POS0603 ANALYSIS OF FACTORS ASSOCIATED WITH THE EFFECTIVENESS OF ABATACEPT IN THE ORIGAMI STUDY

Abstract

Background:The ORIGAMI study is a multicenter, observational study to evaluate the effectiveness, safety, and patient-reported outcomes of abatacept (ABA) in Japanese patients with csDMARD-resistant, Simplified Disease Activity Index (SDAI)-moderate, biologic-naïve rheumatoid arthritis (RA). ABA has shown better effectiveness/efficacy in RA patients with anti-cyclic citrullinated peptide antibody (ACPA) positive (1) and high ACPA titer (2) compared to ACPA negative and low ACPA titer, respectively. However, more accurate predictors of effectiveness in clinical practice are needed than ACPA status.Objectives:This post-hoc analysis is aimed to determine the association between ACPA and ABA effectiveness (disease activity and physical function) or retention rate and to investigate other factors associated with the effectiveness of ABA in patients enrolled in the ORIGAMI study.Methods:Of the 279 patients in the effectiveness analysis set of the ORIGAMI study, 270 patients with baseline ACPA measurement were analyzed. The patients were divided into the ACPA-positive group (ACPA +ve, ≥4.5 U/mL at baseline) and the ACPA-negative group (ACPA –ve, <4.5 U/mL). Patients’ characteristics, changes in disease activity and physical function (Japanese Health Assessment Questionnaire; J-HAQ) through 52 weeks, and retention rates of ABA at week 52 were evaluated. Baseline characteristics and use of concomitant drugs were analyzed as independent variables by multiple regression analysis using a standard linear model adjusted by SDAI at week 0 to identify factors associated with SDAI change at week 52. In addition, the interaction effects among ACPA status, RF status, and the factor that was significantly associated with SDAI change in multiple regression analysis on changes in SDAI were explored.Results:The numbers of ACPA +ve and –ve patients were 226 and 44, respectively. ACPA values (mean ± SD, U/mL) were 280.3 ± 376.8 and 0.9 ± 0.7, and rheumatoid factor (RF) values were 174.8 ± 302.6 and 20.9 ± 61.7 in the ACPA +ve and –ve groups, respectively. Mean (95% confidence interval) changes in SDAI at week 52 were −11.3 (−12.4 to −10.3) and −8.0 (−10.5 to −5.5), and those in J-HAQ were −0.27 (−0.34 to −0.20) and −0.16 (−0.34 to 0.01) in the ACPA +ve and –ve groups, respectively. In the Kaplan–Meier analysis, the retention rates of ABA at week 52 in the ACPA +ve and –ve groups were 72.1% and 58.7%, (discontinuation for any reason), and 91.6% and 75.7% (discontinuation because of lack of effectiveness), respectively. In a multiple regression analysis, the duration of disease (< 1 year) was associated with the change in SDAI at week 52. With respect to SDAI changes, the estimated difference of ACPA +ve and disease duration (< 1 year), ACPA +ve and disease duration (≥1 year), and ACPA –ve and disease duration (< 1 year), versus ACPA −ve and disease duration (≥ 1 year), were −4.26 (p = 0.022), −0.82 (p = 0.618), and −0.93 (p = 0.716), respectively (Fig. 1). The estimated difference of ACPA +ve and RF +ve, ACPA +ve and RF –ve, and ACPA –ve and RF +ve, versus ACPA –ve and RF –ve, were −2.48 (p = 0.060), −2.77 (p = 0.107), and −5.48 (p = 0.087), respectively.Conclusion:A higher retention rate as well as better effectiveness of ABA on disease activity and physical function in ACPA +ve group versus ACPA –ve group were shown in the simple subgroup analysis. ABA effectiveness on the SDAI change was significantly better in patients with disease duration <1 year and ACPA +ve compared to those with ACPA −ve and disease duration ≥ 1 year.