Antibodies to Endothelial Cell Growth Factor and Obliterative Microvascular Lesions in the Synovium of Patients With Antibiotic-Refractory Lyme Arthritis

Autor: Elise E. Drouin, Diana Londoño, Klemen Strle, John M. Aversa, Diego Cadavid, Gail McHugh, Allen C. Steere
Rok vydání: 2014
Předmět:
Zdroj: Arthritis & Rheumatology. 66:2124-2133
ISSN: 2326-5191
DOI: 10.1002/art.38618
Popis: Lyme disease in the northeastern United States is caused by the tick-transmitted spirochete, Borrelia burgdorferi (1). The most common late manifestation of the infection is Lyme arthritis, which often affects one or both knees (2). Most patients can be treated successfully with a 1-month course of oral or intravenous (IV) antibiotic therapy (3,4), called antibiotic-responsive Lyme arthritis. However, a small percentage of patients have persistent synovitis despite treatment with 1–2 months of oral antibiotics and 1 month of IV antibiotics, termed antibiotic-refractory Lyme arthritis (5). After antibiotic therapy, these patients are often treated with disease modifying anti-rheumatic drugs (DMARDs), such as hydroxychloroquine or methotrexate. If the response to DMARDs is incomplete, arthroscopic synovectomy is an option. Antibiotic-refractory Lyme arthritis is associated with infection with highly inflammatory strains of B. burgdorferi (6). However, persistent infection seems not to be the explanation for persistent synovitis after oral and IV antibiotic therapy (7). Culture and PCR results for B. burgdorferi in synovial tissue have been uniformly negative after this therapy (8), and treatment with immunosuppressive DMARDs has not led to reactivation of infection (5). Instead, excessive inflammation in joints (9,10), associations with specific HLA-DR alleles (11), and difficulty down-regulating inflammatory responses (12,13) appear to result in post-infectious inflammatory immune phenomena that lead to persistent synovitis after spirochetal killing. In MyD88−/− mice, which have high pathogen loads, B. burgdorferi antigens are retained near cartilage surfaces after antibiotic therapy, and patellar homogenates induce macrophages to secrete TNF-α (14). However, patients with Lyme arthritis have low pathogen loads (8), and the extensive proliferative synovitis found in patients with antibiotic-refractory arthritis (15–17) is not replicated in mice. Thus, the role of retained spirochetal antigens in post-infectious immune responses in the human disease is not yet clear. We recently identified human, platelet-derived endothelial cell growth factor (ECGF) as the first autoantigen known to be a target of T and B cell responses in about 20% of patients with Lyme arthritis, particularly in those with antibiotic-refractory arthritis (18). In addition, about 15% of patients with erythema migrans (EM), the initial skin lesion of the infection, also had autoantibody responses to ECGF. When archival serum samples were tested from 27 non-antibiotic-treated patients who were followed from EM through the course of arthritis during the 1970s before the cause of the disease was known, 7 (26%) had ECGF antibody responses, which often appeared early in the illness, prior to the onset of arthritis. Moreover, the total duration of attacks of active arthritis in these patients was significantly longer than in those who lacked ECGF reactivity (median, 67 versus 17 weeks, P=0.004). Additionally, ECGF, an IFN-γ-inducible protein (19), was expressed at significantly higher levels in synovial fluid (SF) in patients with antibiotic-refractory arthritis than in those with antibiotic-responsive arthritis, and patients with antibiotic-refractory arthritis often had moderate-to-intense staining for ECGF in the lining and sublining areas of synovial tissue (18). Although the ECGF protein is also found in synovial tissue and SF from patients with rheumatoid arthritis (RA) (20), it seems to be immunogenic only in Lyme arthritis. Thus, anti-ECGF antibodies appear to be a marker for a disadvantageous immune response that is associated with more persistent Lyme arthritis. However, it is not yet known whether autoantibodies to ECGF cause synovial pathology. To investigate this issue, we examined synovial tissue samples, which were obtained at synovectomy, months after antibiotic treatment, from 14 patients with antibiotic-refractory Lyme arthritis. To blind the investigators to diagnosis and clinical data, synovial samples were also included from 6 patients with other forms of chronic inflammatory arthritis, primarily RA. In each of these diseases, synovial tissue often showed synovial hypertrophy, vascular proliferation, immune cell infiltrates, and fibrosis. However, obliterative microvascular lesions were found only in Lyme arthritis. Five of the 14 patients (36%) with this disease had positive antibody responses to ECGF, and all 5 had extensive obliterative microvascular lesions. Moreover, the magnitude of ECGF antibody responses correlated directly with the extent of obliterative microvascular lesions, suggesting that these autoantibodies have specific pathologic consequences in antibiotic-refractory Lyme arthritis.
Databáze: OpenAIRE