Background Rheumatoid arthritis (RA) is a chronic disease with involvement of small and large joints. Treat-to-target strategy aims to improve the outcome and rapidly reduce disease activity by using properly and aggressively all the therapeutic options.

Objectives To estimate the safety of bDMARDs in RA patients, when administered according to international recommendations, compared to delayed initiation of bDMARDs in usual care.

Methods All adverse events (AEs) and their severity were recorded. Cox proportional-hazards models were performed examining the association of treatment groups, with the risk of 1st AE. A separate analysis was performed for new cases of cancer during administration of bDMARDs.

Results There were 250 patients in usual care and 113 patients in T2T group (Table 1). Delayed bDMARD administration was accompanied by more frequent and severe AEs, compared to T2T strategy (Table 2). In the 1st group 155 patients were recorded with at least one AE (56 patients with serious), of which 80 patients had at least one infection (30 a serious one). In T2T group, there were 47, 9, 11 and 2 patients with at least one AE, serious AE, infection and serious infection respectively. The IRR for the T2T group was 0.78 for the 1st AE, 0.56 for the 1st serious AE, 0.41 for the 1st infection and 0.23 for the 1st serious infection. Table 3 illustrates the AEs (irrespective of AEs per person) in both groups.

A greater incidence of cancer was noticed in patients of usual care group. There were 15 new cases of cancer (non-melanoma skin=2 lung=2 lymphoma=5 breast=2 urinary tract=3 blood=1) compared with only one (urinary tract) in T2T group. The IR of the 1st group was 1.99/100PY compared to 0.4/100PY of the 2nd group (p=0.027). Nine of 16 cases had no risk factors for cancer. Mean duration of sDMARDs, before bDMARDs, was 113 months and mean duration of bDMARDs 41 months.

The risk of 1st infection (irrespective of severity) was significantly higher for the usual care group compared to T2T group (p=0.002 for 1st infection, p=0.03 for 1st serious infection, Figure 1). The risk for 1st AE and serious AE was marginally non-significant (p=0.069 and 0.077 respectively) in usual care group.

After controlling for potential confounders, the risk for a serious AE remained significantly higher for usual care group (HR=0.42, p=0.041) (Table 4). Patients in T2T group were more likely to have a reduced risk for infection (HR=0.49 for 1st infection, HR=0.24 for serious one), although the findings were marginally non-significant (p=0.052 and 0.065 respectively).

Extra-articular manifestations increased the risk for any AE (irrespective of severity), progression of age was important for the 1st AE (mild or serious) and 1st serious infection, whereas patients with ESR >40mm/h were more vulnerable to infections.

Conclusions Frequency and severity of AEs, especially infections, were substantially reduced in T2T group. A greater incidence of cancer was noticed in patients with delayed administration of bDMARDs.

Disclosure of Interest None declared