(A-349) Assessing the Association Between COVID-19 Vaccination and Thrombotic Thrombocytopenia Syndrome (ATTEST Study): Analyses of English Data, 2020–2022

Background: A very rare and serious combination of thrombosis events (TE) and thrombocytopenia (TCP), in some cases accompanied by bleeding, was observed following vaccination with AZD1222 post-marketing during the COVID-19 pandemic. Most available epidemiological studies on this issue focused on the first dose of COVID-19 vaccines and covered the time period up to the end of 2021 with limited cases.

Objectives: This was a post-authorisation safety study aimed at evaluating the association of COVID-19 vaccine exposure with thrombotic thrombocytopenia syndrome (TTS) in England from 2020 to 2022 using both primary and secondary care data.

Methods: We used primary care data from the Oxford Royal College of General Practitioners Clinical Informatics Digital Hub (ORCHID) linked to hospitalisation, immunisation and COVID-19 testing data to conduct a matched case-control (mCC) study complemented by a self-controlled case series (SCCS) analysis. TTS was defined algorithmically as the novel co-occurrence of thromboembolism (TE) with thrombocytopenia (TCP) within ± 7 days. Cases with a history of TE/TCP were excluded. Incident TTS cases were matched 1:5 to controls on age, sex and GP practice in mCC study. Individuals were considered exposed to AZD1222 or BNT162b2 (1st and/or 2nd dose) or unexposed (no vaccination) in a 28-day risk window preceding the TTS event or pseudo-event date for matched controls. We used conditional logistic regression in the mCC study to estimate adjusted odds ratios (aOR), and conditional Poisson regression in the SCCS analysis to estimate crude and calendar time adjusted relative incidence ratios (RIR). Subgroup analyses were conducted by age, sex, and anticoagulant use in the prior 90 days. Sensitivity analyses tested different TTS definitions and exposure risk periods.

Results: Out of 7.6 million individuals, 666 algorithm-defined TTS events were identified of which >90% had no COVID-19 vaccine within 28 days prior to TTS. Exposure to AZD1222 dose 1 showed an association with TTS (aOR: 2.12, 95% CI: 1.14-3.92, RIR: 3.49, 95% CI: 2.22-5.49) but dose 2 did not (aOR: 0.78, 95% CI: 0.35-1.76, RIR: 1.36, 95% CI: 0.72 2.54). A stronger association of AZD1222 dose 1 with TTS was observed in the 16-49 age group (aOR: 3.20, 95% CI: 1.04-9.83, RIR: 7.09, 95% CI: 2.78-18.12) but not in the 50+ group. There were no differences by other subgroup variables. No association of TTS with BNT162b2 was observed at any exposure level. Sensitivity analysis results were consistent with the primary analyses.

Conclusions: Both analyses found an increased risk of TTS after the first dose of AZD1222, but not after second dose. No increased risk for TTS was found after BNT162b2 vaccination. These findings are consistent with previous studies.