Multiple myeloma (MM) is a neoplastic plasma cell disorder that accounts for approximately 10% of hematologic malignancies [1]. The introduction of various agents and autologous stem-cell transplantation (ASCT) has improved survival; however, MM is still considered an incurable disease [2,3]. Moreover, patients newly diagnosed with MM might show a heterogeneous prognosis, with survival durations ranging from a few months to more than 10 years [4,5]. Previous studies have attempted to establish a reliable prognostic system based on various factors such as staging and disease-related biology. The International Staging System (ISS) and revised ISS (R-ISS) are representative clinically useful prognostic scoring systems based on serum albumin, beta-2 microglobulin (B2M), and serum lactate dehydrogenase (LDH) levels, as well as high-risk chromosomal abnormalities detected by fluorescence in situ hybridization (FISH) [6,7]. However, these prognostic models still have some pitfalls in that there is no interlaboratory standardization of FISH analysis and cutoff levels for LDH. In addition, a relatively short median follow-up period and exclusion of host-related factors are possible limitations [7].

Magnetic resonance imaging (MRI) is highly sensitive for the detection of focal bone or bone marrow lesions, including those that are not osteolytic. Therefore, the revised International Myeloma Working Group (IMWG) diagnostic criteria for MM include MRI [8]. MRI could predict disease progression in which patients with MM with more than one focal bone lesion had a higher risk of progression to end-organ damage [9-11]. Several studies have classified MRI findings and attempted to identify their clinical significance as a predictive or prognostic modality in patients newly diagnosed with MM [12,13]. Song et al. demonstrated that spine MRI at the time of diagnosis was useful for prognosis in a study of 113 patients [14]. However, the role of MRI in predicting the treatment response and deciding whether to proceed with ASCT remains controversial.

Therefore, this study aimed to identify the novel prognostic role of spine MRI in patients newly diagnosed with MM. In addition, we analyzed the response to standard treatment and ASCT according to the specific patterns of spine MRI.

Osteolytic bone lesions are a hallmark of MM and key factors in the revised IMWG diagnostic criteria [8]. Up to 80% of patients present with bone lesions at the time of diagnosis, and patients with osteolytic bone lesions have an increased risk of skeleton-related events associated with high morbidity and mortality [17]. Myeloma-related bone disease occurs due to an unbalanced bone-remodeling process in which the interaction between myeloma cells and the bone microenvironment leads to the activation of osteoclasts and suppression of osteoblasts, resulting in bone loss [18]. Receptor activator of nuclear factor kappa-Β (RANK)/RANK ligand/osteoprotegerin, Notch, Wnt, and numerous chemokines and interleukins are implicated in these complex intracellular or intercellular signaling cascades [19,20]. However, little research has been conducted on the association between specific MRI patterns in patients with MM and clinical outcomes. In this study, we classified the spine MRI patterns of patients newly diagnosed with MM into five categories (normal, focal, diffuse, combined, and “salt-and-pepper” patterns) based on previous studies [9,21]. Some previous studies have already suggested that patients with focal and diffuse MRI patterns have inferior clinical outcomes compared with those with normal or “salt-and-pepper” patterns [13,22]. In the current study, the focal, diffuse, and combined patterns were significantly associated with older age, advanced disease status, and higher rates of relapse and death. Patients with normal and “salt-and-pepper” MRI patterns were younger, and the majority had a lower disease status in ISS and R-ISS staging.

Generally, X-rays are used to detect bone lesions in patients with newly diagnosed MM. However, plain radiographs have several limitations such as poor quality visualization, low sensitivity, and observer-dependent analysis [23]. Computed tomography (CT) is one of the most commonly used tools for evaluating bone lesions and disease status, with high accuracy. Positron emission tomography and CT (PET-CT) scans are also crucial for the detection of bone lesions in the diagnosis of MM [24]. However, the superiority of clinical use of each modality remains controversial. MRI is regarded as the most sensitive tool for identifying the detailed bone marrow infiltration status [25]. Moreover, Baur-Melnyk et al. [26] demonstrated that whole-body multidetector CT showed a significantly lower detection rate and staging than MRI in patients with MM. In addition, the disease status of MM is mainly based on scoring staging systems such as ISS and R-ISS. The presence of the cytogenetic abnormalities t(4;14), t(14;16), del17p, and add1q21 are known to be associated with poor outcomes [13,22]. Several studies have also suggested an association between cytogenetic abnormalities and MRI patterns of marrow infiltration. Moulopoulos et al. [27] reported that diffused MRI patterns showed a higher incidence of high-risk cytogenetic features, including del17p, add1q21, and del13q, than focal or normal MRI patterns. Moreover, patients with diffuse MRI patterns have adverse myeloma features and increased microvessel density in their trephine biopsies [12]. These results are in accordance with the findings of Hillengass et al. [28], where the presence of 1q21, del17p, and del13q significantly correlated with at least one abnormal finding in bone marrow dynamic contrast-enhanced MRI. In addition, these chromosomal abnormalities can trigger an angiogenic cascade in MM.

High-dose chemotherapy with hematopoietic stem-cell rescue remains the standard of care for transplant-eligible MM patients [29,30]. ASCT is associated with significantly improved PFS and OS [30,31]. However, some cases of recurrence of MM after ASCT have been found in young and low-risk patients in current clinical practice. In contrast, some older patients with high ISS and R-ISS stages who had not undergone ASCT were cured without relapse. Considering transplantation-related mortalities and emerging effective novel agents, performing ASCT in all transplant-eligible patients may not be appropriate. However, the eligibility for ASCT was evaluated through a risk-benefit assessment, including age, comorbidities, and general condition. There are no impactive international guidelines for assessing the progress of ASCT. Imaging can visually provide identifiable information in patients with malignant disease, such as interim PET-CT in malignant lymphoma, and this can change the paradigm of the treatment strategy. Therefore, further research on spine MRI in MM patients should be performed, including interim MRI, regular follow-up MRI, and MRI at the time of diagnosis, to identify the role of MRI as a key factor for establishing a treatment strategy, such as ASCT.

While the present study showed promising results of spine MRI to predict outcomes in newly diagnosed patients, our data should be interpreted cautiously due to certain limitations. First, spine MRI was mainly evaluated rather than whole-body imaging in this study. Second, we could not compare other imaging modalities to MRI in the current analysis. Finally, this study included a diverse patient population and was retrospectively analyzed.

In conclusion, the focal, diffuse, and combined patterns of patients newly diagnosed with MM were significantly associated with older age, advanced disease status, and higher rates of relapse and death, whereas patients with normal and “salt-and-pepper” patterns showed relatively better clinical outcomes. Although data on follow-up MRI and its clinical significance have not been evaluated, spine MRI can play a role in assessing treatment responses. Performing follow-up spine MRI may help clinicians clarify treatment responses and reestablish appropriate treatment plans.