333. Treatment of ANCA-associated vasculitis in the elderly: a single-centre UK experience.

Objective: Despite the high incidence of ANCA-associated vasculitis (AAV) in elderly patients, clinical trials have usually not focused on this patient group and long-term outcome data is lacking. Here, we evaluate the treatment and outcomes in a single-centre cohort of elderly patients with AAV. Methods:Retrospective data were collected on all hospitalised patients over the age of 74 years who underwent induction treatment for AAV between 2014-2021 at Addenbrookes Hospital (Cambridge, UK). Baseline characteristics and outcomes were assessed, and multivariate Cox models were derived to determine factors associated with infection and mortality. Results: 85 patients were included with a median follow-up 21 months (IQR 12-44). Sex distribution was equal. Median age was 79 years (range 75–93); the majority of whom (79%) had newly-diagnosed disease, with predominantly MPA (66.6%) and MPO-ANCA (58.8%). Renal involvement was common: median serum creatinine at presentation was 180 µmol/l; 20% were dialysis-dependent. Induction regimen [cyclophosphamide alone (25.8%), rituximab 500mg x2 (low-dose, 22.3%), rituximab 1g x2 (standard dose, 40%) or rituximab-cyclophosphamide combination (11.9%)] was age-dependent with interaction between age and induction regimen for both infection and mortality outcomes. Younger patients tended to receive the rituximab-cyclophosphamide combination, whereas older patients were more likely to receive low-dose rituximab. 76% of the patients received IV methylprednisolone, 47.1% experienced a serious infection at a median of 12 months [IQR 3-29] and 32.9% died with median time to death of 22 months [IQR 11-43]. After including the interaction effect between age and treatment regimen in the adjusted Cox regression models, methylprednisolone use remained positively associated with increased risk of infection [HR 4.85, CI 1.22-19.2] but negatively associated with mortality [HR 0.23 CI 0.05-0.99]. There was no interaction between methylprednisolone and age or treatment regimen, respectively, in the applied statistical models. Conclusion: Use of intravenous methylprednisolone was common in our unit and was associated with serious infection. Avoidance of high-dose intravenous corticosteroid may reduce the high rates of infections that complicate contemporary induction regimens for AAV, especially in elderly patients. Lack of power due to small sample size limited the conclusions which could be awn regarding the outcomes related to specific induction regimen. These data will be available later. Disclosures: Non