Purpose: Although the role of inflammation in osteoarthritis has been heavily debated, recent evidence has demonstrated macrophage involvement in knee osteoarthritis (OA). Using etarfolatide (EC20) nuclear imaging that detects folate receptors, we have observed activated macrophages in human knee OA joints whose quantity are correlated with both radiographic and symptom severity of knee OA. In addition, we identified synovial fluid (SF) biomarkers, CD163 and CD14, that function as quantitative traits of the etarfolatide-positive phenotype and that predict knee OA progression over 3 years, allowing for the identification of inflammatory OA phenotypes. The purpose of this study was to characterize a molecular biological profile for identifying subsets of subjects with an inflammatory OA phenotype through the investigation of the association of an extensive array of synovial fluid biomarkers with the severity of macrophage infiltration, radiographic knee OA and pain. Methods: The Institutional Review Board approved all procedures and the study was registered at ClinicalTrials.gov (NCT01237405). Participants (n = 25) with radiographic knee OA in at least one knee and knee pain underwent bilateral knee etarfolatide imaging. SF was obtained by either direct aspiration (n = 28 knees) or by lavage (n = 20 knees) when direct aspiration was not possible. Samples were stored at -80 until analyzed. The intensity of etarfolatide imaging of the knee and pain assessment by NHANES criteria were scored semi-quantitatively as normal (score 0), mild (score 1), moderate (score 2), or severe (score 3). Radiographs were scored for Kellgren Lawrence (KL) grade (0-4), joint space narrowing (JSN, 0-3) and osteophyte (OST, 0-3). Biomarkers were quantified by Rules Based Medicine (RBM) using the high sensitivity multiplex immunoassay, Myriad Human InflammationMAP 1.0, including 47 different cytokines, chemokines, and growth factors related to inflammation. Biomarker concentrations in lavage fluid were corrected for dilution by an established urea method. Multivariable linear regression models using generalized estimating equations (GEEs) were performed to evaluate associations of biomarker concentrations with etarfolatide scores, radiographic OA severity, and OA symptoms, adjusting for age, gender and BMI. False discovery rate (FDR) was controlled using the Benjamini-Hochberg method. Spearman correlations were used to test for associations of informative RBM markers with sfCD163 and sfCD14. Results: Participants were 72% female, mean age of 62.4 (range 30-89) and a mean BMI of 29.2 (range 22.5-38.4 kg/m2). The majority (76%) of participants had moderate to severe bilateral radiographic knee OA with 24% having a KL grade 1, 58% KL grade 2-3, and 18% KL grade 4. Significant associations of SF biomarkers included sVCAM-1 and MMP-3 with the presence of activated macrophages (etarfolatide scan), sVCAM-1, sICAM-1, TIMP-1, and VEGF with radiographic OA severity (KL, JSN, and OST), and VEGF, MMP-3, TIMP-1, sICAM-1, sVCAM-1, and MCP-1 with OA symptoms (Table 1). These SF biomarkers were all highly correlated with sfCD163 and sfCD14. Conclusions: This study provides evidence in support of an inflammatory OA subphenotype characterized by the presence of activated macrophages and its association with synovial fluid biomarkers of inflammation and angiogenesis. These processes may interrelate as VEGF, a biomarker of angiogenesis, can increase expression of MMPs and inflammation, and markers of vascular adhesion, VCAM-1 and ICAM-1 correlate with synovitis. This biomarker profile may assist in identifying patients with an inflammatory OA phenotype for targeting of treatment to modulate inflammation and activated macrophages.
Read full abstract