(A-301) Replication of the Association Between Anti-TNF Exposure During Pregnancy and Maternal and Infant Outcomes in the Dorsata Ob-GYN EMR data set and Linked Infant Data

Background: Real-world medication safety studies in pregnant women and linked infants are commonly requested by the U.S. Food and Drug Administration as post-marketing requirements (PMRs). The Dorsata Ob-Gyn EMR data includes maternal medication exposures, pregnancy outcomes, and linked infant data. To date, there has been no replication of a real-world PMR using Dorsata data to evaluate its utility for PMR studies.

Objectives: To evaluate pregnancy, birth, and infant outcomes following anti-tumor necrosis factor agent treatment (anti-TNF) and compare findings to Broms 2020.

Methods: Pregnancies in Dorsata from 2023-2024 for patients with a SNOMED code for inflammatory conditions (rheumatoid arthritis, psoriatic arthritis, psoriasis, ulcerative colitis, and Crohn’s Disease) were selected. Pregnancies with exposure to anti-TNF agents were compared to pregnancies with exposure to non-biologic treatments described in Broms 2020. Logistic regression was used to estimate odds ratios and 95% confidence intervals for birth outcomes (spontaneous and elective terminations (SA and ET), stillbirths, preterm births, ectopic pregnancies) and infant outcomes small for gestational age [SGA]) and major congenital malformations (MCMs). Models were adjusted for maternal age, parity (dichotomous), and pre-pregnancy body mass index.

Results: Among 543 pregnancies, 58 (10.7%) were exposed to anti-TNF agents and 485 (89.3%) received non-biologics, which was proportionally comparable to Broms [Anti-TNF: 1,027 (9.9%); non-biologics: 9,393 (90.1%)]. In the anti-TNF group, there were 53 live births, including 6 preterm and 5 SGA infants; 5 (8.6%) pregnancies had no outcome classification and there were no SA or ET cases. In comparators, there were 432 live births (89.1%), including 40 preterm births (8.2%), and 33 SGA infants (6.8%). Among non-live births, there were 8 SA (1.6%), 3 ET (0.6%), 2 stillbirths (0.4%), 2 (0.4%) unclassified non-live births and 38 (7.8%) pregnancies lacked outcome classification. Anti-TNF exposure suggested an increased risk of preterm birth (OR 1.61, 95% CI: 0.58-4.52) that was elevated as in Broms (OR 2.01, 95% CI:1.87-2.16), but did not achieve statistical significance. Other birth outcomes were not evaluable due to a lack of events in either group or lack of severity definition for comparison (SGA). Modeling for MCMs was not conducted due to zero MCM events in the anti-TNF group.

Conclusions: Anti-TNF agents showed a non-significant association with preterm birth, consistent with published literature, while sample sizes impacted the evaluation of additional outcomes including MCMs. Current application of Dorsata data to FDA PMRs may be feasible in studies with prevalent disease populations.