Introduction
The availability of biologic disease-modifying anti-rheumatic drugs (bDMARDs) and targeted synthetic disease-modifying anti-rheumatic drugs (tsDMARDs) has established treat-to-target strategies in rheumatoid arthritis (RA), leading to improved outcomes. These treatments have enabled more patients to achieve and maintain remission or low disease activity, providing protection against joint deformities and severe complications compared to previous decades. However, despite multiple cycles of biologics or tsDMARDs, there are still RA patients who fail to reach the treatment target, and this condition has recently been termed "difficult-to-treat (D2T) RA" [
1,
2]. The European League Against Rheumatism (EULAR) has provided a definition for D2T RA, which refers to cases where a patient has active or progressive disease while receiving multiple biologics or tsDMARDs with different mechanisms of action [
3]. The prevalence of D2T RA varies depending on the characteristics of the study population and has been reported to range from 5 to 20% of all RA patients.
The concept of “difficult” in D2T RA indicates a state of refractory disease where traditional DMARDs are ineffective. However, it is important to note that the term also encompasses other factors that contribute to the difficulty in achieving treatment success. These factors include limited DMARD options due to adverse events or comorbidities, as well as non-adherence to prescribed treatments [
4]. In an international survey conducted to establish the definition of D2T RA, several prominent characteristics were identified. These included the presence of interfering comorbidities, extra-articular manifestations of RA, radiographic progression indicating joint damage, the presence of synovitis as defined by ultrasonography, and significant side effects from previous treatments [
5,
6]. Recently, the EULAR addressed the considerations and research agendas pertaining to D2T RA [
7]. The EULAR raised several important questions related to D2T RA, such as identifying suspected contributing factors including smoking and obesity, understanding the mechanisms underlying the failure of biologic and tsDMARDs to suppress inflammation in D2T RA, and evaluating the utility of methods like synovial biopsies in selecting appropriate treatment options. Additionally, the assessment of comorbidities and patient compliance with prescribed medications was emphasized. Despite these inquiries, there is still a lack of comprehensive data regarding the proportion of “true” refractory D2T RA and the efficacy and safety of b/tsDMARDs in this specific patient population. To date, there is a scarcity of references that comprehensively outline the clinical characteristics of D2T RA and propose effective management strategies. It is crucial to address this knowledge gap in order to improve treatment outcomes and enhance the overall management of patients facing the challenges of D2T RA.
Therefore, this study aimed to examine the overall treatment outcomes, including drug survival, in patients with D2T RA who received b/tsDMARDs. We attempted to identify the clinical characteristics associated with D2T RA by analysing data from a nationwide and prospective cohort. The study evaluated the effectiveness, safety, and rates of drug discontinuation or switching among patients with D2T and non-D2T RA who were treated with b/tsDMARDs.
Discussion
Patients with D2T RA exhibited distinct characteristics compared to those with non-D2T RA, including younger age, longer disease duration, lower patient global assessment scores, higher SDAI and RAPID3 scores, negative RF, and less prior use of csDMARDs such as methotrexate, sulfasalazine, and leflunomide. However, despite these differences, the overall rate of drug retention did not significantly differ between patients with D2T RA and non-D2T RA after an 8-year period. The proportion of adverse events causing drug withdrawal did not differ between the groups, the proportion of drug withdrawals due to inefficacy was higher in patients with D2T RA compared to those with non-D2T RA.
Regarding the risk factors for D2T RA, it was observed that female RA patients and younger patients were more susceptible to D2T, which is consistent with previous data indicating that female sex and earlier onset of RA contribute to higher disease activity or D2T in RA [
11,
12]. However, neither BMI nor smoking history showed an association with D2T, in contrast to other studies on D2T RA. Previous research has shown that obesity is associated with a lower rate of achieving remission or low disease activity in RA [
13‐
16]. At this point of discrepancy, the BMIs of the study population differed between the previous studies and the KOBIO data. The mean BMI in the KOBIO data was 22.8 ± 3.5 kg/m
2, while the proportions of overweight (defined as BMI 25–30 kg/m
2) or obese (defined as BMI > 30 kg/m
2) patients were more than 32% and 10%, respectively [
17]. Overweight and obesity may contribute to the development of D2T RA, but not in a population with normal BMI values. In addition, the proportion of smoking patients in the KOBIO data was lower than that reported in other cohorts, where it constituted more than 20% of the population, but less than 10% in the Korean RA population including KOBIO data [
11,
18,
19]. It is important to consider that the factors predicting D2T RA may differ depending on the characteristics of the RA population.
Patients with D2T RA are known to have a higher frequency of comorbidities, and CVD is reported to be more prevalent among them [
20,
21]. The comorbidities associated with RA are not only consequences of systemic inflammation or RA itself but also factors that can lead to non-adherence to treatment or limitations in the choice of medications [
22]. The presence of CVD poses challenges in the management of RA, as it restricts the use of certain medications such as GCs, NSAIDs, and several tsDMARDs due to their potential to accelerate atherosclerosis or provoke thrombosis Specifically, tsDMARDs are not recommended for use in patients with CVD risk or in the elderly, as they carry a higher risk of thromboembolism in these high-risk individuals [
23,
24]. The presence of combined CVD creates barriers in selecting suitable drugs for managing disease activity in individuals, thereby suggesting a higher likelihood of progression to D2T RA. It is crucial to consider the impact of comorbidities, particularly CVD, on the treatment options and outcomes in patients with RA. The presence of comorbidities can complicate the management of RA and necessitate careful consideration of the risks and benefits of various treatment approaches. Future studies should further explore the relationship between comorbidities, specifically CVD, and the development of D2T RA to enhance our understanding of these complex interactions and inform the development of tailored treatment strategies for this patient population.
A lower proportion of patients with D2T RA had a history of prior use of methotrexate, sulfasalazine, and leflunomide compared to those with non-D2T RA. Furthermore, prior use of these csDMARDs was independently associated with a reduced risk of developing D2T RA, suggesting that incomplete use of csDMARDs may be associated with a higher risk for D2T RA. In other cohorts, the D2T RA group exhibited a higher proportion of patients with contraindications or intolerance to methotrexate and delayed initiation of methotrexate treatment for more than 12 months compared to the non-D2T RA group [
11,
25]. Standard recommendations for RA emphasize the early initiation of methotrexate followed by sequential use of other csDMARDs [
26,
27]. According to the regulations of the Korean National Health Insurance Services (NHIS), csDMARDs, including methotrexate, should be maintained for at least six months with a minimum of three months of use before initiating b/tsDMARDs. Consequently, a majority of patients in the KOBIO data (92.42% of those with a history of two or more csDMARD treatments) were found to continue using other DMARDs alongside methotrexate. However, there are cases where methotrexate usage may be incomplete due to adverse effects or contraindications, such as combined interstitial lung disease or liver disorders. The incomplete use of methotrexate and other csDMARDs could predict challenges in achieving remission or low disease activity despite sequential use of b/tsDMARDs. These findings underscore the importance of adhering to recommended treatment guidelines, particularly the early and consistent use of methotrexate and other csDMARDs in RA management. Incomplete utilization of these medications may impede the achievement of optimal treatment outcomes, including remission or low disease activity. It is crucial for healthcare providers to address any barriers to csDMARD usage and promote the appropriate and timely initiation of these medications in order to improve treatment responses and prevent the progression to D2T RA.
In contrast to previous studies, the current data did not show an association between GC use and D2T RA.
Giollo et al. reported that GC use for more than 6 months was associated with D2T RA, while
Yoshii et al. suggested that treatment with methotrexate at a dose of ≥ 8.7 mg/week without concomitant GCs might aid in withdrawing from D2T RA [
12,
25]. It is important to note that the control groups in previous studies consisted of RA patients who were treated with csDMARDs only, whereas the non-D2T RA group in the current study received treatment with b/tsDMARDs. All patients in the current study had relatively higher disease activity, which necessitated the use of low-dose GCs, resulting in more than 85% of patients receiving GC therapy. Among patients with RA who have moderate-to-high disease activity and are being treated with b/tsDMARDs, the role of GCs may be limited. While GC use can lead to various complications of varying severity, it has been demonstrated to be effective in relieving joint pain and preventing joint destruction. Furthermore, tapering GCs can pose a risk of disease flare, and the benefits of low-dose GCs, especially in combination with b/tsDMARD therapy, outweigh any negligible harm associated with their use [
28‐
31]. Taken together, the current findings suggest that the use of csDMARDs may impede the progression to D2T RA, whereas the use of GCs does not appear to play a significant role. Comprehensive understanding and patient care implications warrant further research and clinical observation.
The patterns of drug survival differed between patients with D2T RA and non-D2T RA for each b/tsDMARD. In patients with D2T RA, JAK inhibitors had the highest drug survival rate, while rituximab had the lowest. Conversely, in patients with non-D2T RA, rituximab had the highest drug survival rate, while TNF inhibitors had the lowest. Although the population treated rituximab was very small, patients with D2T RA might have a difficulty to tolerate the several months required for rituximab to take effect, as it has a slow onset of action that can extend beyond 4 months. In addition, rituximab targets and depletes B cells and plasma cells involved in the production and action of autoantibodies in RA, so its efficacy might be limited in patients with D2T RA who have uncontrolled disease activity which was derived from factors other than RF production. Therefore, the observed differences in drug survival rates between D2T RA and non-D2T RA patients suggested specific treatment strategies might be necessary for each of the two disparate groups.
In the present study, disease activity markers remained elevated even after 8 years of treatment with b/tsDMARDs in patients with D2T RA. Despite the use of one b/tsDMARD in combination with a csDMARD, a considerable number of patients with D2T RA were unable to adequately control their symptoms. The current recommendations for the management of RA, which are based on extensive knowledge about the disease, have not yielded satisfactory outcomes in the D2T RA population. Furthermore, the reason why medication cannot be changed despite inadequate symptom control is the limited availability of alternative b/tsDMARDs. Due to the limited options of b/tsDMARDs, patients with D2T RA may be unable to switch to a different medication even if their current treatment is not effectively managing their symptoms. To improve the management of D2T RA, it is crucial to adopt a more comprehensive approach that takes into account the unique challenges and characteristics of this subgroup of patients. By focusing on tailored and targeted interventions, including exploring novel treatment options, it may be possible to identify more effective strategies for symptom relief, improved quality of life, and prevention of complications associated with D2T RA. This calls for further research and collaboration to optimize the management of D2T RA and address the unmet needs of these patients [
5,
32].
This study has some limitations. Since the data were extracted retrospectively, it was not possible to control all cases included in the analysis, such as the dosages of methotrexate. Considering the limited number of cases and the characteristics of each medication, we did not separately analyse subcutaneous and intravenous administration methods for drugs offering both options, such as infliximab biosimilar, tocilizumab, abatacept, and golimumab. We were unable to comprehensively categorize the presence or absence of OA and fibromyalgia among the RA patients registered in KOBIO. It is possible that there are RA patients with concomitant OA and fibromyalgia. However, in KOBIO, registration targeted RA patients initiating or switching b/tsDMARDs treatment, leading to the enrollment of RA patients with moderate disease activity or higher. The definition of D2T RA, as per the EULAR criteria, requires patients to be treated according to the EULAR recommendations for RA [
2]. While most patients strictly adhere to these recommendations due to reimbursement conditions set by the Korean NHIS, ensuring a minimum of 6 months of csDMARD use before initiating b/tsDMARD therapy, individual treatment histories were not fully controlled in this retrospective study. However, the insurance regulations did provide some boundaries in terms of available treatment options, leading to a somewhat more controlled study population.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.