Methods
The Sisonke (“Together” in isiZulu) phase 3B study is an open-label, single-arm implementation study of the Ad26.COV2.S COVID19 vaccine among adult (>18 years) HCWs in SA. The trial is sponsored by the SA Medical Research Council (SAMRC) with vaccines provided by Janssen Vaccines & Prevention B.V, a pharmaceutical company of Johnson & Johnson (NCT04838795) (http://sisonkestudy.samrc.ac.za/). The Institutional Review Boards/Ethics Committees of participating Clinical Research Sites approved the study, which was conducted under the oversight of the South African Health Products Regulatory Authority (SAHPRA).
Vaccinations were conducted in collaboration with the routine Provincial Department of Health public and private vaccination centres across all nine provinces of SA and overseen by Good Clinical Practice (GCP)-trained personnel linked to each of the ENSEMBLE trial research sites. Participants underwent informed consent before receiving a single intramuscular (IM) injection of Ad26.COV2.S at a dose level of 5×1010 virus particles. Participants with a previous history of allergic reactions to vaccinations were observed for 30 minutes post-vaccination whereas the rest of the participants were observed for 15 minutes post-vaccination.
Safety monitoring was conducted through a combination of passive reporting and active case finding. An electronic adverse event reporting link was sent via text message on days 1, 7 and 14 post vaccination. Adverse events could also be reported either by calling a toll-free 24-hour safety line or through the completion of an adverse event report form which was available at vaccination sites and hospitals. The safety team reviewed serious adverse events and adverse events of special interest reports daily. Full details of the Sisonke pharmacovigilance and safety reporting processes are detailed elsewhere (11).
The Sisonke safety database was searched for allergic AEFI using a comprehensive list of possible allergy-related search terms (see supplementary appendix). Duplicates and clear non-allergic entries were removed and then all reports were screened by an allergist. In addition to the initial reporting, telephonic contact was attempted with all participants reporting a possible allergy AEFI where details were missing to clarify the event and collect additional information on past medical and allergy history as well as details of treatment and outcomes of mild/moderate allergic AEFI requiring treatment and healthcare contact. Suspected anaphylaxis cases were adjudicated by two physicians using the Brighton Collaboration and NIAID case definition; with cases needing to meet both definitions to be considered confirmed cases (12-14). Immediate reactions were restricted to those occurring within 6 hours post vaccination (15).
Descriptive statistics were performed using counts and proportions for categorical data, and medians and interquartile ranges for continuous variables. All statistical analyses were conducted using STATA version 14 (STATA Corp., College Station, TX, USA).