Whole body diffusion-weighted MRI to predict treatment outcome after one cycle of immunochemotherapy in aggressive non-Hodgkin lymphoma.

Abstract

7534 Background: Treatment adaptation based on early identification of non-Hodgkin lymphoma (NHL) patients not responding to therapy might improve survival. The role of interim fluorodeoxyglucose-positron emission tomography/computed tomography (FDG-PET/CT) after 2-4 cycles of immunochemotherapy (ICT) herein is experimental as it renders false positive results, due to a rituximab-induced inflammatory response. Whole body diffusion-weighted magnetic resonance imaging (WB-DWI/MRI) was evaluated as a radiation-free imaging technique to predict treatment outcome in NHL after one cycle ICT (2-3 weeks). Methods: 47 patients with aggressive NHL (35 DLBCL,2 primary mediastinal BCL,3 unclassifiable BCL, 2 Burkitt,2 MCL, 2 peripheral TCL and 1 extranodal NK-TCL) were enrolled. All had baseline and interim WB-DWI/MRI, and end-of-treatment PET/CT; 39/47 had interim PET/CT. International prognostic index (IPI), immunohistochemical (IHC) markers Ki-67, Bcl-6 and Bcl-2 were evaluated for their predictive value. WB-DWI/MRI was assessed quantitatively with histogram analysis (high b-value signal intensity (SI) and apparent diffusion coefficient (ADC)). Patients were categorized as non-responder when lesions had decreased ADC or insufficient SI decrease between scans. Kaplan-Meier survival analysis was performed with log rank, Cox hazard ratio calculation and multivariate analysis. Outcome measure was disease-free-survival (DFS). Results: Median follow-up time was 43 months (4-70 months). 33 patients had complete remission (CR), 5 progression and 9 recurrent disease. WB-DWI/MRI predicted DFS correctly in 45/47 (96%) [log rank p<0.001; hazard ratio (HR) 52, (CI 95% 6-401)]; end-of-treatment PET/CT was correct in 37/47 (79%) [p=0.003;HR 4.3, (1.5-12.4)], and interim PET/CT in 28/39 (72%) [p=0.016;HR 3.9, (1.2-12.5)]. IPI score and IHC parameters were not significantly predictive. Multivariate analysis showed WB-DWI/MRI as the only independent prognostic factor (p<0.001). Conclusions: WB-DWI/MRI can accurately predict treatment outcome in aggressive NHL after only one cycle of immunochemotherapy without the burden of radiation exposure. Clinical trial information: NCT01231269.