Objectives To investigate the relationship between antidrug antibodies (ADA), adalimumab concentrations and clinical response in patients with psoriatic arthritis (PsA) during 52 weeks of follow-up.
Methods This prospective cohort study included 103 consecutive patients with PsA. Disease Activity Score of 28 joints (DAS28), Erythrocyte Sedimentation Rate, C reactive protein and Psoriasis Area and Severity Index were assessed. Adalimumab concentrations and ADA were measured in serum trough samples, using an ELISA and a radio immunoassay, respectively.
Results Adalimumab concentrations were significantly lower at 28 and 52 weeks in patients with detectable ADA compared with patients without detectable ADA (at week 28: 1.3 mg/L (IQR 0.0–3.2) versus 8.7 mg/L (IQR 5.7–11.5), p<0.001; at week 52: 0.9 mg/L (IQR 0.0–2.9) vs 9.4 mg/L (IQR 5.7–12.1), p=0.0001). DAS28 at 28 weeks (2.16 vs 2.95, p=0.023) and 52 weeks (2.19 vs 2.95, p=0.024) showed a significant difference; patients with detectable ADA had a poorer clinical outcome than patients without.
Conclusions Patients with detectable ADA had lower adalimumab concentrations and a significantly poorer clinical outcome compared with patients in whom ADA were not detected.
Statistics from Altmetric.com
A large registration trial in PsA patients demonstrated an improved clinical outcome for adalimumab compared with placebo.1 However, some patients failed to respond to treatment. Studies in patients with rheumatoid arthritis (RA), ankylosing spondylitis (AS) or Crohn's disease (CD) have shown that absent or diminished response, in some patients, is associated with lower drug concentrations of tumor necrosis factor (TNF) inhibitors due to the formation of ADA.2–5 A small study in 22 PsA patients reported a relationship between the development of ADA (n=4, 18%), lower adalimumab concentrations and diminished clinical response during 12 months of treatment.6 The current study describes adalimumab concentrations, clinical response and the formation of ADA in a cohort of 103 PsA patients treated with adalimumab.
One hundred and three PsA patients with mainly peripheral joint involvement treated with adalimumab were consecutively included in this prospective observational cohort study at the Jan van Breemen Research Institute | Reade, Amsterdam, The Netherlands. Patients included in this cohort failed to respond adequately to at least one disease modifying antirheumatic drug (DMARD), according to the European League Against Rheumatism recommendations for the management of PsA.7
All patients were treated with adalimumab 40 mg injected subcutaneously every other week. Patients were treated with adalimumab monotherapy or adalimumab with concomitant DMARDs with or without additional prednisone.
Patients were included in the analysis if there was a baseline visit and at least one subsequent visit where serum samples and clinical data were obtained.
Approval of this study was obtained from the medical ethics committee. All patients gave their written informed consent.
Disease activity was assessed at baseline and every subsequent visit thereafter at 4, 16, 28, 40 and 52 weeks of treatment. The primary outcome measure for clinical response was DAS28. Secondary outcome measures were: Erythrocyte Sedimentation Rate (ESR), C reactive protein (CRP) and Psoriasis Area Severity Index (PASI).
Measurement of adalimumab concentration and antidrug antibodies against adalimumab
The measurement of adalimumab concentrations has been described previously.8
In short, an ELISA, designed by Sanquin, Amsterdam, The Netherlands, was used to measure trough serum adalimumab concentrations. Adalimumab was captured via its ability to bind TNF-α.
As previously described for the measurement of antidrug antibodies (ADA), IgG, a radio immunoassay (RIA) was used.9
Patients were defined as being positive if ADA titres were above 12 AU/mL in combination with serum adalimumab concentrations below 5 mg/L on at least one occasion, as previously described.10
For differences between groups, we used the independent sample t test, χ2 or Mann–Whitney U test, as appropriate. The independent sample t test or Mann–Whitney U was used to investigate the relationship between the drug concentrations and the detection of ADA or the use of concomitant methotrexate (MTX). A linear regression model was used to investigate the relationship between the primary or secondary outcome measures and the development of ADA. All statistical tests were two-sided, with a threshold for significance set at p<0.05. The following variables were considered potential confounders: MTX use and dose and prednisone use and dose. Variables were included in the regression model if β changed 10% or more. The last observation was carried forward for patients with incomplete data. All statistical analyses were performed using SPSS V.19.0.
Table 1 shows the baseline characteristics of patients included in this study. There was a significant difference in Body Mass Index (BMI), DMARD use, MTX use, prednisone use and dose between patients with and without detectable ADA. Patients with detectable ADA had a higher BMI, were less frequently treated with a DMARD, MTX and prednisone. Also prednisone dosage was lower in patients with detectable ADA.
During 52 weeks of follow-up, 32 (31%) patients dropped out of the study. Of these patients, 15 (47%) stopped treatment due to failure, 9 (28%) were lost to follow-up and 8 (25%) dropped out due to side effects.
Detection of ADA
During follow-up for 52 weeks, 23 (22%) patients developed detectable ADA. In 3 patients, ADA were intermittently detectable. The ADA titre in these patients ranged from 13 to 186 AU/mL and their adalimumab concentrations ranged from 0.8 to 5.9 mg/L.
Adalimumab was not detectable in any of the baseline samples. The adalimumab concentration over time for patients with or without detectable ADA is shown in figure 1A. At week 28, patients who had detectable ADA had significant lower median adalimumab concentrations: 1.3 mg/L (0.0–3.2 IQR) compared with 8.7 mg/L (5.7–11.5 IQR) in patients without ADA, p<0.001. At 52 weeks median, adalimumab concentration in patients with detectable ADA, was significantly lower (p=0.0001) than in patients without detectable ADA (respectively, median 0.9 mg/L, IQR 0.0–2.9 vs median 9.4 mg/L, IQR 5.7–12.1).
The median adalimumab concentration at 28 weeks for patients on monotherapy was significantly lower compared with patients using adalimumab and concomitant MTX (respectively, 0.7 mg/L, IQR 0.0–3.3 vs 8.4 mg/L, IQR 4.9–11.0), p<0.001 (figure 1B). At 52 weeks, patients on monotherapy had a median adalimumab concentration of 1.8 mg/L, IQR 0.0–3.5 compared with 8.8 mg/L, IQR 4.9–11.9 in patients treated with adalimumab and concomitant MTX. This difference was also statistically significant, p<0.001.
Clinical response and ADA
There was a significant difference at 28 and 52 weeks of treatment between patients with and without ADA for DAS28 and CRP, and for ESR only at 52 weeks, as shown in table 2.
All clinical outcome variables showed a consistent pattern: patients with detectable ADA had a poorer clinical outcome at weeks 28 and 52 than patients without detectable ADA.
Concentration effect curve
Figure 1C shows a concentration-effect curve. All patients were sorted from low to high adalimumab concentration. Each dot represents 10 patients (the last dot represents 13 patients) and their mean adalimumab concentration at week 28 and mean DAS28 improvement compared with baseline. Concentrations of approximately 1.0 mg/L already show reasonable efficacy. Adalimumab concentrations between 5–8 mg/L appear optimal. In 48 (47%) patients, adalimumab concentrations exceeded 8 mg/L. Furthermore, 36 (35%) patients had adalimumab concentrations below 5 mg/L. Of these patients 21 (58%) had detectable ADA.
This study shows that detectable ADA was associated with lower adalimumab concentrations in patients with PsA, which is associated with a poorer clinical outcome at 28 and 52 weeks of treatment.
Adalimumab concentrations reflect the amount of unbound drug available in the serum, which can bind to the target. If none or insufficient concentrations of free drug are available, the inflammation cannot be suppressed sufficiently. Therefore, measuring drug concentrations in patients who do not respond adequately could give more insight to the reason why there is an inadequate response. As is shown, one important factor of inadequate drug concentrations is the formation of ADA. However, in several studies, a relationship is found also between high CRP concentrations, which may be considered as an indirect marker of TNF-α, and higher infliximab clearance,11–13 thus indicating that other factors also, such as the severity of the inflammation, influence drug concentrations.
This study identified that the optimal concentration range of adalimumab in PsA patients in this cohort, might be 5–8 mg/L. This is in concordance with the range that has been found in patients with RA.8 With this range, a proportion of the patients receive undertreatment or overtreatment. The adalimumab dose can probably be tapered in a proportion of patients with low disease activity in combination with high adalimumab drug concentrations, without an increase in disease activity.
For RA, it has been shown that some factors can influence ADA development, like concomitant MTX, however, for PsA, data are limited. In a study of PsA patients, an association between ADA development and concomitant MTX use was not found, and the study cautions clinicians in extrapolating the practice of administering combination therapy with MTX.1 However, a study in patients with RA concluded that MTX use reduced the development of ADA in a dose-dependent manner.14 Patients treated with adalimumab and concomitant MTX had higher adalimumab concentrations than patients with adalimumab monotherapy indicating that concomitant MTX can influence adalimumab concentrations and therefore, might be of clinical importance. Some patients used other concomitant DMARDs than MTX. It would be interesting to see whether these DMARDs are also able to reduce ADA formation like MTX.
DAS28 was chosen as a primary outcome measure. This comprises the risk that not all relevant joints are encompassed, because commonly affected joints in PsA, such as distal interphalangeal joints, ankles and toes, are not included in the DAS28. Nevertheless, DAS28 has been shown to discriminate effectively between placebo and infliximab or etanercept in patients with polyarticular PsA.15 Also, in this study, significant differences between groups could be demonstrated using DAS28, which further supports its use. Since there is no standard clinical outcome measure for PsA, other variables were chosen as a secondary clinical outcome measure.16
In conclusion, this study demonstrates that ADA results in lower adalimumab concentrations and poorer clinical outcome. This is a major concern for PsA patients treated with adalimumab.
Further studies regarding measuring drug concentrations would be relevant, since this could give more insight on the cause of inadequate response, especially since treatment options in PsA are limited. Obviously, information on drug concentrations could lead to a more tailored, evidence-based therapy for patients that is potentially cost saving.17
Handling editor Tore K Kvien
Contributors All authors were responsible for study concept and design, clinical revision and drafting of the manuscript for important intellectual content, and approved the final version of the manuscript to be published. EHV and CLMK were responsible for acquisition of data. EHV, ELK, GW and CLMK were responsible for analysis and interpretation of data. MTN and GW obtained funding. ELK, GW and CLMK supervised the study.
Funding The cohort study was partially financed by AbbVie (ACA-NETH-05-012). Measurement of drug concentrations was sponsored by Pfizer (WI180-353). AbbVie and Pfizer had no involvement in the study design; in the collection, analysis, and interpretation of data; in the writing of the manuscript; or in the decision to submit the manuscript for publication.
Competing interests MTN reports having received consultancy fees from Abbott, Roche, Pfizer, MSD, UCB, SOBI and BMS, payment for lectures from Abbott, Roche and Pfizer. TR reports having received payments for lectures from Pfizer and Abbvie. GW reports having received a research grant from Pfizer (paid to the institution) and payments for lectures from Pfizer and Amgen. CLMK reports having received payment for lectures from Pfizer and Abbvie.
Patient consent Obtained.
Ethics approval Medical ethics committee of the Slotervaart Hospital and the Jan van Breemen Research Institute|Reade.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.