Twenty-eight-week results from the REALISTIC phase IIIb randomized trial: efficacy, safety and predictability of response to certolizumab pegol in a diverse rheumatoid arthritis population

Abstract

Introduction: This 28-week, phase IIIb study assessed safety and maintenance of response to certolizumab pegol (CZP) in a diverse population of rheumatoid arthritis (RA) patients, stratified by prior anti-TNF exposure, concomitant methotrexate (MTX) use and disease duration. The ability to predict achievement of low disease activity (LDA) at week 28 from improvements in Disease Activity Score 28 (DAS28), erythrocyte sedimentation rate (ESR), swollen joint count (SJC) and Clinical Disease Activity Index (CDAI) up to week 12 was assessed.

Methods: The 28-week study population included all patients who completed the double-blind (DB) phase and entered the open-label (OL) phase, receiving 200 mg CZP every 2 weeks (Q2W) ≥16 weeks. In the 12-week DB period, patients with active RA and an inadequate response to ≥1 disease-modifying antirheumatic drug (DMARD) were randomized 4:1 to CZP (400 mg at weeks 0, 2 and 4 then 200 mg Q2W) or placebo (Q2W), stratified by prior anti-TNF use, concomitant use of MTX and disease duration (<2 years vs. ≥2 years).

Results: A total of 955 patients entered the OL phase. At week 28, similar clinical improvements were seen in those receiving CZP throughout (CZP → CZP; n = 771) and those receiving placebo during the DB phase and switching to CZP in the OL phase (placebo → CZP; n = 184) (ACR20 response rate = 59.7% vs. 53.3%; ACR50/ACR70 response rates were also similar). Effect of CZP treatment was similar regardless of prior anti-TNF use, disease duration and concomitant DMARDs, based on ACR20 response rates. The percentage of patients achieving DAS28(ESR) LDA at week 28 was calculated for DAS28(ESR), SJC or CDAI responders at earlier time points. Reductions from baseline (Δ) of DAS28(ESR) <1.2, ΔSJC <25% or ΔCDAI <10 by week 12 were associated with <9% chance of achieving LDA at week 28 regardless of prior anti-TNF exposure. Adverse event rates were similar for placebo → CZP and CZP → CZP patients, with no new safety signals identified.

Conclusions: A diverse population of RA patients with varying disease duration showed rapid and sustained clinical improvements on CZP treatment, regardless of prior anti-TNF or concomitant DMARD use. Failure to achieve improvements in DAS28(ESR), SJC or CDAI within the first 12 weeks of CZP therapy was associated with a low chance of achieving LDA at week 28. No new safety signals were observed.

Trial registration: ClinicalTrials.gov, NCT00717236 , 15 July 2008.

Fig. 1

a REALISTIC trial design. b Subject disposition. ^aDB phase. ^bOne week 12 CZP completer discontinued after week 12 due to an AE, did not receive any study medication in the OL phase, and was not included in the OL analysis set. AE adverse event, CZP certolizumab pegol, DMARD disease-modifying antirheumatic drug, OLE open-label extension, Q2W every other week

Fig. 2

Treatment response to week 28. a ACR20 response rates up to week 28. b ACR50 and ACR70 response rates up to week 28 c DAS28(CRP) change from baseline up to week 28. d HAQ-DI change from baseline up to week 28 (OL set, imputed data). Footnote: ACR response rates were calculated using NRI if withdrawal was due to an AE or lack or loss of efficacy, and LOCF in case of any other reason. Least squares means (change from baseline) in DAS28(CRP) and HAQ-DI were analyzed using MMRM. AE adverse event, CRP C-reactive protein, CZP certolizumab pegol, DAS28 Disease Activity Score in 28 joints, HAQ-DI Health Assessment Questionnaire-Disability Index, LOCF last observation carried forward, MMRM mixed-effects model for repeated measures, NRI nonresponder imputation, OL open-label

Fig. 3

Clinical response at week 28, stratified by baseline characteristics. a Week 28 ACR20/ACR50/ACR70 responses stratified by prior anti-TNF therapy. b DAS28(CRP) progression up to week 28 by prior anti-TNF therapy. c Week 28 ACR20/ACR50/ACR70 responses stratified by baseline DMARDs. d Week 28 ACR20/ACR50/ACR70 responses stratified by disease duration (OL set, imputed data). Footnote: ACR response rates were calculated using NRI if withdrawal was due to an AE or lack/loss of efficacy, and LOCF in case of any other reason. Least squares mean (change from baseline) in DAS28(CRP) was analyzed using MMRM. AE adverse event, CRP C-reactive protein, CZP certolizumab pegol, DAS28 Disease Activity Score in 28 joints, LOCF last observation carried forward, MMRM mixed-effects model for repeated measures, NRI nonresponder imputation, OL open-label, TNF tumor necrosis factor

Fig. 4

Clinical improvements in CZP-treated patients to week 12, stratified by prior-anti-TNF exposure. Cumulative proportion of CZP patients achieving (a) ≥1.2 reduction from baseline in DAS28(ESR), (b) ≥25 % change in baseline in SJC and (c) ≥10 reduction from baseline in CDAI (FAS, LOCF imputation). Footnote: Patients could achieve the response at any time point up to and including the relevant week. If all prior weeks were missing the patient was not counted. CDAI Clinical Disease Activity Index, CZP certolizumab pegol, DAS28 Disease Activity Score in 28 joints, ESR erythrocyte sedimentation rate, FAS full analysis set, LOCF last observation carried forward, SJC swollen joint count, TNF tumor necrosis factor