BackgroundIn real-life setting, a greater number of elderly rheumatoid arthritis (RA) patients with impaired glomerular filtration rate (GFR) needs treatment with biologic or target synthetic disease-modifying anti-rheumatic drugs (b/tsDMARD) to achieve disease control and reduce NSAIDs intake. Long-term observational data from the real-life on the use of b/tsDMARD in these patients are scarce.ObjectivesThe aim of this study was to evaluate the retention rate of b/tsDMARD in RA patients with impaired GFR in real-life setting.MethodsData of RA patients treated with at least one b/tsDMARD were retrospectively analyzed form the national Italian GISEA registry from January 2016 to December 2021. Estimated-GFR (eGFR) was calculated with the Cockcroft-Gault equation at the time of any b/tsDMARD prescription. For the purpose of this study, patients were divided in two groups, patients with impaired GFR (eGFR ≤60) and patients with normal GFR (eGFR >60). The retention rate was calculated by the Kaplan-Meier method and compared between these two groups by a log-rank test.ResultsThe study population included 2443 treatment-line with b/tsDMARD from 1888 patients (female 80.4%, age 57±12 years, mean baseline CDAI 17±12, FR/ACPA+ 69.5%) who started a new b/tsDMARD. Disease characteristics are shown in Table 1. 288 treatments with b/tsDMARD were started in patients with impaired eGFR and 2155 in patients with normal eGFR. Compared to patients with eGFR >60, patients with eGFR ≤60 showed higher HAQ-DI (1.3±0.8 vs 1±0.8, p<0.001) at the start of b/tsDMARD treatment. Glucocorticoids were more prescribed in patients with impaired eGFR (80.2% vs 72.8%, p<0.01), while csDMARDs were more prescribed in association with b/tsDMARD in patients with normal eGFR (83.1% vs 76.4%, p<0.01). Of note, CTLA4-Ig treatment was more prescribed in patients with impaired eGFR (26% vs 17.1%, p<0.05), while no difference in b/tsDMARD prescription was observed for other mechanism of actions. Drug survival was similar between RA patients with impaired eGFR [58.2%, mean survival time 35 months (CI95% 31-39)]and RA patients with normal eGFR [55%, mean survival time 34.4 months (CI95% 33-36), log rank: 0.88] (Figure 1). Cox regression model adjusted for age, sex and b/tsDMARD showed no impact of eGFR on drug survival [HR: 0.9 (CI95%: 0.7-1.2).ConclusionOur data show that impaired eGFR seems to not influence the persistence of b/tsDMARD treatment in RA patients.Disclosure of InterestsNone declared