FAVORABLE BALANCE of BENEFIT and HARM of LONG-TERM, LOW-DOSE PREDNISOLONE ADDED to STANDARD TREATMENT in RHEUMATOID ARTHRITIS PATIENTS AGED 65+: The PRAGMATIC, MULTICENTER, PLACEBO-CONTROLLED GLORIA TRIAL

ABSTRACT

Background: Low-dose glucocorticoid (GC) therapy is widely used in RA but the true balance of beneft and harm is still unknown. Objectives: We studied the effects of prednisolone (5 mg/day, 2 years) in RA patients aged 65+, requiring adjustment of antirheumatic therapy (DAS28≥2.60). Methods: Pragmatic double-blind placebo-controlled randomized trial;all co-treatments and changes therein were allowed during the trial except longterm open label GC;Ca/D supplementation was advised in all patients. Minimal exclusion criteria were tailored to seniors. Harm outcome: the number of patients with ≥1 serious adverse event (SAE), or ≥1 'other adverse event of special interest' (other AESI). Other AESI comprised any AE (except worsening of RA) causing study discontinuation, and GC-specifc events (Table 1). Beneft outcomes: improvement in disease activity (DAS28) and joint damage progression (Sharp/van der Heijde). Longitudinal mixed models analyzed the data. Given prior knowledge we report one-sided 95% confdence limit (95%CL) and statistical tests, performed only for the main outcomes. Results: We randomized 451 RA patients in 7 EU countries, 449 received the intervention;of these 63% prednisolone vs 61% placebo patients completed 2 years of follow up. Discontinuations were similar in both groups for AE (14%) and active disease (4%);the remainder mostly for 'trial fatigue' and covid-related access issues (20%). Mean time on study drug was 19 (SD 8) months. 70% of patients were female, mean age was 72 (max 88) years, RA duration 11 years;67% were RF+, 56% ACPA+, 96% had joint damage on radiographs mean score 20, median 8. Mean DAS28 was 4.5. Most patients (79%) were on current DMARD treatment, including 14% on biologics;47% had previously used GC, 14% changed DMARD therapy at baseline. Patients had mean 2.1 active comorbidities, and used median 7 drugs. Beneft Disease activity rapidly declined to stabilize after 1 year (Figure 1), and was lower on prednisolone (adjusted mean difference in DAS28 over 2 years 0.37, 95%CL 0.23, p<0.0001). The contrast in early (3-month) response was larger in 331 patients adherent to protocol on stable treatment mean difference in DAS28 0.62 (95%CL 0.44), more responders on prednisolone (Figure 1). Sig-nifcant time-treatment interaction in secondary analyses suggested a decrease in contrast after the frst year, most likely caused by signifcantly more changes in DMARD treatment on placebo. Joint damage progression over 2 years was signifcantly lower on prednisolone mean 0.6 (SD 1.9) v 1.8 (6.4) score points on placebo, difference 1.2 (95%CL 0.2, p=0.02). Harm 60% prednisolone vs 49% placebo patients experienced the harm outcome: adjusted RR 1.24, 95%CL 1.04, p=0.02;number needed to harm 9.5 (Table 1). During the study 1 vs 2 patients died, and 3 vs 0 died within 5 months of discontinuation. Per 100 patient-years, AE totaled 278 in prednisolone vs 206 in placebo patients, and the difference was most marked for infections (Table 1);these were mostly mild or moderately severe. Other GC-specifc AESI were rare without relevant differences. Conclusion: Add-on low dose prednisolone has benefcial long-term effects on disease activity and damage progression in senior RA patients on standard treatment. The tradeoff is a 24% increase in patients with mostly mild to moderate AE, suggesting a favorable balance of beneft and harm.