Background/Purpose: The management of Rheumatoid Arthritis (RA) has markedly advanced, enabling the achievement of disease control and remission. Nevertheless, a proportion of patients remains refractory to multiple therapies. These individuals have recently been included in the newly proposed Difficult-to-Treat (D2T) RA classification. As this concept gains traction, ongoing research is increasingly aimed at identifying the determinants of treatment resistance across bDMARDs mechanisms of action, including patient characteristics and treatment trajectories. In this study, we aimed to estimate the probability of progressing to D2T RA within our RA UCLouvain cohort and assess whether the initial choice of b- or ts-DMARDs influences this risk.

Methods: Retrospective analysis recording clinical, laboratory, radiologic, and treatment data from the first consultation available through all b/tsDMARD treatments until the last and at becoming D2T visit. Kaplan-Meier analyses and subgroup comparisons were performed based on diagnosis era and treatment strategies.

Results: We included 675 RA patients [M/F, 158/517], with a median duration of follow-up of 217.1 ±9.8. months. 126 (18.7%) RA patients fulfilled the D2T RA definition. As shown in Table 1A, compared to non-D2T patients, the D2T group was diagnosed at a younger age (p=0.032) and had higher functional impairment (HAQ: 1.60 ± 0.12 vs. 1.30 ± 0.04; p = 0.004), and were more frequently exposed to corticosteroids before b or tsDMARD initiation (82.8% vs. 63.8% ;p < 0.001). Kaplan-Meier curves (Figure 1A-E) showed a progressive decline in D2T-free survival. Patients diagnosed after 2000 had earlier access to b or tsDMARDs (p < 0.001) (Table 1B), but D2T rates remained similar, suggesting that earlier treatment did not fully mitigate risk (Figure 1C). Treatment pathway (Figure 2A) revealed frequent transitions between mechanisms of action, particularly from anti-TNF to IL-6 and JAK inhibitors, reflecting a pattern of therapeutic escalation typical of D2T RA. A heatmap (Figure 2B) confirmed that while anti-TNFs dominated the first biologic line (81%), their use dropped markedly in later lines treatments. Analysis of reasons for treatment discontinuation showed that from the third biologic line, inefficacy—either primary or secondary—was by far the most common driver of therapeutic switches (over 80%), outweighing adverse events or other causes, underscoring the refractory nature of the disease in these patients. Importantly, D2T-free survival did not differ significantly based on the class of initial b/tsDMARD (Figure 1D-E) or whether csDMARDs were used in combination.

Conclusion: In this large cohort of 675 RA patients with a longterm follow-up, 127 RA patients developed D2T RA (18.8%). D2T patients are younger and have more functional impairment. Corticosteroid use prior to the first bDMARD was more frequent in the D2T group We did not identify a difference in the outcome (D2T) when comparing different treatment strategies comparing TNFi or other as the first line. Disease severity and persistent inflammation appears to be the primary drivers of D2T progression, rather than specific treatment strategies. Further analyses are ongoing to better define the concept of D2T RA.

Table 1A-B. A) Comparisons of Demographic and Clinical Characteristics Between D2T and Non-D2T RA Groups

B) Comparison of Clinical Characteristics in D2T RA Patients Diagnosed Before and After 2000

Legend: Values are presented as mean ± SD or % (n), unless otherwise indicated.

Abbreviations: ACPA, anti-citrullinated protein antibody; RF, rheumatoid factor; DAS28, 28-joint Disease Activity Score; HAQ, Health Assessment Questionnaire; D2T RA, difficult-to-treat rheumatoid arthritis; csDMARD, conventional synthetic disease-modifying antirheumatic drug; b/tsDMARD, biological or targeted synthetic DMARD; SD, standard deviation; ns, non specified.

Figure 1. Progression to D2T Rheumatoid Arthritis Over Time (Kaplan–Meier Curve)

Legend: Keplen – Meier curve analysis (Outcome = Becoming D2T) for: 1A) Whole cohort up to 50 years of follow up; B) Whole cohort stratified for years of diagnosis (before or after 2000); C) Whole cohort stratified for years of diagnosis (before or after 2000) without considering medical history before year 2000; D) Time to difficult to treat stratified for first line of therapy only including D2T patients; E) Time to difficult to treat stratified for first line of therapy (whole cohort).

Figure 2 A-B. A) Sankey diagram of biologic transitions in D2T RA patients B) Heatmap of treatment classes by biologic line (BIO1-BIO9)

Legend: A) Sankey-Plot showing b/tsDMARDs progression from the first (BIO1) to the ninth (BIO9) line of therapy. B) Heatmap for b/ts DMARDs prescription. Each cell represents the absolute (and relative, %) number of prescriptions for each treatment line

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/does-first-line-b-or-tsdmards-choice-influence-progression-to-difficult-to-treat-rhumatoid-arthritis-insights-from-our-longitudinal-ra-uclouvain-brussels-cohort/