Diffusion-Weighted Magnetic Resonance Imaging (DW-MRI) Is Able to Distinguish Diffuse from Normal MRI Pattern of Marrow Involvement in Patients with Multiple Myeloma

Evangelos Terpos, MD
Department of Clinical Therapeutics
National and Kapodistrian University of Athens
School of Medicine
Athens, Greece

Program: Oral and Poster Abstracts
Session: 653. Myeloma: Therapy, excluding Transplantation: Poster II

Evangelos Terpos1, Vassilis Koutoulidis2*, Sofia Fontara2*, Flora Zagouri1*, Dimitris Matsaridis2*, Dimitrios Christoulas1*, Andreas Koureas2*, Efstathios Kastritis, MD1*, Meletios A. Dimopoulos1 and Lia A. Moulopoulos2*

1Department of Clinical Therapeutics, National and Kapodistrian University of Athens, School of Medicine, Athens, Greece
2First Department of Radiology, National and Kapodistrian University of Athens, School of Medicine, "Aretaieion" Hospital, Athens, Greece

Diffuse MRI pattern of marrow involvement correlates with poor prognosis in patients with multiple myeloma (MM) both in the conventional chemotherapy and novel agent era. Sometimes the differentiation of diffuse from normal MRI marrow involvement in MM patients is difficult, particularly in cases of anemia and of low plasma cell marrow infiltration. Diffusion-Weighted Imaging (DWI) is an MRI technique which is based on differences in the diffusivity of water molecules in the respective tissue. Quantitative analysis of DWI can be achieved by calculating the Apparent Diffusion Coefficient (ADC) values from images with two or more different diffusion weightings. There are very limited data in the literature regarding ADC values in MM patients and almost no data for the ADC values in patients with different MRI patterns of marrow involvement. Thus our objective was to calculate and compare ADC values in newly diagnosed patients with MM, to identify possible differences among diffuse, focal, and normal MM MRI patterns and to establish, if possible, a threshold value which might distinguish diffuse from normal MRI patterns.

We evaluated 44 patients (18M/26F, median age 66 years; range: 37-89 years) with newly diagnosed, untreated, MM patients and 16 healthy, gender and age-matched controls, with MRI of the lumbosacral spine, using a 1.5 Tesla unit. Conventional MR images were obtained according to bone marrow MRI protocols; MRI patterns of involvement were normal in 22, focal in 11 and diffuse in 11 patients. DWI was obtained with an Echo Planar Imaging sequence using 5 b-values (0, 150, 250, 500, 750 sec/mm2). In MM patients with a normal or a diffuse pattern, as well as in healthy controls, region of interest (ROI) measurements were obtained from each of the five lumbar vertebral bodies avoiding the region of the basivertebral vessels and any focal non-myelomatous lesion and the mean ADC value was calculated. In MM patients with a focal pattern, ROIs were placed on several focal lesions and the highest ADC value was recorded; in these patients an additional ROI was placed on an uninvolved vertebra in order to record the ADC value of apparently normal marrow.

Mean ADC values (±SD, x10-3 mm2/sec) were: 0.346±0.107 (range: 0.186-0.544) for the normal MM pattern, 1.188±0.359 (range: 0.715-2.015) for the focal MM pattern and 0.821±0.149 (range: 0.642-1.017) for the diffuse MM pattern [Figure]. Mean ADC value of apparently normal marrow in patients with a focal pattern was 0.403±0.142 (range: 0.113-0.589), while that of healthy controls was 0.325±0.135 (range: 0.152-0.542). One way analysis of variance (ANOVA) showed a significant difference in ADC values within the groups of the study (p<0.0001). Further analysis using t-test revealed significant differences of ADC values between diffuse and focal MM pattern (p=0.003), diffuse and normal MM pattern (p<0.0001), focal and normal MM pattern (p<0.0001), diffuse MM pattern and healthy volunteers (p<0.0001), focal MM pattern and healthy volunteers (p<0.0001), diffuse MM pattern and apparently normal marrow in focal MM pattern (p<0.0001) and finally between focal MM pattern and apparently normal marrow in focal MM pattern (p<0.0001) [Figure]. No significant differences were found between normal MM pattern and healthy volunteers (p=0.442), normal MM pattern and apparently normal marrow in focal MM pattern (p= 0.141), and between apparently normal marrow in focal MM pattern and healthy volunteers (p= 0.175). The 95% confidence intervals of the ADC values for each group were calculated and a receiver operating characteristic (ROC) analysis was performed to determine the cut-off value with the highest accuracy to distinguish a diffuse myeloma pattern from normal patterns (normal MM pattern, apparently normal marrow in focal MM pattern, normal marrow in healthy controls). An ADC value above 0.593x10-3 mm2/sec was found to be diagnostic of diffuse myelomatous infiltration of the bone marrow with a sensitivity and specificity of 100%.

We conclude that normal, focal and diffuse MRI patterns of marrow involvement in patients with MM have distinct ranges of ADC values on DWI. An ADC value above 0.593×10-3 mm2/sec is diagnostic of diffuse myelomatous infiltration of the bone marrow with extremely high accuracy and can be used in cases where diffuse pattern cannot be differentiated from normal pattern in conventional MRI.

Download PDF


Evangelos Terpos, MD, PhD, is an Associate Professor of Hematology, in the Department of Clinical Therapeutics at the University of Athens, School of Medicine. Dr. Terpos is co-chair of the Bone Subgroup of the International Myeloma Working Group and the Guideline Subgroup of the European Myeloma Network. He is a reviewer of scientific papers in more than 50 medical journals and has reviewed abstracts for ASH, EHA & EBMT meetings. He is a member of the editorial board of Haematologica. Visit Dr. Evengelos Terpos’ full biography.

Previous Post
Dr. Evangelos Terpos on VR consolidation post-ASCT (ASH 2014)
Next Post
Dr. Meletios Dimopoulos on the MM-010 Trial (ASH 2014)

Give Where Most Needed

We use cookies on our website to support technical features that enhance your user experience.

We also use analytics & advertising services. To opt-out click for more information.