Exploring the relationship between utilization patterns of biologic disease-modifying anti-rheumatic drugs and disease activity in rheumatoid arthritis patients

Background: The relationship between drug-utilization patterns of biologic Disease-Modifying Anti-Rheumatic Drugs (bDMARDs) and disease activity is poorly explored in real-world studies on rheumatoid arthritis (RA). Objectives: To investigate the disease activity associated with biologic drug initiation and discontinuations in RA patients. Methods: The PATHFINDER study (EUPAS29263) is a retrospective population-based study conducted on Tuscan healthcare administrative database (HAD). We included first-ever bDMARD users for RA with a visit at the Rheumatology ward of Pisa University Hospital from 2013 to 2016. The index date was defined by the first dispensation date of bDMARD. We extracted the drug supplies from HAD and the Disease Activity Score 28 (DAS28) from medical charts. In the 3 years follow-up, we measured discontinuations, as the result of the coverage of each bDMARD dispensation, based on the Defined Daily Dose, and the number of doses supplied plus a grace period of 60 and 30 days in the main and sensitivity analyzes, respectively. We assessed DAS28 before (T0), after (T1) the first bDMARD supply, and before (TD0), after (TD1) discontinuation. We selected patients with at least 3 DAS28 available and stratified by discontinuation occurrence, testing the difference in disease activity recorded at T1. In the second analysis, we evaluated the discontinuation events with DAS28 recorded at both TD0 and TD1, in order to highlight the disease activity trend (i.e. stability or improvement). Results: Out of 95 RA first-ever bDMARD users (73 females, mean age 59.6 standard deviation, SD, 12.1) 70 had at least 3 DAS28 assessments recorded (55 females, mean age 59.3, SD 12.4). Overall at T0, 28 patients had DAS28>3.2 (off-target) and 13 DAS28≤3.2 (in target); otherwise at T1, 28 patients showed DAS28>3.2, while 38 DAS28≤3.2. In the main analysis, we observed 91 discontinuations, corresponding to 33 subjects with at least one discontinuation among the 70 patients. The percentage of patients in target at T1 is significantly higher in those continuing treatment than in those with at least one discontinuation (p-value = 0.053). In the second analysis, we detected 37 discontinuations with both DAS28TD0 and DAS28TD1: at TD0 24/37 discontinuations displayed a DAS28≤3.2, and at TD1 28/37 showed disease improvement or stability. The sensitivity analysis confirmed these findings. Conclusions: Our study observed a disease improvement in the majority of RA patients after starting bDMARDs. Treatment discontinuations often occur when disease activity is stable or improving. These findings are in line with RA clinical guidelines recommending the tapering of drugs upon symptom remission.