Background Early diagnosis and treat-to-target strategies with conventional DMARDs in rheumatoid arthritis (RA) have allowed the achievement of remission in a significant percentage of the cases in daily clinical practice. Whether and in which patients treatments can be suspended with maintenance of health is currently unclear.
Objectives The aim of this study was to investigate the outcomes of methotrexate (MTX) suspension and predictors of disease recurrence in a real life single centre cohort of early RA patients followed prospectively under DMARD- and glucocorticoid-free conditions.
Methods All RA patients included in this current prospective observational study derived from the Pavia's Early Arthritis Clinic (pEAC) and were treated according to a DAS-driven step-up protocol with MTX in monotherapy. Patients achieving stable DAS28 remission and fulfilling the following criteria were eligible for drug suspension: 1) fulfilment of the 2010 ACR/EULAR classification criteria for RA within 12 months from baseline visit; 2) MTX introduced within 12 months from symptoms' onset; 3) ≥24 months of continuative MTX; 4) DAS28 remission documented for ≥6 months in the absence of glucocorticoids. Patients were followed-up at three-months intervals through complete clinical and ultrasonographic (hands-feet) assessments. Radiographs were repeated annually. Treatment was re-introduced in case of DAS28≥3.2 in a single occasion or 3.2<DAS28≥2.6 for >6 months.
Results Of the 70 RA patients fulfilling the inclusion criteria, 63 with at least 6 months of follow-up following DMARDs discontinuation were considered for the present study. Baseline patients' stratification according to remission stringency showed SDAI remission in 51/63 (81%), ACR/EULAR Boolean remission in 46/63 (73%) and absence of clinical-ultrasonographic synovitis (SJC44=0 and power Doppler signal in hands-feet=0) in 22/63 (34.9%) respectively. Treatment restart due to disease recurrence was observed in 28/63 patients (44.4%) over 2 years of follow-up, with a median (IQR) time until retreatment of 6 (6-10.5) months. None of the clinical characteristics at the time of diagnosis showed predictive significance. Worsening of disease activity was more likely to occur in patients not achieving remission according to the SDAI (HR [95% CI] 2.81 [1.24-6.34], p=0.01). Below this threshold, remission stringency failed to show any predictive value, with 10/22 (45%) patients showing disease recurrence despite absence of clinical-subclinical synovitis at the time of drug suspension. By multivariate Cox regression analyses, IgG ACPA were the only predictor of relapse, independent of remission duration and the clinical-ultrasongraphic inflammatory degree at baseline. IgG ACPA-IgM rheumatoid factor (RF) double positivity showed an increased predictive ability (HR [95% CI] 3.69 [1.58-8.61], p=0.003).
Conclusions DMARD suspension/drug-holiday appears a feasible option in a proportion of RA patients achieving stable remission following early treatment and treat-to-target approaches in routine clinical care. The autoimmune status is the strongest predictor of disease reactivation in patients achieving DMARDs-induced clinical-ultrasonographic control of the inflammatory process.
Disclosure of Interest None declared