The impact of infliximab treatment on quality of life in patients with inflammatory rheumatic diseases

Rheumatoid arthritis (RA), psoriatic arthritis (PsA) and ankylosing spondylitis (AS) are common inflammatory rheumatic diseases with severe consequences for patients' health-related quality of life (HRQoL). RA is the most common of the three, and is characterized by chronic, symmetric, and erosive synovitis of peripheral joints. RA is two to three times more prevalent in women than men and has a peak onset between the ages of 40 and 50 years. Compared with RA, PsA also has articular manifestations, but the joint patterns can be different, including the number of affected joints and the type of arthritis [1]. PsA typically includes psoriatic skin lesions and has a similar prevalence in men and women. Patients with PsA are also usually affected at a younger age than those with RA. Both RA and PsA can lead to impaired function and reduced mobility. AS predominantly affects the axial skeleton and sacroiliac joints (sacroiliitis), which typically leads to decreased spinal mobility. However, AS may also include extraspinal manifestations, such as peripheral arthritis and uveitis, and it is often associated with inflammatory bowel disease and psoriasis. In contrast to RA, AS is three times more prevalent in men than women, and symptoms usually begin in adolescence.

Daily pain, stiffness, fatigue, and physical disability are common features of all three rheumatic diseases. Persistent active disease without effective treatment may lead to permanent loss of physical function, reduced productivity, and increased rates of work disability about 10 years after disease onset [2]. Impairments in physical functioning and disability, and the strategies patients use to cope with them, can significantly affect HRQoL. Traditionally, RA is considered to be the most severe of the three diseases, exerting the greatest impact on physical aspects of HRQoL. However, the cohorts of previous studies that compared the HRQoL of patients with RA to that of patients with PsA or AS consisted primarily of patients with mild-to-moderate disease [3, 4]. Disease severity is an important factor influencing the HRQoL of patients. Therefore, the objectives of this study were to compare the HRQoL of patients with moderate-to-severe RA, PsA, or AS, and to compare the effect of treatment with infliximab on the HRQoL of patients with these diseases.

This analysis included 1,987 patients from four clinical trials in which the safety and efficacy of infliximab was evaluated in rheumatic diseases. Two of the trials were conducted in patients with RA: the Anti-TNF Trial in Rheumatoid Arthritis with Concomitant Therapy (ATTRACT) (n = 428) [5–7] and the Safety Trial for Rheumatoid Arthritis with REMICADE Therapy (START) (n = 1083) [8]. The disease severity entry criteria for these studies were similar. All patients must have had active RA with 6 or more tender and swollen joints. In ATTRACT, patients must have also had two or more of the following: morning stiffness ≥45 minutes, erythrocyte sedimentation rate >28 mm/hour, and C-reactive protein >2 mg/dL. Patients in both studies must have demonstrated these symptoms despite receiving stable doses of MTX (≥12.5 mg/week for patients in ATTRACT). Patients in ATTRACT were not permitted to have significant comorbidities or concomitant medications, whereas patients were permitted in START as long as they did not have conditions or concomitant medications that were prohibited by the product labeling.

Data from the Ankylosing Spondylitis Study for the Evaluation of Recombinant Infliximab Therapy (ASSERT) (n = 279) were used to assess HRQoL in AS [9], and data from the Infliximab Multinational Psoriatic Arthritis Clinical Trial II (IMPACT II) (n = 200) [1] were used to assess HRQoL in PsA. Patients were eligible for the ASSERT study if they had a Bath Ankylosing Spondylitis Disease Activity Index score of ≥4 (range 0 to 10) and a spinal pain assessment score of ≥4 on a visual analog scale (range 0 to 10). Patients were eligible for the IMPACT II study if they had ≥5 tender and ≥5 swollen joints and either CRP levels of ≥15 mg/L and/or morning stiffness lasting ≥45 minutes.

All studies were conducted at multiple study sites in North America and Europe. The START study was also conducted at sites in South America.

Detailed descriptions of patient selection criteria and study designs for these trials have been published elsewhere [5–10]. As indicated in these reports, all studies were conducted in compliance with the Declaration of Helsinki, and all patients gave written informed consent to participate in the study. The impact of treatment on the HRQoL of patients in the ATTRACT [6] and IMPACT II [11] studies have also been published previously. In each study, patients were randomly assigned to receive either infliximab or placebo in a 3-dose induction regimen followed by maintenance therapy. Patients in ATTRACT received infliximab at 3 mg per kg of body weight or 10 mg per kg of body weight every 4 or 8 weeks, patients in START received infliximab at 3 mg per kg body weight or 10 mg per kg body weight every 8 weeks, patients in ASSERT received 5 mg per kg body weight every 6 weeks, and patients in IMPACT II received 5 mg per kg body weight every 8 weeks. All patients in RA clinical trials received concomitant methotrexate (MTX). Patients in IMPACT II were allowed to continue therapy with stable doses of MTX if they were receiving it at baseline. None of the patients in ASSERT received concomitant MTX.

HRQoL was assessed using Short Form Health Survey-36 (SF-36) at weeks 0, 10, 30, and 54 in ATTRACT; weeks 0, 6, and 22 in START; weeks 0, 12, and 24 in ASSERT; and weeks 0 and 14 in IMPACT II. The norm-based SF-36 scoring system was used to calculate physical and mental components summary scores (PCS, MCS) as well as the 8 individual scales. Scores for each scale range from 0 to 100 with higher SF-36 scores being indicative of better HRQoL and a score of 50 ± 10 representing the mean ± standard deviation for the general population of the United States. Baseline SF-36 scores and changes from baseline after treatment were adjusted for age, sex, and disease duration when comparing across diseases using analysis of covariance (ANCOVA). Effect size and 95% confidence interval (CI) of change after treatment between treatment groups was estimated using Hedges' method [12].

The burden of RA, PsA, and AS on patients, their caregivers, and society as a whole is significant. Chronic inflammation leads to pain, stiffness, joint erosions, ankylosing features, and impaired mobility, which can result in functional disability, increased health care costs, and reduced employability. These factors and the strategies patients use to cope with disease burden contribute to an impaired HRQoL for affected patients. In this study, we directly compared the HRQoL of patients with three inflammatory rheumatic diseases. The distribution of baseline PCS scores for all three diseases were shifted to the left of the distribution for the general United States population. These results demonstrate that these patients had greatly impaired physical HRQoL relative to the general population.

The data used in this analysis were from patients who participated in clinical trials of the anti-tumor necrosis factor-α (TNFα) agent infliximab. To be eligible for these studies, patients must have had moderate-to-severe disease activity at baseline. The comparison of the mean baseline PCS, as well as scores for the individual scales, showed that patients with PsA and those with AS had similarly impaired HRQoL relative to patients with RA. This finding is different from the reports of previous studies in which patients with RA had lower physical SF-36 summary scores compared with those with AS [3] or PsA [4].

One potential explanation for this discrepancy is the disease severity of patients in each of the cohorts. In the Toronto study that compared the quality of life of patients with RA to that of patients with PsA [4], the mean number of active joints in patients with PsA was 6, while the baseline mean number of tender and swollen joints for patients in the IMPACT II study was 25 and 14 joints, respectively [10]. Similarly, the mean number of active joints in patients with RA in the Toronto study was lower (6) than the mean number of tender and swollen joints for patients in ATTRACT (32 and 22 joints, respectively) and START (24 and 17 joints, respectively). In the analysis of patients with RA or AS in a Dutch study [3], the mean baseline Bath Ankylosing Spondylitis Disease Activity score was 3.9 compared with a mean score of 6.4 for patients in the ASSERT trial [9]. Disease activity assessments for patients with RA in the Dutch study were different from those used in ATTRACT or START studies; however, the mean baseline SF-36 physical component summary score in the RA cohort of the present study (29) was lower than those of patients with RA in the Dutch study (35.7 in men and 34.3 in women).

Results of the present analysis also demonstrate the significant improvement in HRQoL after treatment with the TNFα inhibitor infliximab. After adjustment for age, sex, and disease duration, patients in all the disease cohorts who received infliximab demonstrated significantly greater improvement from baseline in PCS than those who received placebo. Role physical and bodily pain were the domains that had the greatest magnitude of change in each disease cohort. Early improvement was evident at the first assessment point, which ranged from week 6 to week 14 depending on the study design. The magnitude of the difference in the change from baseline in PCS between the infliximab and placebo groups was comparable among the three diseases.

Patients with RA and PsA treated with infliximab also showed greater improvement in MCS than those treated with placebo, although the improvement in MCS between infliximab and placebo groups was not statistically significant in the AS cohort.

Improvements in PCS persisted throughout the placebo-controlled period of each study. The longest placebo-controlled period was in the ATTRACT study, in which improvements in mean PCS were sustained and increased slightly through week 54.

We observed several notable differences in the changes in the placebo group across diseases. Changes in PCS in the placebo group were lower in the PsA and AS cohorts compared with the RA cohort. All patients in the RA cohort, including those in the placebo group, were incomplete responders to MTX at baseline and received concomitant MTX throughout the study. However, eligibility for the trials was determined using clinical signs and symptoms criteria, not quality of life measurements.

There are some potential limitations of this retrospective study, and the results should be interpreted with caution. Although the SF-36 is robust for comparisons across diseases [14], each study assessed quality of life at different time points according to the study design. Thus, the first quality of life assessment ranged from week 6 in START to week 14 in IMPACT II. Moreover, each study had different selection criteria, study design, and infliximab dosages, making comparisons of HRQoL difficult. Although all 3 rheumatic diseases evaluated were immune-mediated inflammatory disorders that share some pathophysiological characteristics, comparisons of patients with different diseases should always be interpreted with caution. In the analysis, we adjusted SF-36 scores for age, sex, and disease duration, to attempt to control for potentially confounding factors.

Patients with rheumatic disease have an impaired physical HRQoL relative to the general population, and the magnitude of that impairment is similar among patients with moderate-to-severe RA, PsA, or AS who were included in clinical trials with infliximab. Treatment with infliximab resulted in significantly improved HRQoL.