Soluble B-Cell Maturation Antigen Levels for Disease Monitoring in Oligo- and Non-Secretory Relapsed Multiple Myeloma

Introduction: With the development of novel therapies that elicit strong and durable responses, the incidence of oligo- or non-secretory (O-S/Non-S) relapsed multiple myeloma (MM) is growing. The lack of reliable biomarkers as alternatives to paraproteins makes it challenging to determine an International Myeloma Working Group (IMWG)-based treatment response, which results in overlooked disease progression. B-cell maturation antigen (BCMA) is selectively overexpressed on the surface of myeloma cells and is shed into the circulation. Soluble BCMA (sBCMA) is an emerging candidate disease-monitoring marker independent of M-proteins for O-S/Non-S subgroups. However, most previous sBCMA studies have focused on its prognostic relevance in secretory diseases, particularly after BCMA-directed therapies, with only a few case series reporting its diagnostic value in the context of O-S/Non-S MM. We investigated whether sBCMA levels correlated with other myeloma tumor volume indicators and its utility in monitoring O-S/Non-S MM. Methods: According to the IMWG criteria, O-S MM was defined as the development of new medullary or extramedullary disease, without measurable disease, while Non-S MM was defined by complete M-protein absence. Serum samples serially collected from MM patients were stored at -40°C until analysis. sBCMA levels were measured using an enzyme-linked immunosorbent assay (Duoset®️ ELISA catalog number: DY193; R&D Systems, Minneapolis, MN, USA) as previously described (Sanchez et al. Br J Haematol 2012). Digital analysis using whole-slide imaging (WSI) was employed to quantify the proportions of bone marrow CD138+ plasma cells (BMPC) clot sections more objectively, using a high-resolution Ultra-Fast Scanner (Philips, Best, The Netherlands) and QuPath software (version 0.5.1; University of Edinburgh, Edinburgh, Scotland). Furthermore, the estimated total tumor volume was semi-automatically quantified as the total diffusion volume (tDV) using whole-body diffusion-weighted magnetic resonance imaging (WB-DW-MRI) (Terao et al. Eur Radiol 2021). The number of circulating tumor cells (CTCs) was assessed using multicolor flow cytometry, with a detection limit of 0.01%. Results: Seventy patients with newly diagnosed MM were retrospectively analyzed. No patient had a history of BCMA-targeted therapy. The median sBCMA levels at diagnosis were 10-fold higher in MM patients (469.8 ng/mL, interquartile range [IQR]: 219.8-1266.9) than in healthy controls (n = 50, 41.4 ng/mL, IQR: 37-46.2; P < 0.001). Of note, sBCMA levels correlated strongly with BMPC percentages (r = 0.71; P < 0.001). Moreover, for tumor volume-related parameters in the imaging analysis, tDV (n = 55) was also moderately correlated with sBCMA levels (r = 0.59; P < 0.001) Another factor reflecting tumor burden was CTC level (n = 68), which tended to be higher with high sBCMA levels (median: 275.2 ng/mL in those with < 0.1%; 771.7 ng/mL in those with 0.1-1.0%; 1705.6 ng/mL in those with > 1.0%). Multiple regression analysis confirmed that these three factors were all significant determinants of sBCMA levels, without significant multicollinearity (adjusted R2 = 0.678, P < 0.001). Based on these findings, we further investigated the usefulness of sBCMA levels for disease monitoring in 17 patients with O-S/Non-S relapsed MM. Remarkably, these patients showed the sBCMA dynamics paralleled with the disease status, with the median of 348.4 ng/mL at diagnosis, declined to 25.4 ng/mL at best response and elevated to 66.6 ng/mL at relapse timing (all comparison P <0.05). Furthermore, even prior to the O-S/Non-S relapse events (6 months before [n = 15], or at the most recent treatment change [n = 2]), sBCMA levels were significantly elevated than those at the best response (median: 42.7 ng/mL vs. 25.4 ng/mL, P = 0.002). The similar pattern of sBCMA change was replicated in secretory relapsed MM (n=25). Finally, sBCMA concentrations at the time of O-SR/Non-SR correlated strongly with the corresponding tDV values (r = 0.81; P < 0.001), while a moderate correlation with %BMPC were also observed (r = 0.67; P < 0.001). Conclusions: In conclusion, we assessed myeloma tumor volume more accurately by comparing sBCMA values with %BMPC using a new method with QuPath software on WSI, correlated it with tDV by WB-DW-MRI, and demonstrated its usefulness in monitoring O-S/Non-S MM patients.