Mass spectrometry compared to serum electrophoresis and immunofixation to detect monoclonal proteins and therapeutic antibodies in plasma cell disorders

Abstract

Abstract Background Serum protein electrophoresis (SPEP) and immunofixation (IFE) are the standard methods for detecting and monitoring monoclonal proteins (M-proteins) in patients with plasma cell disorders. Newer methods with higher sensitivity such as mass spectrometry (MS) have been developed and tested particularly in the setting of minimal residual disease monitoring in multiple myeloma. Here, we aimed to test MS in routine samples from patients with plasma cell disorders of various disease stages and treatment regimens to assess MS test performance compared to standard methods. Methods We analyzed 594 serum samples from individuals who underwent standard-of-care testing with SPEP and IFE over an eight-week period between October and December 2024 at Memorial Sloan Kettering Cancer Center. Residual serum from the same blood draw was used for MS analysis, ensuring that SPEP, IFE, and MS were all performed on the same sample. The matrix-assisted laser desorption/ionization time-of-flight (MALDI-ToF) mass spectrometry EXENT® System (The Binding Site, part of Thermo Fisher Scientific, Birmingham, UK) was used for the detection, isotyping, and quantitative measurement of monoclonal immunoglobulins and light chains. In addition, samples from patients who were on therapeutic monoclonal antibody treatment were analyzed to identify peaks corresponding to the expected mass-to-charge ratio (m/z) of commonly administered therapeutic agents. Results The 594 samples originated from 559 patients, 291 men and 268 women, median age was 68 years. There were 398 Caucasian white patients, 84 African American, 24 Asian, 54 of other or unknown background. At the time of initial diagnosis/work up of the plasma cell disorder, 342 patients had an IgG M-protein, 97 patients had IgA, 16 patients had IgM, 54 had kappa restricted free light chain (FLC) and 39 had lambda restricted FLC disease, 3 had non-secretory multiple myeloma, and 8 individuals had no detectable monoclonal gammopathy. In the collected samples, the median M-protein level was 0.5 g/dL using SPEP. The median kappa FLC level 1.75 mg/dL, lambda FLC level 0.64 mg/dL. Median creatinine level was 0.85 mg/dL. An M-protein was detectable via SPEP in 288 samples, via IFE in 353 samples, and via MS in 424 samples including detectable FLC peaks. In patients who had a detectable M-protein with both SPEP and MS, the correlation between the M-protein level was 0.96. There were 305 samples with no detectable M-protein on SPEP, of these 80 had positive IFE of which 13 were FLC bands (missing=15). In the same 305 SPEP negative samples, 96 were found to have a detectable M-spike on MS. The level of the M-protein in the SPEP negative, MS positive samples were low; two had MS M-protein levels between 0.2-0.6 g/dL, 10 samples had MS M-protein levels 0.1-0.2 g/dL, and the remaining 84 samples had MS M-protein levels <0.1 g/dL. Additionally, 42 of the of the 305 SPEP negative samples had a detectable FLC peak on MS, 37 with the same light chain restriction as baseline. Sixteen samples demonstrated more than one M-protein on SPEP. Two with biclonal disease and two cases where the second IgG M-protein was flagged as a therapeutic monoclonal antibody. The remaining 12 samples with two M-protein bands on SPEP were of IgA subtype and in 100% of these cases, a single IgA peak was observed by MS, likely reflecting dimerization as the cause of two IgA bands by SPEP. Additionally, in three samples from patients with non-secretory disease, a low-level IgG lambda peak was detected by MS in one patient. Furthermore, 221 samples (37%) showed an M-protein with a m/z ratio consistent with a therapeutic monoclonal antibody. As an example, among the 212 samples from patients treated with daratumumab, 182 had a detectable IgG kappa peak on MS flagged as therapeutic antibody. Of these, 117 had an additional M-protein peak by MS corresponding to their endogenous M-protein, 33 were IgG/IgG kappa and 84 were non-IgG peaks. Conclusions This study demonstrates that MS offers enhanced sensitivity detecting and characterizing monoclonal proteins compared to the traditional methods SPEP and IFE. Importantly, MS was able to distinguish therapeutic monoclonal antibodies from endogenous M-proteins, reducing the risk of misclassification and improves accurate disease monitoring. These findings support the use of MS as a complementary and potentially superior method for disease assessment and monitoring of plasma cell disorders.