Rheumatoid arthritis (RA) is one of the most common forms of chronic inflammatory arthritis. This symmetrical polyarthritis mainly affects middle-aged females and leads to progressive joint destruction and loss of function (1). Osteoporosis is characterized by low bone mass, leading to bone fragility as well as a consequent increase in fracture risk (2). It is well known that patients with RA have an increased risk of developing osteoporosis. It has been reported that annual bone loss is greater in patients with active RA than in healthy patients (3). Compared with the general population, a two-fold increase in the frequency of osteoporosis in the spine was observed in RA patients (4). In addition, a meta-analysis revealed that the relative risk for bone fracture was higher among patients with RA than among those without RA (risk ratio 2.25) (5).

In recent years, a greater understanding of immunopathology has facilitated the development of biologic disease-modifying antirheumatic drugs (bDMARDs), which target specific components of the immune response and improve the clinical outcomes of RA (1). Inhibition of pro-inflammatory cytokines, such as tumor necrosis factor-α inhibitor (TNFi), seems to be effective in reducing disease activity and inhibiting bone loss in patients with RA (6–9). CTLA-4 Ig (Abatacept) is a fusion protein that regulates the T-cell co-stimulatory signal and is effective in attenuating disease activity and reducing joint damage in RA patients who have an inadequate response to methotrexate (10, 11). Previous research has demonstrated that abatacept can increase BMD at the femoral neck (FN) and is superior to that of other biologics in patients with RA (12).

In our previous investigation, we demonstrated that three-year biological/targeted synthetic DMARD (b/tsDMARD) treatment can prevent bone loss in RA patients and conventional synthetic DMARD (csDMARD) does not (13). However, the long-term effect in preserving BMD in patients with RA treated with TNFi or abatacept is unknown. The primary aim was to explore the BMD changes in patients with RA treated with csDMARD, TNFi, and abatacept via a three-year, real-world, observational, cohort study. The secondary aim was to investigate the synergistic effect of anti-osteoporosis therapy (AOT) on BMD in patients with RA receiving different therapies.

Current investigation demonstrated that AOT can prevent bone loss irrespective of regimen used, while in patients without AOT, participants who received csDMARD had the most obvious bone loss at all sites. Participants who received abatacept therapy demonstrated stable BMD at all sites, irrespective taking AOT therapy or not. This investigation revealed that abatacept may have a better systemic bone loss protection effect in RA patients than the other two regimens.

The b/tsDMARD not only demonstrated better control of disease activity, but also showed a better bony erosion protection effect than csDMARD in RA patients (17–26). Our previous investigation revealed that long-term b/tsDMARD therapy demonstrated a potentially beneficial effect on the protection of systemic bone loss (13). However, previous investigations regarding the effect of biologics on the prevention of systemic bone loss in RA have only focused on changes in bone turnover markers (18–21), over a short-term observation period (25, 26), and with a lack of an adequate control group (25, 27). In addition, the long-term effects of bDMARDs with different mechanisms of action on systemic bone loss in patients with RA remain obscure.

CTLA-4 is a negative regulator that inhibits antigen-specific immune responses after T-cell activation by interfering with the interaction of CD28 on T-cells and CD80/86 on antigen-presenting cells, including dendritic cells, macrophages, and B-cells (28, 29). Axmann et al. found that CTLA-4 directly binds to osteoclast precursor cells and dose-dependently inhibits receptor activator of nuclear factor-κB ligand (RANKL)-mediated osteoclastogenesis in an animal model (30). However, in human studies, abatacept demonstrated a controversial effect on the protective effect of bone loss in RA patients (12, 31).

Pro-inflammatory cytokines, such as TNF-α, IL-6, and IL-1, play a key role in the pathogenesis of RA and have been approved as treatment targets (7). The interaction between inflammation and osteoporosis has been previously described (32). In synovial fibroblasts, TNF-α upregulates Dickkopf-1 (DKK1), a negative regulator of Wnt signaling, leading to suppression of bone formation (8). In addition, TNF-α directly inhibits osteoblast differentiation and function (9). Therefore, the inhibition of pro-inflammatory cytokines seems to be an effective bone-protecting principle in patients with RA. Indeed, previous studies have demonstrated that TNFi is potentially beneficial for bone loss (17–26). However, the results of these studies were based on their influence on bone turnover markers (18–21). Furthermore, a significant increase in serum parathyroid hormone was found in RA patients undergoing TNFi treatment, which might promote bone resorption and blunt the anti-osteoporotic effect of TNFi (33). Taking together, whether or not there is a substantial improvement in BMD in RA patients treated with TNFi remains controversial (22–24).

As TNFi and abatacept were the first two regimens launched and are the most commonly prescribed biologics in Taiwan, we compared the protective effect to systemic bone loss of these two biologics with csDMARD. The current investigation revealed that abatacept therapy arrested systemic bone loss at FN, TH, and L1-4 in a three-year follow-up period. In addition, it seems that abatacept exhibited bone loss protective effect whether on AOT or not in RA patients. Furthermore, in terms of percent change of BMD, compared with TNFi, abatacept demonstrated a superior effect at FN and L1-4 and at L1-4 in patients who were taking AOT or not taking AOT, respectively.

It remains unclear why abatacept had a more favorable effect on BMD than csDMARD or TNFi. Our previous investigation demonstrated that adequate control of disease activity can protect bone loss in RA patients (34). Meanwhile, it has been demonstrated that positive ACPA status is associated with a differential treatment response to abatacept, but not TNFi (35). Furthermore, abatacept treatment showed differential efficacy in RA patients with higher ACPA titers (36). Previous studies reported that RA patients positive for ACPA had a lower BMD and a higher 10-year probability of fracture as evaluated by FRAX^® (37, 38). Patients with RA receiving abatacept therapy demonstrated a decline in ACPA titers or ACPA seronegative conversion effects (39). In addition, compared with group I or II, the rate of ACPA positivity was significantly higher in the abatacept group (group III) (p = 0.002) in our cohort. Based on the aforementioned findings, we hypothesize that adequate control of disease activity (DAS28-ESR) between baseline and during the observation period (4.9 ± 1.6 and 3.2 ± 1.0, p < 0.001) in the abatacept group and higher rate of ACPA in group III could partly explain the discrepancy in the effect on BMD among the groups. It is also well known that low BMI is one of risk factors of systemic bone loss or osteoporosis (40). In current investigation, abatacept group, although had lower BMI, increased more BMD than other groups after 3 years. After adjusting age, gender, BMI, disease duration, ACPA positivity, and baseline DAS28-ESR, predicted change in BMD at all measured sites in group C remained better than in other groups, suggesting that the better effect of abatacept on prevention of systemic bone loss in RA when compared to other comparison regimens.

As the current study is a real-world investigation, we did not exclude participants who received AOT during the observation period to elucidate the interaction of DMARD therapy and AOT in terms of bone protective effects. Participants who did not receive AOT had significant bone loss at all sites in group A and at the lumbar vertebrae in group B. AOT had a protective effect against bone loss in all groups at all sites. These results suggest that AOT plays the most important role in bone loss protection in RA patients receiving either csDMARD or biologics.

A strength of the current investigation is that it is a longitudinal, real-world, observational, registry, cohort study. We measured and recorded the characteristics of the disease entity at baseline and serial disease activity, which could potentially influence BMD changes during the study period. In addition, most previous studies were single-arm studies without an adequate control group (25, 27). Our initial investigation revealed a significant difference in the rate of glucocorticoid use among groups. As glucocorticoid use is a well-known risk factor for bone loss, we performed a 4:2:1 matching for glucocorticoid use to exclude the confounding effect of glucocorticoids, which had not been done in previous investigations. Furthermore, to adjust confounders of osteoporosis, multiple regression analysis was used to develop a model to predict the change in BMD. Finally, this study is the first investigation to explore the long-term systemic bone loss protective effect among DMARDs with different mechanisms of action to elucidate the effects of biologics on RA patients.

The current study has some limitations. As it was a real-world study, we did not exclude participants who had already received biologics before enrollment. Hence, we could not exclude the residual effect of previous medications on BMD after enrollment. However, we excluded participants with biologic switching during the observation period to avoid additional confounding effects. In addition, we did not compare the bone loss protective effect between abatacept and biologics other than TNFi, eg. tocilizumab, rituximab. So far, we could not know which biologics is the best one to prevent bone loss in RA patients. Fracture prevention is a hard outcome to measure in osteoporosis studies, and we could not compare the effect of fracture prevention among the treatment groups owing to the short duration and relatively small sample size of our study.

RA patients receiving long-term abatacept illustrated a better bone loss protective effect than patients receiving csDMARD or TNFi. Anti-osteoporosis therapy has a vital protective effect on bone loss irrespective of regimens for RA therapy used. Further studies are needed to clarify whether abatacept or other biologics could prevent fragility fractures in patients with RA.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by Chang Gung Memorial Hospital, Kaohsiung and Taipei Veterans General Hospital. The patients/participants provided their written informed consent to participate in this study.

Study concept: M-HC and T-TC. Study design: M-HC, S-FY, and T-TC. Data analysis: M-HC, S-FY, T-TC. M-HC, S-FY, J-FC, W-SC, T-LL, C-TC, C-YH, H-ML, Y-CC, C-YT, and T-TC were responsible for interpretation of the data and for drafting and revising the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by grants CMRPG8F1111 and CMRPG8K0441 from Chang Gung Memorial Hospital (https://www.cgmh.org.tw/), which sponsored the cost of data collection, gathering, processing, and publication. The funder had no role in the study design, data analysis, decision to publish, nor manuscript preparation.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.