34. Post-induction ANCA Titer Does Not Predict Relapse, Mortality, or Renal Outcomes: A Target Trial Emulation Study

Background: ANCA-associated vasculitis (AAV) is associated with disease relapse as well as increased risk of end-stage renal disease (ESRD) and death. In most cases, circulating ANCA targeting proteinase-3 (PR3) or myeloperoxidase (MPO) are present and may be pathogenic. Studies of the association of post-treatment ANCA titers with future outcomes have yielded conflicting results. We evaluated the association of achieving a negative ANCA titer during the first year of treatment with risk of relapse, ESRD, and death.

Methods: Cases were obtained from a consecutive inception cohort of AAV patients who received care at a large multi-hospital system between 2002-2019. All cases were PR3- or MPO-ANCA+. Mortality data were obtained from the National Death Index and ESRD status from the United States Renal Data System. Data regarding baseline covariates and relapses during follow up were obtained from the electronic health record. Missing data was accounted for using multiple imputation. To address confounding and immortal-time biases, we performed a target trial emulation study to examine the association of post-induction ANCA titer with risk of ESRD or death and relapse using a previously described cloning, censoring and weighting approach. We designed a hypothetical trial with two management strategies: “achieve a negative titer” or “does not achieve negative titer” within 180 days of induction. “Clones” of each patient were included in each hypothetical treatment arm and censored when they violated the assigned management strategy. The composite outcome – risk of ESRD or death within five years – and secondary outcome of relapse was estimated using Cox models after accounting for informative censoring using inverse-probability-of-censoring weighting and adjusting for baseline covariates.

Results: The study included 674 patients (mean age: 60 years; 59% female; 87% white) with 46.7 + 19.1 months of mean follow up. The majority were MPO-ANCA+ (68%) and had renal manifestations (64%). Rituximab (RTX)-based induction strategies were used in 52%. 28% achieved a negative titer within one year of induction. In the target trial (Table 1), the HR for the primary outcome of ESRD or death was 0.95 (95%CI 0.68-1.34) in the group that achieved a negative titer compared to the group that did not. In analyses stratified by remission induction strategy, the HR for ESRD or death was 1.05 (95%CI 0.67-1.66) in RTX-based and 0.96 (95%CI 0.54-1.72) in cyclophosphamide (CYC)-based regimen users when comparing those who achieved a negative titer vs those who did not. The HR for ESRD or death was 0.93 (95%CI 0.48-1.80) in the PR3-ANCA+ and 0.93 (95%CI 0.63-1.38) in the MPO-ANCA+ groups that achieved a negative titer. There was no association of achieving a negative ANCA titer with risk of relapse (HR 0.70, 95% CI 0.36-1.34). Our results persisted in analyses examining the association of achieving an ANCA titer within 365 days with these outcomes.

Conclusion: In this target trial using a large cohort of AAV patients, achieving a negative ANCA titer within one year of induction was not associated with improved outcomes. This finding was seen in both RTX- and CYC-treated patients. These findings suggest that post-induction ANCA titers have limited ability to predict mortality and ESRD outcomes in AAV.

Disclosures: ZSW is supported by NIH/NIAMS [K23AR073334]. ZSW reports research support from Bristol-Myers Squibb and Principia/Sanofi and consulting fees from Viela Bio and Medpace.