LAUSR

OBJECTIVES Aiming to assess the best choice of second-line therapy between second-line TNF-inhibitor (TNFi) and biologics of different mode of action (BDMA-rituximab/tocilizumab/abatacept) in rheumatoid arthritis (RA) by assessing their drug-survival spanning more that 10years, after discontinuation of the first-line TNFi. METHODS This retrospective-observational drug-survival study was performed across 2-different hospitals in UK by conventional-statistics and machine-learning approach. RESULTS From a total of 435-patients, 213 [(48.9%); TNFi-n=122 (57.3%), BDMA-n=91(42.7%)] discontinued their second-line biologic [median-drug-survival: TNFi-27months (95%CI 22-32months) vs BDMA-37months (95%CI 32-52months)]. As second-line, BDMA was likely to reduce the risk of treatment-discontinuation [Hazard-ratio/HR-0.63 (95%CI 0.48-0.83)] compared to TNFi, but only in seropositive-patients [HR-0.52 (95%CI 0.38-0.73)], not in seronegative-RA. Uncovered by the survival-tree and adjusted by propensity-score, drug-survival benefit of BDMA over TNFi was not observed if the seropositive-patients were previously exposed to monoclonal-TNFi (HR-0.77, 95% CI 0.49-1.22) versus soluble TNFi (etanercept or its biosimilar) or if first-line TNFi was terminated within 23.9months of initiation (HR-0.97, 95%CI 0.56-1.68). CONCLUSION BDMA, as second-line biologic, is more likely to be sustained in seropositive-patients particularly if they were previously not exposed monoclonal TNFi. Drug-survival benefit of BDMA was not observed in seronegative-patients or if the first-line TNFi was stopped within 2 years.