Advances in Diagnosis and Treatment of Vasculitis
A variety of noteworthy studies pertaining to vasculitis were reported at the 1999 American College of Rheumatology meetings in Boston. The following summary focuses on the major studies related to the diagnosis and treatment of these conditions.
**Two studies focused on the refinement of testing for anti-neutrophil cytoplasmic antibodies (ANCA) in clinical settings. A common problem encountered in the diagnosis of ANCA-associated Vasculitides (AAV) is that numerous common conditions may mimic the presentation of AAV. The test characteristics of ANCA assays in patients with clinical mimickers of AAV are not well-described.
Abstract 627 Prevalence of Antineutrophil Cytoplasmic Antibodies (ANCA) in Patients with Various Pulmonary Diseases or Multi-organ Dysfunction.
D Vassilopoulos, JL Niles, A Villa-Forte, AC Arroliga, EJ Sullivan, PA Merkel, GS Hoffman
Overview of study: This study evaluated the use of immunofluorescence (IF) and enzyme-linked immunosorbent assays (ELISA) in 3 groups of patients: Group I, 29 Pulmonary Clinic outpatients with a variety of lung conditions (interstitial lung disease, cancer, sarcoidosis, actinomycosis, and bronchiectasis); Group II, 99 Intensive Care Unit patients with multi-organ system dysfunction (including 91 with either pulmonary or renal dysfunction, or both); and Group III, 18 patients with biopsy-proven Wegeners Granulomatosis.
Summary of Findings: Seventy-two percent of the Wegeners Granulomatosis patients were positive for ANCA by both IF and ELISA. By comparison, although 28% of the Group I patients and 37% of the Group II patients were positive by IF, none of the patients in these groups who did not have Wegeners Granulomatosis or microscopic polyangiitis were positive for ANCA by ELISA.
Editorial Comment: The results of this study emphasize the added specificity conferred by use of the ELISA technique in the evaluation of patients with possible AAV.
The second study related to ANCA focused on the population of patients who underwent testing for ANCA at a tertiary care center over a period of 3 years, for the purpose of excluding AAV.
Abstract 1473 ANCA Testing in Patients Suspected of ANCA-Associated Conditions: Immunofluorescence and ELISA Techniques in 856 Consecutive Patients
JH Stone, M Talor, J Stebbing, NR Rose, DB Hellmann, CL Burek
Overview of study: A cross-sectional study of stored sera from 856 patients who were tested prospectively for ANCA. Based on the 1994 Chapel Hill Consensus Criteria (CHCC) for nomenclature of the vasculitides, patients were grouped into one of 4 types of AAV (Wegeners granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome, and pauci-immune glomerulonephritis) or into 1 of 6 pre-defined other categories of disease. In a blinded fashion, the investigators then performed ELISAs on the stored sera for antibodies to proteinase-3 (PR3) and myeloperoxidase (MPO).
Summary of Findings: Sixty-nine of the 856 patients (8%) had AAVs based on the CHCC guidelines. The positive predictive value of ELISA was superior to that of IF (83% versus 45%, respectively; P < 0.001 [95% C.I. = 0.20, 0.55]). For patients with both positive IF tests and positive ELISA tests, the predictive value for AAV was 88%. Both IF and ELISA had high negative predictive values (97% and 96%, respectively). Positive ELISA tests had higher likelihood ratios than IF (54.2 [95% C.I. = 26.3, 111.5] versus 9.4 [95% C.I. = 6.9, 12.7]). The likelihood ratio of both a positive IF and a positive ELISA was 82.1 (95% C.I. = 33.3, 202.5).
The investigators concluded that, compared with IF, the use of ELISA tests dramatically increases positive predictive values and likelihood ratios for AAV. The comparable sensitivities of IF and ELISA, combined with the greater costs and time required to perform the more specific ELISA tests, suggest that an effective testing strategy is to perform ELISA tests only on samples that are positive for ANCA by IF.
Editorial Comment: This study is the first to provide both predictive values and likelihood ratios for both IF and ELISA techniques for ANCA. Knowledge of these test characteristics is extremely useful to clinicians in the evaluation of patients with possible AAV.
Together, the studies reported in Abstracts 627 and 1473 illustrate the importance of using the antigen-specific ELISA technique to confirm any report of ANCA found on IF testing.
**In addition to these contributions to improved means of diagnosis in the ANCA-associated Vasculitides, 5 abstracts reported exciting developments in the therapy of the Vasculitides. In the abstract described below, the investigators reported preliminary outcome data from the first trial in vasculitis completed by the European Vasculitis Study group (EUVAS).
Abstract 928 A Multi-Centre Randomised Trial of Cyclophosphamide Versus Azathioprine During Remission in ANCA-Associated Systemic Vasculitis
R Luqmani and D Jayne, for the European Vasculitis Study Group (EUVAS)
Overview of study: The principal aim of this trial was to compare the use of azathioprine (AZA) to that of cyclophosphamide (CYC) in the prevention of disease relapse in patients with Wegeners Granulomatosis and microscopic polyangiitis. Over a 24 month enrollment period, 155 patients were recruited from 40 different medical centers in Europe. Patients with threatened vital organ function and a serum creatinine < 5.7 mg/dL were enrolled. Following a 3-month induction period with daily CYC and prednisolone, patients were randomized to receive either CYC for an additional 12 months, or to switch to AZA. All patients remaining in the trial were converted to AZA 12 months after randomization, and the total study duration for any patient was 18 months. The primary endpoint was the rate of relapse in the two groups.
Summary of Findings: 141 of the 155 patient enrolled (91%) entered remission and were randomized to either CYC (n = 70) or AZA (n = 71). At the conclusion of the trial, relapses had occurred in 17% of the CYC group (12 relapses) and 16% of the AZA group (11 relapses). The relative risk for disease relapse for those randomized to AZA was 0.93; 95% confidence interval 0.39 2.24). Thus, the likelihood of relapse did not differ between the groups. One death occurred in each group during the remission phase.
In terms of adverse events, 20 patients in both groups suffered severe or life-threatening side-effects, and the total number of adverse events did not differ between the groups (88 adverse events for the CYC group, 80 for the AZA group; P = NS).
Editorial Comment: During the 15 month follow-up period of this trial, AZA appeared to be equally effective as CYC in the maintenance of disease remissions after a 3-month induction regimen of daily CYC. The results of this trial provide rationale for the conversion of patients with either Wegeners granulomatosis or microscopic polyangiitis to AZA after a period of CYC use suitable to induce remission. It is conceivable that such a strategy may avoid some of the long-term side-effects of CYC use, particularly the occurrence of malignancies, bladder toxicity, opportunistic infection, and infertility.
**A study from the Mayo Clinic evaluated the long-term outcomes of prednisone-treated patients with GCA in a population-based cohort.
Abstract 834 Long-Term Outcome of Corticosteroid Therapy in a Population-Based Study of Giant Cell Arteritis (GCA), 1950-1991
A Proven, S Gabriel, M OFallon, G Hunder
Overview of study: Between 1950 and 1991, 125 residents of Olmsted County, MN were diagnosed with GCA. In this retrospective review, the cases of 120 patients were evaluated with regard to the course of their corticosteroid (CS) treatment and their frequency of adverse events. Patients in the study were treated according to the best medical judgement of their physicians, rather than a standardized CS tapering regimen.
Summary of Findings: The mean initial dose of prednisone was 60 milligrams/day. Fifteen of the patients died during their CS tapers (3 due to complications of GCA). Among the 105 surviving patients, the median time required to reach 7.5 milligrams/day was 6.4 months. Seventy-five percent of the patients had tapered their doses to 5 milligrams/day by the end of 13 months. Disease relapses occurred in 56 of the patients (53%) during the CS taper. Only half of the patients were able to discontinue their CS completely within 2 years.
Typical CS adverse events were recorded in 86% of the patients, and nearly 60% suffered 2 or more CS-related adverse events.
Editorial Comment: The results of this population-based study emphasize the frequently chronic nature of this condition, the toxicity of the standard therapy, and the importance of developing new approaches to the treatment of GCA.
**A study from another group of investigators at the Mayo Clinic investigating the SCID mouse model of GCA indicated that the anti-inflammatory effects of aspirin may complement those of CS.
Abstract 1279 Complementary Anti-Inflammatory Action of Aspirin and Corticosteroids in Giant Cell Arteritis (GCA)
H Yang, M Kaiser, JJ Goronzy, CM Weyand
Overview of study: Inflamed temporal arteries from patients with GCA were implanted into SCID mice. The mice were divided into 4 treatment groups: saline (placebo), indomethacin, dexamethasone, and aspirin. Explanted grafts were examined for mRNA of inflammatory mediators (interferon-γ, interleukin-1 and 6) after 21 days. In vitro experiments of T-cell clones and monocytes from arterial lesions were also performed.
Summary of Findings: The dexamethasone-treated group demonstrated complete inhibition of both interleukin-1 and 6, but not interferon-γ. In contrast, the group treated with aspirin showed >90% suppression of interferon-γ, but less inhibition of the interleukins. Neither indomethacin nor saline suppressed any of the inflammatory mediators measured. Findings on the comparative efficacy of aspirin and dexamethasone in the inhibition of interferon-γ and interleukin-1 and 6 were confirmed in subsequent in vitro studies of T-cell clones and monocytes from arterial lesions.
Editorial Comment: The complementary anti-inflammatory actions of aspirin and corticosteroids delineated in this animal model raise the question of using these two agents simultaneously in humans with GCA. Extrapolating from their mouse model, the investigators estimate that approximately 1500 milligrams of aspirin would be the equivalent dose in humans. Pertinent follow-up studies to this abstract are planned in humans.
**Investigators from the National Institutes of Health reported the results of a recent open-label trial in Wegeners Granulomatosis.
Abstract 1280 A Staged Approach in the Treatment of Wegeners Granulomatosis: Induction with Glucocorticoids and Daily Cyclophosphamide, Switching to Methotrexate for Remission Maintenance
C Langford and M Sneller
Overview of study: A prospective, open-label trial of daily cyclophosphamide (CYC) and glucocorticoids (GC) as an induction regimen, followed by methotrexate (MTX) for the maintenance of remission. Initial treatment consisted of prednisone 1 mg/kg/day and CYC 2 mg/kg/day. Following the achievement of remission, MTX was begun at 15 mg/week, increased to 20-25 mg/week, maintained for 2 years, and then tapered.
Summary of Findings: Thirty-one patients were enrolled. Sixteen of the patients enrolled (52%) had severe disease, as defined by the presence of significant kidney, lung, or neurological involvement. All 31 patients achieved remission on the induction regimen, with a median time of 3 months (range: 1-10). GC were discontinued at a median time of 8 months (range: 5-19). Twenty-three of the 31 patients (74%) remained on the trial protocol without relapses. Among the 8 trial failures there were 5 disease relapses, 2 withdrawals because of drug toxicity, and 1 voluntary withdrawal. No patients in the study died. At the time of the ACR, 13 patients remained in the maintenance phase of treatment (i.e., were still taking MTX).
Toxicities from the treatment regimen included leukopenia (23% - 3 patients on CYC, 4 patients on MTX); MTX pneumonitis (6%); cystitis (6%); bacterial pneumonia (6%); and herpes zoster (13%).
Editorial Comment: The results of this trial support further investigations of the shorter induction regimen of CYC/GC for patients with Wegeners granulomatosis. Long-term follow-up on this patient cohort is awaited with interest.
**Finally, a group of investigators from Johns Hopkins University and the Cleveland Clinic reported preliminary results of an open-label trial testing the use of a tumor necrosis factor inhibitor (etanercept) in Wegeners Granulomatosis.
Abstract 1467 Etanercept in Wegeners Granulomatosis: Results of an Open-Label Trial
JH Stone, DB Hellmann, ML Uhlfelder, N-M Bedocs, SL Crook, J Holbrook, GS Hoffman
Overview of study: An open-label trial of etanercept (Enbrel; Immunex Corporation; Seattle, WA) added to standard medical therapy in refractory Wegeners Granulomatosis. Interim results of the 6-month trial were reported in this abstract. In addition to standard medications for Wegeners Granulomatosis, etanercept was administered subcutaneously at a dose of 25 mg twice a week. Efficacy of this approach was assessed by a disease-specific modification of the Birmingham Vasculitis Activity Score (BVAS) known as BVAS for Wegeners Granulomatosis (Abstract #1480).
Summary of Findings: Standard medications employed at the time of entry (administered according to severity of disease) included prednisone in 17 patients (85%), cyclophosphamide in 6 (30%), MTX in 9 (45%), azathioprine in 3 (15%), and cyclosporine in 1 (5%). Etanercept was added as the only new treatment variable in 14 (70%) of the patients enrolled.
An interim analysis was conducted at a mean of 21 weeks of treatment (range: 9-28 weeks). At the time of this analysis, all 20 patients (100%) remained on etanercept. Without exception, the medication was well-tolerated. Injection site reactions, so common in trials of RA patients, were uncommon at the time of interim analysis (1 patient had 1 injection site reaction). In addition, despite the substantial use of concomitant immunosuppression, there were few infections. Only 1 major infection occurred a case of pneumococcal pneumonia in a patient with severe subglottic stenosis and concomitant treatment with cyclophosphamide and corticosteroids. This patients infection was uncomplicated, responding promptly to antibiotics. Etanercept was continued. Other adverse events, including all-cause hospitalizations, cytopenias, and hepato-renal dysfunction, did not occur with unusual frequencies, and were not attributed to etanercept.
In terms of efficacy, the results at the interim analysis were as follows. The mean BVAS for Wegeners Granulomatosis at entry was 3.7 (range 1-8)(1 point scored for a minor disease manifestation, 3 for a major). Upon follow-up at a mean of 21 weeks of treatment, the mean BVAS for Wegeners Granulomatosis score was 0.9 (range 0-4)(P < 0.001; 95% confidence interval: 2.1-3.5). Comparing the first to the last BVAS for Wegeners Granulomatosis assessment, 19 (95%) of the patients had improved, including 14 of 14 in whom etanercept was added as the only new treatment variable. Sixteen of the 20 patients (80%) achieved BVAS for Wegeners Granulomatosis scores of zero at some point during follow-up. Improvement in constitutional symptoms was nearly universal among patients in this trial.
Although responses to the treatment regimen (which included both etanercept and standard of care medications) were the rule, limited disease flares and persistent minor features of active disease were also common. Seven minor disease flares, principally upper respiratory tract manifestations, had occurred at the time of the interim analysis. One major disease flare occurred (a case of glomerulonephritis).
Editorial Comment: The results of this unrandomized trial to date appear to confirm the safety of etanercept in Wegeners Granulomatosis (even when the TNF inhibitor is combined with immunosuppressive medications), and provide sound rationale for the conduct of a randomized, placebo-controlled trial of etanercept in this disease. A randomized trial of etanercept in Wegeners Granulomatosis will begin enrollment at 8 American centers early in the year 2000.