Abatacept improves health-related quality of life, pain, sleep quality, and daily participation in subjects with juvenile idiopathic arthritis

Authors


Abstract

Objective

To assess health-related quality of life (HRQOL) in abatacept-treated children/adolescents with juvenile idiopathic arthritis (JIA).

Methods

In this phase III, double-blind, placebo-controlled trial, subjects with active polyarticular course JIA and an inadequate response/intolerance to ≥1 disease-modifying antirheumatic drug (including biologics) received abatacept 10 mg/kg plus methotrexate (MTX) during the 4-month open-label period (period A). Subjects achieving the American College of Rheumatology Pediatric 30 criteria for improvement (defined “responders”) were randomized to abatacept or placebo (plus MTX) in the 6-month double-blind withdrawal period (period B). HRQOL assessments included 15 Child Health Questionnaire (CHQ) health concepts plus the physical (PhS) and psychosocial summary scores (PsS), pain (100-mm visual analog scale), the Children's Sleep Habits Questionnaire, and a daily activity participation questionnaire.

Results

A total of 190 subjects from period A and 122 from period B were eligible for analysis. In period A, there were substantial improvements across all of the CHQ domains (greatest improvement was in pain/discomfort) and the PhS (8.3 units) and PsS (4.3 units) with abatacept. At the end of period B, abatacept-treated subjects had greater improvements versus placebo in all domains (except behavior) and both summary scores. Similar improvement patterns were seen with pain and sleep. For participation in daily activities, an additional 2.6 school days/month and 2.3 parents' usual activity days/month were gained in period A responders with abatacept, and further gains were made in period B (1.9 versus 0.9 [P = 0.033] and 0.2 versus −1.3 [P = 0.109] school days/month and parents' usual activity days/month, respectively, in abatacept- versus placebo-treated subjects).

Conclusion

Improvements in HRQOL were observed with abatacept, providing real-life tangible benefits to children with JIA and their parents/caregivers.

INTRODUCTION

Juvenile idiopathic arthritis (JIA) is the most common chronic rheumatic disease in children and adolescents, with an estimated prevalence ranging between 16 and 400 per 100,000 children (1). The systemic inflammation, chronic arthritis, and possible joint damage that are characteristic of JIA result in pain, stiffness, limitation of motion, and loss of physical function, and may have an impact on growth. This functional impairment limits mobility and participation in daily activities and, consequently, children and adolescents with JIA have poorer health-related quality of life (HRQOL) than their healthy peers (2, 3).

Chronic inflammatory or painful conditions such as JIA can also adversely affect sleep quality (4), and changes in sleep patterns can exacerbate feelings of pain (5). As a result of the chronic pain and physical disability associated with JIA, children and adolescents often miss school, and parents may miss days at work and experience financial burdens (6). Because JIA negatively affects these multiple aspects of HRQOL, there is increasing recognition that the complete assessment of children with JIA requires an understanding of the impact of disease on daily life. Treatment strategies should aim to improve physical function, reduce pain, improve HRQOL, and allow both children and parents or caregivers to resume as many of their normal daily activities as possible.

Pharmacologic therapy for JIA traditionally has involved the use of nonsteroidal antiinflammatory drugs; however, recent advances in disease management include the use of both nonbiologic and biologic disease-modifying antirheumatic drugs (DMARDs) (1, 7). Methotrexate (MTX) is considered the standard treatment and provides clinical benefits in JIA with an acceptable profile of toxic effects (8, 9). For children and adolescents who do not respond to MTX, available biologic therapies, including etanercept, adalimumab, and abatacept, are important treatment options (1, 10–13).

MTX treatment has been shown to produce significant improvements across a wide range of HRQOL components, particularly the physical health concepts, in subjects with JIA (3). However, little information exists about the impact of biologic agents in this population (14).

Abatacept is the first in a class of agents that selectively modulates the CD80/CD86:CD28 costimulatory signal required for full T cell activation. The mechanism of action of abatacept is fundamentally different from biologics such as etanercept and adalimumab, which target the cytokine tumor necrosis factor α. Trials in adult subjects with rheumatoid arthritis (RA) have shown that, in addition to significant improvements in clinical measures of disease activity, abatacept improves HRQOL, physical function, fatigue, sleep problems, and limitation in daily activities (15–17).

The safety, tolerability, and clinical efficacy of abatacept in children and adolescents with JIA have previously been demonstrated in a double-blind, placebo-controlled, randomized withdrawal trial (18). Here we report for the first time, to our knowledge, on the impact of abatacept on HRQOL, pain, sleep quality, and participation in daily activities, including days of paid care, using data from this phase III trial.

SUBJECTS AND METHODS

Subjects and study design.

This was a phase III, randomized, double-blind, placebo-controlled withdrawal trial in subjects with active polyarticular course JIA (extended oligoarticular, polyarticular positive or negative for rheumatoid factor, or systemic without systemic manifestations) conducted in centers of the Paediatric Rheumatology International Trials Organisation (PRINTO) (19) and the Pediatric Rheumatology Collaborative Study Group (PRCSG). Eligible subjects were ages 6–17 years with active disease, defined as at least 2 active joints and 2 joints with a limited range of motion. Subjects had an inadequate response to, or intolerance to, at least 1 DMARD, including biologic agents. All DMARDs except for MTX were withdrawn prior to entry and prohibited during the trial. The trial design has been fully described elsewhere (18). In brief, subjects received infusions of abatacept 10 mg/kg (according to weight at each visit), either as monotherapy with a maximum dose of 1,000 mg or plus MTX, on days 1, 15, 29, 57, and 85 of the 4-month, open-label, lead-in period (period A). Subjects who had improved according to the American College of Rheumatology Pediatric 30 (ACR Pedi 30) definition of improvement (20) were then randomized 1:1 to receive abatacept or placebo every 28 days for 6 months or until disease flare in the double-blind withdrawal period (period B). Background MTX was continued in both treatment arms in period B.

HRQOL assessments.

The national-language version of the parent-administered, validated (21), 50-item Child Health Questionnaire (CHQ) (22–24) was used to assess physical, emotional, and social aspects of HRQOL in children. The questionnaire contains 15 separate health concepts (global health, physical function, role/social limitations–physical, role/social limitations–emotional, pain/discomfort, behavior, global behavior, mental health, self-esteem, general health, change in health, parental emotional impact, parental time impact, family activity, and family cohesion), with higher scores indicating better HRQOL. For all concepts except change in health, raw scores are transformed to a 0–100 scale (change in health remains a categorical score on a scale of 1–5). In addition, 2 summary scores (the physical summary score [PhS] and the psychosocial summary score [PsS]) were also calculated, which were standardized to have a mean ± SD score of 50 ± 10 in the normative population (22). These summary scores are calculated when 10 of the 15 CHQ health concepts are available. For individual CHQ health concepts, data are also presented for healthy children to aid interpretation of the results. Data for these healthy children were collected as part of the PRINTO cross-cultural adaptation and psychometric evaluation of the Childhood Health Assessment Questionnaire (C-HAQ) and CHQ, for which children were defined as healthy after examination by a physician and/or based on the parent's observation (24).

Pain assessments.

The level and frequency of a child's pain/discomfort over the previous 4 weeks were rated by parents/caregivers as part of the CHQ. Additionally, pain was rated by parents/caregivers according to how much average pain they thought their child had experienced in the past week due to his/her illness as part of the C-HAQ. Pain was measured by a 100-mm visual analog scale (VAS) (24–26), with higher scores indicating more severe pain.

Sleep assessments.

Sleep quality was measured using the Children's Sleep Habits Questionnaire (CSHQ), which includes 33 items grouped into 8 subscales representing key clinical sleep symptoms: bedtime resistance, sleep-onset delay, sleep duration, sleep anxiety, night wakings, parasomnias, sleep-disordered breathing, and daytime sleepiness (27). Parents rated each item on a 3-point scale, depending on the occurrence of the sleep behavior over an “average” week. Total scores range from 0–100, with higher scores indicating more sleep problems.

Participation in daily activity assessments.

Participation in daily activities for the child and the parent were assessed by a parent questionnaire containing the following questions:

“In the past 30 days, on how many days did your child's JIA keep him/her from attending school (excluding vacation/holidays)? ___ days”

“In the past 30 days, on how many days did your child's JIA keep you from doing your usual activities? ___ days”

“In the past 30 days, on how many days did you pay for child care for your son/daughter with JIA in order for you to engage in your usual activities? ___ days”

Data collection and statistical analysis.

CHQ, pain, sleep quality, and participation in daily activity assessments were performed on days 1, 57, and 133 of period A, and on days 1, 85, and 169 of period B. Data collected during period A were summarized using descriptive statistics for actual scores and mean changes from baseline. For period B data, descriptive statistics were provided for actual scores and mean changes from baseline were analyzed with an analysis of covariance model, with treatment as the main factor and baseline value as the covariate. A P value for the treatment difference between the randomized treatment groups in period B was provided. All analyses were based on available data at the designated time point (as-observed analysis). For individual CHQ health concepts and activity limitation assessments in period A, data are shown for the entire population. For CHQ component scores, pain, and sleep assessments in period A, data are given for both ACR Pedi 30 responders and nonresponders. All period B assessments are given for subjects randomized to abatacept or placebo.

RESULTS

Subject disposition and clinical characteristics.

Of the 190 subjects who entered the trial, 170 (89%) completed period A; subject disposition through period A has been previously reported (18). At the end of period A, 123 subjects (65%) had achieved an ACR Pedi 30 response and were eligible for randomization in period B; 1 subject left the study and 122 were randomized and treated with either abatacept (n = 60) or placebo (n = 62). In the abatacept group, 10 subjects discontinued due to lack of efficacy and 1 withdrew consent; 49 completed the 6-month double-blind period (period B). In the placebo group, 31 subjects discontinued, all due to lack of efficacy; 31 completed period B.

At entry to the study, subjects were predominantly female (72.1%) and white (77.4%), with a mean age of 12.4 years and a mean disease duration of 4.4 years (Table 1). The mean MTX dosage at baseline was 13.2 mg/week, and 57 subjects (30%) had received biologic therapy prior to study entry. At the start of the double-blind period (period B), baseline demographics and clinical characteristics were comparable between the abatacept and placebo groups (Table 1).

Table 1. Baseline demographics and summary of key variables over time*
CharacteristicDay 1 of period A (4-month, open-label), abatacept (n = 190)Day 1 of period B (6-month double-blind period)End of period B (6-month double-blind period)
Abatacept (n = 60)Placebo (n = 62)Abatacept (n = 49)Placebo (n = 31)
  • *

    Values are the mean ± SD unless otherwise indicated. Sample sizes are for the given group unless otherwise stated. JIA = juvenile idiopathic arthritis; VAS = visual analog scale; C-HAQ = Childhood Health Assessment Questionnaire; ESR = erythrocyte sedimentation rate; CSHQ = Children's Sleep Habits Questionnaire; MTX = methotrexate; N/A = not applicable.

  • N = 189.

  • N = 30.

  • §

    N = 107 subjects with available CSHQ data (subject characteristics were generally comparable to the overall group).

  • N = 44 with available CSHQ data (subject characteristics were generally comparable to the overall group).

  • #

    N = 41 with available CSHQ data (subject characteristics were generally comparable to the overall group).

  • **

    N = 36 with available CSHQ data (subject characteristics were generally comparable to the overall group).

  • ††

    N = 22 with available CSHQ data (subject characteristics were generally comparable to the overall group).

  • ‡‡

    N = 140.

  • §§

    N = 48.

  • ¶¶

    N = 47.

Age, years12.4 ± 3.012.6 ± 312.0 ± 312.2 ± 312.1 ± 2.7
Female, %72.171.772.669.474.2
White, %77.476.779.075.580.6
Duration of JIA at baseline, years4.4 ± 3.83.8 ± 3.73.9 ± 3.53.7 ± 3.84.2 ± 3.4
No. of joints with active arthritis16.2 ± 12.75.4 ± 5.53.9 ± 5.92.0 ± 3.03.6 ± 4.4
No. of joints with limitation of motion16.3 ± 14.58.8 ± 10.77.4 ± 12.66.4 ± 10.18.2 ± 13.3
Physician global evaluation of disease activity (0–100-mm VAS)54.2 ± 20.315.9 ± 12.512.5 ± 12.59.2 ± 12.212.4 ± 15.9
Parent global assessment of child's overall well-being (0–100-mm VAS)44.5 ± 24.617.2 ± 16.217.1 ± 16.811.2 ± 15.217.7 ± 18.7
C-HAQ disability index score (range 0–3, best to worst)1.3 ± 0.80.8 ± 0.80.7 ± 0.60.5 ± 0.70.7 ± 0.7
ESR, mm/hour32.0 ± 26.822.2 ± 20.823.1 ± 25.023.2 ± 27.124.6 ± 28.6
C-reactive protein level, mg/dl3.2 ± 4.41.6 ± 2.41.8 ± 3.31.4 ± 2.32.0 ± 4.2
CSHQ total score45.9 ± 7.4§45.8 ± 9.342.4 ± 5.3#42.8 ± 5.8**45.0 ± 6.0††
MTX dosage, mg/m2/week13.2 ± 4.7‡‡13.2 ± 4.4§§12.3 ± 3.6¶¶N/AN/A
Prior biologic therapy, no. (%)57 (30.0)8 (13.3)13 (21.0)5 (10.2)5 (16.1)

At the end of the 6-month, double-blind treatment period, improvements were observed in the 6 JIA core set variables in the abatacept group, whereas these parameters remained similar to baseline or worsened in the placebo group (Table 1). The impact of abatacept versus placebo on the core set variables and other measures of clinical efficacy in this trial have been reported fully elsewhere (18).

HRQOL.

Scores at baseline and at the end of period A for the 15 health concepts of the CHQ are shown in Figure 1A; reference values from the group of healthy children are also shown. At baseline, subjects had lower scores than healthy children in all health concepts. In particular, global health, physical function, role/social limitations–emotional, role/social limitations–physical, and pain/discomfort were more than 2 SDs below the mean scores for healthy children (3). After 4 months of abatacept treatment in period A, substantial improvements were seen across all of the CHQ health concepts (Figure 1A). The health concepts of global health, role/social limitations–emotional, role/social limitations–physical, and pain/discomfort improved to levels that were within 2 SDs of the mean scores for healthy children (only physical function remained more than 2 SDs below the healthy population). At the end of the 6-month double-blind period B, subjects who were randomized to continue receiving abatacept treatment had further mean improvements in all health concepts except for global behavior. In contrast, subjects randomized to placebo either remained stable or worsened (although differences were not statistically significant between the treatment groups) (Figure 1B). In both the abatacept and placebo groups, all of the health concepts reached values within 2 SDs from the mean of healthy controls at the end of period B.

Figure 1.

Change over time in Child Health Questionnaire (CHQ) health concepts scores. A, Mean scores in the individual subscales of the CHQ from baseline to the end of open-label period A. The black lines on the green bars show −2 SDs. For abatacept-treated patients in period A, n = 172 except for global health (n = 161); physical function (n = 173); pain/discomfort (n = 170); self-esteem and change in health (n = 169); and parental time impact, family activity, and family cohesion (n = 171). B, Adjusted mean change from the beginning of double-blind period B to the end of period B for patients who achieved an American College of Rheumatology Pediatric 30 improvement criteria response at the end of period A. N = 52 for abatacept and n = 34 for placebo, except for parental emotional impact, parental time impact, family activity, family cohesion, change in health, and global behavior (n = 33 for placebo); change in health and global behavior (n = 51 for abatacept); and global health (n = 49 for abatacept and n = 32 for placebo). P > 0.05 for abatacept versus placebo for all subscales. The adjusted mean change in role/social limitation–physical from baseline in placebo-treated patients was 0.01.

Improvements in the CHQ summary scores (PhS and PsS) were seen with abatacept treatment in period A in ACR Pedi 30 responders (from 30.1 to 41.8 for the PhS and from 43.6 to 49.2 for the PsS) (Figures 2A and B). The greatest improvements were seen in the PhS, with subjects entering the trial at approximately 2 SDs below the standardized mean score of 50 for the normative population, and improving to values within the range of the healthy normative population (mean ± SD 50 ± 10) (Figure 2A). During period B, improvements in both summary scores were maintained in subjects treated with abatacept (from 42.9 to 43.6 for the PhS and from 49.6 to 51.7 for the PsS), whereas reductions were observed in subjects who were randomized to placebo (P = 0.666 and P = 0.056 for the PhS and PsS, respectively) (Figures 2A and B).

Figure 2.

Change over time in Child Health Questionnaire (CHQ) component physical (PhS) and psychosocial (PsS) summary scores. A, Mean CHQ PhS scores over open-label period A and double-blind period B. B, Mean CHQ PsS scores over open-label period A and double-blind period B. The grey dotted line shows 2 SDs below the standardized mean score of 50 for the normative population. Error bars show the SEM. At the end of period B, P = 0.666 for the PhS and P = 0.056 for the PsS for abatacept versus placebo. ACR Ped 30 = American College of Rheumatology Pediatric 30 improvement criteria.

A reduction in pain was observed with the CHQ pain/discomfort health concept (Figure 1A) in period A for all subjects and in period B for subjects continuing to receive abatacept. Similarly, pain levels assessed by parents/caregivers (100-mm VAS, as part of the C-HAQ) in period A were reduced from baseline by day 15 with abatacept treatment in ACR Pedi 30 responders (−8.6), and were further reduced by the end of period A (−23.8) (Figure 3). Reductions in pain were also observed in the subjects who did not achieve an ACR Pedi 30 response by the end of period A, although these were of a lower magnitude (reduction of −5.6 from baseline to the end of period A) (Figure 3).

Figure 3.

Change over time in parent global assessment of pain (100-mm visual analog scale [VAS]). Mean global parent assessment of pain (0–100-mm VAS) scores over open-label period A for American College of Rheumatology Pediatric 30 (ACR Ped 30) improvement criteria responders and nonresponders, and over double-blind period B for responders randomized to abatacept or placebo. Abatacept responders are defined as those subjects achieving at least an ACR Ped 30 response by the end of period A. Error bars show the SEM. P = 0.105 for the abatacept versus placebo group at the end of period B.

Over 6 months in period B, pain was further reduced in the abatacept group, whereas pain scores worsened in subjects randomized to placebo (P = 0.105 for the abatacept versus placebo groups) (Figure 3).

After 4 months of abatacept treatment in period A, sleep problem index scores were decreased by 2.66 points from baseline in ACR Pedi 30 responders. The elements of sleep problems that showed the greatest improvement were bedtime resistance (−0.51), parasomnias (−0.46), and daytime sleepiness (−1.19). During period B, sleep quality improved further in abatacept-treated subjects (CSHQ scores decreased by 0.72 during period B). The elements of sleep problems that improved the most in period B were bedtime resistance (−0.36), sleep anxiety (−0.29), and daytime sleepiness (−0.48). In subjects randomized to placebo in period B, sleep quality deteriorated (CSHQ scores increased by 1.66 during period B; P = 0.076 for abatacept versus placebo at the end of period B).

Participation in daily activities.

The mean number of missed school days at baseline for the ACR Pedi 30 nonresponders was lower when compared with the ACR Pedi 30 responders. By the end of open-label period A, the number of missed school days per month was reduced by 2.6 days in abatacept-treated ACR Pedi 30 responders (Figure 4A). The number of days of normal activity per month missed by parents, including work and non-work activities, was also reduced by 2.3 days in ACR Pedi 30 responders (Figure 4B). The number of days when paid care was required was reduced by 1.14 days per month with abatacept treatment in ACR Pedi 30 responders in period A. For ACR Pedi 30 nonresponders, improvements in limitation of daily activities were either not observed or were of a lower magnitude (Figure 4). In period B, subjects randomized to abatacept experienced a further 4-fold gain in school days (1.9 days); subjects randomized to placebo also experienced a further gain in school days (0.9 days), but this was of a lower magnitude than the abatacept group (P = 0.033 for the difference between groups). There were also further gains in days of usual activity for parents of subjects in the abatacept group (0.2 days) but, in contrast, parents of subjects in the placebo group experienced a loss of days of usual activity (−1.3 days; P = 0.109 for the difference between groups). In period B, the number of days when paid care was required remained stable in both groups.

Figure 4.

Reduction in children's missed school days and parents' missed usual activity days. A, Mean number of children's missed school days per month at baseline, the end of open-label period A, and the end of double-blind period B (P = 0.033 for abatacept versus placebo). B, Mean number of parents' missed usual activity days per month at baseline, the end of open-label period A, and the end of double-blind period B (P = 0.109 for abatacept versus placebo). ACR Ped 30 = American College of Rheumatology Pediatric 30 improvement criteria.

DISCUSSION

Significant clinical efficacy benefits with abatacept have previously been demonstrated in this phase III trial in children and adolescents with JIA (18). HRQOL assessments provide valuable information regarding aspects of disease that are important to the patient, but that may not be captured by standard clinical assessments. Furthermore, recent evidence has demonstrated that children with no clinical symptoms or only mild symptoms can still have impaired HRQOL, highlighting the importance of assessing HRQOL independently of clinical status (28). Here we demonstrate that the clinical efficacy benefits previously demonstrated with abatacept in children with JIA are coupled with significant improvements in multiple aspects of HRQOL.

At entry to the trial, subjects had a high level of disease activity (evidenced by the disease parameters shown in Table 1), which was accompanied by a poor HRQOL. The CHQ (22, 24) is an instrument designed to assess HRQOL in children, and comprises 15 health concepts covering different physical, emotional, and social aspects of HRQOL. At baseline, the health concepts of the CHQ were lower than those observed in healthy children, and 5 of these health concepts were more than 2 SDs below the mean scores for the healthy population, suggesting a clinically significant impairment in HRQOL. The only health concepts that did not appear to be impacted by JIA, compared with healthy children, were those of behavior, family activity, and family cohesion, which is consistent with previous reports (29). After 4 months of open-label treatment with abatacept, the mean scores in all 15 health concepts improved, with the greatest improvements observed in pain and discomfort, global health, and physical function. For subjects randomized to abatacept during the 6-month double-blind period, scores continued to improve in all health concepts of the CHQ with the exception of global behavior. Interestingly, improvements in the double-blind period B were greatest in family cohesion and self-esteem, suggesting that early gains in physical aspects of HRQOL may then result in subsequent improvement in emotional/social aspects. In subjects randomized to placebo in period B, CHQ scores either deteriorated (in 8 health concepts) or improvements were of a lower magnitude (in 7 health concepts) than those observed in subjects who continued on abatacept therapy.

In addition to the CHQ individual health concepts, 2 summary scores covering the physical and psychosocial aspects of HRQOL were calculated (22), and in keeping with previous reports in children with JIA, baseline values for the PhS were considerably lower than for the PsS (3). However, similar trends were observed for both summary scores in response to treatment, with the greatest improvements seen in the PhS. In subjects who were ACR Pedi 30 responders at the end of the open-label period A, values improved through 4 months of open-label abatacept treatment, whereas ACR Pedi 30 nonresponders showed either no improvement or a low magnitude of change. For the PhS in particular, there was a marked difference between scores in the ACR Pedi 30 responders versus nonresponders, suggesting a strong link between the physical aspects of HRQOL and clinical efficacy as assessed by the ACR Pedi 30 criteria. In subjects who remained on abatacept therapy through period B, scores were maintained or improved further, whereas withdrawal of abatacept resulted in deterioration of scores, albeit of a moderate magnitude. The PhS and PsS improvements were similar across the different JIA subtypes, as also reported by Oliveira et al in a large cohort of more than 3,000 JIA patients (2), and in a trial of JIA patients newly treated with MTX (3). This was expected because although the study enrolled subjects with different onset subtypes (oligoarthritis, polyarthritis, and systemic), all had polyarticular course JIA that was unresponsive to MTX and/or a biologic agent.

Pain can have a strong influence on HRQOL, and has been associated with a poor PhS following 6 months of treatment with MTX (3, 9) and with sleep disturbances in subjects with JIA (30). The two parameters are probably intrinsically linked; it has been proposed that pain interferes with sleep and that, conversely, sleep disturbance can exacerbate feelings of pain (31). Changes in pain and sleep problems with abatacept treatment followed a similar pattern to those observed with HRQOL assessments. Improvements were observed during the 4-month open-label period; these improvements were further increased in subjects who remained on treatment during the 6-month double-blind period, whereas subjects who stopped abatacept either remained stable or worsened.

The occurrence of JIA in school-age children can significantly impact their education, with effects ranging from a few missed days of school each year to prolonged time off, which may require tutoring to keep up with missed work (6). Time off from school can result in the interruption of parents' usual daily activities, or can mean that paid care is required. Abatacept treatment resulted in a reduction in the mean number of school days missed per month from an average of more than 4 days at baseline to approximately 0.5 days at the end of period B after 10 months of treatment. This was mirrored by similar reductions in the number of usual activity days missed by the parent or caregiver due to the child's JIA; reductions in the number of days of paid care required were also observed with abatacept treatment.

Pediatric conditions such as JIA have a substantial impact on patients' HRQOL, the effects of which impact beyond the child to affect the parents and the entire family. The improvements observed in HRQOL measures were accompanied by increased participation in daily activities, suggesting that the HRQOL benefits provided by abatacept are clinically meaningful. This is consistent with the significant and clinically meaningful HRQOL improvements observed in adult RA subjects treated with abatacept (32, 33).

When we compared our results with published pediatric data we found many similarities. Oliveira et al examined a large cohort of more than 3,000 children with JIA compared with an equal sample of healthy children, and found that patients with JIA have a significant impairment of their HRQOL compared with healthy peers, particularly in the physical domains (2). Also, Cespedes-Cruz et al (3), in a JIA trial in patients newly treated with MTX, found that treatment produces a significant improvement across a wide range of HRQOL components, particularly in the physical domains. To the best of our knowledge, only one recent prospective study of a Dutch JIA biologics registry has examined HRQOL outcomes in subjects treated with etanercept; statistically significant improvements were seen in 11 CHQ health concepts and in PhS scores, which had started 2.5 SDs below the scores of healthy children and improved to 1.5 SDs below, whereas the PsS remained substantially unchanged (14). Accounting for the known differences in study populations, this brief comparison with the published literature demonstrates how second-line treatment with either MTX or biologic agents is able to significantly ameliorate mainly the physical well-being of children, whereas the psychosocial aspects of the disease do not appear to be much affected in the JIA population. However, no firm conclusions can be drawn at this stage about which second-line treatment is preferable in terms of HRQOL improvement.

Several limitations of this study should be considered. First, the HRQOL assessments performed included parent- or caregiver-reported evaluations. Because previous findings have suggested a discordance between self-reports and proxy reports of HRQOL (34, 35), these data should be interpreted within this context. In addition, the data presented here are as observed, including only subjects who had assessments available at the visit of interest, with analysis limited to main periods of the trial (periods A and B). This raises the possibility that the subset of subjects with available data could be different from the overall trial population in some way. However, for measures with a considerable number of missing assessments (e.g., the CHSQ), the baseline characteristics of patients included in the analyses were comparable with the rest of the trial population.

In conclusion, the results presented here demonstrate that abatacept provided significant and meaningful improvements in HRQOL, which translate into real-life tangible benefits for children and adolescents with JIA and their caregivers. It will be important to evaluate if the improvements in HRQOL documented here will be maintained over time in the children who continued to receive abatacept in the long-term extension of this trial (36).

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Ruperto had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Ruperto, Lovell, Li, Sigal, Block, Martini, Giannini.

Acquisition of data. Ruperto, Lovell, Li, Sztajnbok, Goldenstein-Schainberg, Scheinberg, Penades, Fischbach, Alcala, Hashkes, Hom, Jung, Oliveira, Wallace, Alessio, Quartier, Cortis, Eberhard, Simonini, Lemelle, Chalom, Sigal, Martini.

Analysis and interpretation of data. Ruperto, Lovell, Li, Sigal, Block, Covucci, Nys, Martini, Giannini.

ROLE OF THE STUDY SPONSOR

The study, analysis, manuscript preparation, and submission were conceived and driven by the Paediatric Rheumatology International Trials Organisation and the Pediatric Rheumatology Collaborative Study Group. The study was funded by Bristol-Myers Squibb. Medicus International provided editorial assistance. Bristol-Myers Squibb approved the final version of the manuscript.

Acknowledgements

The study, analysis, and manuscript preparation were conceived and driven by PRINTO and PRCSG. We thank Medicus International for their editorial assistance. The authors wish also to thank the following additional investigators who participated in the trial: Christian Huemer, MD (Austria); Clovis A. Silva, MD, Claudia Saad-Magalhães, MD (Brazil); Chantal Deslandre, MD, Richard Mouy, MD, Brigitte Bader Meunier, MD (France); Ruben Burgos-Vargas, MD, Carlos Abud Mendoza, MD, Alejandro Flores Nunez, MD, Nadina Rubio Perez, MD (Mexico); Ivan Foeldvari, MD, Gerd Horneff, MD, Kirsten Minden, MD, Hans Iko Huppertz, MD (Germany); Valeria Gerloni, MD, Anna Loy, MD, Angelo Ravelli, MD, Fernanda Falcini, MD (Italy); Eliana Paz, MD, Jose Chavez, MD (Peru); Jose A. Melo-Gomes, MD (Portugal); Francisco J. Blanco, MD (Spain); Michael Hofer, MD (Switzerland); Alan Kivitz, MD, Marilynn Punaro, MD, Nancy Olson, MD (US).

Ancillary