Background
In rheumatoid arthritis (RA), bone damage on radiographs is visible as erosions and periarticular osteoporosis. Substantial data support that both erosions and osteoporosis in RA share a common cellular pathway which involves stimulation of the osteoclast. This osteoclast activation depends on stimulation from receptor activator of nuclear factor-κ ligand (RANKL) which binds to the receptor activator of nuclear factor-κ (RANK) on the osteoclast. The expression of RANKL is stimulated by pro-inflammatory cytokines (i.a. TNF-α, interleukin-1 (IL-1), IL-6 and IL-17). In addition recent data also suggest decreased osteoblast activation through the Wnt system [
1].
In comparison to disease modifying anti-rheumatic drugs (DMARDs) including methotrexate (MTX), anti-TNF therapy has been shown to be superior in reducing the rate of both radiographic joint damage [
2‐
4] and hand bone loss [
5,
6]. Recently, the rate of radiographic joint progression was reported to be reduced independent of a patient's clinical response to anti-TNF therapy [
7,
8]. This may suggest an additional positive effect of anti-TNF therapy on bone in RA independent of its anti-inflammatory effect. This has previously not been examined for periarticular bone loss.
The objective of this study was to examine if treatment with the TNF-α inhibitor adalimumab also could reduce periarticular bone loss in RA patients independent of disease activity.
Methods
The PREMIER study cohort was used to examine the relationship between periarticular bone loss and clinical response in RA patients treated with MTX and anti TNF-therapy.
In this cohort, radiographic joint progression has recently been reported to be reduced independently of patients' clinical responses to anti-TNF therapy with adalimumab [
7].
The clinical, radiographic and bone density data from this 2-year, multi-centre, double-blind, randomised controlled study has previously been described in detail [
6,
9] In short, the efficacy and safety of adalimumab plus MTX was compared with adalimumab monotherapy and with MTX monotherapy in 799 adult patients with early (< 3 years, mean disease duration 9.1 month), aggressive RA (inclusion criteria: ≥8 swollen joint; erythrocyte sedimentation rate ≥28 or C-reactive protein (CRP) ≥1.5 mg/dl; erosions or rheumatoid factor positive), who previously had not been treated with MTX [
9].
Digital X-ray radiogrammetry (DXR) (Sectra, Linköping, Sweden) was used to measure hand metacarpal cortical index (MCI) on the same digitised hand X-rays used for assessment of radiographic joint damage. DXR-MCI is defined as the combined cortical thickness divided by the bone width and is a relative bone measure independent of bone size and bone length [
10,
11]. In the literature short-time in-vivo precision (CV%) has been reported to range from 0.31-0.64% for DXR-MCI [
10,
12,
13].
DXR-BMD (def:
cxVPA
comb
x(1-
p), where c is a density constant, VPA is volume per area, and p is porosity) was intended to be the main outcome measure in this study. However, many radiographs could not be analysed for BMD because of unknown image resolution. The equation for DXR-BMD is based on volume per area and requires a known resolution. Thus, DXR-MCI, which is a relative measure less dependent of image resolution, was used as the primary outcome measure [
6]. DXR-MCI has been shown to be highly correlated with hand bone mineral density, both measured as DXA hand and DXR-BMD [
14]. Detailed information on the DXR-MCI measurement of the PREMIER RA cohort has previously been reported [
6].
This sub analysis involved 188 patients from the MTX group and 214 patients from the combination group with available DXR data and disease activity measured by DAS28 at 52 weeks [
15]. The combination group received adalimumab 40 mg subcutaneously every other week plus weekly oral MTX (rapidly increased to 20 mg/week), and the monotherapy group received weekly oral MTX plus placebo injections.
Disease activity was assessed by the disease activity score using a 28 joint count (DAS28) at 52 weeks. Levels achieved for DAS28 scores were stratified according to the EULAR improvement criteria [
16] for remission (DAS28 ≤ 2.60), low disease activity (DAS28 2.61-3.20), moderate disease activity (DAS28 3.21-5.10), and high disease activity (DAS28> 5.10).
DXR-MCI bone loss at 52 weeks follow- up were compared for patients in remission and low disease activity vs. moderate and high disease activity according to treatment groups, as well as for patients in remission, low, moderate and high disease activity.
DXR-MCI loss dependent on inflammation, assessed by CRP during follow-up, was also analysed according to treatment groups. Mean CRP was calculated from values at baseline, 26 weeks and 52 weeks follow-up, and a mean CRP of <10 was defined as low inflammation.
Statistics
Since the data were skewed, non-parametric analyses were conducted. No imputations were performed. Baseline values were compared between treatment groups with the Mann-Whitney method for continuous variables and the chi-squared method for categorical variables. DXR-MCI loss is described as negative values. Group analyses were performed with the Mann-Whitney method for two independent samples and Kruskall-Wallis for more than two independent samples. Spearman correlation analyses were conducted in an attempt to correlate changes in DXR-MCI with disease activity measured by DAS28. Analyses that came out with a p-value ≤ 0.05 were considered as statically significant.
Study ethics
The PREMIER study was approved a central institutional review board or independent ethics committee at each participating site approved the study, and all patients provided written informed consent [
9].
Discussion
The main finding in this study was that adalimumab in combination with MTX reduces hand bone loss independently of clinically assessed disease activity. This has previously been shown for radiographic joint damage [
7,
8], but the present study is the first to show this for hand bone loss. This disconnection between inflammation and bone loss was not seen in MTX-treated patient. Patients with a poor clinical response or elevated CRP are at high risk of developing bone damage when treated with MTX monotherapy, while the bone will be protected independent of clinical response among users of MTX in combination with anti-TNF-α therapy.
These results support the hypothesis that TNF-α can influence bone loss not only by stimulating RANKL by inflammation. The most probable additive mechanism is that TNF-α activates the osteoclast directly by binding to osteoclasts precursors through TNF-α receptor. Blocking of this receptor will down regulate the osteoclast formation [
1,
17]. Another theoretical mechanism of TNF-α could be inflammation independent stimulation of RANKL, however this has not yet been clinically proven. The fact that TNF-α does have more than one way of stimulating the osteoclast may explain the positive effect on bone despite any clinical improvement [
7,
8].
Despite that bone loss was independent of disease activity in the combination group, there were still a bone loss observed. This loss may have been a result of the patient selection in the PREMIER study. The included patients had high disease activity and poor prognosis in terms of bone damage as Rheumatoid Factor-positive and erosive disease [
9,
18].
This is a retrospective study and we did not have any influence on the technical conditions regarding the radiographs. Due to difficulties analysing DXR-BMD the relative measure DXR-MCI was used, as described in detail previously [
6]. However, DXR has improved the precision of MCI, and there is a strong correlation between DXR-BMD and DXR-MCI (r > 0.9) [
6,
14]. Another limitation was our inability to retrieve information on the use of bisphosphonates. This may be of importance as treatment with bisphosphonates increases DXR bone density [
19].
Competing interests
M. Hoff, T. K. Kvien, A. Kavanaugh and G. Haugeberg have received consulting fees as speakers from Abbott Laboratories. Tore K. Kvien, A. Kavanaugh and G. Haugeberg have received founding for independent research from Abbott Laboratories. Aake Elden is employed by Abbott Laboratories. Johan Kälvesten is employed by Sectra.
Authors' contributions
MH performed statistical analyses and prepared the manuscript. TKK contributed to statistical analysis and interpretation of results. JK analysed the radiographs. AE prepared the collection of radiographs. AK contributed to statistical analysis and interpretation of results. GH contributed to statistical analysis and interpretation of results and was the main responsible investigator of this project. All the authors substantially contributed to the manuscript. All authors read and approved the final manuscript.