Article Text

THU0104 Limited Safety Signals, but No Advantage of Cobra-Light over Cobra Combination Therapy 4 Years after Initiation of The Cobra-Light Trial in Early Rheumatoid Arthritis
  1. N. Konijn1,
  2. L. van Tuyl1,
  3. M. Boers1,2,
  4. D. den Uijl1,
  5. M. ter Wee1,2,
  6. L. van der Wijden1,
  7. P. Kerstens3,4,
  8. A. Voskuyl1,
  9. D. van Schaardenburg3,5,
  10. M. Nurmohamed1,3,
  11. W. Lems1,3
  1. 1Rheumatology, Amsterdam Rheumatology and immunology Center, VU University Medical Center
  2. 2Epidemiology & Biostatistics, VU Medical Center
  3. 3Rheumatology, Amsterdam Rheumatology and immunology Center, Reade, Amsterdam
  4. 4Rheumatology, Westfriesgasthuis, Hoorn
  5. 5Rheumatology, Amsterdam Rheumatology and immunology Center, Academic Medical Center, Amsterdam, Netherlands


Background COBRA and COBRA-light combination therapy are equally effective and safe treatments for early rheumatoid arthritis (RA) after 26 and 52 weeks [1,2], but long-term safety of COBRA-light therapy is unknown.

Objectives To investigate survival and comorbidities of initial COBRA and COBRA-light combination therapy after a 4 year follow-up period. A detailed analysis of long-term efficacy will be presented separately.

Methods In the COBRA-light trial, 164 patients with recent-onset RA were randomized to either COBRA (prednisolone 60 mg/day, tapered to 7.5 mg/day in 6 weeks; MTX 7.5 mg/wk and sulfasalazine 2 g/day; n=81) or COBRA-light therapy (prednisolone 30 mg/day, tapered to 7.5 mg/day in 8 weeks and MTX escalated to 25 mg/wk in 8 weeks; n=83). Between 26 and 52 weeks; treatment intensification of MTX and addition of etanercept was protocollized (treat-to-target). After 52 weeks, treatment was continued without protocol. Four year after trial initiation, all patients who had initiated therapy (n=162) were invited to participate in the COBRA-light extension study. All patients were interviewed and clinical records were protocollized examined for new comorbidities developed during the follow-up period.

Results A total of 149 out of 162 original trial patients participated in the extension study (77 COBRA and 72 COBRA-light). During the follow-up period, 5 five patients had died: 2 COBRA (both cancer) and 3 COBRA-light (1x dementia; 2x unknown); 6 were not able or willing to participate, and 2 were in drug-free remission and out of care. After the 52 weeks trial period, 43% COBRA vs. 42% COBRA-light patients used prednisolone for ≥90 days, 34% vs. 44% switched to other DMARDs, and 39% vs. 28% used biologicals during the follow-up period. After mean 4 years follow-up, 118 patients (79% in each treatment group) had developed new comorbidities. Cardiovascular events, hypertension, hypercholesterolemia and diabetes mellitus were equally prevalent in both treatment groups (Table 1), notably angina pectoris and heart failure were not seen. Bone quality tended lower in the COBRA-light group, as evidenced by prevalence of osteoporosis on DXA (1% COBRA vs. 6% COBRA-light, p=0.20); clinical fractures (8% vs. 18%, p=0.06), and especially vertebral and rib fractures (0% vs. 6%, p=0.05). Seven patients developed cancer during follow-up: 2 COBRA (both lung cancer, both deceased during follow-up), and 5 COBRA-light (3x lung cancer, 1x basal cell carcinoma and 1x meningioma).

Conclusions Over a 4 year follow-up period, early RA patients initially treated with COBRA and COBRA-light therapy showed limited and similar safety signals in terms of survival and major comorbidities, except for the higher incidence of clinical fractures in the COBRA-light group, which could not be explained by a significant higher prevalence of osteoporosis. Overall, attenuating the original COBRA regimen did not further improve its favorable safety profile.

  1. Den Uyl, ARD 2014;

  2. Ter Wee, ARD 2015

Disclosure of Interest N. Konijn: None declared, L. van Tuyl: None declared, M. Boers Consultant for: MundiPharma, Pfizer, D. den Uijl: None declared, M. ter Wee: None declared, L. van der Wijden: None declared, P. Kerstens: None declared, A. Voskuyl: None declared, D. van Schaardenburg: None declared, M. Nurmohamed: None declared, W. Lems Grant/research support from: This study was supported by an unrestricted grant from Pfizer

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.