In patients with rheumatoid arthritis (RA) treated with anti-TNF drugs (anti-TNF). A correlation has been found between drug levels and clinical response1. For example, in infliximab-treated patients, the presence of anti-drug antibodies (ADA) provoke loss of response, ifuional reactions and increased treatment withdrawal2. These data are not exclusive to RA and have also been described in spondyloarhritis (SpA), where anti-infliximab and anti-adalimumab antibodies are found in non-responders and in the case of infliximab, they are also related to infusional reactions, motivating a switch to another drug3.

Various drug level measurement techniques are available with different standardisation4, creating uncertainty as to the value of measuring drug levels. A consensus article has recently been published with the aim of reducing this uncertainty in the Spanish setting5.

The aims of this study were to compare anti-TNF levels in RA patients vs. SpA, in patients in remission/low disease activity (LDA) vs. non-remission/non-LDA according to rheumatic disease, and to determine the influence of the use of conventional synthetic disease-modifying drugs (DMARDs) on these levels.

A total of 474 patients were being treated with anti-TNF drugs at our centre, including 51 patients with infliximab, 198 patients with adlmumab, and 225 patients with etanercept. A retrospective, observational study was conducted to measure anti-TNF levels and the presence of ADA in a group of consecutively selected patients who attended consultations at the rheumatology department of the Miguel Servet University Hospital.

A total of 53, 73, and 78 patients on treatment with infliximab, adalimumab, and etanercept, respectively, were studied. In all cases, a single sample extraction was performed. Eleven patients were excluded due to AFA and absence of drug levels: 3 (5.6%) treated with infliximab, all with SpA, none treated with DMARDs; 8 (11%) treated with adalimumab, 2 with RA on leflunomide, 6 patients with SpA, one treated with leflunomide, the remaining 5 without DMARDs. No anti-etanercept antibodies were detected.

No differences were found in drug levels according to disease type and disease activity. Patients with RA had a median DAS28: 2.79 (RI: 2.27−3.62) and those with SpA had a BASDAI of 3.47 (SD: 2.2) and an ASDAS of 1.70 (1.40−2.27). The variables studied are displayed in Table 1.

The percentage of remission/LDA in cases with RA and SpA was 74.3% and 77.8%, respectively. The time of disease progression was 12.9 years (RI: 7.49−22.5), with no differences between diseases (Table 1); adherence was ≥80% in 88.6% of cases.

Drug levels in patients on standard-dose anti-TNF were 2.2 g/mL (1.4–5.2), 4.9 g/mL (0.8−8.9), and 3.1 g/mL (2.3−4.4) for infliximab, adalimumab and etanercept, respectively (Table 2A), with no differences according to rheumatic disease (data not shown). We also found no differences in the group of cases with standard drug dosages, segmented by disease type and activity (Table 2B). When dividing patients with standard drug dosages according to treatment with/without DMARDs, we obtained significant differences in serum levels of infliximab and etanercept; adalimumab levels were similar in both groups (Table 2C), and subdividing these cases according to rheumatic disease, only etanercept levels exhibited differences in patients with SpA (Table 2C1).

The drug level achieved after the standard treatment regimen may differ from one patient to another; however, the drug level required to achieve remission may not differ, hence the importance of evidence to support monitoring of anti-TNF6 levels.

The patients studied were fairly homogeneous, even between those in remission and non-remission; for instance, RA patients had generally low activity indices, which may have contributed to differences in drug levels not being conspicuous when comparing patient subgroups.

With respect to infliximab, the study by Jurado T et al. found that most RA patients in remission at week 54 after perfusion displayed levels of 4.4 g/mL at week6,7 whereas in our study at week 12, we logically found lower levels of infliximab, with no differences in the remission/LDA vs. non-remission/non-LDA groups. Our results show no difference between infliximab levels between RA and SpA; nevertheless, Padilla-Martínez EM et al. detected different infliximab trough levels between the 2 diseases8, but did not exclude cases with ADA from the analysis; similarly, they had a smaller sample size.

Sanmarti R et al. found adalimumab levels of 6.9 g/mL in patients with RA in remission9. Pouw MF et al. suggest that they should have levels of between 5−8 g/mL10, similar to what was found in our study in patients in remission. Finally, Kneepkens EL et al. obtained an etanercept level of 3.8 g/mL (2.5–5.2) in SpA patients with ASDAS < 2.111 and Sanmarti R et al. obtained etanercept levels of 2.3 g/mL in RA patients in remission9. Our results were similar in both diseases excluding patients with suboptimal doses.

Anti-TNF levels can be affected by different factors such as comorbidity, route of drug administration, ADA formation, or the use of cSFs12. In our study, we detected differences in drug levels between patients with/without cAMP for infliximab and etanercept; nonetheless, we found no differences for adalimumab levels. However, Vogelzang EH et al. found higher levels of adalimumab in patients with RA or psoriatic arthritis treated with methotrexate13.

In general, we did not find differences in drug levels according to disease activity, perhaps because of the homogeneity of the sample in this aspect, but if we add a cSF-MEF, then differences do emerge, mainly in SpA. Does this suggest the need to always use cSF-MEFs in association with anti-TNFs? Probably not routinely, albeit this opinion might be different if we assessed other biomarkers, such as the CXCL12 genotype or HLA-DQA1*0514,15.

One of the limitations of our study is that as it is observational, retrospective, and conducted in clinical practice. As such, we do not aways have the results for all the variables studied. In addition, when the sample was subdivided, it was reduced, in some cases losing representativeness of the population under study.

In summary, we purport that anti-TNF drug levels will be modified by treatment with cAMP.

This research has not received any specific public or private funding.

None.