FRI0119 Effect of discontinuing tnf inhibitors during pregnancy on the course of rheumatoid arthritis and juvenile idiopathic arthritis

Background Treatment changes at early pregnancy can be followed by a disease worsening. 1 Objectives To investigate whether the discontinuation of tumour necrosis factor inhibitors (TNFi) use during pregnancy is associated with any changes of disease activity at the third trimester in women with rheumatoid arthritis (RA) and juvenile idiopathic arthritis (JIA). Methods A prospective cohort study was conducted using the Organisation of Teratology Information Specialists (OTIS) Autoimmune Diseases in Pregnancy Project in the U.S. and Canada. Pregnant women with RA and JIA were enrolled between 2005 and 2017. Information about medication and disease activity were collected by telephone-based interviews prior to gestational week 20 and at gestational week 32. Disease activity was assessed by the Health Assessment Questionnaire Disability Index (HAQ-DI), the patient’s pain scale and the patient’s global scale. The composite tool Patient Activity Scale (PAS) was calculated in retrospect. Results In the OTIS cohort, data were available for 490 women of whom 397 had RA and 93 had JIA. Of all patients, 323 (65.9%) used TNFi during pregnancy of whom 122 (24.9%) patients discontinued TNFi before gestational week 20 (the mean time of discontinuation was gestational week 6 (SD ±5.03)) and 201 (41.0%) used TNFi beyond week 20. There were 167 (34.1%) patients not taking TNFi during pregnancy. At the time of enrollment, disease activity was low to minimal in 357 (72.9%) patients as defined by PAS scores below 3.7. From the first to the third trimester, women using TNFi beyond week 20 showed a decrease of the PAS scores (p=0.02, figure 1) whereas women not using TNFi and those discontinuing TNFi before gestational week remained stable. The univariate regression analysis, but not the adjusted model, revealed that TNFi use beyond week 20 was associated with improved HAQ scores at the third trimester (coefficient B −0.142, 95% CI −0.258 to −0.026) and with improved PAS scores (coefficient B −0.423, 95% CI −0.843 to −0.002). However, the various TNFi treatment modes during pregnancy were not associated with any minimum clinically important difference at the third trimester. When selecting for 58 patients with active disease (PAS score ≥3.71) at the first trimester, the discontinuation of TNFi before gestational week 20 was not associated with any clinically important worsening of the disease at the third trimester. Conclusions In patients with RA and JIA who enter pregnancy with well controlled disease, the discontinuation of TNFi before gestational week 20 is possible without a risk of disease flares at the third trimester. Reference [1] van den Brandt S, Zbinden A, Baeten D, Villiger PM, Ostensen M, Forger F. Risk factors for flare and treatment of disease flares during pregnancy in rheumatoid arthritis and axial spondyloarthritis patients. Arthritis research & therapy. 2017;19(1):64. Disclosure of Interest None declared