PET/MR in the Assessment of Pediatric Histiocytoses

Abstract

The aim of this study was to analyze the feasibility and diagnostic performance of 18F-FDG PET/MR compared with 18F-FDG PET/CT in a cohort of pediatric histiocytosis patients with regard to image quality, lesion detection, and FDG quantification. Children with a diagnosis of Langerhans cell histiocytosis or Rosai-Dorfman disease were prospectively recruited. Seventeen PET/CT and PET/MR examinations were performed on 9 patients (mean age, 6.2 years) following a single-injection dual-imaging protocol. The indication was Langerhans cell histiocytosis in 10 examinations and Rosai-Dorfman disease in 7 examinations. The anonymized data were evaluated for image quality and lesion detection. SUV quantification of each lesion was compared between modalities. All PET/MR examinations had good or excellent image quality and were deemed clinically acceptable. There was substantial agreement of PET image quality among readers of PET/MR images. Per patient, PET/MR and PET/CT had complete concordance in identifying active disease. PET/MRI correctly classified 74 (96%) of 77 foci of disease identified on PET/CT. The per-lesion maximum SUVs were strongly correlated between modalities with a Spearman correlation coefficient of 0.73 (P < 0.001). The use of PET/MR is clinically feasible in pediatric patients with a histiocytic disease. PET/MR demonstrates comparable image quality and lesion detection to PET/CT while maintaining strongly correlated quantitative performance. Given the significant radiation dose savings, PET/MR represents an appealing alternative to PET/CT in the care of children with histiocytic disorders.