Objectives To determine drug trough serum levels and anti-drug antibodies (Ab) in rheumatoid arthritis (RA) or polyarticular psoriatic arthritis (PsA) patients in clinical remission (CR) or low disease activity (LDA) receiving TNF inhibitors [adalimumab (ADA), etanercept (ETN) and infliximab (IFX)], to correlate trough serum levels and Ab with disease flare after a 1-year follow-up, and to determine predictive factors of disease flare.
Methods Prospective, multicentre study of RA and PsA patients attended in 3 hospital outpatient clinics (Catalonia, Spain) treated with ADA, ETN or IFX for ≥3 months in CR or LDA measured by DAS28-ESR at ≥2 consecutive visits. Ab and trough serum levels (ELISA Kit Promonitor®, Progenika SA, Spain) were determined at 0, 4, 8 and 12 months. Disease flare was defined as DAS28>3.2 and a delta DAS28 >0.6 during the 1 year follow-up compared with visit 0. Variables collected: demographic data; disease activity, diagnosis; disease duration; biologic drug; reduced dose, and concomitant csDMARD therapy. Differences between patients with and without disease flare were studied by univariate analysis. Multivariate Cox regression analysis was used to establish associations between baseline variables and disease flares.
Results 187 patients (RA 103 [57%], PsA 81 [43%]), 66% female, mean age 57±12 years, were included. 138 (74%) patients were in CR, 49 (26%) with LDA, 69 ADA, 83 ETN, and 35 IFX. Mean treatment duration was 60±41 months. 34% and 58% of RA and PsA patients were on reduced doses of TNFi, and 75% and 44% on concomitant csDMARD, respectively. Thirty-two patients (17.1%: 9 ADA, 13 ETN, 10 INF) had a disease flare during follow-up: 15 at month 4, 12 at month 8 and 5 at month 12. Ab (2ADA, 2IFX) were detected in 12.5% of these patients. Patients with disease flares had lower mean baseline trough serum TNFi levels (2.36±0.45 μ/ml vs. 3.92±0.29 μ/ml p<0.01), although the difference was significant only for ADA (2.2±1.4 μ/ml vs. 6.6±4.1 μ/ml p=0.02), higher baseline DAS28 (2.42 vs. 2.04 p<0.001), and CDAI (5.80 vs. 4.15 p=0.001) and a lower frequency of remission (53% vs. 78% =0.003). During flares, trough serum levels of the three TNFi decreased, although the different was significant compared with baselines only in ETN (0.9±1.2 μ/ml vs. 2.3±1.26 μ/ml p=0.02). Cox multivariate analysis showed associations between disease flares and serum levels of TNFi (hazard ratio (HR) =0.84), glucocorticoids use (HR=2.11), and higher DAS28 (HR=2.19) and CDAI (HR=1.14) levels. Other baseline variables were not associated with disease flares.
Conclusions Over a one-year follow up disease flares were observed in 17% of RA and PsA patients treated with TNFi for a prolonged period who achieved clinical remission/low disease activity. Anti-drug antibodies explained only 12.5% of disease flares. Low drug trough serum levels and high disease activity levels at baseline were associated with disease flares.
Sanmarti et al. Ann Rheum Dis. 2015 Aug;74(8):e42.
Disclosure of Interest R. Sanmarti Grant/research support from: This Study was supported by an Investigator Initiated Research (IIR) from Pfizer (WS2392358), J. Inciarte-Mundo: None declared, P. Estrada-Alarcόn: None declared, M. García-Manrique: None declared, J. Narvaez Grant/research support from: This Study was supported by an Investigator Initiated Research (IIR) from Pfizer (WS2392358), A. Gόmez-Centeno Grant/research support from: This Study was supported by an Investigator Initiated Research (IIR) from Pfizer (WS2392358), J. Rodríguez-Moreno Grant/research support from: This Study was supported by an Investigator Initiated Research (IIR) from Pfizer (WS2392358), M. Pascal: None declared, J. Yagüe: None declared