Background The anti-tumor necrosis factor (anti-TNF) therapy has shown to be effective in rheumatoid arthritis (RA). Recent publications have shown the association between drug levels (DL) and clinical response to the anti-TNF therapies. For now it is not clear if the DL monitoring to the 1st anti-TNF in patients who discontinued due to inefficacy, can be useful to evaluate the clinical efficacy to the 2nd anti-TNF.

Objectives To evaluate in a RA cohort (who stopped to the 1st anti-TNF due to inefficacy), the influence of the presence/absence of DL [infliximab (Ifx) y adalimumab (Ada)] at discontinuation, on the clinical response to the 2nd anti-TNF during the 1st year of therapy.

Methods Seventy seven out of 182 RA patients who discontinued the Ifx or Ada therapy were recruited. All these patients dropped out the first anti-TNF due to inefficacy and switched to a 2nd anti-TNF (48 patients: 2 with Ifx, 4 with Ada, 40 with etanercept and 2 with certolizumab) or other biologics (OB) different to anti-TNF (29 patients: 17 with rituximab, 6 with tocilizumab and 6 with abatacept). Clinical activity was measured by DAS28 and clinical improvement by Delta-DAS28 adjusted to baseline clinical activity (baselineDAS-DAS (24weeks or 52weeks)/baselineDAS) at baseline, 24 weeks (24w) and 52 weeks (52w). DL were measured prior to drug administration at 24w and 52w by ELISA.

Results Of the 77 RA patients, 82% (63/77) were women and the mean disease duration was 21±13 years. At the end of the 1st antiTNF, 54/77 (70%) had undetectable DL and 23/77 (30%) had detectable DL. At w24 of the 2nd biologic, patients treated with OB tended to be less active than patients treated with a 2nd anti-TNF but these differences were not significant at w52 (DAS28 at w24: 3.5±1.3 with OB vs 4.1±1.1 wirh 2nd anti-TNF, p=0.053; Delta-DAS28 at w24: 0.3±0.3 with OB vs 0.2±0.3 with 2nd anti-TNF, p=0.326; DAS28 at w52: 3.3±1.2 with OB vs 3.8±1.1 with 2nd anti-TNF, p=0.166; Delta-DAS28 at w52: 0.3±0.2 with OB vs 0.2±0.3 with 2nd anti-TNF, p=0.501). Patients with undetectable DL to first anti-TNF treated with a 2nd anti-TNF had better clinical response during the first year of therapy (DAS28 at w24: 3±1.1 with non DL vs 4.3±1.2 with DL, p=0.004; Delta-DAS28 at w24: 0.3±0.3 with non DL vs 0.2±0.3 with DL p=0.074; DAS28 at w52: 2.9±1.1 with non DL vs 4±1.2 with DL, p=0.015; Delta-DAS28 at w52: 0.4±0.2 with non DL vs 0.2±0.3 with DL, p=0.210). However, no differences in the clinical activity were observed between patients with undetectable or detectable DL who switched to OB (DAS28 at w24: 4.4±1.4 with non DL vs 3.8±1 with DL, p=0.247; Delta-DAS28 at w24: 0.2±0.2 with non DL vs 0.2±0.2 with DL, p=0.382; DAS28 at w52: 4.1±1.2 with non DL vs 3.5±0.7 with DL, p=0.122; Delta-DAS28 at w52: 0.2±0.2 with non DL vs 0.3±0.1 with DL, p=0.376).

Conclusions In our cohort of RA patients with inefficacy to the 1st anti-TNF, a high proportion of patients discontinued the therapy with undetectable DL. The DL monitoring during the 1st anti-TNF seems to be a useful tool to define what patients have a better clinical response to the 2nd anti-TNF, although this effect does not influence on those patients who switch to OB.

Disclosure of Interest None declared