High-Throughput Plasma Proteomics to Define the Precursor Multiple Myeloma Proteome and Identify Candidate High-Risk Disease Biomarkers of Progression

Introduction

Multiple Myeloma (MM) develops from well-defined precursors Monoclonal Gammopathy of Undetermined Significance (MGUS) and Smoldering Multiple Myeloma (SMM), where patients remain stable or may unknowingly rapidly progress. Bone marrow (BM) biopsies are not routine for precursor disease management, and precursor patients are limited to monitoring few proteins within peripheral blood (PB) for signs of progressive disease. Deep proteome profiling of PB circulating proteins may help track disease; however, the dynamic range of the plasma proteome has limited the depth of detection for MS-based proteomics without first depleting abundant proteins and fractionating the samples after digestion to peptides. Technological advancements in multiplex immunoassays with low cross-reactivity and off-target events have enabled plasma profiling for disease stage classification, defining high-risk disease features, and novel therapeutic target discovery. Here, we perform the first comprehensive plasma proteomic profiling study on patients across the MM disease continuum and longitudinal sequential samples from progressive and stable disease.

Methods

We carried out high-throughput plasma proteomic profiling for approximately 3000 proteins simultaneously using the Olink® Explore 3072 library and Proximity Extension Assay (PEA) technology. Targeted proteins are recognized by multiplexed, matched pairs of antibodies labelled with unique DNA oligonucleotides that upon binding come into proximity, hybridize, and are extended to generate a unique sequence for protein identification with NextGen DNA sequencing.

We profiled 423 PB plasma samples from 348 individuals, including MGUS (n=67), SMM (n=179), MM (n=44), and healthy controls (n=58). Sequential samples from patients with progressive disease (n=27) and patients with stable disease with matched clinical follow-up time were also profiled. Precursor defined samples from progressors ranged 1.03-5.88 yrs prior to diagnosis and were untreated in the precursor setting, while MM disease samples were also collected prior to any active disease therapy. Patients had a median clinical follow-up time of 7.05 years. T-tests, ANOVAs, and a linear mixed effect (LME) model were used to identify proteins that change across disease stages, progression status, and time. Results were adjusted for multiple testing using the Benjamini-Hochberg Method.

Results

We identified 751 significantly dysregulated proteins with the majority upregulated in progressive disease. We captured circulating levels of proteins highly expressed on the surface of plasma cells, including CD38, SDC1, BCMA and SLAMF7, highlighting the utility of PEA technology to monitor clinically-relevant targets for which monoclonal antibodies, antibody-drug conjugates, CAR-T and BiTE therapies are being developed. We identified proteins that significantly distinguished MGUS, SMM, and NDMM from healthy donors (n=222, 423 and 494), where increasing B-cell maturation antigen (BCMA) was a significant classifier and positive control. Consistent with previous findings, baseline BCMA levels were also significantly elevated in SMM-MM progressors vs. SMM non-progressors, further supporting the potential utility of BCMA measurements during routine blood tests of precursor MM patients. Proinflammatory cytokines were also identified, including IL1, IL5, IL6, IL16, and IL18, known to create a BM environment that promotes malignant cell development by suppressing the microenvironment, promoting cellular adhesion, or increasing angiogenesis. Four novel proteins that are vital for calcium homeostasis and integrin-mediated cell adhesion significantly increased from healthy to MM and were also significantly elevated in SMM progressors vs. SMM non-progressors, nominating these proteins as candidate biomarkers of high-risk disease.

Conclusion

We performed the most comprehensive plasma proteomics study to date, which characterized disease stage proteomes and identified candidate high-risk disease biomarkers in longitudinal progressor samples. Further advancements are underway to validate the accuracy levels of the novel candidates, test the performance of a classification model that recognizes disease stage-specific proteins, and determine how best to integrate proteins into current MM risk stratification models.