Background Long-term prospective observational studies are complementary to controlled clinical trials to explore effectiveness and safety of biological therapies in clinical practice.

Objectives To study abatacept survival, reasons of discontinuation and clinical responses in everyday clinical practice of patients with rheumatoid arthritis (RA).

Methods Prospective, observational single center study at the Rheumatology Clinic, University Hospital of Heraklion, Crete. At baseline, patient demographics, co-morbidities and disease characteristics are being recorded, while during follow-up, discontinuations, disease activity and adverse events are collected. For this analysis, all patients who received Abatacept intravenously from 6/2007 till 6/2016 were included. Kaplan-Meier curves and Cox regression analysis were used to determine drug survival and predictors thereof. Linear regression was used to compare DAS difference at 12 months between different lines of bDMARD therapy.

Results A total of 224 patients (women: 87%, seropositive: 34%) were included. Median (IQR) age was 63 (56–70) years, disease duration 7.4 (4–13.4) years and baseline DAS28 5.9 (5.2–6.5). Abatacept was the 1st bDMARD in 59 (26%), 2nd in 71 (32%) and ≥3rd in 94 (42%) patients. During follow-up [total: 508 patient-years, median (IQR): 1.7 (0.7–3.3) years], 54% patients discontinued therapy (87% for treatment failure, 10% for adverse events). Two-year treatment persistence was 59%. In multivariable regression analysis, predictors of longer Abatacept survival were lower baseline HAQ [HR (95% CI) for unit increase =2.29 (1.5–3.48), p<0.001], longer disease duration [HR (95% CI) for ≥8 vs.<8 years=0.51 (0.30–0.88), p=0.016], Abatacept as 2nd vs 1st or ≥3rd bDMARD [HR =0.52 (0.30–0.91), p=0.022] and a more recent year of therapy start [HR=0.39 (0.16–0.95), p=0.022].

DAS28<3.2 and remission at 6 (12) months were achieved by 12% (18%) and 5% (7%) of patients respectively. DAS28 difference at 12 months was greater in patients who received Abatacept as the ≤2nd than those on ≥3rd bDMARD (p=0.009).

A total of 312 adverse events were registered, of which 65 were serious (SAE). Incidence of total (serious) adverse events was 61 (13) /100 patients/year. SAE included 5 cases of cancer, 10 cardiovascular events and 24 infections, mainly of the respiratory tract.

Conclusions In the present study, Abatacept survival at 2 years was 59%. The majority of patients discontinued therapy due to inadequate response. Use as a 2nd line biologic agent and lower baseline HAQ predicted better survival. Improvement in DAS was higher when Abatacept was used as the ≤2nd bDMARD. Rates of remission or low disease activity in clinical practice are rather low, while the safety profile was excellent.

Disclosure of Interest None declared