L. Nalysnyk1, Y. Xu1, K. Williams1, S. Sharma1, I. Villanueva2, V. Sallerin3, T. Koncz3. 1Health Care Analytics, United BioSource Corporation, Medford; 2Outcomes Research, Bristol-Myers Squibb, Wallingford, United States; 3Global Medical Affairs, Bristol-Myers Squibb, Rueil-Malmaison, France

Background: Tumor necrosis factor (TNF) antagonists have revolutionized the treatment of active rheumatoid arthritis (RA).1 However, a substantial proportion of patients have inadequate response to these agents: No response (primary failure), loss of initial response (secondary failure), partial response, or unacceptable side effects (intolerance failure).2 With the increased number of therapeutic options for these patients and a lack of randomized controlled trials with the anti-TNFs, physicians face challenges as to what treatment to prescribe next.
Objectives: To estimate the efficacy of 2nd and 3rd anti-TNF agents in treating RA patients after failing a first anti-TNF.
Methods: A systematic review of English literature published during 1/1995-11/2007 was performed and 2004-2007 EULAR and ACR meetings abstracts were also included. Interventional and observational studies were eligible and evaluated for quality. The proportion of ACR and EULAR responders and the mean changes in DAS28 and HAQ were assessed. Meta-analytic pooling of outcomes across studies was performed using a restricted-maximum likelihood random-effects model. Efficacy was evaluated by type of failure, number of previous anti-TNFs failed, and stratified by study duration. Heterogeneity was examined using Cochrane's Q-statistic.
Results: This review identified 31 primary studies evaluating 5306 patients (16 full publications, 15 meeting abstracts). The majority (77%) were prospective cohorts, conducted in Europe (81%). There was only one randomized controlled trial. Most studies (60%) were of short duration (<6m). The mean age of patients was 55y (45-68) and the vast majority were female (∼80%). The primary reason for switching was due to efficacy failure (66%). Meta-analytic results are presented in the table.
Conclusion: In this analysis, patients who switched due to primary failure with first anti-TNF agent have lower response compared to those switching due to secondary or intolerance failure. Similarly, patients who failed two or more anti-TNF agents have lower responses compared to those who failed only one anti-TNF agent. As the data suggest decreasing benefit when switching TNF inhibitors, therapies with different mode of action may need to be considered for patients failing their first anti-TNF.
References: 1. Bennett AN et al: Rheumatology 2005;44: 1026-31;
2. van Vollenhoven RF. Ann Rheum Dis. 2007;66(7):849-51;

Ann Rheum Dis 2008;67(Suppl II):326

Table 1. Efficacy of 2nd and 3rd anti-TNF by Type of Failure
% or Δ [95%CI] t (N)PrimarySecondaryOverall Efficacy*Intolerance1 TNF Failed2+TNF Failed
ACR2047.8 [37,59]62.2 [50,74]58.5 [50,67]66.3 [60,73]62.0 [56,68]43.4 [35,52]
5 (233)6 (399)13 (666)4 (199)11 (1257)2 (156)
ACR5025.3 [20,31]33.3 [26,40]31.9 [25,39]46.6 [31,63]40.1 [29,51]24.0 [12,36]
5 (233)4 (363)11 (630)5 (236)10 (1258)2 (156)
ACR709.7 [6,14]13.4 [10,17]12.7 [10,15]28.9 [13,45]20.2 [10,31]10.8 [5,16]
3 (207)3 (344)8 (585)4 (231)8 (908)1 (120)
EULAR (Good/moderate)65.5 [56,75]72.5 [67,78]69.9 [65,74]71.5 [59,84]66.0[61,71]71.3[27,100]
11 (340)10 (637)24 (1058)7 (297)27 (1666)4 (185)
10 (298)10 (489)25 (1018)6 (287)27 (1447)4 (169)
4 (512)3 (319)7 (831)3 (216)6 (1252)2 (138)
t = groups, N = patients; *primary, secondary and unspecified efficacy failure.

Session: RA – Anti-TNF therapy


Close Window