Elsevier

Joint Bone Spine

Volume 85, Issue 1, January 2018, Pages 59-64
Joint Bone Spine

Original article
Early non-response to certolizumab pegol in rheumatoid arthritis predicts treatment failure at one year. Data from a randomised phase III clinical trial

https://doi.org/10.1016/j.jbspin.2017.01.011Get rights and content

Abstract

Objectives

To compare different early clinical criteria of non-response determined at three months as predictors of clinical failure at one year in patients with rheumatoid arthritis starting therapy with certolizumab pegol.

Methods

Data were derived from a randomised Phase III clinical trial in patients with rheumatoid arthritis who failed to respond to methotrexate monotherapy. Patients included in this post-hoc analysis were treated with certolizumab pegol (400 mg qd reduced to 200 mg qd after one month) and with methotrexate. The study duration was twelve months. Response at three months was determined with the American College of Rheumatology-50, Disease Assessment Score-28 ESR, Health Assessment Questionnaire and the Clinical Disease Activity Index. The performance of these measures at predicting treatment failure at twelve months defined by the American College of Rheumatology-50 criteria was determined, using the positive predictive values as the principal evaluation criterion.

Results

Three hundred and eighty two patients were available for analysis and 225 completed the twelve-month follow-up. At Week 52, 149 (38.1%) patients met the American College of Rheumatology-50 response criterion. Positive predictive values ranged from 81% for a decrease in Health Assessment Questionnaire- Disability index score since baseline > 0.22 to 95% for a decrease in Disease Assessment Score-28 score since baseline  1.2. Sensitivity was   70% in all cases. Performance of these measures was similar irrespective of the definition of treatment failure at 12 months.

Conclusions

Simple clinical measures of disease activity can predict future treatment failure reliably and are appropriate for implementing treat-to-target treatment strategies in everyday practice.

Introduction

Rheumatoid arthritis (RA) is a progressive destructive autoimmune disease of the joints with a prevalence of around 0.5% in the general population [1], [2], [3]. The disease course is typically characterised by episodic acute exacerbations separated by periods of relative disease stability. Recovery from exacerbations is rarely complete, resulting in progressive irreversible structural damage to the affected joints [4].

The goals of treatment are to relieve symptoms, to prevent exacerbations and to slow structural joint damage. Long-term maintenance therapy is based on the use of disease-modifying antirheumatic drugs (DMARDs), a class that includes both synthetic drugs and biological agents. Considerable between-patient heterogeneity in response to DMARDs appears to exist, with ACR50 response rates in randomised clinical trials being generally less than fifty percent [5] and even lower in everyday clinical practice [6]. The recommended treatment strategy is to initiate treatment with DMARDs as early as possible after diagnosis, to monitor patients regularly using composite measures of disease activity and, in case of inadequate response, to switch to another DMARD in order to achieve or maintain tight control of disease activity [7], [8]. Such strategies have been shown to be highly effective [9], [10], [11], [12], [13]. Current practice guidelines recommend initiating treatment, typically with methotrexate, and re-evaluating the treatment at three-monthly intervals, and in case of non-response, switching to another synthetic DMARD or to a biological agent [7], [14] or moving to combination therapy [15] in order to achieve tight disease control.

In order to follow such a strategy, rheumatologists need to know the most appropriate clinical information to assess treatment response or failure. Although the American College of Rheumatology (ACR) [14] and the European League Against Rheumatism (EULAR) [7] both emphasise in their practice guidelines the importance of timely adaptation of treatment in case of inadequate response, neither provide any explicit guidance on the most appropriate criteria for defining treatment failure. In everyday practice, large improvements in disease status are unambiguous and the physician is reassured about maintaining the current treatment. However, if clinical evolution following treatment initiation is modest, physicians may adopt a strategy of ‘watchful waiting’ which may be detrimental to long-term prognosis rather than taking a more proactive approach [16].

A number of studies have investigated the association between early measures of clinical outcome and long-term treatment response [17], [18], [19], [20], [21], [22]. However, from the therapeutic perspective of clinical practice, the issue is not so much to predict treatment response but to predict treatment failure as accurately and as early as possible so that therapy can be adjusted accordingly and the overall long-term therapeutic outcome optimised. The sensitivity and specificity of a given predictive markers will not be the same with respect to treatment failure as with respect to treatment response. In particular, a marker that will predict treatment response with high sensitivity but with low specificity will predict failure with low sensitivity and high specificity. For this reason, it is important to identify reliable and practical markers of treatment failure.

To our knowledge, there is little information available on the comparative performance of potential early measures of treatment failure. A recent study of infliximab has suggested that the combination of disease activity after six weeks and infliximab serum trough levels may be a useful as a predictor of treatment failure after six months of therapy [23]. A post-hoc analysis of a study of certolizumab pegol [24] has also demonstrated that a change in DAS-28 score of < 1.2 during the first three months of the study was predictive of failure to achieve low disease activity at one year. It remains to be demonstrated what is the most appropriate early clinical measure to determine treatment failure. The objective of the present study was to quantify and compare the positive predictive values (PPV) of different validated clinical criteria of non-response determined at three months with respect to clinical failure at one year in patients with RA starting biological anti-TNFα therapy with certolizumab pegol.

Section snippets

Design

This study corresponds to a post-hoc analysis of data from the RAPID 1 [25] prospective, randomised, placebo-controlled, double-blind Phase III clinical trial, conducted between February 2005 and October 2006, which evaluated the efficacy and safety of certolizumab pegol in adult patients with RA. The methodology of this trial has been described in detail elsewhere [25] and is briefly summarised below.

The RAPID 1 trial included 982 adult patients fulfilling the 1987 ACR criteria for RA [26] who

Study population

Of the 393 patients who were originally randomised to the 200 mg maintenance dose of certolizumab pegol, 11 were excluded since they had been previously exposed to a TNFα inhibitor. The study sample thus consisted of 382 patients randomised into the certolizumab pegol 200 mg maintenance dose group and with no prior exposure to TNFα inhibitor. Of these 382 patients, 96 (25.1%) were withdrawn from the study due to inadequate efficacy. These included 72 patients who failed to meet the ACR20

Discussion

In this study, we demonstrate that the values of different early disease activity measures can be used to predict treatment failure in RA patients starting treatment with an anti-TNFα. Although all the measures performed acceptably in correctly predicting treatment failure, some differences were observed, notably with the HAQ-DI performing least well.

As has been observed previously by Curtis et al. [19], several of the early measures tested are not entirely suitable for use in everyday clinical

Funding

This publication was funded by UCB Pharma.

Disclosure of interest

FB is a consultant for UCB Pharma. TP is a consultant for UCB Pharma. PC received consulting fees, speaking fees and/or honoraria from Abbott, Wyeth-Pfizer, Roche, Chugai, Bristol-Myers Squibb, Schering-Plough, UCB Pharma, and was investigator for Roche Chugai, Sanofi Aventis, Abbott, Wyeth-Pfizer and BMS. TDC was an employee of UCB Pharma at the time of the study. JMJ is an employee of UCB Pharma, and holds UCB Pharma stock options. CS is an employee of UCB Pharma. LRF is a consultant for UCB

Acknowledgments

The authors would like to thank Foxymed (Paris, France) and Costello Medical Consulting for editorial support.

References (37)

  • J.S. Smolen et al.

    EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs

    Ann Rheum Dis

    (2010)
  • Y.P. Goekoop-Ruiterman et al.

    Clinical and radiographic outcomes of four different treatment strategies in patients with early rheumatoid arthritis (the BeSt study): a randomized, controlled trial

    Arthritis Rheum

    (2005)
  • V.P. Nell et al.

    Benefit of very early referral and very early therapy with disease-modifying anti-rheumatic drugs in patients with early rheumatoid arthritis

    Rheumatology (Oxford)

    (2004)
  • M.F. Bakker et al.

    Tight control in the treatment of rheumatoid arthritis: efficacy and feasibility

    Ann Rheum Dis

    (2007)
  • L.G. Schipper et al.

    Meta-analysis of tight control strategies in rheumatoid arthritis: protocolized treatment has additional value with respect to the clinical outcome

    Rheumatology (Oxford)

    (2010)
  • K.G. Saag et al.

    American College of Rheumatology 2008 recommendations for the use of nonbiologic and biologic disease-modifying antirheumatic drugs in rheumatoid arthritis

    Arthritis Rheum

    (2008)
  • J.R. O’Dell et al.

    Therapies for active rheumatoid arthritis after methotrexate failure

    N Engl J Med

    (2013)
  • D. Aletaha et al.

    Disease activity early in the course of treatment predicts response to therapy after one year in rheumatoid arthritis patients

    Arthritis Rheum

    (2007)
  • View full text