Katja Natascha De Paepe MD, Abstract Co-Author: Nothing to Disclose

Frederik De Keyzer MSc, Abstract Co-Author: Nothing to Disclose

Pascal Wolter MD, Abstract Co-Author: Nothing to Disclose

Raymond Oyen MD, PhD, Abstract Co-Author: Nothing to Disclose

Gregor Verhoef MD,PhD, Abstract Co-Author: Nothing to Disclose

Vincent Vandecaveye, Abstract Co-Author: Nothing to Disclose

Katrijn Michielsen PhD, Presenter: Nothing to Disclose

The aim of this study was to evaluate 3 Tesla (T) whole body diffusion-weighted imaging (WB-DWI) for staging of Non-Hodgkin lymphoma (NHL).

Sixteen patients with NHL underwent 3T WB-DWI using 2 b-values (0-1000 s/mm2). Coronal reformatted images served for visual analysis in which lesions showing high signal intensity (SI) on b1000 images were regarded suspect of lymphomatous involvement. Regions of interest (ROIs) were drawn on transverse b0 and b1000 s/mm2 images from which the apparent diffusion coefficient (ADC) was calculated. A Mann-Whitney-U test was performed for different tissue subtypes to assess the discriminatory ability of b1000-SI and ADC between malignant and benign lesions. When these lesions demonstrated a significant difference, a receiver operating characteristic (ROC) analysis was performed from which area under the curve (AUC), positive predictive value (PPV), negative predictive value (NPV) and accuracy were calculated. Based on the optimal thresholds obtained for b1000-SI and ADC for lesion differentiation, patients were staged by the Ann Arbor staging system in comparison with positron emission tomography – computed tomography (PET-CT). Correlative imaging and histopathology were used as reference standard to correlate PET-CT and WB-DWI findings.

Per lesion based analysis showed a significant difference in signal intensity and ADC between malignant lesions and benign tissue (p<0,001). Highest accuracy for characterizing lymph nodes was attained using ADC with an AUC, PPV and NPV of 91%, 80% and 90%, respectively. Bone and soft tissue lesions were most accurately characterized by b1000-SI with an AUC of 97%, a PPV of 91% and a NPV of 98% while for soft tissue lesions an AUC, PPV and NPV of 100% was obtained. When applied on a per patient basis, Ann Arbor staging using WB-DWI agreed with PET-CT in 13 of 16 patients. WB-DWI correctly downstaged a patient with sternal osteomyelitis from stage IV to II, correctly upstaged a patient from stage I to II and incorrectly downstaged a patient with testicular lymphoma from stage II to I who underwent WB-DWI after orchidectomy.

Staging of NHL is feasible with 3T WB-DWI showing similar results to PET-CT.

Accurate staging of NHL patients is crucial for prognosis and treatment planning and is currently based on PET-CT findings. WB-DWI may be of complementary value in lesion differentiation and staging.

De Paepe, K, De Keyzer, F, Wolter, P, Oyen, R, Verhoef, G, Vandecaveye, V, Michielsen, K, Staging Non-Hodgkin Lymphoma with 3 Tesla Whole-Body Diffusion-weighted Imaging (WB-DWI).  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11014595.html