P203 Incidence, characteristics and determinants of severe infections in patients with anti-neutrophil cytoplasm antibody-associated vasculitis receiving rituximab therapy: a single centre study

Abstract

Abstract Background/Aims The efficacy of rituximab in managing anti-neutrophil cytoplasm antibody (ANCA)-associated vasculitis (AAV) is well-established; however its associated complications, including risk of developing serious infections, are less well characterised. Identifying infection risk factors may help to improve care of patients with AAV receiving rituximab. We characterised severe infections in a cohort of patients over a three-year period and identified factors affecting their risk of developing severe infections. Methods Electronic patient records were interrogated over a retrospective period (August 2016-August 2019) to compile baseline data and episodes of severe infection. Differences between groups were determined using appropriate parametric or non-parametric methods. Risk factors for severe infections were identified through multivariate binomial logistic regression analysis. Variables with an event count of ≥ 5 and a p-value ≤0.1 at univariate level were entered into the final multivariate analysis, where p ≤ 0.05 was taken as significant. Results Fifty patients were included. 24 (48%) were male, and 26 (52%) female. Average age was 60 years (range 25-90). 36 (72%) had GPA, 2 (4%) MPA, 1 (2%) EGPA, and 11 (22%) had overlap or undefined AAV. 14 (28%) patients developed at least one severe infection (≥grade 3, CTCAE criteria), giving an incidence of 15.4 severe infections per 100 person-years. The 18 severe infection events included 10 (56%) respiratory tract, 5 (28%) sepsis/neutropenic sepsis, 1 (6%) cellulitis, 1 (6%) complicated UTI and 1 (6%) recurrent wound infection. Patients who developed severe infections had lower immunoglobulin levels (IgG<6g/L, 36% vs 6%, p = 0.009), concomitant COPD (21% vs 3%, p = 0.029), and lower rates of concomitant co-trimoxazole use (7% vs 44%, p = 0.012) compared to patients not developing severe infections. Regression analysis of demographic, baseline blood markers and concomitant therapy data was performed to identify risk factors for developing severe infections (Table.1). Hypogammaglobulinaemia increased risk of infection (OR = 8.782, 95%CI=1.194-64.605, p = 0.033), while co-trimoxazole decreased risk of infection (OR = 0.096, 95%CI=0.009-0.996, p = 0.050). P203 Table 1:Univariate analysisMultivariate analysisVariablesOR95% CIP valueOR95% CIP valueAge (years)1.0320.994-1.0720.105Sex (male)0.3200.084-1.2130.0940.2860.057-1.4350.128Creatinine1.0050.987-1.0240.577eGFR0.9750.945-1.0060.109CRP1.0010.989-1.0120.899ESR1.0030.960-1.0460.908Neutrophils0.9730.789-1.2010.802WBC0.9590.780-1.1800.694Lymphocytes0.8700.420-1.8050.709CD190.4780.000-2000.9670.862CD19%0.9870.906-1.0750.767CD30.3140.055-1.7750.190CD3%0.9960.941-1.0550.902CD40.3920.034-4.4870.451CD4%1.0110.959-1.0660.675CD80.0530.001-4.7440.200CD8%0.9850.914-1.0610.689IgG0.8900.749-1.0570.185Hypogammaglobulinemic?8.6111.423-52.0910.0198.7821.194-64.6050.033IgM0.8360.350-1.9970.686IgA1.0890.787-1.5070.607BMI0.9460.858-1.0430.262COPD*9.5450.899-101.3380.061DM0.8330.147-4.7230.837Hypertension0.4330.082-2.2910.325CV disease0.000-0.999AKI0.4580.088-2.3780.353Latent TB0.3710.022-6.3830.495DI2.6920.157-46.2640.495Hypothyroidism0.000-1.000RA5.8330.484-70.2440.165OP2.6920.157-46.2640.495OA0.000-1.000SLE0.000-1.000AIH0.000-1.000CKD0.000-1.000Co-trimoxazole use0.0960.011-0.8150.0320.0960.009-0.9960.050Cumulative dose of RTX0.6690.392-1.1410.140Did patient have prior RTX before Aug 2016?0.6940.194-2.4870.575Yearly influenza vaccine between 2016-2019 (patients for whom vaccine records exist on EPR)1.3850.188-16.2770.796Average length of follow-up (days)1.0010.999-1.0030.360Prednisolone cumulative dose1.0001.000-1.0000.843Prednisolone average dose0.9920.991-1.0800.852MTX cumulative dose1.0000.999-1.0010.701MTX average dose1.0200.696-1.4960.918HCQ cumulative dose1.0001.000-1.0000.331HCQ average dose0.9850.964-1.0060.162LFN cumulative dose1.0001.000-1.0000.855LFN average dose0.8510.605-1.1990.357MMF cumulative dose1.0010.999-1.0020.263MMF average dose3.9960.402-39.7260.237AZT cumulative dose1.0001.000-1.0000.372AZT average dose0.9440.878-1.0140.116*too few events for progression to multivariate analysis. Conclusion The incidence of severe infections in patients with AAV receiving rituximab is significant. Our results support the monitoring of IgG levels to identify patients who may be more susceptible to infection, as well as the prescription of prophylactic co-trimoxazole to reduce overall severe infection risk. Disclosure F. Dernie: None. N. Ahmad: None. R. Luqmani: None. J. David: None.