Loss of Chemokine Receptor Expression and Tetraspanins Is Correlated with Extramedullary Disease In Multiple Myeloma

Extramedullary disease (EMD) portends poor prognosis in multiple myeloma (MM). Human MM cells show different chemokine receptor profiles at different sites of disease activity; down regulation in advanced stages results in impaired retention of MM cells in the bone marrow (BM) thus promoting EMD. Tetraspanins (Tspns) are cell surface proteins involved in cellular adhesion, motility, tumor invasion, etc. Tspn expression is often deregulated in cancer and correlates with metastasis in various cancers. In vitro studies show down regulation of Tspn CD82 in human myeloma cell line (HMCL) in end stage disease and reduced cell motility and invasive potential with re-induction of CD81/CD82. Epigenetic silencing of CD9 is reported to correlate with disease progression and cell survival. It is likely that decreased expression of these membrane proteins is related to disease dissemination in MM.

Based on the known function of chemokine receptors and Tspns, we hypothesize that 1) EM spread of myeloma is mediated, at least in part by down regulation of Tspns CD81, CD82, and CD9, and chemokine receptors CXCR4 and CCR2 on the BM MM cells 2) The decreased expression is likely limited to a subpopulation of neoplastic cells.

In this study we examine the expression of Tspns CD81, CD82, and CD9, chemokine receptors CXCR4, CCR2, and intercellular antigen CD56 on surface of BM myeloma cells, and correlate expression of these antigens with presence of EMD, and with the 70-gene expression profiling (GEP) determined risk category.

We studied consecutive BM aspirate specimens from patients with MM with at least 10% aneuploid plasma cell population, or clearly light chain restricted if <10% or diploid. Four-color flow cytometry was performed on the BD FACSCanto II. Using CD138, CD38 and CD45 antibodies to gate, the bright CD138 cells were selected for further analysis using fluorescent-tagged antibodies to CD81, CD82, CD9, CD184 (CXCR4), and CD192 (CCR2). GEP was performed on the Affymetrix U133Plus 2.0 microarray platform (Santa Clara, CA) using CD138-enriched cells. Risk categorization was done using the 70 gene-discriminatory model: High Risk score >0.66 (Blood 2006; 108: 2020–2028). The molecular subtypes: HY, CD-1, CD-2, LB, MY, MS, MF, and PR, are as previously described. EMD was determined by PET scans. Active disease was defined as presence of ≥ 3 FDG-avid lesions on PET scan. The analysis was done using STATA and test of significance was done using chi2.

77 patients were analyzed; Age: 41–88 years; M/F: 1.6:1; EMD: 12 (15.6%) without EMD: 65 (84.4%). Sites of EMD: soft tissue 8; lymph node 3; spleen 2; pleura 2; breast 1, muscle 1; nasopharynx 1; CNS 1; plasma cell leukemia 1. The GEP data for risk categorization was available in 75/77 patients. Table 1 shows the distribution of disease status, risk category and antigen expression in the EMD and no EMD groups.

Summary: EMD is not uncommon (15.4% of patients in this study). High-risk disease was present in 58% of patients with EMD compared to only 22% in those without EMD. Expression of CD56, CD81, CD82, CD9, CXCR4, and CCR2 on individual patient samples was heterogeneous with respect to proportion of positive cells and the mean fluorescence intensity (MFI).

Loss of surface CXCR4 was seen in 50% of EMD vs. 18% without EMD (p = 0.01), while loss of surface CD82 occurred in 75% of EMD vs. 58% without. Loss of CXCR4 expression was noted in 43% of high-risk group compared to 17% in the low risk (p = 0.04). Preliminary results suggest that EMD may be correlated with loss of surface antigens especially CXCR4 and CD82. It is likely that combinations of antigen loss are more relevant than single antigen losses. MFI analysis shows that loss of CXCR4 and CD82 is not related to decreased antigen density on individual cells but loss of a subset of plasma cells with antigen expression, suggesting clonal evolution. Down regulation of cytokine receptors or Tspns may be related to genomic events that occur with disease progression.

No relevant conflicts of interest to declare.