Whole-body MRI, FDG-PET/CT, and bone marrow biopsy, for the assessment of bone marrow involvement in patients with newly diagnosed lymphoma

Purpose: To compare whole-body MRI (WB-MRI) with diffusion-weighted imaging (DWI), FDG-PET/CT, and bone marrow biopsy (BMB), for the evaluation of bone marrow involvement (BMI) in patients with newly diagnosed lymphoma. Materials and Methods: This retrospective study was approved by our Institutional Review Board. Two independent radiologists and one nuclear medicine specialist reviewed all WB-MRI and FDG-PET/CT scans prospectively performed on 104 patients with newly diagnosed lymphoma (53 males; 47 Hodgkin; mean age: 44 years; range, 15–86 years) between 2013 and 2015. The delay between imaging scans and BMBs was up to 10 days. The diagnostic accuracy of WB-MRI (1.5 Tesla MR scanner, with T1w, T2w-STIR, and DWI sequences) was evaluated using BMB and FDG-PET/CT as the reference standard. We applied Cohen's kappa coefficient to assess the inter-observer agreement in WB-MRI interpretation and to compare WB-MRI, FDG-PET/CT and BMB. The Student's t test was done to compare pelvic marrow ADC values of patients with positive and negative BMB. A P-value of < 0.01 was considered significant. Results: Inter-observer agreement was excellent (k = 0.937). Agreement between WB-MRI and FDG-PET/CT was excellent, with a k = 0.935. Agreement between WB-MRI and BMB was moderate (k = 0.489), and fair between FDG-PET/CT and BMB (k = 0.370). WB-MRI and FDG-PET/CT were falsely negative in four indolent non-Hodgkin lymphomas with BMI < 30% of marrow cellularity. Conversely, WB-MRI and FDG-PET/CT detected all cases with a BMI>30% of marrow cellularity. Mean ADC values in patients with positive and negative BMB were not significantly different (P = 0.049). Conclusion: WB-MRI and FDG-PET/CT are valuable tools for the assessment of BMI. Level of Evidence: 3. J. Magn. Reson. Imaging 2017;45:1082–1089.