ACR Meeting Abstracts

ACR Meeting Abstracts

Abstract Number: 1541

Does Leflunomide Have a Place As Remission Maintenance Therapy in ANCA-Associated Vasculitis? A Bayesian Network Meta-Analysis with Hypothesis Driven Sensitivity Analyses to Adjust for Potential Biases

Glen S. Hazlewood1, Claudia Metzler2, George A. Tomlinson3, Wolfgang L. Gross4, Brian M. Feldman5, Loic Guillevin6 and Christian Pagnoux7, 1Medicine, University of Toronto, Toronto, ON, Canada, 2University of Lubeck, Bad Branstedy, Germany, 3University of Toronto, Toronto, ON, Canada, 4Dept of Clinical Rheumatology, Medical University at Lubeck, Lubeck, Germany, 5Division of Rheumatology, The Hospital for Sick Children, Toronto, ON, Canada, 6Internal Medicine, Cochin University Hospital, Paris, France, 7Rheumatology, Mount Sinai Hospital, Toronto, Canada, Toronto, ON, Canada

Meeting: 2012 ACR/ARHP Annual Meeting

Keywords: ,

Session Information

Title: Vasculitis

Session Type: Abstract Submissions (ACR)

Background/Purpose: Primary: To determine the relative treatment effects of maintenance therapy in adult patients with ANCA-associated vasculitis who have achieved remission, using a Bayesian network meta-analysis of randomized controlled trials (RCT). Secondary: To model the impact of a priorihypotheses about potential biases in a RCT comparing leflunomide (LEF) to methotrexate (MTX) on the relative treatment effects.

Methods:  Study selection: RCTs identified from an existing systematic review and updated PUBMED and MEDLINE searches comparing at least 2 of the following maintenance agents: MTX, LEF, azathioprine (AZA) or mycophenolate mofetil (MMF). Population: Adult patients (age>18) with ANCA-associated vasculitis who have achieved clinical remission. Outcome: Relapse-free survival. Data analysis: A Bayesian arms-based fixed-effects network meta-analysis was performed using hazard ratio data. Sensitivity analyses were performed by down-weighting the effect of LEF in the LEF-MTX RCT because of the early trial termination (using a published meta-analysis of the impact of early termination) and by modeling the removal of early methotrexate relapses, as the initial dose titration of MTX in this trial was slow.

Results:   Three trials were available (LEF-MTX; MTX-AZA; AZA-MMF). In the primary analysis, LEF was superior to MMF (HR:0.26 [0.08, 0.87]) and showed a trend towards superiority to AZA (HR:0.43 [0.14, 1.36]) and MTX (HR:0.47 [0.18, 1.22]). The probability that each treatment was the best was: LEF 90%, AZA 6%, MTX 4% and MMF <1%. The probability that LEF was the best decreased to 74% after the treatment effect for LEF/MTX was down-weighted for early trial termination in the LEF-MTX RCT. LEF remained the highest ranked treatment unless >6/13 of the initial MTX relapses (all relapses within the first 7 months) were censored. 

Conclusion:  Based on indirect evidence, there is a high probability that LEF is an effective maintenance therapy for ANCA-associated vasculitis after adjusting for potential study biases. Further RCTs of LEF should be considered to provide direct evidence.

Disclosure:

G. S. Hazlewood,
None;

C. Metzler,

Sanofi-Aventis Pharmaceutical,

9;

G. A. Tomlinson,
None;

W. L. Gross,
None;

B. M. Feldman,
None;

L. Guillevin,
None;

C. Pagnoux,
None.

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/does-leflunomide-have-a-place-as-remission-maintenance-therapy-in-anca-associated-vasculitis-a-bayesian-network-meta-analysis-with-hypothesis-driven-sensitivity-analyses-to-adjust-for-potential-biase/