AB0836 Drug retention rate of the first tnf inhibitor in radiographic and non radiographic axial spondyloarthritis: data from a multicenter study. (12th June 2018)

• Record Type: Journal Article
• Authors: Lopalco, G.; Venerito, V.; Cantarini, L.; Emmi, G.; Salaffi, F.; Tafuri, S.; Bianchi, F.P.; Di Carlo, M.; Gentileschi, S.; Di Scala, G.; Nivuori, M.; Perniola, S.; Fornaro, M.; Anelli, M.G.; Cacciapaglia, F.; Scioscia, C.; Fabiani, C.; Galeazzi, M.; Lapadula, G.; Iannone, F.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 1546
• Page End: 1546
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.3069
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Good survival rates of TNF inhibitors (TNFi) have been shown in patients affected with axial Spondyloarthritis (axSpA) treated in a real-life setting, irrespective of administered anti-TNF agent. 1 Although the use of these drugs in patients with non radiographic axSpA (nr-axSpA) remains topic of debate, 2 several RCTs support their employment in these patients. 3 Objectives: To assess the retention rate of the first TNF inhibitor between axSpA and nr-axSpA patients. Additional aims: i) to evaluate any difference in persistence rates between anti-TNF monoclonal antibodies and etanercept (ETN) ii) to assess any impact of clinical, therapeutic and demographic features as well as radiographic findings on drug survival. Methods: We retrospectively assessed 221 patients with axSpA, all fulfilling the ASAS criteria, who underwent first line therapy with TNFi from January 1st 2012 to September 30th 2016. Clinical, therapeutic and demographic features as well as radiographic findings were recorded at baseline and at the moment of therapy discontinuation. 126/221 patients (57.01%) were treated with Adalimumab, 45/221 with ETN, 22/221 with Infliximab, 20/221 (9.05%) with Golimumab, whereas 8/221 (3.62%) with Certolizumab Pegol. Drug retention rates were analysed using Kaplan-Meier curves; log-rank test was performed to demonstrate differences in the survival function. Cox regression models were used to estimate the inference of several clinical and disease characteristics on drug discontinuation. Goodness of fit of the final model was assessed comparing Cox-Snell residuals to Nelson-Alen cumulative hazard function. Results: Drug survival in first line therapy was significantly lower in patients who had nr-axSpA than in those with radiographic sacroiliitis (p=0.005, figure 1) whereas survival rate did not differ significantly between patients treated with ETN and anti-TNFα monoclonal antibodies (p=0.057). Multivariate Cox model shown that nr-axSpA (HR 1.90), higher BMI (HR 1.18), higher HAQ, (HR 1.82) and higher BASDAI (HR 1.25), were predictors of drug discontinuation, Nelson–Alen hazard function following very closely Cox-Snell residuals for drug persistence in first line therapy showed that the final model fitted well the data, except for large values of time (figure 2). Conclusions: Effectiveness of TNFi, estimated by drug survival, seems to be lower in patients with nr-AxSpA than those affected with axSpA. The reason of these findings remain to be elucidated. However, a possible explanation may be searched in the limit of the classification criteria for nr-axSpA. In addition overweight and high disease activity negatively impact the persistence on first line anti-TNFα treatment in axSpA patients in real life setting. References: [1] Favalli EG, et al. Arthritis Care Res (Hoboken). 2017Jun;69(6):867–874. [2] Vidal C, et al. Joint Bone Spine. doi:10.1016/j.jbspin.2017.08.005 [3] . F Sieper Jet al. Ann Rheum Dis72(6):815–822. doi:10.1136/annrheumdis-2012–201766 Disclosure of Interest: None declared