Background:Glucocorticoids (GCs) are a milestone of Rheumatoid Arthritis (RA) treatment; EULAR recommendations on the management of medium to high dose glucocorticoids remind to evaluate comorbidities and risk factors for adverse events when planning GCs treatment. Tofacitinib and Baricitinib are Janus kinases inhibitors (JAKi) registered for RA treatment. About 60% of patients enrolled in randomized clinical trials with JAKi were co-treated with GCs; however, little is known about tapering and percentage of withdrawal both in clinical trials and real life.Objectives:To evaluate the steroid-sparing effect of JAKi in patients with RA.Methods:We prospectively enrolled consecutive adult patients with RA starting JAKi. At baseline and after 4, 12 and 24 weeks we calculated C-Reactive Protein based Disease Activity score 28 (DAS28CRP). Daily dose of GCs was recorded at each visit as prednisone (PDN)-equivalent dose. Data are expressed as median (IQR). Continuous variables were compared by Mann Whitney test while dichotomous ones by Chi-square test. P values < 0.05 were considered statistically significant.Results:Between January 2018 and January 2020, 108 patients started JAKi: 67 patients Baricitinib, 41 patients Tofacitinib. The analysis was restricted to 64 RA patients (50 female, 14 male) who had at least 6 months of follow-up. Table 1 shows the demographic, clinical and clinimetric characteristics of the cohort. Patients treated with baricitinib and tofacitinib were comparable for age, disease duration, PDN dose and previous number of csDMARDS and bDMARDS; 30 patients (47.6%) were treated with JAKi in monotherapy. At baseline, the median daily PDN dose was 5 (7.25) mg; after 4, 12 and 24 weeks the median daily dose significant decreased to 5 (6.25) mg, 2.5 (5) mg and 0 (5) mg, respectively (p<0.0001). The percentage of patients treated with GC decreased from 81.5% to 63.5% at week 4, and to 48.4% at week 12 and 24. After 4, 12 and 24 weeks we detected a significant reduction of DAS28 (p<0.00001 compared to baseline). A similar percentage of patients who withdrew PDN compared to those who were still on PDN achieved remission after 12 and 24 months. Similarly, the reduction in DAS28 was comparable between the two groups at week 4 [4.8 (4.2) in those who withdrew vs 4.1 (1) in those who did not] at week 12 [4.8(1.6) for both] and at week 24 [3.7 (1.4) in those who withdrew vs 2.3 (0.7) in those who did not].Table 1.Demographic, clinic and clinimetric characteristics of the 64 patientsBaricitinib= 41Tofacitinib=23P valueFemale:male33:817:6P=nsAge, median (IQR), years58 (15)66 (14.5)P=nsDisease duration, median (IQR), months144 (144)150 (120)P=nsN° of previous csDMARDS3 (3)3 (1)P= nsN° of previous bDMARDs2 (3)1 (3)P= nsDAS28CRPat baseline4.7 (1.6)4.8 (2)P=nsPDN dose at baseline, median (IQR), mg5 (7.5)5 (5)P=nsPDN dose at 4 weeks, median (IQR), mg5 (7.5)5 (5)P=nsPDN dose at 12 weeks, median (IQR), mg2 (5)2.5 (4.7)P=nsPDN dose at 24 weeks, median (IQR), mg0 (5)2.5 (0)P=nsIQR: interquartile range; DAS28CRP: Disease Activity Score 28 using C-Reactive Protein, csDMARDS: conventional synthetic Disease Modyfing anti-rheumatics drugs, bDMARDS: biotheconological Disease Modifying anti-rheumatics drugs PDN= prednisoneConclusion:The rapid reduction of disease activity determined by JAK inhibitors allows a fast tapering of PDN, as suggested by the last EULAR recommendations for the management of RA.Disclosure of Interests:Ilaria Duca: None declared, Francesca Romana Spinelli Grant/research support from: Pfizer, Speakers bureau: Lilly, BMS, Celgene, Fulvia Ceccarelli: None declared, Cristina Garufi: None declared, Silvia Mancuso: None declared, cristiano alessandri Grant/research support from: Pfizer, Rossana Scrivo: None declared, Roberta Priori: None declared, Valeria Riccieri: None declared, Manuela Di Franco: None declared, fabrizio conti Speakers bureau: BMS, Lilly, Abbvie, Pfizer, Sanofi