3 vs 567 ± 243 cm3, p < 0.001) while subcutaneous, intramuscular and intermuscular fat increased (476 ± 255 cm3 vs 607 ± 254 cm3, p < 0.001; 63 ± 20 cm3 vs 76 ± 26 cm3, p < 0.001; 58 ± 19 cm3 vs 71 ± 23 cm3, p < 0.001); the metabolic activity of the main nodal lesion, muscles, and subcutaneous fat significantly decreased (p < 0.05, each). None of the examined variables acted as predictors of the response to treatment (p = 0.283). A strong correlation between BMI and subcutaneous fat volume at diagnosis (r = 0.675, p < 0.001) and restaging (r = 0.600, p < 0.001) emerged. Conclusions The proposed method demonstrated that pediatric patients with lymphoma undergo muscle loss and an increase of subcutaneous fat during treatment. Clinical relevance statement The proposed automatic and volumetric MR-based assessment of body composition in children with lymphoma can be used to monitor the effect of chemotherapy and may guide tailored exercise programs during chemotherapy. Key Points T1w Dixon images can be used for the automatic segmentation and quantification of body composition. Muscle and subcutaneous fat volume do not act as predictors of the response to treatment in children with lymphoma. Chemotherapy induces changes in body composition in children with lymphoma.