Background: Fibrocytes are circulating, pro-collagen expressing myeloid cells that contribute to wound healing. Emerging data implicates these cells in cancer progression. Our data supports a role for fibrocytes in the progression of multiple myeloma (MM).

Methods: Peripheral blood and bone marrow samples were collected according to an IRB approved protocol (IRB Number: 09-0768). Fibrocytes were identified in the PBMC fraction by flow cytometry as CD45+, COL1A1+, and CD115-cells. These results were confirmed by counting fibrocytes after five days of culture in serum-free conditions. Statistical analysis was performed by dividing the sample set into four subgroups: Monoclonal Gammopathy of Unknown Significance (MGUS), Smoldering Myeloma (SM), treated myeloma (TrMM), and untreated or relapsed myeloma (U/RMM). Significance was determined using non-parametric methods with the Bonferroni correction for multiple group comparisons.

The effects of MM cells on fibrocyte number were studied in vitro using a transwell system that co-cultured healthy donor PBMCs with MM cells from U/RMM patients. MM samples from patients with a greater than average number of circulating fibrocytes were compared with samples from patients with fewer than average. Results were reported as the fold increase in fibrocyte number compared with donor PBMCs alone.

Differences in gene expression between fibrocytes from five U/RMM and five MGUS patients were assessed using a RT2Profiler PCR Custom Array (Qiagen). Fibrocytes were isolated by culturing PBMCs in serum free conditions followed by immunomagnetic bead purification.

Results: Eighty-eight patient samples were analyzed. There were no differences in age, WBC count, monocyte count, hemoglobin, albumin, or creatinine. Measures of disease burden including β2M, percentage of bone marrow plasma cells, and concentration of the M-protein were consistent with the patients’ strata.

U/RMM patients had an increased number of circulating fibrocytes (4.7 ± 0.6 x107 cells/L) compared with TrMM (2.8 ± 0.5 x107 cells/L), SM (2.0 ± 0.5 x107 cells/L), MGUS patients (2.0 ± 0.6 x107 cells/L), or healthy controls (0.8 ± 0.2 x107 cells/L) (p < 0.05). These differences were confirmed using the cell culture techniques. Twenty-nine patients with MGUS or SM were not treated when entering the study. Seven of these developed progressive disease that required therapy. This group had a significantly greater number of circulating fibrocytes compared with the 22 patients without progression (2.4 ± 0.7 vs 1.3 ± 0.4 x 107 cells/L, p<0.05).

Primary MM cells were isolated from U/RMM patients with greater than average circulating fibrocytes (6.5 ± 1.2 x 107 cells/L) and those with less than average fibrocytes (1.8 ± 0.6 x 107 cells/L). The co-culture of these cells with healthy donor PBMCs increased the number of fibrocytes by 3.9 ± 0.9 fold. This increase was significantly greater using MM cells from patients with more circulating fibrocytes compared with those with less (5.4 ± 1.5 vs 2.5 ± 0.3 fold; p<0.05).

Twenty-eight potential mediators of this process were compiled from the literature and were assessed using bioinformatics applied to public databases. Of these, SPARC was the most promising candidate based on its differential expression in MM samples compared with normal plasma cells. Real time PCR measurement of SPARC expression in our U/RMM samples was positively correlated with the number of circulating fibrocytes (r=0.982, p=0.018).

We began to explore the ability of fibrocytes to promote MM by screening these cells using a custom real time PCR array. Twenty genes were selected for analysis based on our work using murine models. Of these, SPP1 was expressed at least one log greater than any of the other 19 genes (p < 0.01). Furthermore, fibrocytes from U/RMM patients had significantly greater expression of SPP1 than MGUS patients (1.82 fold, p<0.05).

Proposed Model: Our data support a model in which MM cells produce SPARC. This signal causes an expansion of fibrocytes. These cells, in turn, promote MM progression by expressing SPP1, a gene that can activate osteoclasts leading to a release of MM growth factors. In this way, fibrocytes can participate in a positive feedback loop leading to progression of myeloma.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution