Macrophage activation syndrome is characterized by an overwhelming inflammatory reaction driven by excessive expansion of T cells and hemophagocytic macrophages. Levels of soluble interleukin-2 receptor α (sIL-2Rα) and soluble CD163 (sCD163) may reflect the degree of activation and expansion of T cells and macrophages, respectively. This study was undertaken to assess the value of serum sIL-2Rα and sCD163 in diagnosing acute macrophage activation syndrome complicating systemic juvenile idiopathic arthritis (JIA).
Enzyme-linked immunosorbent assay was used to assess sIL-2Rα and sCD163 levels in sera from 7 patients with acute macrophage activation syndrome complicating systemic JIA and 16 patients with untreated new-onset systemic JIA. The results were correlated with clinical features of established macrophage activation syndrome, including ferritin levels.
The median level of sIL-2Rα in the patients with macrophage activation syndrome was 19,646 pg/ml (interquartile range [IQR] 18,128), compared with 3,787 pg/ml (IQR 3,762) in patients with systemic JIA (P = 0.003). Similarly, the median level of sCD163 in patients with macrophage activation syndrome was 23,000 ng/ml (IQR 14,191), compared with 5,480 ng/ml (IQR 2,635) in patients with systemic JIA (P = 0.017). In 5 of 16 patients with systemic JIA, serum levels of sIL-2Rα or sCD163 were comparable with those in patients with acute macrophage activation syndrome. These patients had high inflammatory activity associated with a trend toward lower hemoglobin levels (P = 0.11), lower platelet counts, and significantly higher ferritin levels (P = 0.02). Two of these 5 patients developed overt macrophage activation syndrome several months later.
Levels of sIL-2Rα and sCD163 are promising diagnostic markers for macrophage activation syndrome. They may also help identify patients with subclinical macrophage activation syndrome.
In pediatric rheumatology the term “macrophage activation syndrome” refers to a set of symptoms caused by excessive activation and proliferation of T cells and well-differentiated macrophages (1, 2). Such activation leads to a potentially fatal overwhelming inflammatory reaction. The pathognomonic feature of macrophage activation syndrome, the presence of numerous, well-differentiated macrophages (histiocytes) actively phagocytosing hematopoietic elements, is often found in bone marrow aspirates. Although macrophage activation syndrome has been increasingly recognized in association with almost any rheumatic disease, it is by far most common in the systemic form of juvenile idiopathic arthritis (JIA) (1, 2).
The pathologic mechanisms of macrophage activation syndrome are not fully understood. Previous studies have focused on natural killer (NK) cell and cytotoxic T cell function, the rationale being that there are striking similarities between macrophage activation syndrome and inherited forms of hemophagocytic lymphohistiocytosis (3, 4). In hemophagocytic lymphohistiocytosis there is uncontrolled proliferation of T cells and macrophages linked to decreased NK cell and cytotoxic T cell function, often due to mutations in the gene encoding perforin (5, 6). Recent observations suggest that, as in hemophagocytic lymphohistiocytosis, patients with macrophage activation syndrome have profoundly depressed NK cell function, often associated with abnormal perforin expression (2, 3). Moreover, a subgroup of patients with systemic JIA, without apparent macrophage activation syndrome, have similar immunologic abnormalities (3).
Clinical evidence and observations of animal models of hemophagocytic lymphohistiocytosis indicate that one of the early events in hemophagocytic lymphohistiocytosis and macrophage activation syndrome is an uncontrolled, and persistent, expansion of activated CD8+ T lymphocytes that produce proinflammatory cytokines, including interferon-γ, which then drive the expansion of hemophagocytic (CD163+) macrophages (7, 8). The emerging consensus is that serum levels of soluble interleukin-2 receptor α (sIL-2Rα) and soluble CD163 (sCD163) reflect the degree of activation and expansion of T cells and phagocytic macrophages, respectively (9–12). These laboratory markers appear to be very useful tools in diagnosing hemophagocytic lymphohistiocytosis and may also provide a means to monitor treatment response.
The IL-2 receptor complex is a trimer, consisting of α-, β-, and γ-chains, that interacts with IL-2. The α subunit of this complex, also known as Tac antigen and as CD25, is a 55-kd transmembrane glycoprotein with only 13 of 351 amino acids located on the cytoplasmic side of the membrane. A soluble form of IL-2R appears in serum and plasma, concomitantly with increased cell surface expression. Increased levels of sIL-2Rα in biologic fluids correlate with the activation of T and B lymphocytes. The findings of a number of studies suggest a correlation between serum IL-2Rα levels and disease activity in autoimmune and infectious disorders as well as in transplant rejection. The increase in the levels of sIL-2Rα is particularly striking in hemophagocytic syndromes, including hemophagocytic lymphohistiocytosis (10, 11).
The hemoglobin-haptoglobin scavenger receptor (CD163) is a monocyte/macrophage-restricted transmembrane protein of the cysteine-rich scavenger receptor family. CD163 expression identifies macrophages undergoing the alternative pathway of differentiation that is associated with enhanced phagocytic activity (12). The extracellular portion of CD163 is shed from the cell surface in the form of sCD163 when cells are appropriately activated (12). Extensive expansion of CD163+ macrophages has been demonstrated in the bone marrow of a patient with macrophage activation syndrome (13), and sCD163 appears to be a valuable diagnostic marker in hemophagocytic syndromes (9).
In this study, we assessed the serum levels of sIL-2Rα and sCD163 in patients with acute macrophage activation syndrome presenting as a complication of systemic JIA. We compared these findings with those in patients with systemic JIA without apparent macrophage activation syndrome.
PATIENTS AND METHODS
The patients included in this study were enrolled in a multicenter, prospective study of gene expression profiles in childhood arthritis. Diagnosis of systemic JIA was based on International League of Associations for Rheumatology criteria (14). Clinical information was prospectively recorded on study forms, and peripheral blood samples were obtained from patients with new-onset systemic JIA, prior to the initiation of treatment with disease-modifying antirheumatic drugs (DMARDs). Sixteen patients with new-onset systemic JIA were studied; 9 were male, and the median age was 11.5 years. Disease duration at the time of enrollment ranged from 2 weeks to 12 weeks.
Macrophage activation syndrome was diagnosed based on the combination of cytopenias affecting at least 2 cell lines, coagulopathy, and liver dysfunction (Table 1), according to the guidelines proposed by Ravelli et al (15). Seven patients with macrophage activation syndrome were studied; 4 of these had hemophagocytosis in the bone marrow. All patients with macrophage activation syndrome had systemic JIA (2 had new-onset systemic JIA but were not included in the group of 16 patients with new-onset systemic JIA described above).
Table 1. Clinical characteristics of the patients with systemic JIA during the acute phase of macrophage activation syndrome*
The study was approved by the institutional review boards of the participating clinics. Written informed consent was obtained from parents of all patients.
Whole blood was collected using sodium heparin as an anticoagulant. Serum was separated from cells, divided into aliquots, frozen, and stored at −80°C at the coordinating center until used. When possible, a separate sample was collected for NK cell functional studies. When collected at centers other than the coordinating center, serum was separated within 2 hours of blood collection, divided into aliquots, frozen, and transported overnight on dry ice. A separate sample for NK cell functional studies was shipped at room temperature.
Soluble CD163 assay.
Levels of sCD163 in serum were measured using a noncompetitive sandwich enzyme-linked immunosorbent assay (ELISA; Cedarlane Laboratories, Hornby, Ontario, Canada). Briefly, a monoclonal antibody specific for sCD163 was precoated onto a 96-well microplate. Standards, and serum samples from patients, were pipetted into the wells (50 μl of serum diluted to 1 ml in the assay buffer). Eighteen hours later, the plate was washed and a chicken anti-human antibody against sCD163 was added. Soluble CD163 was sandwiched by the immobilized antibody and polyclonal antibody specific for sCD163. Peroxidase-conjugated anti-chicken antibody was added. Following the incubation and 3 washes, a substrate solution was added and color developed. The color development was stopped and the intensity of color was measured.
Soluble IL-2R α-chain assay.
Levels of sIL-2Rα were measured using a Quantikine ELISA kit (R&D Systems, Minneapolis, MN). Briefly, a monoclonal antibody specific for sIL-2Rα was precoated onto a 96-well microplate. Standards, and serum samples from patients and controls, were pipetted into the wells. After incubation, the plate was washed and horseradish peroxidase–conjugated polyclonal antibody against sIL2-Rα was added. Soluble IL-2Rα was sandwiched by the immobilized antibody and the polyclonal antibody specific for sIL-2Rα. Following the incubation and 3 washes, a substrate solution was added to the wells and color developed. The color development was stopped and the intensity of color was measured.
NK cell cytotoxicity analysis.
NK cell activity was assessed after coincubation of peripheral blood mononuclear cell preparations (effector cells) with 51Cr-labeled target cells at various ratios of effector cells to target cells, as previously described (3). The NK cell–sensitive K562 line was used as a source of target cells.
Descriptive analysis was performed and, given the skewness of the data, medians and interquartile ranges (IQRs) were reported as measures of central tendency. Spearman's correlation coefficient (rs) was used to assess the relationship between clinical, laboratory, and cell markers. Groups of patients were compared for significant differences in laboratory values and cell markers using Wilcoxon's rank sum test and t-test approximations for minimal group sizes >5; for minimal group sizes <5 a median 2-sample test was done with z-test approximation. Two-sided P values are reported.
Soluble IL2-Rα and sCD163 levels during the acute phase of macrophage activation syndrome and in new-onset systemic JIA.
The clinical characteristics of the patients with acute macrophage activation syndrome are summarized in Table 1. Notably, in the 4 patients with macrophage activation syndrome in whom both C-reactive protein (CRP) level and erythrocyte sedimentation rate (ESR) were assessed, the degree of CRP elevation was out of proportion compared with the ESR. The serum levels of sIL-2Rα and sCD163 were measured prior to the initiation of specific treatment of macrophage activation syndrome. As shown in Table 1, most patients had very high levels of both sIL-2Rα and sCD163. In fact, the levels were comparable with those reported in patients with inherited hemophagocytic lymphohistiocytosis (9, 11).
To determine whether sIL-2Rα and sCD163 levels may help distinguish acute macrophage activation syndrome from active systemic JIA, the levels of these markers were assessed in serum samples obtained from 16 patients with new-onset systemic JIA prior to the initiation of treatment with DMARDs. The median level of sIL-2Rα in the patients with macrophage activation syndrome was 19,646 pg/ml (IQR 18,128), compared with 3,787 pg/ml (IQR 3,762) in those with systemic JIA (P = 0.003). Similarly, the median level of sCD163 in patients with macrophage activation syndrome was 23,000 ng/ml (IQR 14,191), compared with 5,480 ng/ml (IQR 2,635) in those with systemic JIA (P = 0.017).
A subgroup of 5 (31%) of the 16 patients with new-onset systemic JIA had levels above the 75th percentile for at least 1 of the 2 markers at baseline, in the absence of overt macrophage activation syndrome. In fact, the serum levels of sIL-2Rα and/or sCD163 in these patients were comparable with those observed in patients with acute macrophage activation syndrome. The clinical characteristics of these patients are summarized in Table 2.
Table 2. Clinical characteristics of patients with new-onset systemic JIA with abnormal levels of sIL-2Rα and sCD163*
Patients with high levels of macrophage activation syndrome markers
Patients with low levels of macrophage activation syndrome markers, median (IQR) (n = 11)
NK = natural killer; LU = lytic units (see Table 1 for other definitions).
Disease duration, weeks
Serum ALT, units/liter
Serum AST, units/liter
NK cell cytolytic activity, LU
Time to development of macrophage activation syndrome, months
All 5 patients had a high level of inflammatory activity associated with a marked elevation of ESR and CRP levels. Notably, 4 had a normal platelet count. This was in contrast to the other patients with systemic JIA, in whom elevated platelet counts paralleled the increase in overall inflammatory activity, as is typically encountered in a generalized inflammatory status. There was a trend toward lower levels of hemoglobin in these 5 patients compared with the remaining 11 systemic JIA patients, although the difference was not statistically significant (P = 0.11). D-dimers were measured in 2 of the 5 patients and were found to be moderately elevated. Serum ferritin levels were highly elevated in all 5 patients (median 2,950 ng/ml [IQR 1,001]), whereas only 3 of the remaining 11 patients with systemic JIA had an abnormal ferritin level. The difference in ferritin levels between the 2 groups of patients with systemic JIA was significant (P = 0.02). NK cell cytolytic activity was assessed in 4 of the 5 patients and was lower than in the remaining 11 patients (P = 0.05). Notably, 2 of the 5 patients developed overt macrophage activation syndrome (18 months and 6 months, respectively, after the initial evaluation) (Table 2).
Correlation of ferritin levels with sIL-2Rα and sCD163 levels.
In correlation analyses including all patients with new-onset systemic JIA (n = 16), strong-to-moderate correlations were noted between serum levels of ferritin and sIL-2Rα (rs = 0.58, P = 0.004). There was a moderate correlation between serum ferritin and sCD163 levels (rs = 0.47, P = 0.02).
Longitudinal assessment of sIL-2Rα and sCD163 levels in patients with macrophage activation syndrome.
Soluble IL-2Rα and sCD163 were assessed longitudinally in patient 5, who had particularly severe and protracted macrophage activation syndrome (Table 1). The levels of sIL-2Rα and sCD163 paralleled the established clinical features of macrophage activation syndrome, such as cytopenia, ferritin levels, liver dysfunction, and coagulopathy. Figure 1 illustrates the relationship between sIL-2Rα and sCD163 levels and 2 clinical features of macrophage activation syndrome, thrombocytopenia and elevation in serum ferritin levels.
In patients 1, 2, 3, and 7 (Table 1), the assessment of sIL-2Rα and sCD163 was repeated after the resolution of cytopenia, liver dysfunction, and coagulopathy. In all of these patients, the levels of sIL-2Rα and sCD163 returned to the normal range.
We demonstrated that high levels of sIL-2Rα and sCD163 are present during the acute phase of macrophage activation syndrome associated with systemic JIA. The levels were comparable with those reported in both inherited (primary) and reactive (secondary) forms of hemophagocytic lymphohistiocytosis (9–11). Levels of sIL-2Rα and sCD163 returned to the normal range after the resolution of macrophage activation syndrome. In 1 patient with a protracted course of macrophage activation syndrome, serial measurements of sIL-2Rα and sCD163 paralleled other established clinical features of macrophage activation syndrome, including cytopenia and serum ferritin levels, suggesting that these markers may reflect changes in response to therapeutic measures. Although the number of patients with macrophage activation syndrome in this study was small, the data suggest that sIL-2Rα, reflective of T cell activation, and sCD163, corresponding to macrophage (histiocyte) activation, might be useful as diagnostic markers and helpful in monitoring disease activity and response to treatment.
Unexpectedly, 5 of 16 patients with new-onset systemic JIA had levels of at least 1 marker comparable with those in patients with acute macrophage activation syndrome. Interestingly, despite highly active disease, these patients had normal platelet counts, in contrast to other patients with new-onset systemic JIA, in whom platelet counts were elevated parallel to the increase in other markers of inflammation. This would indirectly suggest that these patients had had relative thrombocytopenia. Additionally, these patients had lower hemoglobin values and very high levels of serum ferritin. Two had moderately elevated levels of D-dimers, suggestive of ongoing coagulopathy. NK cell function in these patients was lower than in the entire group of patients with new-onset systemic JIA.
Taken together, these observations suggest that these 5 patients in fact had subclinical macrophage activation syndrome. Assessment of sIL-2Rα and sCD163 levels may help identify such patients. Since overt macrophage activation syndrome is life-threatening, treatment modifications and closer followup may be required in these patients. Consistent with these findings, 2 of the 5 patients later developed overt macrophage activation syndrome. This raises the possibility that these 5 patients represent a distinct subtype of systemic JIA with a potentially distinct outcome. Alternatively, it is feasible to consider that any patient with systemic JIA may develop similar abnormalities during the evolution of the disease, for example, following infections. A prospective study addressing this issue is in progress.
Interestingly, in the 4 patients with macrophage activation syndrome in whom both CRP levels and ESR were assessed, the degree of CRP elevation was out of proportion compared with the ESR. CRP and the proteins that influence the ESR (fibrinogen, in particular) are produced mainly by the liver as a part of the acute-phase response. In macrophage activation syndrome, ESRs (disproportionately low considering the degree of inflammatory activity) parallel the degree of hypofibrinogenemia caused by decreased production of fibrinogen due to liver dysfunction and increased consumption of fibrinogen due to coagulopathy. In contrast, the CRP levels in macrophage activation syndrome appear to reflect the degree of inflammatory activity more accurately, suggesting that the impact of the liver dysfunction on CRP synthesis is relatively low. These considerations raise the question of whether the combination of persistently high CRP levels and relatively low ESR might be another marker of macrophage activation syndrome.
The main limitation of the present study was the small number of patients. A larger study including both patients with systemic JIA and patients with other febrile illnesses associated with immune activation may help define the true diagnostic value of these markers. Such studies may lead to the identification of diagnostic cutoff levels to determine which patients are at risk of developing macrophage activation syndrome.
Dr. Bleesing had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study design. Bleesing, Ilowite, Ramanan, Filipovich, Grom.