Apparent diffusion coefficient measurements of bone marrow infiltration patterns in multiple myeloma for the assessment of tumor burden - a feasibility study.

Background: The purpose of our study was to explore and compare the tumor burden of different bone marrow infiltration patterns and evaluate the feasibility of apparent diffusion coefficient (ADC) value to identify patterns in multiple myeloma (MM).

Patients and methods: Ninety-three patients with newly diagnosed multiple myeloma and 23 controls had undergone routine magnetic resonance imaging (MRI) and diffusion-weighted MRI (DWI) from January 2019 to November 2020. Five bone marrow (BM) infiltration patterns were allocated according to routine MRI. The laboratory data and ADC values of patterns were analyzed and compared. ROC analysis was used to establish the best diagnostic ADC threshold value for identifying these patterns and distinguishing normal pattern from controls. Besides, the correlation between the ADC values of diffuse pattern and the plasma cells ratio was assessed.

Results: The values of hemoglobin, beta-2 microglobulin (β2-MG), plasma cell, M protein, the percentages of stage, high-risk fluorescence in situ hybridization, and ADC values showed significant difference among patterns. ADC_mean at a specific value (368.5×10^-6 mm²/s) yielded a maximum specificity (95.5%) and sensitivity (92.0%) in diagnosing MM. A specific value (335.5×10^-6mm²/s) yielded a maximum specificity (84.7%) and sensitivity (88.0%) in discriminating visually normal pattern in MM from controls. There was a moderate positive correlation between the plasma cells ratio and ADCs of diffuse infiltration patterns (r = 0.648, P < 0.001).

Conclusions: The bone marrow infiltration patterns in MM patients can indicate the tumor burden and ADC value has the ability to discriminate these patterns objectively.