Baseline and Increases in Serum B-Cell Maturation Antigen Levels Rapidly Indicate Changes in Clinical Status Among Relapsed/Refractory Multiple Myeloma Patients Undergoing New Treatments

Introduction: B-cell maturation antigen (BCMA) is a protein that is expressed on malignant plasma cells from patients (pts) with multiple myeloma (MM). Our group has previously shown that MM pts have higher levels of serum (s) BCMA than healthy subjects and that sBCMA levels can be used to monitor the disease course of MM pts. Notably, we have shown that the half-life of soluble BCMA is only 24-36 hours which is much shorter than sM-protein, and its levels are independent of renal function. With the expanding therapeutic options for the treatment (Tx) of MM, there is a need for more rapid and accurate ways to assess the efficacy of new therapies. In this study, we analyzed the relationship between progression free survival (PFS) and changes in weekly levels of sBCMA, sM-protein and serum free light chain (sFLC) during the first cycle of a new treatment among relapsed/refractory (RR)MM patients.

Methods: Serum was obtained weekly during each pt's first cycle of a new therapy (C1) and the first day of their second cycle (C2D1) from all RRMM pts (n = 122) at a single clinic from August 2016 to December 2018. sM-protein and sFLC levels were measured, and sBCMA levels were determined using an enzyme-linked immunosorbent assay (R&D Systems; Minneapolis, MN). Percentage changes in sBCMA, sM-protein and sFLC (difference between involved and uninvolved FLCs) throughout Tx were determined relative to levels measured at the start of treatment (C1D1). Kaplan-Meier analysis was used to assess differences in PFS based on the percentage changes from baseline in these levels at cycle 1 day 8 (C1D8), day 15 (C1D15), day 22 (C1D22), and day 29 (C2D1). All pt samples were obtained following proper informed consent in accordance with the Declaration of Helsinki.

Results: The analysis involved 122 Tx in 75 RRMM pts undergoing new treatment (IgG [n = 33], IgA [n= 15], IgM [n = 1], and light chain only [n = 26]). All pts were evaluable by sBCMA (above the lower limit of detection of the assay [16 ng/mL]). Pts whose baseline sBCMA was higher than the median (217.6 ng/mL) had a significantly shorter PFS (n = 61, median PFS = 2.5 mo) than those below the median (n = 61, median = 7.3 mo, p = 0.0003). When baseline sBCMA levels were quartiled, there was an inverse relationship between sBCMA and PFS (Q1: n = 30, median = 8.0 mo; Q2: n = 31, median = 7.2 mo; Q3: n = 31, median = 3.2 mo; Q4: n = 30, median = 1.9 mo; p = 0.0002). Pts whose sBCMA increased ≥ 25% on C1D8 (n = 8) had a shorter PFS (median = 1.6 mo) than those that did not (n = 114, median = 3.9 mo, p = 0.0042). Those whose sBCMA increased ≥ 25% at any time during C1 (n = 31) also had a shorter PFS (median = 1.6 mo) than those that did not (n = 91, median = 5.2 mo, p < 0.0001).

sM-protein was only evaluable by IMWG criteria (> 1.0 g/dL) in 45 (37%) of the pts and only 2 and 5 pts showed an increase of > 25% on C1D8 and anytime during cycle 1, respectively. In the remaining 77 pts who were not evaluable by sM-protein, 51 (68%) were evaluable by sFLC using IMWG criteria (> 100 mg/L difference between the involved and uninvolved FLC) and only 13 showed a > 25% increase during the first cycle.

Next, we determined if changes in sBCMA could be used for pts who were not evaluable by either sM-protein or sFLC. Among pts that could not be followed by sM-protein, those with a ≥ 25% increase in sBCMA at any point in C1 (n = 18, median = 2.2 mo) had a significantly shorter PFS than those who did not (n = 59; median = 4.4 mo; p = 0.0218). Among pts that could not be followed by sFLC, those with a ≥ 25% increase in sBCMA at any point in C1 (n = 12) had a significantly shorter PFS (median = 1.6 mo) than those who did not (n = 30; median PFS = 8.7 mo; p = 0.0072). Among pts that could not be followed by either sFLC or sM-protein, a > 25% in sBCMA at any point in C1 (n = 8) showed a shorter PFS (median = 1.6 mo) than those who did not (n = 17; median = 19.3 mo; p = 0.0154). Notably, among pts whose sFLC or sM-protein were below evaluable levels, ≥ 25% increases in sFLC or sM-protein did not predict PFS.

Conclusions: We have shown that serum BCMA was evaluable in all RRMM pts at the start of new therapy in contrast to the standard sM-protein and sFLC biomarkers. Baseline levels of sBCMA predict PFS in RRMM pts undergoing new therapy. Additionally, increases of sBCMA > 25% during the first cycle occur in more pts than sM-protein or sFLC, and predict for a shorter PFS. These results suggest that baseline sBCMA and monitoring its levels weekly during the first cycle of a new treatment can help improve predicting outcomes for RRMM pts.