Background TNF-inhibitors (TNFi) have brought new perspectives for AS patients to control disease activity and improve quality of life. However, some concerns still exist about their safety, especially in long term use.
Objectives To evaluate persistence of AS patients on anti-TNF therapy, to identify the most common reasons of TNFi discontinuation, and to find out any difference in survival curves among these drugs
Methods AS patients with TNFi in Slovakia were retrospectively analysed since March 3rd 2003 until December 31st 2013. Duration of anti-TNF therapy was calculated on patient-years (PYs). Disease activity was evaluated by BASDAI, ESR, CRP and global patient assessment of disease activity. Adherence to therapy was evaluated using Kaplan-Meier analysis and results were expressed as cumulative survival. Logrank analysis was used to compare survival curves among TNFi. Hazard ratio (HR) for drug discontinuation was calculated for each individual TNFi. Cox proportional-hazard regression analysis was used for evaluation of risk factors of treatment discontinuation. Adverse events leading to drug discontinuation were closely monitored. Statistical analyses were performed using MedCalc Software, version 12.5, Ostend, Belgium.
Results 385 patients with AS on TNFi were analyzed, of them 327 men (85%) and 58 women (15%). Mean age of patients at the time of initiation of TNFi was 39.0 (±10.7) years. Total time on anti-TNF therapy was calculated on 1129 patient-years. 81 patients were exposed to infliximab for 274 PYs, 102 patients to etanercept for 394 PYs, 155 patients to adalimumab for 390 PYs, and 47 patients to golimumab for 72 PYs. Baseline disease activity was high with mean BASDAI 6.1 (±1.3), ESR 53.7 (±27.6) mm/h, CRP 37.3 (±28.5) mg/L, and significantly dropped during the treatment. 98.1% patients remained on therapy with TNFi in 3rd month, 87.1% in 1st year, 74.5% in 2nd year, 59.9% in 3rd year, and 47.8% in 4th year. Patients with infliximab remained on anti-TNF therapy significantly shorter (median 50.0 months), than patients with etanercept or adalimumab (both equally 78.0 months, p=0.0257). HR for discontinuation of infliximab was 1.65-times higher than for etanercept, and 1.8 times higher than for adalimumab. From baseline parameters only ESR (p=0.0004, 95% CI 1.0049-1.0172) and CRP (p=0.0005, 95% CI 1.0041-1.0146) influenced drug survival in AS patients, other parameters like age, gender, disease duration, HLA-B27+, BASDAI and BASFI had no influence. The most common reason of drug discontinuation was lost of efficacy (43.2%), following adverse events (24.8%), extra-articular manifestations (16%), incompliance (5.6%), death (4.8%), remission (1.6%) and other reasons (4%). Patients with infliximab had to stop their treatment mostly due to adverse events (38.7%), with etanercept due to worsening of extra-articular manifestations (34.2%) and with adalimumab due to loss of efficacy (61.1%).
Conclusions Patients with etanercept, following with adalimumab tend to remain longer on their therapy, compared to infliximab. The most common reason of drug discontinuation was loss of efficacy following adverse events. Type of drug discontinuations might reflect the character of TNFi used.
Disclosure of Interest None declared