Clinical Application Of Diffusion-weighted Whole-body Imaging With Background Body Signal Suppression(DWIBS)in Lymphoma

Part 1 Comparison of DWIBS and 18F-fluoro-2-deoxyglucose positronemission tomography/computed tomography(18F-FDG PET/CT)for initial staging of lymphoma patientsObjective1.To evaluate the clinical value of diffusion-weighted whole-body imaging with background body signal suppression(DWIBS)by comparing the accuracy of DWIBS and 18F-fluoro-2-deoxyglucose positron emission tomography/computed tomography(18F-FDG PET/CT)in the staging of newly diagnosed lymphoma patients.2.To compare the coincidence of DWIBS staging and clinical staging,PET/CT staging and clinical staging.MethodsNewly diagnosed lymphoma patients prospectively underwent DWIBS,conventional coronal magnetic resonance imaging(T2WI-STIR)on a Siemens 3.0T magnetic resonance(MR)scanner and 18F-FDG PET/CT on a Discovery STE16 PET/CT.The 14 nodal regions(left and right cervical,left and right axillary,left and right infraclavicular mediastinal,hilar,periaortic,messentary,left and right pelvic and left and right inguinal)and the 12 extranodal regions((Waldeyer ring,lungs,liver,spleen,stomach,Small intestine,colon,right kidney,left kidney,bone marrow,soft tissues(skin/fat/muscle),and other organs/tissues(e.g.,thyroid,pancreas,breast,et.))were evaluated.When one or more lesions were found in a nodal region,the nodal region was recorded as positive and was counted as one region for statistical analysis(ie,we would count the lymph node regions rather than add up the number of lymph nodes).The staging of DWIBS and PET/CT was performed according to the Ann Arbor staging system.The staging of lymphoma diagnosed by surgical pathology was based on the distribution of postoperative lesions.Clinical staging was based on histopathology,imaging and follow-up results performed by clinicians,according to Ann Arbor stage.All statistical analyses were performed using Statistical Package for Social Science(SPSS 22.0)software.Kappa coefficients were calculated to estimate agreement between the two imaging techniques.Agreement was considered excellent at a k value of 0.81?1;good at 0.61?0.80;moderate at 0.41?0.60;fair at 0.21?0.40;poor at<0.20.The PET/CT was considered as the reference standard when assessing the sensitivity and specificity of DWIBS for detecting lesions.ResultsFrom August 2017 to December 2018,A total of 54 newly diagnosed patients(male:25;female:29)were enrolled.The median age was 53(13-81)years.Thirty-two patients were diagnosed with diffuse large B-cell lymphoma,1 patient with Burkitt's lymphoma,1 patient with mantle cell lymphoma,6 patients with follicular lymphoma,1 patient with marginal zone lymphoma,3 patients with peripheral T-cell lymphoma,3 patients with T-lymphoblastic lymphoma,1 patient with NK/T cell lymphoma and 6 patients with Hodgkin's lymphoma.Clinical stage 0 was observed in 2 patients,stage ? in 3 patient,stage ? in 15 patients,stage ? in 10 patients,and stage IV in 24 patients.Two hundred and seventy-three lesions(222 nodal lesions,51 extranodal lesions)and 271 lesions(207 nodal lesions,64 extranodal lesions)were detected by DWIBS and 18F-FDG PET/CT,respectively.Two hundred and twenty-seven lesions(188 nodal lesions,39 extranodal lesions)were positive in both DWIBS and 18F-FDG PET/CT.The agreement of detecting nodal lesions between the two imaging methods was excellent(k=0.828,p<0.001).The sensitivity and specificity of DWIBS detecting nodal lesions were 90.8%and 93.8%,respectively.The agreement of detecting extra-nodal lesions between the two imaging methods was good(k=0.674,p<0.001).The sensitivity and specificity of DWIBS detecting extranodal lesions were 60.9%and 100%,respectively.The agreement of detecting overall lesions(nodal lesions and extra-nodal lesions)between the two imaging methods was good(k=0.795,p<0.001).The sensitivity and speciificity of DWIBS detecting overall lesions were 83.8%and 95.9%,respectively.The two methods were concordant in the staging in 45(83.3%)patients(k=0.755,p<0.001).DWIBS staging and clinical staging were concordant in 45(83.3%)patients(?=0.757,p<0.001).18F-FDG PET/CT staging and clinical staging were concordant in 47(87.0%)patients(?=0.809,p<0.001).ConclusionsDWIBS in the detection of nodal lesions was superior to the detection of extranodal lesions.our results showed good agreement between DWIBS and PET/CT in staging of newly diagnosed lymphoma patients.Although the concordance rate between DWIBS staging and clinical staging was slightly lower than that of between PET/CT staging and clinical staging,DWIBS has the advantages of safety and efficiency,affordability and utilizing nonionization radiation.Therefore,the value of DWIBS in lymphoma is promising.Part ? Correlation between DWIBS ADCmin and prognosis indexes in patients with lymphomaObjectiveTo investigate the correlation between minimum apparent diffusion coefficient(ADCmin)and maximum standard uptake value(SUVmax),proliferation index(Ki-67)expression,international prognostic index(IPI)score,clinical staging,serum lactate dehydrogenase(LDH)and p2-microglobulin(p2-MG)level.MethodsThe study objects were the same as part 1.Because no lesions were detected in both DWIBS and PET/CT,two patients with stage 0 clinical stage were excluded from the analysis.For each patient,the ADCmin of the lesion was measured on the ADC map and SUVniax was measured in the lesion with high FDG uptake.The mean value of ADCmin for all measurable lesions per patient was indicated by ADC-min.Pathological specimens were immunohistochemically stained to obtain Ki-67 positive rates.Serum LDH and ?2-MG level results were collected.Clinical staging standards were the same as part 1.The IPI score of each patient at the time of initial diagnosis was obtained according to the IPI scoring criteria.All statistical analyses were performed using SPSS 22.0 software.To analyze whether there are statistical differences in ADCmin between different clinical stages and IPI scores.If the data was belong to normal distribution and homogeneity of variance,One-Way ANOVA would be performed,or else,K Independent Samples Test(Kruskal-Wallis Test)would be used.Spearman correlation statistic was used to analyze the association between ADCmin/ADC min and SUVmax,Ki-67,LDH,?2-MG.p<0.05 was taken to indicate statistical significance.ResultsA total of 52 newly diagnosed patients were analyzed.One hundred and ninety lesions simultaneously obtained ADCmin and SUVmax.Based on these 190 lesions,ADCmin and SJVmax were inversely correlated(r=-0.222,p=0.002).The ADCmin of lymphoma patients with ?-? stages were 691.57±184.62x 10-6mm2/s,698.92±171.56×10-6mm2/s,670.3±140.86x 10-6nmm2/s,693.68±146.48×10-6mm2/s,respectively.There is no statistical difference between different stages by One-Way ANOVA.The ADCmin of low risk group,low-intermediate risk group,high-intermediate risk group,high risk group were 707.04±156.66x 10-6mm2/s,704.48±187.09x10-6mm2/s,615.05±89.06×10-6mm2/s,722.73±84.7×10-6mm2/s,respectively.No signifieant differences were found between different IPI scores.There were no correlations between ADCmin and IPI,clinical staging,LDH,?2-MG.Among 42 patients with biopsy,a significant inversely correlation was observed between ADCmin and Ki-67 of the biopsy lesions(r=-0.401 p=0.009).ConclusionsIn lymphoma,an inversely correlation between ADCmin and SUVmax means that the glucose metabolism of tumor tissue is inversely correlated with the diffusion of water molecules of that.An inversely correlation between ADCmin and the proliferation index Ki-67 means that the lower the ADC value of lymphoma lesions,the higher the malignant degree of lymphoma.These results showed that DWIBS combined with ADC values can non-invasively evaluate the malignant degree of lymphoma,and has great prospects in evaluating lymphoma prognosis.