Background Smoking is a known environmental risk factor for developing rheumatoid arthritis (RA), and is also a marker of a worst clinical and radiological prognosis. Despite this relation, the effect of smoking on RA treatment effectiveness is still controversial, with the available reports suggesting a deleterious effect.

Objectives To analyse the effect of smoking on treatment response in RA patients under biologics, addressing possible differences in the degree of response between those being treated with anti-TNF alpha agents and those under other biologics.

Methods A retrospective analysis of all patients with RA receiving biologic therapy of a rheumatology department of a Portuguese University Hospital. Were excluded from the analysis all the patients without data on smoking history. For this analysis, were considered “smokers” all patients with current or past smoking habits. The demographic and clinic baseline data, disease activity at baseline, 6, 12 and 18 months of therapy (DAS28 and EULAR response) and HAQ at baseline were collected. The variations in DAS28 and EULAR response were determined, as well as their difference according to current exposure to anti-TNF alpha agents or to other biologics (Rituximab, Tocilizumab, Abatacept). The effect of smoking on DAS28 was adjusted to age and gender. Statistical analysis was done using SPSS® 22.

Results In January 2015, 145 patients met the inclusion criteria; 88.3% were women, with median disease duration of 16.5 years. Rheumatoid factor and/ or anti-cyclic citrullinated peptide (anti-CCP) antibodies were positive in 82.8% of patients. Median baseline DAS28 was 6.09 and median baseline HAQ was 1.86. There were 35 patients with past or current smoking history. Smoker patients were younger than non-smoker patients (median age 52.4 vs. 57.2, p=0.047), and have a lower proportion of women (62.6% vs. 94.5%, p<0.001).

The median DAS28 reduction at 6 months was significant different between smokers and non-smokers, favouring the latter (-1.41 vs. -1.68, p=0.029), even after adjusting for age; after adjusting for gender the relation lost significance. It was observed a tendency for a smaller proportion of patients achieving EULAR response (moderate or good) among smokers at 6 months (56.5% vs. 75.3%, p=0.083). At 12 and 18 months of treatment was also observed the same trend to a smaller reduction of DAS28 among smokers (-1.79 vs. -1.84, p=0.892 and -1.98 vs. -2.21, p=0.140 respectively).

Amongst smoking patients, those treated with anti-TNF agents seem to have worse response when compared with patients under biologics with other mechanism of action after 6 months (-0.75 vs. -1.43, p=0.220), 12 months (-1.63 vs. -2.16, p=0.790) and 18 months of treatment (-1.40 vs. -2.30), p=0.159). The same trend was observed in those achieving a moderate or good EULAR response.

Conclusions This analysis suggests a negative effect of smoking on therapeutic response in Rheumatoid Arthritis patients under Biologics, and that other biologics than anti-TNF alpha agents may be better choices for smoker rheumatoid arthritis patients.

Disclosure of Interest None declared