Magnetic Resonance Imaging in Multiple Myeloma: Diagnostic and Clinical Implications

Abstract

Magnetic resonance imaging (MRI) permits the detection of diffuse and focal bone marrow infiltration in the absence of osteopenia or focal osteolysis on standard metastatic bone surveys (MBSs). Both baseline MBS and MRI were available in 611 of 668 myeloma patients who were treated uniformly with a tandem autologous transplantation-based protocol and were evaluated to determine their respective merits for disease staging, response assessment, and outcome prediction. MRI detected focal lesions (FLs) in 74% and MBS in 56% of imaged anatomic sites; 52% of 267 patients with normal MBS results and 20% of 160 with normal MRI results had FL on MRI and MBS, respectively. MRI- but not MBS-defined FL independently affected survival. Cytogenetic abnormalities (CAs) and more than seven FLs on MRI (MRI-FLs) distinguished three risk groups: 5-year survival was 76% in the absence of both more than seven MRI-FLs and CA (n = 276), 61% in the presence of one MRI-FL (n = 262), and 37% in the presence of both unfavorable parameters (n = 67). MRI-FL correlated with low albumin and elevated levels of C-reactive protein, lactate dehydrogenase, and creatinine, but did not correlate with age, beta-2-microglobulin, and CA. Resolution of MRI-FL, occurring in 60% of cases and not seen with MBS-defined FL, conferred superior survival. MRI is a more powerful tool for detection of FLs than is MBS. MRI-FL number had independent prognostic implications; additionally, MRI-FL resolution identified a subgroup with superior survival. We therefore recommend that, in addition to MBS, MRI be used routinely for staging, prognosis, and response assessment in myeloma.