Article Text

Download PDFPDF

Golimumab trough levels, antidrug antibodies and clinical response in patients with rheumatoid arthritis treated in daily clinical practice
  1. Eva L Kneepkens1,
  2. Chamaida Plasencia2,
  3. Charlotte LM Krieckaert1,
  4. Dora Pascual-Salcedo3,
  5. Desiree van der Kleij4,
  6. Michael T Nurmohamed1,5,
  7. M Teresa López-Casla2,
  8. Roeland Wieringa4,
  9. Theo Rispens6,
  10. Gertjan Wolbink1,6
  1. 1 Department of Rheumatology, Jan van Breemen Research Institute | Reade, Amsterdam, The Netherlands
  2. 2 Department of Rheumatology, La Paz University Hospital, Madrid, Spain
  3. 3 Immunology Unit, La Paz University Hospital, Madrid, Spain
  4. 4 Sanquin Diagnostic Services, Amsterdam, The Netherlands
  5. 5 Department of Rheumatology, VU University medical centre, Amsterdam, The Netherlands
  6. 6 Department of Immunopathology, Sanquin Research and Landsteiner Laboratory Academic Medical Centre, Amsterdam, The Netherlands
  1. Correspondence to Eva-Linda Kneepkens, Department of Rheumatology, Jan van Breemen Research Institute | Reade, Dr Jan van Breemenstraat 2, Amsterdam 1056 AB, The Netherlands; e.kneepkens{at}

Statistics from

Request Permissions

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.

Tumour necrosis factor inhibitors (TNFi) are effective in the majority of patients with rheumatoid arthritis (RA),1 however, an important reason for non-response is low drug level due to immunogenicity.2 To our knowledge, no data collected during a prospective observational study is currently available regarding the relationship between golimumab level, immunogenicity and response in RA.

This prospective observational cohort consisted of 37 consecutive adult patients with RA, according to the American College of Rheumatology 1987 revised criteria,3 in whom golimumab 50 mg subcutaneously once monthly was initiated according to the judgment of the rheumatologist, and who were recruited from two departments (Spain and The Netherlands). The study was approved by both Medical Ethics Committees. Clinical response was defined as Disease Activity Score using 28 joint count (DAS28) <3.2, calculated with erythrocyte sedimentation rate (ESR) (mm/h). Patients were eligible for inclusion when clinical data and sera of baseline with ≥ one follow-up visit were available.

Clinical measurements and trough-level sera were collected at baseline and 4, 16, 28 and 52 weeks (The Netherlands), or half yearly (Spain), thereafter. Golimumab levels were measured analogously to adalimumab4 using TNF for capture and rabbit antigolimumab for detection (lower limit of quantification (LLOQ) 5 ng/mL; accuracy 103%, precision 12%). Antidrug antibodies (ADA) were measured, using an ADA radio-immune assay, described previously.4 ,5 Cut-off (mean+3 SD) was based on a serum panel of 80 healthy donors and 15 sera containing anticyclic citrullinated peptide (anti-CCP), ANA, and/or rheumatoid factor. All baseline samples were ADA against golimumab negative.

For statistical analysis SPSS V.17.0 and Graph Pad Prism 5 for windows were used. Threshold for significance was set at p<0.05. To analyse the association between golimumab level and response at 1 year, last observation carried forward was used for patients who discontinued golimumab treatment prematurely.

For baseline characteristics, see table 1.

Table 1

Baseline characteristics

At week 52, 15 patients (40.5%) were responder and 22 (59.5%) non-responder. Nineteen patients (51.4%) discontinued golimumab treatment prematurely due to inefficacy (11), side effects (7) or other reasons (1); with a median drug survival of 16 weeks. Median golimumab level (mg/L) at week 52 was 0.55 (0.27–1.48), and was significantly higher, analysed with a χ2 test, in responders, 1.36 (0.5–1.82), compared with non-responders, 0.43 (0.23–0.84) (p=0.023) (figure 1A). Generalised estimating equation analysis demonstrated, after adjustment for baseline values, a statistically significant inverse association between golimumab level and C-reactive protein (CRP) (mg/L)/ESR. After correction, a trend remained visible for DAS28 (figure 1B).

Figure 1

(A) Median golimumab trough level (mg/L) (with IQR) was higher in patients with a DAS28 <3.2 versus ≥3.2 at week 52 of treatment (p=0.023). (B) The association between golimumab trough level and disease activity over time during one year of follow-up, analysed with a generalised estimating equation. (C) Percentage of patients with DAS28 <3.2 and ≥3.2 stratified according to the golimumab level at 52 weeks of treatment. Each group contains 9 patients (25% of all patients) and the last quartile 10.

All patients were stratified according to the golimumab level at week 52 and divided into quartiles (figure 1C). The lowest quartile (golimumab <0.25 mg/L) comprised 32% of all non-responders, while, the highest (golimumab >1.4 mg/L) comprised 47% of all responders.

During 52 weeks, 3 patients were ADA positive (ADA >12 AU/mL one ≥1 occasion in combination with golimumab levels <0.1 mg/L).4 All three patients discontinued golimumab prematurely due to inefficacy. One patient used concomitant methotrexate. However, the assay used to detect ADA can be influenced by drug interference, resulting in an underestimation of ADA.6 The percentages of patients with ADA against golimumab found in prior studies varied between 2.1% to 13%.7–10 However, head-to-head comparison of ADA percentages is complicated, since several factors can influence immunogenicity2 and clinical non-response is multifactorial.

In conclusion, responders had a significantly higher golimumab trough level at 1 year of treatment. ESR and CRP were statistically significantly inversely associated with golimumab level over time. Three patients had high ADA titres resulting in undetectable golimumab levels, and thus in a poor clinical outcome. These results can be used to further optimise golimumab treatment in RA.

There are some limitations to this study: limited patient number, the majority of patients used prior TNFi, and golimumab discontinuation rate was relatively high.


The authors are grateful to the research nurses and medical doctors for performing clinical assessments. Finally, the authors wish to thank the technicians of Sanquin Diagnostics Services for performing the assays, and Sanquin Reagents for the preparation of the biotinylated rabbit antigolimumab antibodies.



  • Contributors Study concept and design: ELK, CP, CLMK, DP-S, DvdK, MTN, MTL-C, RW, TR, GW. Acquisition of data: ELK, CP, DP-S. Analysis and interpretation of the data: ELK, CP, GW. Clinical revision and drafting of the manuscript for important intellectual content: ELK, CP, CLMK, DP-S, Dv dK, MTN, MTL-C, RW, TR, GW. Obtained funding: none. Study supervision: MTN, GW. Final approval: ELK, CP, CLMK, DP-S, DvdK, MTN, MTL-C, RW, TR, GW.

  • 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. GW reports having received a research grant from Pfizer (Wyeth) (paid to the institution) and payments for lectures from Pfizer, UCB, Abbvie and Amgen. CLMK reports having received payment for lectures from AbbVie and Pfizer. TR reports having received payment for lectures from AbbVie and Pfizer. CP reports having received a research grant from Pfizer. DP-S reports having received payments for lectures from Pfizer and a research grant from Pfizer. ELK reports having received payments for lectures from Pfizer. DvdK, MTL-C and RW have no disclosures

  • Patient consent Obtained.

  • Ethics approval The study was approved by the Medical Ethics Committee of the Slotervaart Hospital and Jan van Breemen Research Institute | Reade, Amsterdam (The Netherlands) and La Paz Hospital, Madrid (Spain).

  • Provenance and peer review Not commissioned; externally peer reviewed.