Background/Purpose:

Many evidences including clinical trials of biologics lead us earlier and more aggressive treatment strategy for patients with rheumatoid arthritis (RA). It is stated as “treat-to-target”, that is remission. It is critical issue in clinical practice how to evaluate risks and to balance between safety and efficacy of treatment, which could be related to experience.

The aims of this study are to explore the learning curve of treatment with biologics during the first 7 years in Japan (2003-2010) based on safety results associated with the changes in baseline characteristics according to treat-to-target strategy in RA patients treated with biologics using multicenter registry; Tsumumai Biologics Communication Registry (TBCR).

Methods:

We analyzed changes in baseline characteristics by  initiation year of 1^st biologics and the incidence rate of adverse event (AE) caused discontinuation of 1^st biologics in patients who were registered in TBCR from 2003 to 2008 and followed up to 2010.  Predictive factors at baseline for incidence of adverse event were determined using multivariable Cox’s proportional-Hazards regression model.

Results:

A total of 874 cases (2052 person-years) were observed. Patients with younger age (≤53 ys) had significantly shorter disease duration (<2 years), less dysfunction, joint damage, and disease activity (DAS28-CRP) at 2008, compared to 2003-2005 while patients with older age(>64 ys) had no significant changes. During total observation, 122 AEs (6.0 events/100 person-years) were occurred.  The incidence rate of AEs were significantly higher in older age group than in younger age group (p<0.001, log-rank test). Kaplan-Meier curves showed that the incidence rate of AEs was significantly decreased with year of initiation, especially in older age group and that the differences was remarkable up to 6 months.   Actually, multivaliable analysis showed that, during 2 years observation, older age [OR 1.8, 95%CI (1.1-3.0)], worse physical function [OR 2.1, 95%CI (1.4-3.1)], MTX use [OR 0.4, 95%CI (0.3-0.7)], and etanercept use (vs infliximab use) [OR 0.4, 95%CI (0.3-0.7)] had significant impacts on incidence of the AEs. Interestingly, initiation year had significant impacts on incidence of AEs up to 6 months [initiation at 2008: OR 0.3, 95%CI (0.1-0.7), compared to initiation 2003-2005] and while the impact disappeared up to 2 years observation. The impact of MTX use and etanercept use was not detected up to 6 months observation. Incident rate of the AEs in pateints with older age who were initiated biologics at 2008 was comparable to patients with younger age. These results suggested that, during first 3 years (2003-2005), we could not evaluate risk factors properly, especially in respiratory system, with inadequate experience.

Conclusion:

We clearly demonstrated improving of safety of 1^st biologics with first 7-years experiences in real clinical practice using multicenter registry of biologics in Japan.