Purpose
To determine the peripheral levels of 20 immune mediators in serum samples from patients with birdshot retinochoroidopathy (BSRC).
Design
Single-center prospective case-control study.
Methods
The serum of 17 BSRC patients during different phases of disease activity and therapy were analyzed with a quantitative multiplex sandwich enzyme-linked immunosorbent assay–based microarray to determine the levels of 20 immune mediators (T cell and proinflammatory). The serum of 12 healthy volunteers was used as controls.
Results
Serum levels of interleukin (IL)-21 ( P = .0005), IL-23 ( P = .0005), and transforming growth factor (TGF)-β1 ( P = .0011) were elevated in BSRC patients with active disease naïve to systemic therapy compared with that of controls. There was no significant difference in the serum levels of immune mediators between controls and BSRC patients who had a current or past history of IMT or who were in remission. The levels of IL-21, IL-23, and TGF-β1 were positively correlated (IL-23/IL-21, r = 0.91; TGF-β1/IL-21, r = 0.97; TGF-β1/IL-23, r = 0.87; for all, P < .0001).
Conclusions
BSRC patients with active disease naïve to systemic therapy have elevated serum levels of 3 key immune mediators known to promote T helper 17 (Th17) cells in autoimmune disease. Our results suggest that IL-21, IL-23, and TGF-β1 may play an important role in the development of site-specific Th17 cell–mediated inflammation in BSRC, which underscore the importance of systemic therapy and offer new insights into the potential of targeted treatments.
Birdshot retinochoroidopathy (BSRC) is a potentially blinding chronic inflammatory disease characterized by bilateral vitritis, retinal vascular leakage, and distinctive multiple cream-colored or hypopigmented postequatorial chorioretinal lesions. Patients are predominantly white, presenting in the sixth decade. There is strong association with human leukocyte antigen (HLA)-A29, but unlike other HLA-associated diseases there are no known predominant extraocular manifestations. The pathophysiology of BSRC remains to be elucidated, but an autoimmune etiology of the disease is supported by clinical evidence that immunomodulatory therapy (IMT) can preserve visual function and retinal physiology. It has been theorized that prior microbial infections could play a role through molecular mimicry, and that HLA-A29 cross-reactive proliferative responses directed against retinal antigens could lead to loss of immunologic tolerance and autoimmunity in BSRC. This is supported by the finding that transgenic mice expressing HLA-A29 spontaneously develop a retinopathy that resembles BSRC, and sequences from retinal soluble antigen (S-antigen) bind efficiently to HLA-A29. In vitro immune responsiveness to S-antigen, although not unique to BSRC, has been demonstrated in a high percentage of BSRC patients. In addition, the histopathologic findings in an eye from a BSRC patient with immune responsiveness to S-antigen were similar to those found in monkeys with S-antigen-induced uveitis. Cyclosporine A (CsA), a potent inhibitor of T cell function, not only has been shown to be effective in the treatment of BSRC patients but also inhibits S-antigen-induced experimental uveitis, suggesting that BSRC may be a T cell–mediated disease. Indeed, the presence of lymphocytic infiltration in the choroid, retinal blood vessels, and prelaminar optic nerve head was found on histopathologic sections from the eye of a BSRC patient. Certain allelic combinations of the killer cell immunoglobulin-like receptor (KIR) gene, which modify the receptors expressed on natural killer (NK) cells and T lymphocytes, are strongly associated with risk of developing BSRC. Moreover, T helper 17 (Th17) cells, which are a subset of CD4+ T lymphocytes that mainly produce interleukin (IL)-17 and are known to play an important role in autoimmune diseases, have been implicated in BSRC. Kuiper and associates recently showed that the levels of IL-17 were both elevated and concentrated in the aqueous humor of BSRC patients. To better understand the potential role of peripheral cytokines in BSRC, we used a quantitative multiplex immunoassay to investigate the serum profile of 20 immune mediators in BSRC patients, with emphasis on clinical disease activity and status of systemic therapy.
Methods
Subjects
The New England Institutional Review Board reviewed and approved this single-center prospective case-control study and the informed consents. All patients and healthy volunteers were recruited and consented at Massachusetts Eye Research and Surgery Institution, and the study was carried out in accordance with HIPAA regulations. Patients and volunteers who were pregnant or suffering from concurrent systemic disease such as autoimmune disease, malignant neoplasm, microbial infection, or recent surgery within 1 month of recovery were excluded. The diagnosis of BSRC was based on the research criteria of an international consensus conference, and all BSRC patients were HLA-A29 positive. BSRC patients naïve to systemic therapy were defined as having no history of using systemic corticosteroids or IMT. For BSRC patients undergoing treatment, only those treated with first-line IMT without systemic corticosteroids were recruited. At the study site, first-line treatment for BSRC was combination mycophenolate mofetil (MMF) and CsA, and second-line treatment included infliximab. Patients on second-line therapy were not included in the scope of this study, because chimeric monoclonal antibodies likely have significant effects on multiplex immunoassays that remain to be elucidated. For BSRC patients in remission off IMT, only those successfully treated with combination MMF and CsA were included. Every attempt was made to recruit healthy individuals with a similar age range and sex distribution as the BSRC group. The diagnosis of active disease was based on subjective visual complaint and examination findings, such as vitreous haze, macular edema, and vasculitis. Examination findings were confirmed with fluorescein angiography and optical coherence tomography. In addition, Humphrey visual field (HVF) and full-field electroretinogram (ffERG) were used to assess the level of visual field deficits and retinal pathology, respectively. Examination and testing were repeated during the course of IMT, and the diagnosis of disease remission was based on the resolution of vitreous haze, macular edema, and vasculitis, and normalization or a plateau in the improvement of HVF and ffERG results. Control volunteers and BSRC patients underwent phlebotomy, and the sera were immediately stored at −80 C until analysis.
Multiplex Immunoassay
Undiluted serum samples were analyzed for 20 soluble immune mediators by a commercially available quantitative multiplex sandwich enzyme-linked immunosorbent assay (ELISA)–based microarray from Raybiotech (Norcross, Georgia, USA). First, interfering heterophilic immunoglobulins were removed from serum using protein-L-coated beads (NAb Protein L Spin Columns; Thermo Scientific Pierce, Rockford, Illinois, USA). The immunoglobulin-extracted sera were stored at −80 C and shipped on dry ice to Raybiotech for analysis using the QAH-TH17-1 microarray, which quantifies levels of IL-1β, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p70, IL-13, IL-17, IL-17F, IL-21, IL-22, IL-23, IL-28, interferon (IFN)-γ, macrophage inflammatory protein (MIP)-3α, transforming growth factor (TGF)-β1, tumor necrosis factor (TNF)-α, TNF-β, and granulocyte macrophage colony-stimulating factor (GM-CSF).
Statistical Analysis
Statistical analysis was performed using Excel (Microsoft Corporation, Redmond, Washington, USA) and Igor Pro 6.22A (WaveMetrics Inc, Lake Oswego, Oregon, USA). Group differences for the serum levels of 20 cytokines were analyzed using 2-tailed Mann-Whitney U test corrected for unequal sample sizes and Bonferroni correction for multiple comparisons (α = 0.05/20 = 0.0025). Geometric mean was used to represent the data in the table to provide a more meaningful average, given unequal numeric ranges. Correlations between immune mediator concentrations were analyzed according to the Pearson product-moment correlation coefficient, and a correlation P value less than .05 was chosen as statistical significance. Immune mediator concentrations that were below the detectable limit were given a value half that of the lowest detectable concentration. No immune mediator levels exceeded the maximum detectable concentration.
Results
Our study cohort included 12 controls, 4 BSRC patients with active disease naïve to systemic therapy, and 13 BSRC patients with a current or past history of IMT. Among the 13 BSRC patients with a history of IMT, 3 still had active disease after starting IMT, 6 had disease remission while on IMT, and 4 had sustained disease remission after completing 2 years of IMT. The mean age of BSRC patients was 52 ± 10 years, ranging from 26-69 years, whereas the mean age of control volunteers was 49 ± 14 years, ranging from 29-70 years. Of the BSRC patients with active disease naïve to systemic treatment, the mean age was 50 ± 2 years, ranging from 47-52 years. The female-to-male ratio was 8:9 in BSRC patients and 9:3 in controls. Only the BSRC subgroup with active disease naïve to IMT had significantly elevated serum levels of IL-21 ( P = .0005), IL-23 ( P = .0005), and TGF-β1 ( P = .0011) compared with controls; none of the other 17 immune mediators were significantly elevated, specifically IL-1β, IL-2, IL-6, IL-17, TNF-α, and IFN-γ. There was no significant difference in the serum levels of immune mediators between the control group and the other BSRC subgroups. The mean serum levels of the 20 immune mediators in the BSRC patients and controls are reported ( Table ). Statistical analysis comparing individual BSRC subgroups with each other were limited by sample size. However, scatterplots of IL-21, IL-23, and TGF-β1 serum levels for the BSRC subgroups show a distribution of lower serum cytokine levels in patients undergoing IMT, with the lowest levels in patients who were in remission ( Figure 1 ). Patients on IMT with active disease were still undergoing titration and optimization of MMF and CsA, which is reflected in a distribution of intermediate-range cytokine levels. Serum cytokine levels for the 2 subgroups of BSRC patients in remission were within the range of controls. Among the patients who were in remission and off IMT, the range of time off IMT was 3-96 months. Correlation analysis revealed a positive correlation among the levels of IL-21, IL-23, and TGF-β1 (IL-23/IL-21, r = 0.91; TGF-β1/IL-21, r = 0.97; TGF-β1/IL-23, r = 0.87; for all, P < .0001; Figure 2 ).
| Mediator | Control Geometric Mean (Range) | BSRC Geometric Mean (Range) | |||
|---|---|---|---|---|---|
| Active/Naïve | Active/IMT | Remission/IMT | Remission | ||
| IL-1β | 2.6 (1.1-14.3) | 7.3 (1.5-483.9) | 1.4 (1.1-2.4) | 4.7 (4.2-7.1) | 4.3 (4.2-4.7) |
| IL-2 | 58.0 (17.3-311.3) | 108.0 (5.0-369.1) | 102.9 (68.9-229.7) | 15.4 (2.2-68.9) | 14.9 (2.2-74.0) |
| IL-4 | 12.1 (4.7-20.7) | 18.7 (4.7-131.7) | 7.1 (4.7-16.5) | 10.0 (3.9-15.2) | 10.8 (3.9-15.2) |
| IL-5 | 8.9 (1.8-82.1) | 18.0 (1.8-109.1) | 6.5 (1.8-84.5) | 2.6 (0.9-3.8) | 3.5 (0.9-12.1) |
| IL-6 | 16.4 (5.4-43.5) | 28.6 (5.4-53.3) | 11.0 (5.4-45.1) | 10.4 (5.4-14.2) | 11.3 (5.7-14.2) |
| IL-10 | 3.6 (1.9-8.6) | 10.8 (5.5-37.2) | 6.2 (1.9-19.4) | 3.3 (1.7-4.5) | 3.5 (1.7-4.5) |
| IL-12p70 | 2.9 (1.7-4.9) | 9.5 (3.5-159.4) | 4.2 (1.8-9.4) | 3.8 (1.8-4.9) | 4.2 (2.7-4.9) |
| IL-13 | 13.4 (5.9-24.1) | 38.1 (19.3-167.8) | 9.1 (5.9-21.8) | 13.0 (5.9-18.0) | 19.8 (14.6-36.2) |
| IL-17 | 14.9 (7.7-26.4) | 12.8 (7.7-29.5) | 13.1 (7.7-38.0) | 14.2 (7.7-17.5) | 15.8 (11.5-17.5) |
| IL-17F | 13.6 (1.3-54.0) | 69.6 (10.0-741.0) | 30.4 (10.8-54.7) | 8.7 (1.8-49.1) | 14.3 (1.8-249.2) |
| IL-21 b | 613.5 b (145.1-1475.2) | 3699.6 b (1621.0-8828.0) | 540.4 (145.1-2012.3) | 429.1 (145.1-726.9) | 683.3 (493.2-1298.4) |
| IL-22 | 14.5 (9.6-42.7) | 64.7 (32.0-282.5) | 15.9 (9.6-43.4) | 11.9 (8.5-13.6) | 12.1 (8.5-13.6) |
| IL-23 c | 139.9 c (36.7-1059.8) | 1905.6 c (1143.7-3028.0) | 129.2 (36.7-1602.4) | 62.0 (36.7-98.2) | 172.2 (63.0-451.1) |
| IL-28 | 26.3 (11.3-41.8) | 19.0 (11.3-29.1) | 15.2 (11.3-27.7) | 26.3 (9.7-41.8) | 29.0 (9.7-41.8) |
| IFN-γ | 107.2 (13.4-508.3) | 113.8 (13.4-483.9) | 37.6 (13.4-295.7) | 49.3 (9.7-102.6) | 78.5 (9.7-371.0) |
| TGF-β1 d | 639.3 d (130.7-1274.8) | 3703.6 d (1135.9-11 860.0) | 850.3 (130.7-2269.0) | 513.6 (130.7-691.5) | 827.9 (614.4-1599.1) |
| TNF-α | 9.4 (3.1-45.5) | 17.9 (3.1-64.6) | 6.7 (3.1-31.5) | 4.4 (2.7-5.4) | 4.5 (2.7-5.4) |
| TNF-β | 7.2 (2.6-33.2) | 17.8 (2.6-51.9) | 6.4 (2.6-38.2) | 4.4 (2.6-5.1) | 5.1 (4.8-6.1) |
| GM-CSF | 19.3 (4.6-43.0) | 19.6 (4.6-78.7) | 14.6 (7.3-20.7) | 28.1 (20.7-30.7) | 20.4 (6.0-30.7) |
| MIP-3α | 1.3 (0.5-28.8) | 1.7 (0.5-6.1) | 1.1 (0.5-5.3) | 0.6 (0.5-0.6) | 0.6 (0.5-0.6) |
a Serum of BSRC patients was analyzed for 20 immune mediators. Healthy age-matched volunteers served as controls. Data are presented as geometric mean and range (pg/mL).
b Statistically significant difference between Active/Naïve and Control groups for IL-21 ( P = .0005).
c Statistically significant difference between Active/Naïve and Control groups for IL-23 ( P = .0005).
d Statistically significant difference between Active/Naïve and Control groups for TGF-β1 ( P = .0011).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
