AB0226 PREDICTORS OF response AFTER CESSATION OF JAKI THERAPY IN PATIENTS WITH RA

Abstract

BackgroundIn a previous study based on data from the Belgian register “Tool for Administrative Reimbursement Drug Information Sharing” (TARDIS)[1], IL6 inhibitors showed a better clinical response after JAKi cessation compared to other b/tsDMARDs. However, considerable baseline differences in demographic and clinical variables, and previous exposure to advanced therapies were found between treatment groups, which could have influenced our results.ObjectivesWe aimed to find predictors of short-term response after cessation of JAKi therapy in patients with RA.MethodsPatients were selected from TARDIS for this analysis if they had stopped JAKi therapy and initiated a subsequent therapy between register inception and December 2021. Patients were grouped per subsequent therapy: TNFi, Abatacept, IL6 inhibition or JAKi. Rituximab was excluded as patients are reimbursed on RA flare in Belgium.Linear and logistic regression models were constructed to establish predictors of clinical response. DAS28 change between baseline and first follow-up, DAS28 remission (DAS28<2.6) and DAS28 low disease activity (LDA, DAS28≤3.2) at first follow-up were taken as dependent variables. Following variables were inserted in the models: age, disease duration, b/tsDMARD naive, treatment group and baseline values for HAQ, swollen/tender joint count, PGA, ESR, and CRP. Final models included only variables reaching statistical significance of 0.05 (by backwards elimination).ResultsIn total, 193 RA patients, who had stopped JAKi therapy and had complete baseline and follow-up data, could be included. Table 1 shows baseline demographic and clinical characteristics. TNFi, Abatacept, IL6 inhibition or JAKi therapy was initiated in 33% (63/193), 11% (20/193), 18% (34/193) and 39% (76/193) of patients respectively. Remission and LDA were reached in 42% (81/193) and 63% (122/193) of patients respectively. Common predictors for achieving remission and LDA were baseline HAQ and not receiving abatacept (Figure 1). Variables positively associated with a higher DAS28 change were baseline PGA (Relative Risk (RR) 0.02 (0.01, 0.03)), TJC (0.06 (0.01, 0.12), SJC (RR 0.10 (0.03, 0.16)) and CRP (RR 0.02(0.01, 0.03)), while baseline HAQ (RR -0.60 (-0.88, -0.31) and not receiving abatacept (RR -0.70 (-1.25, -0.15)) were negatively correlated with DAS28 change in the adjusted multivariate linear regression model.Table 1.Patient characteristics at the start of next treatment after failing the first JAKi therapyTotal populationNumber193Age (mean ±SD, years)57 ±13Disease duration (mean ±SD, years)10 ±8ESR (mean ±SD, mm/hour)24 ±18CRP (mean ±SD, mg/L)11 ±18SJC28 (mean ±SD)6 ±5TJC28 (mean ±SD)8 ±5DAS28 (mean ±SD, years)4.6 ±1.1HAQ-DI (mean ±SD, 0-3)*1.3 ±0.7PGA (mean ±SD, 0-100)64 ±20Used JAKi as first line advanced therapy76/193 (39%)Legend: Number given are mean ± SD or number, proportion. TNFi = tumour necrosis factor inhibitor, JAKi = Janus Kinase inhibitor, HAQ= health assessment questionnaire, PGA= Patient Global assessment; CRP= C-reactive protein; ESR= erythrocyte sedimentation rate; TJC= tender joint count; SJC= Swollen joint Count; DAS28 = disease activity score based on the 28joints. *Missing HAQ(0-3) scores were imputed by regression using age and HAQ(0-60) scores, the sum of the scores (0-3) on all (20) individual HAQ questions.Figure 1.Predictors of remission or LDAConclusionThis explorative analysis reveals that after adjustment, patients who have worse functionality and receive abatacept compared to other b/ts DMARD show worse clinical outcomes in the short-term. This treatment effect should be further explored, but caution is warranted as the sample size was relatively small and indication bias cannot be excluded.Reference[1] De Cock et al. ARD. Volume 81, supplement 1, year 2022, page 626 (POS0694)Acknowledgementson behalf of the Royal Belgian Society for Rheumatology.Disclosure of InterestsNone Declared.