Background Observational studies have already reported the risk of severe infections in rheumatoid arthritis (RA) treated with tocilizumab but in limited samples.
Objectives The aim of this study was to investigate the predictive risk factors of severe infections in the largest European registry of patients treated with tocilizumab for RA.
Methods 1491 RA patients included in the French REGistry –RoAcTEmra (REGATE) were analysed to calculate incidence rate of severe infections and to identify independent factors associated with severe infections.
Kaplan -Meier method was used to assess probability of remaining severe infection-free. Cox models were performed to identify independent factors associated with severe infections. Variables showing bivariate association with the dependent variable with a P value less than 0.15 were entered into the multivariate model. Multiple imputation was used to compensate missing data in some variables. A P value <0.05 was considered statistically significant.
Results Exposure was 2,606 person-years (mean age±SD, 56.6±13.6). 125 serious infections occurred in 122 patients (4.7/100 patient-years). Most frequent infections were lung and respiratory tract and skin/soft tissue in 35 (28%) and 32 (26%) cases respectively. Three deaths were related to one septic chock secondary to pyelonephritis, one pneumocystosis, and one Haemophilus influenzae infection. Favourable outcome was observed in most cases. Four opportunistic infections were reported: 1 Pneumocystosis, 1 tuberculosis, 1 Haemophilus influenzae infection and 1 infection related to Klebsiella pneumoniae. A pathogen was identified in 41 cases. TCZ was definitively stopped for 35 patients. The incidence of serious infection was stable over the first 3 years of follow-up in the registry. Bivariate analysis identified initial ACPA positivity as the only factor associated with a lower risk of severe infection (HR 0.55 CI95% 0.35–0.86). Factors significantly associated with a risk of severe infections were DAS28-CRP, DAS28-ESR, and number of polymorphonuclear neutrophils (PMN) at baseline. Initial PMN above 4.5 G/L (HR 1.62 CI95% 1.06–2.5, p=0.02) and negative ACPA (HR 0.63 CI95% 0.41–0.95, p=0.03) remains significantly associated with severe infections in multivariate analysis after imputation for missing data.
Conclusions The rate of severe infections in current practice is similar to that reported in clinical trials. High PMN above 4.5 G/L at baseline and negative ACPA are predictive factors of serious infection requiring in this case a tighter surveillance.
Acknowledgement The French registry REGATE was supported by an unrestricted grant of Roche Chugaï. We thank Isabelle Pane for assistance with database management.
Disclosure of Interest J. Morel Consultant for: ROCHE CHUGAI, NORDIC PHARMA, A. Constantin Consultant for: ROCHE CHUGAI, Speakers bureau: NORDIC PHARMA, E. Dernis: None declared, S. Rist Speakers bureau: ROCHE CHUGAI, R. M. Flipo Consultant for: ROCHE CHUGAI, NORDIC PHARMA, Speakers bureau: ROCHE CHUGAI, NORDIC PHARMA, T. Schaeverbeke Consultant for: ROCHE CHUGAI, NORDIC PHARMA, Speakers bureau: ROCHE CHUGAI, NORDIC PHARMA, O. Vittecoq Speakers bureau: ROCHE CHUGAI, M. Soubrier: None declared, A. Saraux Grant/research support from: ROCHE CHUGAI, Consultant for: ROCHE CHUGAI, Speakers bureau: ROCHE CHUGAI, B. Combe Consultant for: ROCHE CHUGAI, Speakers bureau: ROCHE CHUGAI, M. Dougados Consultant for: ROCHE CHUGAI, Speakers bureau: ROCHE CHUGAI, A. Pinta Employee of: ROCHE CHUGAI, G. Baron: None declared, J.-E. Gottenberg: None declared, X. Mariette: None declared, P. Ravaud: None declared, J. Sibilia Consultant for: ROCHE CHUGAI, Speakers bureau: ROCHE CHUGAI