Diffusion-weighted MRI for staging and evaluation of response in diffuse large B-cell lymphoma: a pilot study

Abstract

The aim of this study was to compare diffusion-weighted MRI (DW-MRI) with positron emission tomography/computed tomography (PET/CT) for the staging and evaluation of the treatment response in patients with diffuse large B-cell lymphoma (DLBCL). Institutional review board approval was obtained for this study; all subjects gave informed consent. Twelve patients were imaged before treatment and eight of these were also imaged after two cycles of chemotherapy using both DW-MRI and PET/CT. Up to six target lesions were selected at baseline for response assessment based on International Working Group criteria (nodes > 1.5 cm in diameter; extranodal lesions > 1 cm in diameter). For pretreatment staging, visual analysis of the numbers of nodal and extranodal lesions based on PET/CT was performed. For interim response assessment after cycle 2 of chemotherapy, residual tumor sites were assessed visually and the percentage changes in target lesion size, maximum standardized uptake value (SUVmax ) and apparent diffusion coefficient (ADC) from pretreatment values were calculated. In 12 patients studied pretreatment, there were 46 nodal and 16 extranodal sites of lymphomatous involvement. Agreement between DW-MRI and PET/CT for overall lesion detection was 97% (60/62 tumor sites; 44/46 nodal and 16/16 extranodal lesions) and, for Ann Arbor stage, it was 100%. In the eight patients who had interim assessment, five of their 49 tumor sites remained abnormal on visual analysis of both DW-MRI and PET/CT, and there was one false positive on DW-MRI. Of their 24 target lesions, the mean pretreatment ADC value, tumor size and SUVmax were 772 µm(2) /s, 21.3 cm(2) and 16.9 g/mL, respectively. At interim assessment of the same 24 target lesions, ADC values increased by 85%, tumor size decreased by 74% and SUVmax decreased by 83% (all p < 0.01 versus baseline). DW-MRI provides results comparable with those of PET/CT for staging and early response assessment in patients with DLBCL.

Keywords: apparent diffusion coefficient; diffuse large B-cell lymphoma; diffusion-weighted MRI; positron emission tomography/computed tomography (PET/CT); response to chemotherapy; standardized uptake value.

Figure 1

Study flow diagram.

Figure 2

56-year-old woman with stage IV DLBCL and hepatic involvement. DW-MRI and PET/CT for lesion detection. (A) Pre-treatment DW-MRI with b value 50 sec/mm² showing two hyperintense hepatic lesions (arrows). (B) DW-MRI with b value 800 sec/mm² showing the same tumors. (C) Corresponding axial ADC coefficient map demonstrating hyperintense lesions (ADC mean 889 μm²/s). (D) Corresponding PET/CT image with intense FDG uptake in the lesions (SUV_max 16.9).

Figure 3

59-year-old woman with stage IV DLBCL and a false-negative mesenteric lymph node on baseline MRI. (A) Pretreatment DW-MRI (b value 50 sec/mm²⁾ and (B) DW-MRI (b 800 sec/mm²⁾ demonstrate a lesion with restricted diffusion in the splenic hilum (arrow), which was assumed to be a splenule. (C) Corresponding pretreatment PET/CT shows increased FDG uptake (much greater than that in the normal spleen), consistent with involvement of a lymph node near the splenic hilum.

Figure 4

39-year-old woman with stage IV DLBCL. DW-MRI and PET/CT used in treatment response monitoring. There is widespread nodal involvement as well as involvement of the lungs. (A) Pretreatment DW-MRI (b value 50 sec/mm²⁾ (top left), DW-MRI (b value 800 sec/mm²⁾ (top right), axial ADC coefficient map (bottom left), and fused PET/CT (bottom right) demonstrate lymphadenopathy in right axillary, subpectoral and paratracheal lymph nodes, as well as a 1.5-cm left upper lobe pulmonary nodule (arrow). FDG-avid right paratracheal lymphadenopathy is not well demonstrated on the DWI as the lymph node was slightly below the selected DW-MRI imaging plane. (B) Pretreatment whole-body coronal b-50 and b-800 images (left and middle panels) and PET (right panel) show subpectoral and paratracheal lymph nodes and left upper lobe pulmonary nodule as well as a splenic mass. (C) Post therapy whole-body coronal. b-800 (left panel) and PET (right panel) images show resolution of subpectoral, mediastinal and lung abnormalities and a residual splenic mass with mildly increased FDG uptake (arrow). The b-50 DW-MR image had a similar appearance.

Figure 4

39-year-old woman with stage IV DLBCL. DW-MRI and PET/CT used in treatment response monitoring. There is widespread nodal involvement as well as involvement of the lungs. (A) Pretreatment DW-MRI (b value 50 sec/mm²⁾ (top left), DW-MRI (b value 800 sec/mm²⁾ (top right), axial ADC coefficient map (bottom left), and fused PET/CT (bottom right) demonstrate lymphadenopathy in right axillary, subpectoral and paratracheal lymph nodes, as well as a 1.5-cm left upper lobe pulmonary nodule (arrow). FDG-avid right paratracheal lymphadenopathy is not well demonstrated on the DWI as the lymph node was slightly below the selected DW-MRI imaging plane. (B) Pretreatment whole-body coronal b-50 and b-800 images (left and middle panels) and PET (right panel) show subpectoral and paratracheal lymph nodes and left upper lobe pulmonary nodule as well as a splenic mass. (C) Post therapy whole-body coronal. b-800 (left panel) and PET (right panel) images show resolution of subpectoral, mediastinal and lung abnormalities and a residual splenic mass with mildly increased FDG uptake (arrow). The b-50 DW-MR image had a similar appearance.

Figure 5

53-year-old man with stage IV DLBCL and a false positive right axillary lymph node. (A) Pretreatment DW-MRI (b value 50 sec/mm²⁾ demonstrates diffusion restriction in right axillary lymph nodes. (B) Corresponding pretreatment PET/CT shows increased FDG uptake. (C) Following two cycles of R-CHOP chemotherapy, there is residual diffusion signal in a right axillary lymph node. (D) This node did not demonstrate increased activity on PET/CT.

Figure 6

Average pre- and posttreatment ADC values. To better illustrate the distribution of the data points; they are spread horizontally to minimize overlapping. The ends of the boxes are the 25th and 75th quartiles (quartiles). The lines across the middles of the boxes are the medians. The interquartile range is the difference between the quartiles. The lines (whiskers) extend from the boxes to the outermost points that fall within the distance computed as 1.5 (interquartile range).