Tocilizumab may slow radiographic progression in patients with systemic or polyarticular-course juvenile idiopathic arthritis: post hoc radiographic analysis from two randomized controlled trials

Abstract

Background: Few clinical trials have investigated the prevention of radiographic progression in children with juvenile idiopathic arthritis treated with antirheumatic drugs. This study aimed to investigate radiographic progression in patients with systemic juvenile idiopathic arthritis (sJIA) and patients with polyarticular-course juvenile idiopathic arthritis (pcJIA) treated with the anti-interleukin-6 receptor antibody tocilizumab for 2 years in the TENDER and CHERISH randomized controlled trials, respectively.

Methods: Standard radiographs of both wrists and both hands in the posteroanterior view were obtained within 4 weeks of baseline and were repeated at weeks 52 ± 4 and 104 ± 4 in both trials. All films were scored by two independent readers using the adapted Sharp-van der Heijde (aSH) and Poznanski scoring methods. Although the Poznanski score indicates bone growth limitation or cartilage growth decrease, which are not the same as joint space narrowing in rheumatoid arthritis, its change reflects damage to cartilage. Therefore, impairment in the Poznanski score as well as the aSH score was considered as a measure of structural joint damage. Radiographic progression was defined as worsening of radiographic scores beyond the smallest detectable difference.

Results: Poznanski and aSH scores were available at baseline and at one or more postbaseline time points for 33 and 47 of 112 sJIA patients and 61 and 87 of 188 pcJIA patients, respectively, providing a representative subset of the study populations. The inter-reader and intra-reader agreement intra-class correlation coefficient was > 0.8. Median baseline Poznanski and aSH scores, respectively, were - 2.4 and 24.6 for sJIA patients and - 1.5 and 8.0 for pcJIA patients. Compared with baseline, aSH scores remained stable for all sJIA patients at week 52, whereas 9.4% of sJIA patients had radiographic progression according to Poznanski scores at week 52; at 104 weeks, radiographic progression according to aSH and Poznanski scores was observed in 5.4% and 11.5%, respectively. In pcJIA patients, radiographic progression from baseline at 52 weeks and at 104 weeks was 12.5% and 2.9%, respectively, using aSH scoring and 6.5% and 4%, respectively, using Poznanski scoring.

Conclusion: Tocilizumab may delay radiographic progression in children with sJIA and children with pcJIA.

Trial registration: Trial registration numbers and dates: TENDER, NCT00642460 (March 19, 2008); CHERISH, NCT00988221 (October 1, 2009).

Keywords: Biologicals; Disease-modifying antirheumatic drugs (DMARDs); Polyarticular-course juvenile idiopathic arthritis; Systemic juvenile idiopathic arthritis; Tocilizumab.

Conflict of interest statement

CM has received consulting fees (< $10,000) from AbbVie.

NR has received honoraria for consultancies or speaker bureaus (< $10,000) from the following in the past 2 years: Ablynx, AbbVie, Boehringer, Bristol-Myers Squibb, Eli Lilly, EMD Serono, F. Hoffmann-La Roche, Janssen, Merck, Novartis, Pfizer, Rewind, R-Pharma, Sanofi, Servier, Sinergie, and Takeda. The Gaslini Hospital, where NR works as a full-time public employee, has received contributions (> $10,000) from the following industries in the last 2 years: Abbott, Bristol-Myers Squibb, The Angelini Group “Francesco Angelini,” GlaxoSmithKline, F. Hoffmann-La Roche, Italfarmaco, Janssen, Novartis, Pfizer, Sanofi Aventis, Schwarz Biosciences, Sobi, Xoma, and Wyeth. This funding has been reinvested for the research activities of the hospital in a fully independent manner without any commitment to third parties.

SP, EP, AP, BF, SG, and GG declare no competing interests.

CWouters declares unrestricted research grants to her institution from GlaxoSmithKline, Pfizer, and Roche.

PD has received speaker or consultant fees from AbbVie, Novartis, Pfizer, F. Hoffmann-La Roche, Medac, Eli Lilly, Sanofi, and Merck Sharp & Dohme (< $10,000).

CWells is an employee of Roche Products Ltd.

WD is an employee of and owns stock in Roche Products Ltd.

HIB has received consulting fees (< $10,000) from Pfizer, EMD Serono, and AstraZeneca and (> $10,000) from Bristol-Myers Squibb and Janssen and research grants from Bristol-Myers Squibb and Pfizer.

FDB has received consulting fees (< $10,000) from F. Hoffmann-La Roche and an unrestricted research grant from F. Hoffmann-La Roche.

AR has received consulting fees (< $10,000) from AbbVie, Bristol-Myers Squibb, Pfizer, F. Hoffmann-La Roche, Novartis, Centocor, Francesco Angelini, and Reckitt Benckiser.

Figures

Fig. 1

Bland and Altman plots of radiographic progression based on SDD in patients with sJIA. a, b Mean adapted SH progression scores across two reviewers for each patient with radiographic scores at baseline and week 52 (a, n = 45) and baseline and week 104 (b, n = 37). c, d Mean Poznanski score across two reviewers for each patient with Poznanski scores at baseline and week 52 (c, n = 32) and baseline and week 104 (d, n = 26). SDD thresholds are represented by vertical dashed lines. Patients with adapted SH progression are represented to the right of the vertical line in a and b. Patients with Poznanski progression are represented to the left of the vertical line in c and d. Horizontal dashed lines represent mean difference between readers ± SDD. In case of discrepancy between readers, radiographs were adjudicated and reread independently so consensus could be reached. SDD, smallest detectable difference; SH, Sharp–van der Heijde; sJIA, systemic juvenile idiopathic arthritis

Fig. 2

Bland and Altman plots of radiographic progression based on SDD in patients with pcJIA. a, b Mean adapted SH progression scores across two reviewers for each patient with radiographic scores at baseline and week 52 (a, n = 76) and baseline and week 104 (b, n = 70). c, d Mean Poznanski scores across two reviewers for each patient with Poznanski scores at baseline and week 52 (c, n = 55) and baseline and week 104 (d, n = 44). SDD thresholds are represented by vertical dashed lines. Patients with adapted SH progression are represented to the right of the vertical line in a and b. Patients with Poznanski progression are represented to the left of the vertical line in c and d. Patients randomly assigned to tocilizumab in part 2 are represented by filled circles; those randomly assigned to placebo in part 2 are represented by empty circles. Horizontal dashed lines represent mean differences between readers ± SDD. In case of discrepancy between readers, radiographs were adjudicated and reread independently so consensus could be reached. pcJIA, polyarticular-course juvenile idiopathic arthritis; SDD, smallest detectable difference; SH, Sharp–van der Heijde; TCZ, tocilizumab

Fig. 3

Radiographic scores at baseline, week 52, and week 104 for patients with sJIA and pcJIA. a Adapted SH scores and b Poznanski scores for patients with sJIA. c Adapted SH scores and d Poznanski scores for patients with pcJIA. SH score: higher score indicates greater damage. Poznanski score: the more negative a Poznanski score, the more severe the radiographic damage. IQR, interquartile range; pcJIA, polyarticular-course juvenile idiopathic arthritis; SH, Sharp–van der Heijde; sJIA, systemic juvenile idiopathic arthritis

Fig. 4

Efficacy for patients who did not experience radiographic progression in the pcJIA trial. a Efficacy response for patients who did not experience adapted SH progression based on SDD at week 52 compared with response rates for all patients in the continuous TCZ population. b Efficacy response for patients who did not experience adapted SH progression based on SDD at week 104 compared with response rates for all patients in the continuous TCZ population. JIA ACR, juvenile idiopathic arthritis American College of Rheumatology response; pcJIA, polyarticular-course juvenile idiopathic arthritis; SDD, smallest detectable difference; SH, Sharp–van der Heijde; TCZ, tocilizumab