FRI0079 LONG-TERM CLINICAL EFFICACY OF CYCLING VS SWITCHING TO A NON-TNF INHIBITOR IN RA PATIENTS WHO FAILED TO A FIRST TNF INHIBITOR

Abstract

Background In rheumatoid arthritis (RA) patients (pts) after discontinuation of tumour necrosis factor α inhibitors (TNFi) physicians can choose an alternative TNFi or a biological agent with another mode of action (non-TNFi), however little guidance on choosing one or another exists. Objectives To compare the long-term clinical response, survival and safety of TNFi versus non-TNFi after discontinuing a previous TNFi in pts with RA, both in the global cohort and in the subpopulations stratified by reason of discontinuation of 1st TNFi. Methods Observational study including 127 pts from La Paz University Hospital biological RA registry, who discontinued a first TNFi between 1999 and 2016 and subsequently were treated with a second biologic. Disease activity was assessed by DAS28 at the beginning of the second biologic and at 6 (m-6), 12 (m-12) and 24 months (m-24) follow-up. Primary outcome was the proportion of pts with good or moderate EULAR response (E-Resp). Sensitivity analysis to evaluate clinical response according to the reasons for discontinuation of the first TNFi was performed too. Pts were classified into primary and secondary failure based on the characteristics of the non-response and if possible on the measure of drug and anti-drug antibodies, which were measured by ELISA after 6 months of the first biologic treatment and before switching. Mann–Whitney U test and Fisher’s exact test were used to test statistical differences. Factors associated with clinical response were assessed using univariable and multivariable logistic regression analysis. Drug retention was compared using Cox proportional hazards models. Results Seventy-seven (61%) pts received a TNFi and 50 (39%) a non-TNFi as second biologic therapy. Mean age was 56 years, 84% were women and 58% were also treated with methotrexate. At baseline, no significant differences between groups were found, except for significantly higher CRP levels in non-TNFi group (Table 1). No statistical differences were observed in E-resp between groups at m-6 and m-12. Nevertheless, at m-24 more pts achieved E-resp in non-TNFi group (80% vs 52%; p=0.001, Figure 1). Likewise, 100% (n=6) of the pts who achieve remission with the second biologic were treated with a non-TNFi. In the univariable analysis, higher baseline DAS28 (OR=1.65, p=0.06), being ACPA positive (OR=3.6, p=0.02) and treatment with a non-TNFi (OR=3.53, p=0.01) were associated with E-resp at m-24. In the multivariable analysis, baseline DAS28 (OR=1.65, p=0.01) and non-TNFi treatment (OR=3.7, p=0.02) remained associated. Drug survival was similar in both groups (1.6±0.7 in TNFi vs 1.6±0.6 years in non-TNFi, p=0.5). In the subgroup of pts who dropped out due to secondary inefficacy (n=76) no differences in response to the second biologic were found, however, pts who stopped the first TNFi for other reasons (primary inefficacy and adverse events) achieved more frequently E-resp after switching to non-TNFi since the beginning of treatment (m-6: 54% in TNFi vs 86% in non-TNFi, p=0.04; m-12: 56% in TNFi vs 70% in non-TNFi, p=0.4; m-24: 47% in TNFi vs 86% in non-TNFi, p=0.04). The overall incidence of AEs was similar between TNFi and non-TNFi groups (21% in TNFi vs 14% in non-TNFi (p=0.9). Conclusion In our cohort of patients with RA and failure to a first TNFi, and mainly after primary failure and adverse events, treatment with a non-TNFi was more effective than a second TNFi at 24 months after switch, without significant differences in the survival of the two treatments. In the subgroup of patients who failed the first TNFi due to secondary inefficacy, both treatment options were equally effective. These data reflect how the reason for discontinuation of a first biological can guide the response to a second treatment. Acknowledgement to Chamaida Plasencia, Victoria Navarro-Compan and Alejandro Balsa Disclosure of Interests Patricia Bogas Grant/research support from: non restricted grant from Sanofi, Chamaida Plasencia Speakers bureau: Pfizer, MSD, Victoria Navarro-Compan: None declared, Laura Nuno: None declared, Irene Monjo: None declared, Carolina Tornero: None declared, Juan Molina Collada: None declared, Alejandro Balsa Grant/research support from: Abbvie, Pfizer, Novartis, BMS, Nordic, Sanofi, Consultant for: Abbvie, Pfizer, Novartis, BMS, Nordic, Sanofi, Sandoz, Lilly, Paid instructor for: Pfizer, Speakers bureau: Pfizer, Novartis, UCB, Nordic, Sanofi, Sandoz, Lilly