key: cord-0759196-i0kixcep
authors: Takuva, S.; Takalani, A.; Seocharan, I.; Yende-Zuma, N.; Reddy, T.; Engelbrecht, I.; Faesen, M.; Khuto, K.; Whyte, C.; Bailey, V.; Trivella, V.; Peter, J.; Opie, J.; Louw, V.; Rowji, P.; Jacobson, B.; Groenewald, P.; Dorrington, R. E.; Laubscher, R.; Bradshaw, D.; Moultrie, H.; Fairall, L.; Sanne, I.; Gail-Bekker, L.; Gray, G.; Goga, A.; Garrett, N.
title: Safety of the single-dose Ad26.CoV2.S vaccine among healthcare workers in the phase 3b Sisonke study in South Africa
date: 2021-12-21
journal: nan
DOI: 10.1101/2021.12.20.21267967
sha: 26b5813226f8b06e7d59ee74c8cb4f4de43f8137
doc_id: 759196
cord_uid: i0kixcep

Background: The Sisonke openlabel phase 3b implementation study aimed to assess the safety and effectiveness of the Janssen Ad26.CoV2.S vaccine among health care workers (HCWs) in South Africa. Here, we present the safety data. Methods: We monitored adverse events (AEs) at vaccination sites, through self reporting triggered by text messages after vaccination, health care provider reports and by active case finding. The frequency and incidence rate of non serious and serious AEs were evaluated from day of first vaccination (17 February 2021) until 28 days after the final vaccination (15 June 2021). COVID 19 breakthrough infections, hospitalisations and deaths were ascertained via linkage of the electronic vaccination register with existing national databases. Findings: Of 477,234 participants, 10,279 (2.2%) reported AEs, of which 139 (1.4%) were serious. Women reported more AEs than men (2.3% vs. 1.6%). AE reports decreased with increasing age (3.2% for 18 to 30, 2.1% for 31 to 45, 1.8% for 46 to 55 and 1.5% in >55 year olds). Participants with previous COVID 19 infection reported slightly more AEs (2.6% vs. 2.1%). The commonest reactogenicity events were headache and body aches, followed by injection site pain and fever, and most occurred within 48 hours of vaccination. Two cases of Thrombosis with Thrombocytopenia Syndrome and four cases of Guillain Barre Syndrome were reported post-vaccination. Serious AEs and AEs of special interest including vascular and nervous system events, immune system disorders and deaths occurred at lower than the expected population rates. Interpretation: The single-dose Ad26.CoV2.S vaccine had an acceptable safety profile supporting the continued use of this vaccine in our setting.

vaccines.(7) Adverse events following COVID-19 vaccination are generally mild, and include local reactions, such as injection site pain, redness, swelling, and systemic reactions, like fever, headache, fatigue, nausea, vomiting, diarrhea. (8, 9) As reported by the US Center for Disease Control, severe or potentially life-threatening AEs are rare, and after 12.6 million doses of the Ad26.CoV2.S vaccine, 38 cases of Thrombosis with Thrombocytopenia Syndrome (TTS) and 98 cases of Guillain-Barré Syndrome (GBS) were reported, while after 141 million second mRNA vaccine doses, 497 cases of myocarditis were reported. (10) Following the precautionary pause instituted by the Food and Drug Administration (FDA) in April 2021, the SA Health Products Regulatory Authority recommended a similar 2-week pause for the Sisonke study. (11) The study recommenced with additional safeguards including screening and monitoring of participants at high risk of thrombosis and implementing measures to safely manage participants with TTS. Participant information sheets and informed consent forms were updated to include the newly identified AEs. Identification of such rare events illustrated that continued evaluation of the safety profile of vaccines post licensure is crucial to accurately characterize safety and to identify very rare adverse events that may not be reported in clinical trials.

The Sisonke study enrolled almost half a million HCWs, providing an opportunity to further evaluate the safety of the Ad26.COV.S2 vaccine in an expanded population.

The Sisonke study is a multi-centre, open-label, single-arm phase 3b implementation study among HCWs (>18 years) in South Africa, which is conducted in collaboration with the 

Participants received appointments for vaccination through the EVDS or were invited via employer lists. Vaccinations were conducted in collaboration with the National Department of Health public or private vaccination centres across all nine South African provinces and overseen by Good Clinical Practice-trained personnel linked to one of the ENSEMBLE trial research sites. Participants received a single intramuscular injection of Ad26.COV2.S at a dose of 5×10 10 virus particles and were observed for AEs for 15 minutes post-vaccination, and for 30 minutes, if they had a previous history of allergic reactions to vaccinations.

Adverse events were reported into the study database via multiple streams using a hybrid surveillance system that combined passive with active reporting.(12) Firstly, we designed an electronic case report form (eCRF). After vaccination every participant received a text message with COVID-19 infection prevention measures, that also listed common signs and symptoms of reactogenicity and provided an AE reporting web link, which allowed participants access to the form for AE reporting. Second, health care providers were able to complete paper-based CRFs that were available at healthcare and vaccination facilities, which were then submitted to the Sisonke safety desk and captured in the AE database. Third, the study . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint In addition, we actively linked EVDS data via national identification numbers with national patient-level disease databases, COVID-19 case notifications and the national population registry to identify vaccinees with COVID-19 infections, COVID-19-related hospitalisations and deaths. COVID-19 is a notifiable medical condition in South Africa and tests conducted across laboratories are reposited in the National Health Laboratory Service data system, which was used to identify seropositive Sisonke participants via active linkage. A death notification form must be submitted to the Department of Home Affairs to obtain a death certificate. Therefore, in addition to case reports and active tracing, mortality was ascertained via linkage with the national population registry. After identification of deaths, the safety staff contacted next of kin, primary health care providers and solicited medical records to ascertain causes of death.

Adverse event reports were processed daily and screened for serious AEs (SAE), which met International Conference on Harmonization criteria and AEs of special interest (AESI) per Brighton Collaboration list (9) . Seven days after reporting an AE, participants received a followup text message with a link to the eCRF. Participants reporting worsening and non-resolving symptoms were followed up by safety desk staff. After the FDA lifted the cautionary 2-week pause in vaccinations, two additional follow-up text messages were sent to all participants seven and 14 days after vaccination. The texts highlighted signs/symptoms associated with TTS and provided a link to the eCRF. Safety staff made attempts to obtain medical records and supporting information from health care providers for all reported SAEs.

The protocol safety review team comprising principal investigators, safety physicians and subject matter experts (haematologists, neurologist, allergy expert, infectious disease specialists) provided oversight by weekly safety data review. An independent safety monitoring committee provided additional safety oversight.

For descriptive statistics, counts and proportions were used for categorical variables, and medians and interquartile ranges for quantitative variables. Participants reporting and not reporting AEs were compared by baseline characteristics. SAEs were summarized by . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint and European population databases. (10, (15) (16) (17) (18) Person-time was accrued from the date of vaccination until death or dataset closure on 15 June 2021 (28 days after the last vaccination).

The incidence rates per 100,000 person-years were calculated using a Poisson model with person-years as an offset. Deaths were excluded from the SAE analysis and examined as a separate entity. COVID-related deaths were excluded in this report and published in a separate effectiveness report. All statistical analyses were conducted using STATA version 14 (STATA Corp., College Station, TX, USA).

The Sisonke study enrolled and vaccinated 477,234 participants from all nine South African provinces between 17 February 2021 and 17 May 2021. The majority were women (74.9%), the median age was 42 years (IQR 33-51) and 16.3% were older than 55 years. Previous COVID-19 infection was self-reported by 14.5% of participants. The most prevalent comorbidities were hypertension (15.6%), HIV infection (8.3%) and diabetes mellitus (5.9%). Most reported AEs (n=9,021, 81%) were reactogenicity events; the commonest were headaches and body aches (including arthralgia, myalgia and fatigue) which occurred within . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21267967 doi: medRxiv preprint the first seven days of vaccination, followed by mild injection site pain and fever ( Figure 1 ).

These events occurred predominately on the day of vaccination, reducing in frequency by day two. Self-reported severity was mild to moderate in 67% (n=7,157), i.e. the event did not result in loss of ability to perform usual social and functional activities, while 32% (n=3,375) reported being unable to perform usual activities and 2% (n=213) reported that they needed to visit the emergency room or were hospitalised. Follow up at day seven post vaccination indicated that 92% of participants reporting AEs had either completely recovered or were recovering. The remaining 8% of participants were contacted by the safety team and, if required, referred for care. One in five (19%) AEs were not consistent classified as reactogenicity events (Supplementary Table 1 ). Table 2) . SAE outcomes were: 48 (34.8%) recovered, 36 (26.1%) recovering, 45 (32.6%) ongoing and 9 (6.5%) deceased.

The commonest vascular disorders were pulmonary emboli (n=10, 10.6 per 100,000 person years, 95%CI 5.7-19.6) and ischaemic strokes (n=10, 10.6 per 100,000 person years, 95%CI 5.7-19.6) followed by deep vein thrombosis (n=4, 4.2 per 100,000 person-years, 95%CI 1.6-11.3). Three participants had both pulmonary embolism and deep vein thrombosis. There were two cases classified as TTS. The first case was a woman in the 45-50 year age group presenting with pulmonary embolism, thrombocytopenia and positive anti-platelet factor 4 antibodies nine days after vaccination. She had a history of injectable contraceptive use, underlying chronic respiratory, neurological condition and was being investigated for an autoimmune disorder. The second case was a woman in the 25-30 year age group who was admitted to hospital unconscious after experiencing a severe headache, restlessness and confusion from 33 days after vaccination. A CT brain scan with venogram was in keeping with superior sagittal sinus thrombosis. Anti-platelet factor 4 antibodies assay were negative, and she had marginally low platelets. She was a current smoker but had no other significant medical history. Both participants recovered.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint Figure 1 illustrates the frequency of SAE reporting from day of vaccination. As expected, there was a drop off in events when we analysis was restricted to 28 days post-vaccination, but however there was not a significant change in the observed vs. expected ratios of SAEs.

A total of 157 non-COVID-19 related deaths (167 per 100,000 person-years) were identified via the active linkage system with the national population registry. Of these deaths, 67/157 (42.7%) causes of death were adjudicated and ascertainment continues for the remainder. Mortality Surveillance Report and pre-COVID-19 local employee group life assurance data for a similar age-structured working population. (19) The mortality rate in Sisonke was similar to the working population mortality data with similar ages, and well below that of the overall population mortality rate (Figure 2 ).

The Sisonke study, a large single arm, open-label phase 3b implementation study aimed to assess the safety and effectiveness of the single-dose Ad26.CoV2.S vaccine among almost half a million HCWs in South Africa. With over 10,000 AE reports, this was the largest safety analysis of the Ad26.CoV2.S vaccine from a low-and middle-income country. As observed in phase 3 trials, similar patterns of AEs were found and were mostly expected reactogenicity signs and symptoms. Furthermore, SAEs were rare and occurred below expected rates.

However, we did observe very rare events of TTS and GBS in this study. Nevertheless, overall, this study provides additional real-world evidence that the vaccine is safe and well tolerated, supporting its continued use in this setting.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21267967 doi: medRxiv preprint Adverse events were more often reported by women than men. While this may illustrate a stronger immune response seen in females compared to males (20) (21) (22) , behavioural factors such as reduced reporting among men may have played a role, but were not measured. The prevalence of reported AEs reduced with increasing age. A number of studies show that vaccine-related AEs and reactogenicity are less prevalent in older people due to waning of innate immune defence mechanisms, lower systemic levels of IL-6, IL-10, C-reactive protein and lower neutralising antibody titres after vaccination as compared to younger individuals.(23-25) Individuals reporting previous COVID-19 infection seemed to have more AE reporting rates. Some studies suggest that there is increased immunogenicity in the setting of past infection leading to higher antibody titres (26, 27) and therefore higher reactogenicity rates.

Thrombosis with Thrombocytopenia Syndrome and GBS occurred at very low rates in this study, however the disproportionality analysis showed a higher event rate than expected in the population. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21267967 doi: medRxiv preprint commonest cause of death during the study period, which highlights the urgent need for specialist oncology services, it is concerning that among HCWs advanced HIV and tuberculosis remain common causes of death. It shows that despite gains in access to HIV testing and treatment, HIV and TB care require further improvement. Local data highlight that the COVID-19 epidemic heavily impacted HIV testing and treatment initiations.(30) Motor vehicle accidents and homicides were also common causes of death, a reflection of the injury and trauma burden in South Africa. One death was related to TTP, which has previously been reported post Ad26.COV2-S vaccination(31) and warrants further evaluation in other studies.

The Sisonke study had several limitations. Firstly, the surveillance system was primarily passive relying on self-reporting, thus some AEs may have gone unreported. It is likely that the system was better suited to detect SAEs than milder AEs, which participants may have ignored rather than reported. As active contact with participants was up to two-weeks postvaccination, it is probable that SAEs other than deaths and COVID-19 events were more likely to be reported during this period leading to underestimation of SAEs which occurred. The active linkage of the unique identifier in EVDS with the deaths on the national population registry and COVID-19 laboratory system ensured identification of nearly all possible deaths and COVID-19 events in the study. Additionally, considering the large number of participants in the study, not all self-reported AEs could be verified and only SAEs and AEs of medical concern were investigated further. Interpretation of the disproportionality analysis should be cautious given the uncertainties in both the observed and expected event rates, variable follow-up time, non-South African reference data for some groups, and potential differences in age-sex distributions between Sisonke and reference data. However, while disproportionality analysis in the context of safety signal detection is mainly exploratory, it has the utility to identify potentially important associations between AE and vaccine. In this study, the analysis confirmed current reports on safety risk of the Ad26.CoV2.S vaccine with the TTS and GBS.(10) Lastly, It is also important to note that without a placebo group, open label single arm studies are subject to measurement bias with the potential of overreporting of AEs and hence some caution in interpreting safety data. No other safety concerns were found in this study. (10, 11) In conclusion, this study affirms that the single dose Ad26.COV2.S vaccine is safe and well tolerated when administered to adults in South Africa. Few SAEs were observed and they were successfully managed with prompt identification. The Sisonke study underscores the value of setting up robust pharmacovigilance systems for prompt identification, evaluation and reporting of AEs to enable continued assessment of the risk-benefit profiles of COVID-19 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The OE analysis compares the observed and expected numbers of cases. This may be expressed as the ratio

The rates are for adults (male/females) and are not stratified by age -group.

Where a range is given in the literature on incidence, the mid-point was used.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21267967 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

Safety and Efficacy of Single-Dose Ad26.COV2.S Vaccine against Covid-19

Detection of a SARS-CoV-2 variant of concern in South Africa

COVID-19 Sentinel Hospital Surveillance Weekly Update on Hospitalized Horsham, PA

Our World in Data

Safety monitoring in the Vaccine Adverse Event Reporting System (VAERS)

Vaccines and Related Biological Products Advisory Committee Meeting FDA Briefing Document Janssen Ad26 . COV2 . S Vaccine for the Prevention of COVID-19 Sponsor

Safety Platform for Emergency vACcines

COVID-19 Vaccines in Adults : Benefit-Risk Discussion Current COVID-19 vaccine policy

SAHPRA. SAHPRA Statement -Update on Sisonke Phase 3B Implementation Study

Thromboembolic Events in the South African Ad26.COV2.S Vaccine Study. The New England journal of medicine

Safety Platform for Emergency vACcines Guide for 1st Tier AESI Acute Myelitis

Epidemiology of acute coronary syndromes in a Mediterranean country; aims, design and baseline characteristics of the Greek study of acute coronary syndromes (GREECS). BMC Public Health

Discovery Health Medical Scheme. Deep venous thrombosis and pulmonary embolism rates

Visual representation of national institute of allergy and infectious disease and food allergy and anaphylaxis network criteria for anaphylaxis

Assessment Of An Adverse Event Following Immunization

Rapid Mortality Surveillance Report

Sex-based differences in immune function and responses to vaccination

Sex and Gender Differences in the Outcomes of Vaccination over the Life Course

SeXX matters in immunity

Safety and immunogenicity of ChAdOx1 nCoV-19 vaccine administered in a primeboost regimen in young and old adults (COV002): a single-blind, randomised, controlled, phase 2/3 trial

Interim Results of a Phase 1-2a Trial of Ad26.COV2.S Covid-19 Vaccine

Safety and Immunogenicity of Two RNA-Based Covid-19 Vaccine Candidates

Antibody response to first BNT162b2 dose in previously SARS-CoV-2-infected individuals

Covid-19: People who have had infection might only need one dose of mRNA vaccine

We acknowledge the South African Medical Research Council for sponsorship and oversight, Janssen Vaccines and Prevention for the supply and transport of the study product to South Africa, Right to Care for supporting the safety monitoring and reporting infrastructure and the National Department of Health.We thank all the health care workers that participated in this study, the investigators and staff members at the vaccination centers and clinical research sites, the protocol team, protocol safety review team and members of the independent data and safety monitoring committee 

vaccines. This has the potential to improve public confidence in vaccine safety and reduce vaccine hesitancy.