Quantitative Analysis of Whole‐Body MRI for Accessing the Degree of Diffuse Infiltration Patterns and Identifying High Risk Cases of Newly Diagnosed Multiple Myeloma

Abstract

BackgroundAccurate identification of high‐risk multiple myeloma (HRMM) is important for prognostication. The degree of diffuse infiltration patterns on magnetic resonance imaging (MRI) is associated with patient prognosis in multiple myeloma. However, objective indexes to determine the degree of diffuse infiltration patterns are unavailable.PurposeTo investigate whether qualitative and quantitative evaluations of diffuse infiltration patterns on MRI could identify HRMM.Study TypeRetrospective.SubjectsTotally, 180 patients (79 HRMM and 101 standard‐risk MM) were assessed. The presence of del(17p), t(4;14), t(14;16), t(14;20), gain 1q, and/or p53 mutations was considered to indicate HRMM.Field Strength/Sequence3.0 T/diffusion‐weighted whole‐body imaging with background body signal suppression (DWIBS), modified Dixon chemical‐shift imaging Quant (mDIXON Quant), and short TI inversion recovery (STIR).AssessmentQualitative analysis involved assessing the degree of diffuse marrow infiltration (mild, moderate, or severe), and quantitative analysis involved evaluating apparent diffusion coefficient (ADC), fat fraction (FF), and T2* values. Clinical data such as sex, age, hemoglobin, serum albumin, serum calcium, serum creatinine, serum lactate dehydrogenase, β2‐microglobulin, and bone marrow plasma cells (BMPCs) were also included.Statistical TestsUnivariate and multivariate analyses, receiver operating characteristic (ROC) curve. P < 0.05 was considered statistically significant.ResultsThe high‐risk group had significantly higher ADC and T2* and lower FF compared with the standard‐risk group. Multivariate analysis indicated BMPCs as a significant independent risk factor for HRMM (odds ratio (OR) = 1.019, 95% CI 1.004–1.033), while FF was a significant independent protective factor associated with HRMM (OR = 0.972, 95% CI 0.946–0.999). The combination of BMPCs and FF achieved the highest areas under the curve (AUC) of 0.732, with sensitivity and specificity of 70.9% and 68.3%, respectively.Data ConclusionCompared with qualitative analysis, FF value was independently associated with HRMM. The quantitative features of diffuse marrow infiltration on MRI scans are more effective in detecting HRMM.Level of Evidence3Technical EfficacyStage 2