TRIAL DESIGN AND PATIENTS
The trial design has been described previously, and the protocol is available with the full text of this article at NEJM.org.20 Patients were enrolled at 143 centers in an international, randomized, double-blind, double-dummy, controlled trial. The aim was to evaluate whether avacopan could replace a glucocorticoid-tapering regimen used in the treatment of ANCA-associated vasculitis.21 Avacopan (30 mg twice daily) or matching placebo was given for 52 weeks, with 8 weeks of follow-up. Prednisone or a matched placebo was given on a tapering schedule for 20 weeks (60 mg per day tapered to discontinuation by week 21) (Table S1 in the Supplementary Appendix, available at NEJM.org).
The screening period for trial eligibility was not to exceed 14 days. Eligible patients had newly diagnosed or relapsing granulomatosis with polyangiitis or microscopic polyangiitis, according to the Chapel Hill Consensus Conference definitions,2 for which treatment with cyclophosphamide or rituximab was indicated; had tested positive for antibodies to either proteinase 3 or myeloperoxidase; had an estimated glomerular filtration rate (eGFR) of at least 15 ml per minute per 1.73 m2 of body-surface area; and had at least one major or three nonmajor items or at least two renal items of hematuria and proteinuria on the Birmingham Vasculitis Activity Score (BVAS), version 3 (a composite of signs and symptoms in nine organ systems; total range, 0 to 63, with higher scores indicating more disease activity) (see the Supplementary Appendix).22 All use of immunosuppressants had to cease before trial entry. Patients were excluded if they had received more than 3 g of intravenous glucocorticoids within 4 weeks, or more than 10 mg per day of oral prednisone (or equivalent) for more than 6 weeks continuously, before screening. Complete inclusion and exclusion criteria are described in the Supplementary Appendix.
The trial was performed in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Ethics committees and institutional review boards at participating sites approved the research protocol. All the patients or a parent or guardian gave written informed consent before entry. ChemoCentryx sponsored the trial and provided trial medication. Medpace conducted the trial with guidance from ChemoCentryx. The first two authors had confidentiality agreements with the sponsor. All the authors vouch for the completeness and accuracy of the data, the complete reporting of adverse events, and the adherence of the trial to the protocol. All the authors participated with the sponsor in the design of the trial, the analysis of the data, and the writing of the manuscript.
RANDOMIZATION AND TREATMENT
Patients were randomly assigned in a 1:1 ratio to receive 30 mg of avacopan twice daily orally plus prednisone-matching placebo or a tapering oral regimen of prednisone plus avacopan-matching placebo in a double-dummy design. Randomization was stratified according to vasculitis disease status (newly diagnosed or relapsing), ANCA status (antiproteinase 3 positive or antimyeloperoxidase positive), and immunosuppressive treatment (cyclophosphamide or rituximab, assigned at the discretion of the investigators at the inception of the trial for each patient and continued throughout the trial). Randomization was performed centrally through an interactive Web-response system with the use of a minimization algorithm to maintain balance between the treatment groups.23 Glucocorticoid treatment during the screening period had to be tapered to 20 mg or less of prednisone equivalent before the patient began the trial, and this open-label glucocorticoid treatment was further tapered to discontinuation by the end of week 4 of the trial. Patients in either trial group who had a worsening of disease that involved a major item in the BVAS could be treated with rescue therapy consisting of intravenous glucocorticoids (typically 0.5 to 1 g of methylprednisolone per day for 3 days), oral glucocorticoids, or both, tapered according to the patient’s condition.
All the patients received one of three regimens: cyclophosphamide intravenously at a dose of 15 mg per kilogram of body weight up to 1.2 g on day 1 and at weeks 2, 4, 7, 10, and 13; cyclophosphamide orally at a dose of 2 mg per kilogram up to 200 mg per day for 14 weeks (see the Supplementary Appendix, including Table S2); or intravenous rituximab at a dose of 375 mg per square meter of body-surface area per week for 4 weeks. From week 15 onward, cyclophosphamide was followed by oral azathioprine at a target dose of 2 mg per kilogram per day. No rituximab was given beyond the first 4 weeks. Investigators were instructed that the use of additional glucocorticoids, not supplied as trial medication, was to be avoided as much as possible (see the Supplementary Appendix). Prophylactic therapy for infection, including for Pneumocystis jirovecii, was required according to the protocol.
Patients, trial personnel, and sponsor representatives involved in the conduct of the trial were unaware of the trial group assignments. The trial drugs and their matching placebos were provided to trial centers in identical bottles.
END POINTS
The two primary efficacy end points were clinical remission at week 26, defined as a BVAS of 0 and no receipt of glucocorticoids for 4 weeks before week 26, and sustained remission, defined as remission at week 26 and at week 52 and no receipt of glucocorticoids for 4 weeks before week 52. Patients were not considered to be in sustained remission if they had remission at week 26 but a relapse thereafter; relapse was defined as a return of vasculitis activity on the basis of at least one major BVAS item, at least three minor BVAS items, or one or two minor BVAS items for at least two consecutive trial visits.
Secondary end points were glucocorticoid-induced toxic effects according to the Glucocorticoid Toxicity Index (GTI) during the first 26 weeks (measured by both the Cumulative Worsening Score [GTI-CWS], which ranges from 0 to 410, and the Aggregate Improvement Score [GTI-AIS], which ranges from –317 to 410; on both scales, higher scores indicate greater severity of toxic effects) (Table S3)24,25; a BVAS of 0 at week 4; change from baseline in health-related quality of life, assessed with the 36-Item Short Form Health Survey (SF-36), version 2, and the EuroQoL Group 5-Dimensions 5-Level Questionnaire (EQ-5D-5L) (range, 0 to 100 for both, with higher scores indicating better quality of life)26,27; relapse (assessed in a time-to-event analysis); change from baseline in the eGFR; urinary albumin:creatinine ratio; urinary monocyte chemoattractant protein 1:creatinine ratio; and the Vasculitis Damage Index (range, 0 to 64, with higher scores indicating more damage).28 Details of the trial assessments are provided in the Supplementary Appendix.
STATISTICAL ANALYSIS
We calculated that a planned sample size of 150 patients per group would provide the trial with at least 90% power to show the noninferiority of avacopan to prednisone with respect to the primary end point of remission at week 26, assuming a noninferiority margin of −20 percentage points and an incidence of remission in the prednisone group of 60%. The primary efficacy analyses were conducted in the modified intention-to-treat population, defined as all randomly assigned patients who received at least one dose of trial medication.
With respect to the primary end points, if the lower boundary of the two-sided 95% confidence interval for the difference (avacopan minus prednisone) in the incidence of remission was greater than −20 percentage points, avacopan would be considered not inferior to prednisone. If the lower boundary of the 95% confidence interval was greater than 0.0 percentage points, avacopan would be considered superior to prednisone. Summary score estimates of the common difference in the incidences of remission were calculated with the use of inverse-variance stratum weights (see the Supplementary Appendix). For the primary end points, missing data at week 26 and week 52 were imputed as no remission. To preserve the type I error, the two primary end points were tested in a gatekeeping procedure in the following sequence: noninferiority at week 26, noninferiority at week 52, superiority at week 52, and superiority at week 26. Data were analyzed after all the patients had completed the 52-week treatment period. No interim analyses were performed.
Secondary end-point analyses of continuous variables were performed with the use of mixed-effects models for repeated measures. Least-squares means, standard errors, and confidence intervals are from models incorporating treatment group, visit, treatment-by-visit interaction, and stratification factors as covariates. Longitudinal measurements from the same patients were considered as repeated-measure units in the model. The Kaplan–Meier method was used to estimate the time to relapse of vasculitis, and the proportionality assumption was upheld. There was no prespecified plan for adjustment of confidence intervals for multiplicity of the secondary end points; point estimates and 95% confidence intervals only are presented, and no definite conclusions can be drawn from these data. Prespecified subgroup analyses were performed, but the trial was not powered to make conclusions from these data.
最近のコメント