Risk considerations

The AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine uses a non-replicating and non-integrating platform. This means the vaccine component does not interact with the genome and does not carry a risk for infection [Antrobus RD, 2014 ].

Adenoviral vectors are frequently used for gene transfer because of their high cellular transduction efficiency in vitro and in vivo. Expression of viral proteins and the low capacity for foreign DNA limits the clinical application of first- and second-generation adenoviral vectors. Adenoviral vectors with all viral coding sequences offer the prospect of decreased host immune responses to viral proteins, decreased cellular toxicity of viral proteins and increased capacity to accommodate large regulatory DNA regions [Lundstrom K, 2021 ].

In most cases, third generation replication-deficient adenoviral vectors with deletions in genes E1 and E3 have been use for the expression of the SARS-CoV-2 spike (S) protein or their receptor binding domain (RBD) [Lundstrom K, 2021 ].

The ChAdY25/ChAdOx1 is a recombinant E1 E3-deleted vector developed a decade ago based on the chimpanzee adenovirus Y25. The vector was constructed in a bacterial artificial chromosome and demonstrated comparable immunogenicity to that of other species of chimpanzee adenovirus vectors. The prevalence of virus neutralizing antibodies against ChAdY25 in serum samples collected from human populations was particularly low compared to published data for other chimpanzee adenoviruses [Dicks MD, 2012 ].

Efficacy of preclinical studies on the vaccine

Van Doremalen et al. [van Doremalen N, 2020 ] conducted a study in mice and rhesus macaques. The vaccine elicited a robust humoral and cell-mediated response in mice, predominantly Th1. In rhesus macaques, vaccination induced a balanced Th1/Th2 humoral and cellular immune response. Vaccinated animals challenged with SARS-CoV-2 did not develop pneumonia and presented a reduced viral load in bronchoalveolar lavage fluid and lower respiratory tract tissue compared with control animals. There was no difference in nasal shedding between vaccinated and control animals.

Lambe et al. [Lambe T, 2021 ] conducted a study using rhesus macaque and ferret challenge models. In rhesus macaques, the lung pathology caused by SARS-CoV-2 mediated pneumonia was reduced by prior vaccination which induced neutralizing antibody responses after a single intramuscular administration. In ferrets, the vaccine reduced both virus shedding and lung pathology. Antibody titers were boosted by a second dose.

Silva-Cayetano et al. [Silva-Cayetano A, 2020 ] demonstrated that the vaccine induced both cellular and humoral immunity in adult and aged mice using in-depth immunophenotyping to characterize the innate and adaptive immune response induced upon intramuscular administration of the vaccine in mice. A single dose generated spike-specific Th1 cells, Th1-like Foxp3+ regulatory T cells, polyfunctional spike-specific CD8+ T cells and granzyme-B-producing CD8 effectors. A second dose enhanced the immune response in aged mice.

Graham et al. [Graham S.P., 2020 ] compared the immunogenicity of one or two doses of the vaccine in both mice and pigs. A single dose induced antigen-specific antibody and T cells responses. A booster immunization enhanced antibody responses with a significant increase in SARS-CoV-2 neutralizing titers.

Van Doremalen et al. [Neeltje van Doremalen, 2021 ] conducted a study in Syrian hamsters and rhesus macaques. Intranasally vaccinated hamsters exposed to SARS-CoV-2 virus with the D614G mutation in the spike protein, significantly decreased viral load in comparison to controls. No viral RNA or infectious virus was found in lung tissue, both in a direct challenge and a transmission model. 

Efficacy of the vaccine in clinical trials

Main Immunogenicity outcomes

The combined results of 17,178 participants analyzed in the COV001, COV002, COV003 and COV005 trials (8597 receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine and 8581 receiving control vaccine) [Voysey M, 2021 ] showed antibody levels were maintained by day 90 (geometric mean ratio [GMR] 0.66 [95% CI 0.59 to 0.74]). In the participants who received two standard doses the binding antibody response after an interval of 12 or more weeks in those who were aged 18-55 years was GMR 2.32 [2.01 to 2.68]).

The D8110C00001 [Falsey AR, 2021 ], was a phase 3 randomized trial, that included 34,117 participants who received the AstraZeneca / Oxford COVID-19 vaccine or placebo. Participants that were seronegative at baseline showed a large vaccine-induced serum IgG response to spike protein. Neutralizing antibody levels were higher than baseline at different measurement points in the vaccine group, increasing further after the second dose. However, they remained low throughout the trial in the placebo vaccine group.

ICMR / IBS-COVISHIELD was a phase 2/3 randomized trial, which included 1601 participants who received the AstraZeneca/Oxford COVID-19 vaccine or the Serum Institute of India (ISS) COVID-19 vaccine. The trial included the immune response of the participants. Results showed that in the immunogenicity/reactogenicity cohort (n = 401) baseline anti-S IgG antibody titers were <100 (AU)/ml in the SII and AstraZeneca/Oxford COVID-19 vaccine groups. One month after the first dose (day 29), the GMTs were >6600 AU/ml in both groups, and one month after the second dose, the GMTs were >28500 AU/ml in both groups, this means that there was a 325-fold increase in antibody titers in both groups compared to baseline. In both vaccine groups, >91% of participants demonstrated seroconversion after the first dose and ≥98% demonstrated seroconversion after the second dose. These means that ISS COVID-19 vaccine showed a non-inferior immune response compared to the AstraZeneca/Oxford COVID-19 vaccine [Prasad S. Kulkarni, 2021 ].

Sobieszczyk ME et al was a phase 3 randomized trial that included 26,145 participants included in double-blind period: 17,617 vaccine group and 8,828 placebo group. The primary efficacy end point was confirmed SARS-CoV-2 reverse-transcriptase PCR-positive symptomatic COVID-19 at 15 or more days after a second dose in baseline SARS-CoV-2-seronegative participants. Participants who received AstraZeneca and were seronegative at baseline showed strong vaccine induced serum antibody responses to the spike protein. Median SARS-CoV-2 neutralizing antibodies titers at day 1 were 20.0 for vaccine group and 20.0 for placebo group. Median SARS-CoV-2 neutralizing antibodies titers at day 57 were 254.5 for vaccine group and 20.0 for placebo group. [Sobieszczyk ME, 2022 ]

Key messages

AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine reduces the risk of contracting COVID-19

AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine probably reduces the risk of contracting severe COVID-19

Main efficacy outcomes of AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine

Contracting COVID-19 (measured at least 14 days after the second injection)

The relative risk of contracting COVID-19 in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 0.42 (95% CI 0.34 to 0.52). This means AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine reduced the risk of contracting COVID-19 in 58%, compared with placebo vaccine.

Figure - Forest plot of risk ratio meta-analysis. Outcome: contracting COVID-19. Comparison: AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus placebo vaccine

In the trial identified in this review, 530 people not receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine out of 17109 presented this outcome (31 per 1000) versus 302 out of 26214 in the group that did receive it (13 per 1000). In other words, 18 less people per 1000 did not develop the outcome because of the vaccine. This is the same as saying that the intervention led to an absolute risk reduction of 1.8%, or that the intervention reduced the risk of contracting COVID-19 by 1.8 percentage points. Another way of presenting the same information about the absolute effects is the number needed to treat for an additional beneficial/harmful outcome (NNTB/H), the number of participants who need to receive the intervention for one of them to experience the outcome. In this case, the NNTB is 56. Which means that 56 people need to receive the vaccine for one of them to not contract COVID-19.

Applying the GRADE approach [The GRADE Working Group, 2013 ], we assessed the certainty of the evidence for this outcome as high certainty evidence.

Contracting severe COVID-19 (measured at least 14 days after the second injection)

The relative risk of contracting severe COVID-19 in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 0.08 (95% CI 0.01 to 0.48). This means AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine reduced the risk of contracting severe COVID-19 in 92%, compared with placebo vaccine.

Figure - Forest plot of risk ratio meta-analysis. Outcome: contracting severe COVID-19. Comparison: AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus placebo vaccine

In the trial identified in this review, 11 people not receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine out of 20252 presented this outcome (5 per 10000) versus 0 out of 29638 in the group that did receive it (0 per 10000). In other words, 5 less people per 10000 did not develop the outcome because of the vaccine. This is the same as saying that the intervention led to an absolute risk reduction of 0.05%, or that the intervention reduced the risk of contracting severe COVID-19 by 0.05 percentage points. Another way of presenting the same information about the absolute effects is the number needed to treat for an additional beneficial/harmful outcome (NNTB/H), the number of participants who need to receive the intervention for one of them to experience the outcome. In this case, the NNTB is 2000. Which means that 2000 people need to receive the vaccine for one of them to not contract severe COVID-19.

Applying the GRADE approach [The GRADE Working Group, 2013 ], we assessed the certainty of the evidence for this outcome as moderate certainty evidence.

Mortality

The existing evidence does not allow to assess the impact of AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine on the risk of death attributable to COVID-19. The information provided by randomized trials was not adequately powered to estimate a difference in this outcome. Deaths can occur in the intervention and control group for reasons unrelated to COVID-19 or the vaccine. Establishing that there is a reduction (or increase) in the risk of death attributable to Moderna COVID-19 vaccine would require trials with a higher statistical power.

Efficacy of the vaccine in subgroups

Contracting COVID-19 (>65y) (measured at least 14 days after the second injection)

The relative risk of contracting COVID-19 in > 65-year-old participants in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 0.18 (95% CI 0.06 to 0.49). This means AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine reduced the risk of contracting COVID-19 in > 65-year-old participants in 82%, compared with placebo vaccine.

Figure - Forest plot of risk ratio meta-analysis. Outcome: contracting COVID-19 in >65 year old participants. Comparison: AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus placebo vaccine

In the trial identified in this review, 19 people not receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine out of 1816 presented this outcome (10 per 1000) versus 6 out of 3696 in the group that did receive it (2 per 1000). In other words, 8 less people per 1000 did not develop the outcome because of the vaccine. This is the same as saying that the intervention led to an absolute risk reduction of 0.8%, or that the intervention reduced the risk of contracting COVID-19 in >65 year old participants by 0.8 percentage points. Another way of presenting the same information about the absolute effects is the number needed to treat for an additional beneficial/harmful outcome (NNTB/H), the number of participants who need to receive the intervention for one of them to experience the outcome. In this case, the NNTB is 125. Which means that 125 people need to receive the vaccine for one of them to not contract COVID-19.

Applying the GRADE approach [The GRADE Working Group, 2013 ], we assessed the certainty of the evidence for this outcome as low. The certainty of the evidence is based in the following judgments: Risk of bias: no concerns; Inconsistency: no concerns; Indirectness: the information available is based on a short term follow-up; Imprecision: the information provides from a small sample; Publication bias: no concerns.

Contracting COVID-19 (female) (measured at least 14 days after the second injection)

The relative risk of contracting COVID-19 in female subgroup in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 0.41 (95% CI 0.29 to 0.59). This means AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine reduced the risk of in 59%, compared with placebo vaccine.

Figure - Forest plot of risk ratio meta-analysis. Outcome: contracting COVID-19 in female subgroup. Comparison: AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus placebo vaccine

In the trial identified in this review, 46 people not receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine out of 3721 presented this outcome (12 per 1000) versus 32 out of 7740 in the group that did receive it (4 per 1000). In other words, 8 less people per 1000 did not develop the outcome because of the vaccine. This is the same as saying that the intervention led to an absolute risk reduction of 0.8%, or that the intervention reduced the risk of contracting COVID-19 in female subgroup by 0.8 percentage points. Another way of presenting the same information about the absolute effects is the number needed to treat for an additional beneficial/harmful outcome (NNTB/H), the number of participants who need to receive the intervention for one of them to experience the outcome. In this case, the NNTB is 125. Which means that 125 people need to receive the vaccine for one of them to not contract COVID-19.

Applying the GRADE approach [The GRADE Working Group, 2013 ], we assessed the certainty of the evidence for this outcome as low. The certainty of the evidence is based in the following judgments: Risk of bias: no concerns; Inconsistency: no concerns; Indirectness: the information available is based on a short term follow-up; Imprecision: the information provides from a small sample; Publication bias: no concerns.

Contracting COVID-19 (male) (measured at least 14 days after the second injection)

The relative risk of contracting COVID-19 in male subgroup in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 0.35 (95% CI 0.26 to 0.46). This means AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine reduced the risk of contracting COVID-19 in male subgroup in 65%, compared with placebo vaccine.

Figure - Forest plot of risk ratio meta-analysis. Outcome: contracting COVID-19 in male subgroup. Comparison: AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus placebo vaccine

In the trial identified in this review, 118 people not receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine out of 4814 presented this outcome (25 per 1000) versus 84 out of 9885 in the group that did receive it (9 per 1000). In other words, 16 less people per 1000 did not develop the outcome because of the vaccine. This is the same as saying that the intervention led to an absolute risk reduction of 1.6%, or that the intervention reduced the risk of contracting COVID-19 in male subgroup by 1.6 percentage points. Another way of presenting the same information about the absolute effects is the number needed to treat for an additional beneficial/harmful outcome (NNTB/H), the number of participants who need to receive the intervention for one of them to experience the outcome. In this case, the NNTB is 63. Which means that 63 people need to receive the vaccine for one of them to not contract COVID-19.

Applying the GRADE approach [The GRADE Working Group, 2013 ], we assessed the certainty of the evidence for this outcome as low. The certainty of the evidence is based in the following judgments: Risk of bias: no concerns; Inconsistency: no concerns; Indirectness: the information available is based on a short term follow-up; Imprecision: the information provides from a small sample; Publication bias: no concerns.

Summary of findings (iSoF)

Efficacy and effectiveness of the vaccine in subgroups

Sex
Randomized trials
The proportion of females in the combined analysis of COV002 and COV003 was 60.5% (7045 out of 11636 participants) [Voysey, Merryn, 2021 ].
The magnitude of the effect was similar between the subgroups, and there was no statistical evidence of a subgroup effect by sex 
     
Age
Randomized trials
The proportion of patients >55 years of age in the efficacy population of the combined analysis of COV002 and COV003 was 12.2% (1418 out of 11636 participants) [Voysey, Merryn, 2021 ].
The efficacy in the 18-55 years group in the different trials, doses and administration schedule ranged from 59.3% to 65.6%. These results were no statistically different from the efficacy of the vaccine in the overall group [Voysey, Merryn, 2021 ].
Efficacy estimates for different age have not been yet reported. The researchers plan to present them in future analyses when a larger dataset is available [Voysey, Merryn, 2021 ].
   
Other comparative studies
Lopez Bernal et al. conducted a case-control design study in the United Kingdom [Jamie Lopez Bernal, 2021 ] evaluating all adults aged 70 years and older (over 7.5 million) from December 2020 to February 2021. Vaccination with a single dose of the AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine was associated with a significant reduction in symptomatic SARS-CoV2 positive cases in older adults with even greater protection against severe disease. The vaccine effects were seen from 14-20 days after vaccination reaching effectiveness of 60% (95% CI 41-73%) from 28-34 days and further increasing to 73% (95% CI 27-90%) from day 35 onwards.
          
Vasileiou et al [Eleftheria Vasileiou, 2021 ] found that the effect of a single dose of the AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine would prevent hospitalization related to COVID-19 when restricting the analysis to those aged ≥80 years (81%; 95% CI 65 to 90 at 28-34 days post-vaccination).

The VIVALDI study included 1317 samples from 402 residents and 632 staff fully vaccinated  (AstraZeneca=593; Pfizer=534). Data were collected 21 days after the second dose. The study showed that two-dose of AstraZeneca or Pfizer COVID-19 vaccine elicits robust and stable antibody responses in older Long-Term Care Facility residents, consistent with overall levels of vaccine-induced immunity. [Oliver Stirrup, 2022 ]

Elsie MF Horne et al was a comparative cohort study conducted in United Kingdom that estimated vaccine effectiveness of two doses of the Pfizer and AstraZeneca vaccines (versus no vaccine) in participants aged 18 to 64 and over 65 years. Hazard ratio for positive SARS-CoV-2 test (40-64 years years) was 0.92 (95% CI, 0.89-0.96). Hazard ratio for positive SARS-CoV-2 test (>65  years) was 0.23 (0.53 (95% CI, 0.39-0.71). [Elsie MF Horne, 2023 ]
    
Children and adolescents
Randomized trials
Children were excluded from the COV001, COV002, COV003 and COV0005 trials, so no data are available for this subgroup [Voysey, Merryn, 2021 ]. 
COV006 was a phase 1/2 participant-blinded, randomized controlled trial evaluating the immunogenic response in children 6-17 years old. Participants were randomly assigned in a 4:1:4:1 ratio to receive two doses of AstraZeneca COVID-19 vaccine or control. Anti-spike antibodies reached geometric means (AU/mL) of 43,280 (95%CI 35,852 to 52,246) in the 12-17 year old short interval group, 73,371 (95%CI 58,685 to 91,733) in the 12-17 year old long interval group and 108,924 (95% CI 84,852 to 139,823) in the 6-11 year old long interval group. [Li G, 2022 ].

Other comparative studies
COVA is an ongoing non-randomized study [The University of Hong Kong, 2021 ] evaluating the efficacy in children 11-16 years old.
           
Pregnancy
Randomized trials
Pregnant females have been excluded from the COV001, COV002, COV003 y COV005 trials, so no data are available for this subgroup [Voysey, Merryn, 2021 ]. 
                  
Breast-feeding
Randomized trials
Breastfeeding females were excluded from the COV001, COV002, COV003 and COV005 trials, so no data are available for this subgroup [Voysey, Merryn, 2021 ]. 
             
Immunocompromised people
Randomized trials
The phase 2, randomized, single-blinded study Boost-TX is currently evaluating the efficacy/safety of the vaccine in kidney transplant recipients [Medical University of Vienna, 2021 ].

COV005 was a phase 1/2 randomized controlled trial conducted in Africa. In this ongoing trial, people with HIV and HIV-negative participants aged 18−65 years were enrolled at seven South African locations. Between June 24 and Nov 12, 2020, 104 people with HIV and 70 HIV-negative individuals were enrolled. For SARS-CoV-2 seronegative participants at baseline, full-length spike geometric mean concentrations (GMC) at day 28 was 163.7 (95% CI, 89.9–298.1) in the HIV group, and 112.3 (95% CI, 61.7–204.4) in HIV-negative participants, with a rising response in GMC at day 42 in both groups [Madhi SA, 2021 ].

Other comparative studies
The cohort study COVATRANS is currently evaluating the effectiveness/safety of the vaccine in kidney transplant recipients aged 15 years and older who receive Pfizer, Moderna and Astra-Zeneca vaccines [University Hospital, Strasbourg, France, 2021 ].

The non-randomized study COVAC-IC is currently evaluating the effectiveness/safety of the vaccine in immunocompromised and immunocompetent haematology patients aged 18 years and older [University Hospitals of North Midlands NHS Trust, 2021 ].

The prospective, cohort, non-interventional, single-center clinical study ImmunoHaema-COVID-VAX-21 is currently evaluating the effectiveness/safety of the vaccine in patients with haematological malignancies 18 years of age and older [Ospedale di Circolo - Fondazione Macchi, 2021 ].

Spensley K et al. included 1121 patients on haemodialysis. All patients underwent weekly screening for SARS-CoV-2 infection via RT-PCR testing between December 1, 2021 and January 16, 2022. The study showed that partial vaccination did not provide protection against infection. Vaccine effectiveness against Omicron infection in patients who had received a booster vaccine was 58%. Analysing vaccine effectiveness in the 747 patients who had been boosted, significant effectiveness was seen in both patients who received AstraZeneca COVID-19 vaccine (47%)and Pfizer COVID-19 vaccine (66%).

Lin et al conducted a sub-analysis of a previous non-randomized clinical trial [Sheng WH, 2022 ] which compare the immunogenicity of heterologous ChAdOx1/mRNA-1273 versus standard homologous ChAdOx1 and mRNA-1273 vaccination. 399 participants were enrolled, there were 100, 100, 100, and 99 participants undergoing ChAdOx1/ChAdOx1 8 weeks apart (Group 1), ChAdOx1/mRNA-1273 8 weeks apart (Group 2), ChAdOx1/mRNA-1273 4 weeks apart (Group 3), and mRNA-1273/mRNA-1273 4 weeks apart (Group 4), respectively. The majority of the enrolled participants were ≤50 years with 74.7% being women.  Among participants undergoing prime vaccination with ChAdOx1, compared with healthy participants aged ≤50 years, participants with immunocompromising conditions, had similar anti-SARS-CoV-2 spike IgG titers before booster vaccination (geometric means, 75.36 vs. 82.87 BAU/mL; P = 0.429). However, numerically lower anti-SARS-CoV-2 spike IgG titers before booster vaccination were found in participants with autoimmune diseases compared to those without (geometric means, 34.76 vs. 84.25 BAU/mL; P = 0.173). The participants receiving hydroxychloroquine, low-dose steroid, methotrexate, and/or sulfasalazine had statistically significantly lower anti-SARS-CoV-2 spike IgG titers before booster vaccination compared with those not receiving (geometric means, 36.39 vs. 83.84 BAU/mL; P = 0.001), especially in those receiving hydroxychloroquine (geometric means, 38.48 vs. 82.97 BAU/mL; P = 0.009) and sulfasalazine (geometric means, 21.96 vs. 82.90 BAU/mL; P < 0.001). The participants receiving NSAIDs also had statistically significantly lower anti-SARS-CoV-2 spike IgG titers before booster vaccination compared with those not receiving (geometric means, 39.04 vs. 83.15 BAU/mL; P = 0.007), especially in those receiving COX-2 selective NSAIDs (geometric means, 27.88 vs. 83.49 BAU/mL; P < 0.001) [Lin KY, 2022 ].

Chambers et al. conducted a test-negative study to estimate vaccine effectiveness (VE) against SARS-COV-2 infection among a population-based cohort of people with HIV in Ontario, Canada. Among 21,023 adults living with HIV, there were 801 (8.3%) test positive cases and 8,879 test-negative controls. Adjusted VE against any infection was 77% (95% CI 34-92) for ChAdOx1 primary schedule and 80% (95% CI 49-92) for heterologous primary schedule (ChAdOx1 - BNT16b2 or mRNA-1273). [Chambers C, 2022 ]

Vaccine effectiveness (other comparative studies)

Contracting COVID-19
Ghosh S et al. was a comparative cohort study conducted in India. The study enrolled 1,595,630 participants (1,312,938 Vaccine group; 282,692 Control group). Based on data from the existing surveillance system among Health Care Workers and Frontline Workers of the Indian Armed Forces from January 16 until May 30, 2021. The cohort transitioned from unvaccinated to fully vaccinated, serving as its own internal comparison. The outcome was measured from 14 days after vaccination. The result of the study found vaccine effectiveness of 94.9% (95% CI 92.5 to 96.6) [Ghosh S, 2021 ]

Pritchard E et al. was a cohort study conducted in the United Kingdom that enrolled 373,402 participants (99,267 in the vaccine group and 274,135 in the control group). Based on data from the existing surveillance system among Health Care Workers and Frontline Workers of the Indian Armed Forces from January 16th to May 30th 2021. The cohort transitioned from 'Unvaccinated' to 'Fully Vaccinated', serving as its own internal comparison. The outcome was measured from 21 days after the second dose. The result of the study found vaccine effectiveness of 79% (95% CI, 65% to 88%) [Pritchard E, 2021 ].

Lopez Bernal J et al. was a case-control study conducted in United Kingdom. The study enrolled 156,930 participants (76,385 received AstraZeneca vaccine). Based on a test-negative design including 70 years or older in England from 6 October 2020 to 21 February 2021. The outcome was measured from 14 days after vaccination. The result of the study found vaccine effectiveness of 73% (95%CI 27 to 90). [Lopez Bernal J, 2021 ].

Hitchings M et al. was a case-control study conducted in Brazil. The study enrolled 61,164 participants (30,680 Vaccine group; 30,680 Control group). Based on test negative control design using individual-level information from adults ≥60 years of age who had a residential address in São Paulo State on 9 July 2021 from the SES-SP laboratory testing registry, the national surveillance databases for acute respiratory illness (ARI) and severe ARI, and the SES-SP vaccination registry. The outcome was measured from 14 days after vaccination. The result of the study found vaccine effectiveness of 77.9% (95%CI 69.2 to 84.2). [Matt Hitchings, 2021 ].

Alireza Mirahmadizadeh et al. was a cohort study conducted in Iran. The study enrolled 1,882,148 participants: 881,638 vaccine group; 1,000,510 control group. Based on data derived from administrative repositories during mass-vaccination campaigns or programs between February 09, 2021 and the end of follow-up in October 22, 2021; the study results showed a vaccine effectiveness of 84.4% (95%CI 83.5 to 85.3%) against infection, 81.5% (95%CI 79.5 to 83.4%) against hospitalization and 91.8% (95%CI 88.2 to 95.4%) against death [Alireza Mirahmadizadeh, 2022 ]

Chadeau-Hyam M et al. was a cohort study conducted in United Kingdom. The study enrolled 172,862 participants: 76,291 vaccine group; 96,571 control group. Based on data from a  series of random cross-sectional surveys in the general population of England aged 5 years and older, between May 2020 and September 2021, the study results showed a vaccine effectiveness of 44.8% (95%CI 22.5 to 60.7) against infection in vaccinated individuals [Chadeau-Hyam M, 2022 ].

Rearte A et al. was a case control study conducted in Argentina. The study included 358,431 individuals in the analysis of ChAdOx1 nCoV-19 vaccine: 126,531 in the case group and 231,900 in the control group. The study assessed the effectiveness of three vaccines (rAd26-rAd5, ChAdOx1 nCoV-19, and BBIBP-CorV) on SARS-CoV-2 infection and risk of death in people with RT-PCR confirmed COVID-19, using data from the National Surveillance System (SNVS 2.0). All individuals aged 60 years or older reported to SNVS 2.0 as being suspected to have COVID-19 who had disease status confirmed with RT-PCR were included in the study. The odds ratio of the ChAdOx1 nCoV-19  vaccine on the risk of SARS-CoV-2 infection in individuals with only one dose was OR 0.6 (95% CI 0.59 to 0.61). After the second dose was OR 0.32 (95% CI 0.31 to 0.33) [Rearte A, 2022 ].

Fano V et al was a comparative cohort study conducted in Italy. The study included 371,423 participants: 221,000 received Pfizer vaccine and 150,423 received AstraZeneca vaccine. The study estimated the effectiveness of Pfizer and AstraZeneca vaccines since 27/12/2020, and followed until diagnosis of SARS-CoV-2 infection or 25/4/2021, whichever came first. Adjusted hazard ratios of SARS-CoV-2 infection at weekly intervals since the first dose were estimated through a Cox regression model using 0-13 days as reference time-interval. Adjusted AstraZeneca vaccine effectiveness between 49 and 55 days was 63.2% ( 95%CI 24.8 to 82) [Fano V, 2022 ].

Horne E et al was a cohort study conducted in England, including 7,594,195 participants, amongst which 1,951,866 were Pfizer-BioNTech recipients, 3,219,349 AstraZeneca recipients, and 2,422,980 remained unvaccinated. Rates of COVID-19-related hospital admission and death were substantially lower among vaccinated than unvaccinated adults up to 26 weeks after the second dose, with estimated vaccine effectiveness ≥75% for ChAdOx1. By weeks 23-26, rates of positive SARS-CoV-2 test in vaccinated people were similar to or higher than in unvaccinated people (adjusted hazard ratios up to 1.86 (1.79 to 1.93) for ChAdOx1). COVID-19-related hospitalizations: Estimated adjusted hazard ratios for ChAdOx1 recipients compared with unvaccinated people in the ≥65 subgroup were 0.13 (0.08 to 0.23) during weeks three to six after the second dose, 0.25 (0.21 to 0.29) during weeks 23-26. [Horne EMF, 2022 ].

Costa Clemens SA was a case-control study (test-negative design) conducted in Brazil. The study included data from 7,958 participants: 2,109 cases and 5,849 controls. Odds of being vaccinated among cases vs controls were calculated to estimate vaccine effectiveness. All individuals aged 18–60 who received a PCR test after May 16th and were unvaccinated prior to this date were included in the analysis. Vaccine effectiveness against any COVID-19 illness was 74.5% (95% CI, 23.7% to 51.5%). Vaccine effectiveness against any COVID-19 illness was 76.0% (95% CI, 49.5% to 87.8%) against Delta Variant and 81.3% (95% CI, −298.0% to 97.9%) against the Gamma variant. [Costa Clemens SA, 2022 ]

Chung et al. conducted a case-control study with a test-negative design to estimate vaccine effectiveness (VE) against SARS-CoV-2 infection after the primary schedule of any combination of BNT162b2, mRNA-1273, and ChAdOx1 between January 11th and November 21st 2021 in Ontario, Canada. It included 261,360 test-positive cases (of any SARS-CoV-2 lineage) and 2,783,699 individuals as test-negative controls. VE for a homologous ChAdOx1 schedule 7-59 days after second dose was 82% (95% CI, 79% to 85%) against any infection and 86% (95% CI, 80% to 90%) against symptomatic infection. VE 120-179 days after second dose was 67% (95% CI, 63% to 71%) against any infection and 80% (95% CI, 76% to 83%) against symptomatic infection. [Chung H, 2022 ]

Elsie MF Horne et al was a comparative cohort study conducted in United Kingdom that estimated vaccine effectiveness of two doses of the Pfizer and AstraZeneca vaccines (versus no vaccine) in participants aged 18 to 64 and over 65 years. Hazard ratio for positive SARS-CoV-2 test (40-64 years years) was 0.92 (95% CI, 0.89-0.96). Hazard ratio for positive SARS-CoV-2 test (>65  years) was 0.23 (0.53 (95% CI, 0.39-0.71). [Elsie MF Horne, 2023 ]

Bello-Chavolla OY et al was a retrospective cohort study conducted in Mexico, including 793,487 vaccinated and 4,792,388 unvaccinated individuals. This study estimated vaccine effectiveness (VE) against infection and hospitalization, based on data from the COVID-19 surveillance system between December 2020 and September 2021. Among 60,250 participants vaccinated with AstraZeneca, VE against infection was 80.79% (95% CI, 80.4% to 81.1%), and VE against hospitalization was 80.23% (95% CI, 79.29% to 81.1%) [Bello-Chavolla OY, 2023 ].

Contracting severe COVID-19
Lopez Bernal J  et al. was a case-control study conducted in the United Kingdom. The study enrolled 156,930 participants (76,385 received the AstraZeneca vaccine). Based on a test-negative design including 70  years or older in England from 6  October  2020  to  21 February 2021. The outcome was measured 14 days after vaccination. The result of the study found vaccine effectiveness of   37% (95%CI 13 to 59) [Lopez Bernal J, 2021 ].

Alencar CH et al. was a comparative cohort study conducted in Brazil. The study enrolled 313,328 participants:159,970 Vaccine group (27,193 received AstraZeneca vaccine) and 40,941 Control group. Based on a retrospective data analysis from people aged 75 years or older from the National Mortality System (SIM) and from the Immunization Program (SIPNI), between 17 January and 11 May 2021. The outcome was measured from 21 days after vaccination. The result of the study found vaccine effectiveness of 99.8 (95%CI 99.6 to 99.9). [Alencar CH, 2021 ].

Ghosh S et al. was a comparative cohort study conducted in India. The study enrolled 1,595,630 participants (1,312,938 Vaccine group; 282,692 Control group). Based on data from the existing surveillance system among Health Care Workers and Frontline Workers of the Indian Armed Forces from January 16 until May 30, 2021. The cohort transitioned from unvaccinated to fully vaccinated, serving as its own internal comparison. The outcome was measured from 14 days after vaccination. The result of the study found vaccine effectiveness of 98.53% (95% CI: 0.00 to 99.9) [Ghosh S, 2021 ].

Hitchings M et al. was a case-control study conducted in Brazil. The study enrolled 61,164 participants (30,680 Vaccine group; 30,680 Control group). Based on test negative control design using individual-level information from adults ≥60 years of age who had a residential address in São Paulo State on 9 July 2021 from the SES-SP laboratory testing registry, the national surveillance databases for acute respiratory illness (ARI) and severe ARI, and the SES-SP vaccination registry. The outcome was measured from 14 days after vaccination. The result of the study found vaccine effectiveness of 87.6% (95%CI 78.2 to 92.9). [Matt Hitchings, 2021 ].

Sheikh A 2021, was a cohort study conducted in Scotland. This study used data from a national surveillance database to estimate vaccine effectiveness against death from delta variant infection between April 1 and August 16, 2021, among adult participants who were followed up to September 27, 2021. Mortality analysis was based on 114,706 adults who tested positive for SARS-CoV-2 and a total of 201 deaths were reported. Among participants between 40 and 59 years of age, vaccine effectiveness against death from COVID-19 was 88% (95% CI, 76 to 93) for ChAdOx1 nCoV-19 and 95% (95% CI, 79 to 99) for BNT162b2. Among participants aged 60 years or older vaccine effectiveness was 90% (95% CI, 84 to 94) and 87% (95% CI, 77 to 93), respectively. Overall, vaccine effectiveness against death from the delta variant 14 or more days after the second vaccine dose was 90% (95% CI, 83 to 94) for BNT162b2 and 91% (95% CI, 86 to 94) for ChAdOx1 nCoV-19. [Sheikh A, 2021 ]

Suah JL et al. was a cohort study conducted in Malaysia. The study enrolled 1,239,445 participants: 489,921 Vaccine group; 749,524 Control group and was based on data from four secondary data sets constructed from national COVID-19 surveillance: (1) the COVID-19 casesline listing, (2) the ICU admissions register, (3) the COVID-19 deaths line listing, and (4)the COVID-19 vaccine recipients line listing, linked deterministically with the case andpersonal identification numbers, between 1 April 2021 and 15 September 2021. The outcome was measured starting 14 days after the second dose. The study results showed a vaccine effectiveness of 95.6% (95%CI 88.3 to 98.4) for ICU admission and 95.3% (95%CI 91.3 to 97.4) for Death in vaccinated individuals [Suah JL, 2021 ]

Alireza Mirahmadizadeh et al. was a cohort study conducted in Iran. The study enrolled 1,882,148 participants: 881,638 vaccine group; 1,000,510 control group. Based on data derived from administrative repositories during mass-vaccination campaigns or programs between February 09, 2021 and the end of follow-up in October 22, 2021; the study results showed a vaccine effectiveness of 84.4% (95%CI 83.5 to 85.3%) against infection, 81.5% (95%CI 79.5 to 83.4%) against hospitalization and 91.8% (95%CI 88.2 to 95.4%) against death [Alireza Mirahmadizadeh, 2022 ]

Lytras T et al. included a total of 14,676,605 vaccine administered doses ( Pfizer= 11,427,784; Moderna=1,161,905; AstraZeneca=1,505,334; Janssen= 581,582). Data were collected between 11 January 2020 and 8 December 2021. The study showed that two doses of Pfizer, Moderna, or AstraZeneca COVID-19 vaccines offered vaccine effectiveness >90% against both intubation and death across all age groups. The effectiveness of the Janssen COVID-19 vaccine ranged between 61-81%. There was some waning over time but vaccine effectiveness remained >80% at six months, and three doses increased vaccine effectiveness again to near 100%. Vaccination prevented an estimated 19,691 COVID-19 deaths (95% CI 18,890 to 20,788) over the study period [Theodore Lytras, 2022 ].

Taherian Z et al. conducted a cohort study in Iran which included 583,434 vaccinated individuals and 2,551,140 unvaccinated persons.Based on data of all vaccinated individuals from March 2020 to August 2021. Data of unvaccinated individuals were collected from the census for this period. Vaccine effectiveness as risk reduction rate for infection was 99, for hospitalization was 99.4 and for death it was 100 [Taherian Z, 2022 ].

Arregocés-Castillo et al was a retrospective cohort study conducted in Colombia. The study included 2,828,294 participants: 1,414,147 fully vaccinated (any vaccine) and 1,414,147 unvaccinated. The study evaluated the effectiveness of vaccines against COVID-19-related hospitalization and death in people aged 60 years and older. Participant follow-up was done between March 11, 2021, and Oct 26, 2021.It was estimated the overall effectiveness of being fully vaccinated, as well as effectiveness for each vaccine. The aim results showed that vaccine effectiveness against hospitalization without death was 90.8% (95% CI 85.5 to 94.2) in adults > 60 years. Effectiveness against death after hospitalization was 97.5% (95% CI 95.8 to 98.5) and against death without hospitalization was 93.9% (95% CI 89.3 to 96.6) in adults > 60 years [Arregocés-Castillo L, 2022 ].

Elsie MF Horne et al. was a comparative cohort  study conducted in England. The study included data from the OpenSAFELY-TPP database. The study included data from the OpenSAFELY-TPP database. The study included 1,773,970 individuals with BNT162b2 vaccine,  2,961,011 individuals with ChAdOx1 vaccine and 2,433,988 unvaccinated individuals. The study compared individuals who had received two doses of BNT162b2 or ChAdOx1 with unvaccinated individuals during six 4-week comparison periods, separately by age subgroups. The Hazard Ratio versus unvaccinated for COVID-19 hospitalization ranged from (≥ 65 years) 1.16 (95% IC 0.98 to 1.37) to 1.20 (95% IC 1.14 to 1.27) for ChAdOx1 vaccine. Rates of COVID-19 hospitalization and COVID-19 death at week 26 showed HRs <0.26 (>74% vaccine effectiveness) for ChAdOx1 [Elsie MF Horne, 2022 ].

Chanda et al conducted a retrospective cohort study among admitted patients at 8 COVID-19 treatment centers across Zambia from April 2021 through March 2022, when Delta and Omicron variants were circulating. 1821 patients had COVID-19 vaccination status documented. Adjusted Vaccine Effectiveness (AVE) Against Progression to In-Hospital Mortality—Zambia: 65.2% (95% CI 33.5-83.1) for ChadOx1-S. [Chanda D, 2022 ]

Chung et al. conducted a case-control study with a test-negative design to estimate vaccine effectiveness (VE) against SARS-CoV-2 infection after the primary schedule of any combination of BNT162b2, mRNA-1273, and ChAdOx1 between January 11th and November 21st 2021 in Ontario, Canada. It included 261,360 test-positive cases (of any SARS-CoV-2 lineage) and 2,783,699 individuals as test-negative controls. VE for a homologous ChAdOx1 schedule 7-59 days after second dose was 95% (95% CI, 89% to 98%) against severe outcomes (hospitalization/death). VE 120-179 days after second dose was 97% (95% CI, 94% to 98%) against severe outcomes (hospitalization/death). [Chung H, 2022 ]

Nittayasoot et al was a test negative case-control study to examine the effectiveness of COVID-19 vaccines during January to April 2022 in Thailand. They analyzed secondary data from four main national health data bases: Co-Lab, Co-Ward, COVID-10 Death and MOPH-IC, using the national identification numbers of each individual as a unique identifier to link the same person across databases. They obtained a total of 3,059,616 records including: 1,015 cases of COVID-19 pneumonia requiring invasive ventilation from 652,854 cases with SARS-CoV-2 detection and 2,046,762 controls or non-SARS-CoV-2 detection. Vaccine Effectiveness against pneumonia requiring invasive ventilation for ChAdOx1 + ChAdOx1 schedule was 58.41% (95% CI, 39.95% to 71.19%). [Nittayasoot N, 2022 ] 

Bello-Chavolla OY et al was a retrospective cohort study conducted in Mexico, including 793,487 vaccinated and 4,792,388 unvaccinated individuals. This study estimated vaccine effectiveness (VE) against infection and hospitalization, based on data from the COVID-19 surveillance system between December 2020 and September 2021. Among 60,250 participants vaccinated with AstraZeneca, VE against infection was 80.79% (95% CI, 80.4% to 81.1%), and VE against hospitalization was 80.23% (95% CI, 79.29% to 81.1%). [Bello-Chavolla OY, 2023 ]

Bouillon K et al was a comparative cohort study conducted in France that included data from 28,611,967 vaccinated participants: 7,161,658 with the Pfizer vaccine, 856,599 with the Moderna vaccine and 3,238,575 with the AstraZeneca vaccine. The aim was to estimate the effectiveness of two doses of Pfizer, Moderna and AstraZeneca vaccines. Vaccine effectiveness against hospitalization was 91% (95% CI, 91% to 92%), 95% (95% CI, 93% to 96%) and 91% (95% CI, 89% to 94%), for the Pfizer, Moderna and AstraZeneca vaccines, respectively. [Bouillon K, 2022 ]

de Arriba Fernández et al was a retrospective cohort study conducted in Spain that included data from 110,726 vaccinated participants: 50,639 with the Pfizer vaccine, 27,914 with the Moderna vaccine, 7,551 with the Janssen vaccine and 8,065 with the AstraZeneca vaccine. The aim was to assess the risk of developing persistent COVID-19 or SARS-CoV-2 virus reinfection. Risk of reinfection or persistent COVID-19 was OR 0.09 (95% CI, 0.07-0.12), OR 0.09 (95% CI, 0.06-0.13), OR 0.15 (95% CI, 0.09-0.25) and OR 0.08 (95% CI, 0.04-0.16), for the Pfizer, Moderna, Janssen and AstraZeneca vaccines, respectively. [de Arriba Fernández A, 2023 ]

Transmission
Zaidi A et al. was a retrospective cohort study conducted in the United Kingdom that 1,779,448 index cases and 4,110,051 contacts.  Based on data from the HOSTED dataset and the National Immunisation Management System in England. Data from February to September, 2021. The adjusted hazard ratio for contracting COVID-19 among unvaccinated contacts of AstraZeneca-vaccinated individuals was 1.06 (95% CI, 1.04-1.08). [Zaidi A, 2022 ]

Efficacy and effectiveness against SARS-CoV-2 variants

Immunogenicity outcomes
Alpha (B.1.1.7)
Emary KRW et al. was a randomized clinical trial conducted in the UK (COV002 phase 2/3). The study included 8,534 participants ≥ 18 years old (4,244 vaccinated group and 4,290 control group). The study measured neutralizing antibody responses using a live virus microneutralization assay against lineage B.1.1.7 and a non-canonical lineage B.1.1.7 (Victoria). The neutralization activity of the laboratory virus by the antibodies induced by the vaccine was lower against variant B.1.1.7 than against the Victoria lineage (Geometric mean ratio 8 · 9 (95% CI: 7 · 2–11 · 0 )) [Emary KRW, 2021 ].

Suntronwong N et al. included 458 fully vaccinated participants (2 doses Sinovac=248; 2 doses Sinovac plus AstraZeneca booster=210). Data were collected 34 days after infection. The study showed that binding antibody levels in sera from patients with breakthrough infection were significantly higher than those in individuals who had received AstraZeneca as a booster dose. However, neutralizing activities against wild-type and variants including alpha, beta, and delta were similar between patients with breakthrough infections and those who received a booster vaccination with AstraZeneca. Omicron was neutralized less effectively by serum from breakthrough infection patients, with a 6.3-fold reduction compared to delta variants [Nungruthai Suntronwong, 2022 ].

Behrens et al. conducted a longitudinal monitoring of ChAd/ChAd (n = 41) and ChAd/BNT (n = 88) vaccinated individuals and the impact of a third vaccination with BNT. Whilst the majority of participants had neutralizing antibodies against the Wuhan strain in plasma before the third vaccination, neutralizing antibodies against the Alpha variant were particularly in the ChAd/ChAd group less frequent or virtually absent. In the ChAd/ChAd group, the third immunization profoundly increased neutralization of the Alpha variant, which was low after prime/boost. Whilst initial ChAd/BNT immunization had induced neutralizing antibodies at high levels against Alpha variant after ChAd/ChAd vaccination, the following decline was more than restored by the third vaccination. All BNT-boosted ChAd/BNT vaccines had efficient neutralizing activity against Alpha variant and titers were mostly above those identified after the second vaccination [Behrens GMN, 2022 ].

Beta (B.1.351)
Madhi SA et al. was a multicenter, randomized, controlled trial conducted in South Africa (COV005 phase 1/2). The study included 2,021 participants ages 18 to 65 (1,011 in the vaccine group and 1,010 in the placebo group). The study measured neutralizing antibody titers and anti-SARS-CoV-2 spike-binding antibody (bAb) levels 28 days after two standard doses (5 × 1010 viral particles). Vaccination recipients had greater resistance to variant B.1.351 than placebo recipients. They showed an antibody response after the first dose with a mean titer 131.57 (95% CI 20 to 403.72) and after the second dose a mean titer 276.61 (95% CI 123.96 a 525.46) The geometric mean titers fell from 297 against the parent virus to 85 against the RBD only mutant and 74 against the B.1.351 variant [Madhi SA, 2021 ].

Suntronwong N et al. included 458 fully vaccinated participants (2 doses Sinovac=248; 2 doses Sinovac plus AstraZeneca booster=210). Data were collected 34 days after infection. The study showed that binding antibody levels in sera from patients with breakthrough infection were significantly higher than those in individuals who had received AstraZeneca as a booster dose. However, neutralizing activities against wild-type and variants including alpha, beta, and delta were similar between patients with breakthrough infections and those who received a booster vaccination with AstraZeneca. Omicron was neutralized less effectively by serum from breakthrough infection patients, with a 6.3-fold reduction compared to delta variants [Nungruthai Suntronwong, 2022 ].

Niyomnaitham et al. was a comparative study conducted in Thailand. The study recruited 210 participants equally divided (n=30) to receive Sinovac, AstraZeneca and Pfizer-BioNTech vaccines either as a first or second dose. Participants who received only Sinovac or AstraZeneca first and second doses also received a booster dose with Pfizer-BioNTech. The study found antibody levels were highest among the groups that received Pfizer-BioNTech as a second dose, levels were similar between homologous and heterologous regimens. These levels were significantly higher compared with the groups who received AstraZeneca or Sinovac as a second dose.  The groups who were given Pfizer-BioNTech as second dose had significantly higher neutralizing titers against Delta and Beta than the groups that received AstraZeneca or Sinovac as the second dose. Neutralizing titers against the Beta variant were reduced by 2 to 5-fold compared to the Delta variant. Overall, neutralization against Omicron was low across the groups and were 28- to 229-fold lower than Delta, depending on the vaccine schedules. Neutralization against both the Delta and Omicron variants were significantly lower among groups who received Sinovac as a second dose compared to the other groups. Seropositivity rate for the Omicron variant was : 80% (45/56), 50% (30/60) and 21% (21/58) for the groups that received  Pfizer-BioNTech, AstraZeneca and: Sinovac as a second dose respectively [Suvimol Niyomnaitham, 2022 ].
 

Behrens et al. conducted a longitudinal monitoring of ChAd/ChAd (n = 41) and ChAd/BNT (n = 88) vaccinated individuals and the impact of a third vaccination with BNT. Whilst the majority of participants had neutralizing antibodies against the Wuhan strain in plasma before the third vaccination, neutralizing antibodies against Beta variant were particularly in the ChAd/ChAd group less frequent or virtually absent. In the ChAd/ChAd group, the third immunization profoundly increased neutralization of the Beta variant, which was virtually absent after prime/boost. All BNT-boosted ChAd/BNT vaccines had efficient neutralizing activity against Beta variant and titers were mostly above those identified after the second vaccination [Behrens GMN, 2022 ].

Delta (B.1.617.2)
Suntronwong N et al. included 458 fully vaccinated participants (2 doses Sinovac=248; 2 doses Sinovac plus AstraZeneca booster=210). Data were collected 34 days after infection. The study showed that binding antibody levels in sera from patients with breakthrough infection were significantly higher than those in individuals who had received AstraZeneca as a booster dose. However, neutralizing activities against wild-type and variants including alpha, beta, and delta were similar between patients with breakthrough infections and those who received a booster vaccination with AstraZeneca. Omicron was neutralized less effectively by serum from breakthrough infection patients, with a 6.3-fold reduction compared to delta variants [Nungruthai Suntronwong, 2022 ].

Niyomnaitham et al. was a comparative study conducted in Thailand. The study recruited 210 participants equally divided (n=30) to receive Sinovac, AstraZeneca and Pfizer-BioNTech vaccines either as a first or second dose. Participants who received only Sinovac or AstraZeneca first and second doses also received a booster dose with Pfizer-BioNTech. The study found antibody levels were highest among the groups that received Pfizer-BioNTech as a second dose, levels were similar between homologous and heterologous regimens. These levels were significantly higher compared with the groups who received AstraZeneca or Sinovac as a second dose.  The groups who were given Pfizer-BioNTech as second dose had significantly higher neutralizing titers against Delta and Beta than the groups that received AstraZeneca or Sinovac as the second dose. Neutralizing titers against the Beta variant were reduced by 2 to 5-fold compared to the Delta variant. Overall, neutralization against Omicron was low across the groups and were 28- to 229-fold lower than Delta, depending on the vaccine schedules. Neutralization against both the Delta and Omicron variants were significantly lower among groups who received Sinovac as a second dose compared to the other groups. Seropositivity rate for the Omicron variant was : 80% (45/56), 50% (30/60) and 21% (21/58) for the groups that received  Pfizer-BioNTech, AstraZeneca and: Sinovac as a second dose respectively [Suvimol Niyomnaitham, 2022 ].
 

Behrens et al. conducted a longitudinal monitoring of ChAd/ChAd (n = 41) and ChAd/BNT (n = 88) vaccinated individuals and the impact of a third vaccination with BNT. Whilst the majority of participants had neutralizing antibodies against the Wuhan strain in plasma before the third vaccination, neutralizing antibodies against the Delta variant were particularly in the ChAd/ChAd group less frequent or virtually absent. In the ChAd/ChAd group, the third immunization profoundly increased neutralization of the Delta variant, which was low after prime/boost. Whilst initial ChAd/BNT immunization had induced neutralizing antibodies at high levels against Delta variant, the following decline was more than restored by the third vaccination. All BNT-boosted ChAd/BNT vaccines had efficient neutralizing activity against Delta variant and titers were mostly above those identified after the second vaccination[Behrens GMN, 2022 ].
 

Buathong et al conducted an observational study to determine the levels of neutralizing antibodies against the SARS-CoV-2 ancestral strain, Delta and Omicron variants of concern (VOCs), in 125 healthcare workers who received CoronaVac as their primary vaccination and later received either a single ChAdOx1 or a combination of two consecutive boosters using either two ChAdOx1 doses or a ChAdOx1 or BNT162b2 as the primary and second boosters, respectively, or two doses of BNT162b2. Coronavac Coronavac + ChAdOx1 - 8.5% of subjects were NT positive against the Delta VOC at day 0 post-booster, with a MI of 1.9% (95% CI: 0.7–3.1%), but the percentage increased significantly (p < 0.0001) to 91.5% at week 1 post-booster with an MI of 76.8% (95% CI: 70.9–82.7%) and again to 98.1% at week 4 post-booster with an MI of 81.3% (95% CI: 76.9–85.7%). Coronavac Coronavac + ChAdOx1 ChAdOx1 - The rates of NT positivity against the Delta VOC, increased from 56% with an MI of 49% (95% CI: 35.3–62.7%) at Day 0 post second booster to 76% with an MI of 70.4% (95% CI: 58.9–81.9%, p = 0.005) at week 1 post second booster. Coronavac Coronavac + ChAdOx1 Pfizer - The rates of NT positivity against the Delta VOC, % rates increased from 62.9% with an MI of 56.4% (95% CI: 45.5–67.3%) at Day 0 to 97% with an MI of 80.9% (95% CI: 76.8–85.1%) at week 1 post second booster. [Buathong R, 2022 ]

Sira et al. (TCTR20210722003) conducted a randomized clinical trial in Thailand, with 422 adults with a mean age of 44 years old. The aim of the study was to evaluate the immunogenicity and reactogenicity of standard-dose and low-dose boosters of Oxford-AstraZeneca vaccine, after the full CoronaVac schedule in healthy adults. 206 individuals received low dose and 208 individuals standard dose. At baseline, geometric means (GM) of sVNT against delta variant and anti-S-RBD IgG were 18.1% inhibition (95% CI 16.4-20.0) and 111.5 (105, 1-118.3) BAU/ml. GM of sVNT against delta variant and anti-S-RBD IgG in DS were 95.6% inhibition (95% CI 94.3-97.0) and 1975.1 (1841.7-2118.2) BAU /ml at day 14, and 89.4% inhibition (86.4-92.4) and 938.6 (859.9-1024.4) BAU/ml at day 90, respectively. The GMRs of sVNT against delta variant and anti-S-RBD IgG in LD compared to SD were 1.00 (95% CI: 0.98-1.02) and 0.84 (0.76-0, 93) at day 14, and 0.98 (0.94-1.03) and 0.89 (0.79-1.00) at day 90, respectively. [Sira Nanthapisal, 2022 ]

Nawfal et al conducted a observational study, in Duhok City (Iraq) between April to October 2021 to investigate the COVID-19 breakthrough infection rate and safety profile of AstraZeneca vaccine. The average age of the 265 participants was 43.85 years and 169 (63.77%) were male. The participants were followed up for 104 days after receiving the second dose. After the second dose, 18 (6.71%) participants contracted the infection. The SARS-CoV-2 delta variant was responsible for all infections but no participants required hospitalization. They found a significant correlation between post-vaccination IgG levels and infection (P = 0.001; OR = 0.959; 95%CI 0.944–0.974). A history of previous COVID-19 infection was significantly associated with lower post-vaccination infection rates (P = 0.005; OR = 0.1; CI = 0.009–0.6). IgG levels were significantly higher in women than in men (p = 0.006) and were significantly higher in patients who developed side effects after vaccination than in those without side effects (P = 0.04). Significant association was found between a history of COVID-19 infection prior to vaccination and IgG levels (P = 0.001). [Hussein NR, 2022 ]

Niyomnaitham et al was a phase 2 randomized clinical trial conducted in Thailand. The study included 1,243 participants with CoronaVac primary schedule: 312 in the AstraZeneca booster (half dose) group, 307 in the AstraZeneca booster (full dose) group, 316 in the Pfizer booster (half dose) group and 308 in the Pfizer booster (half dose) group. Vaccination-induced immunogenicity to Ancestral, Delta and Omicron BA.1 strains were evaluated by assessing anti-spike (‘anti-S’), anti-nucleocapsid antibodies, pseudovirus neutralization (‘PVNT’), micro-neutralization titers, and T-cells assays. Within platforms and irrespective of dose or platform, seroconversions were greater than 97%, and greater than 90% for neutralising antibodies against pseudovirus, but similar against SARS-CoV-2 strains. Anti Spike RBD IgG Geometric means concentration (U/mL) at day 28 were 8237.0 (95% CI, 7679.3-8835.2) for the AstraZeneca half dose group, 8973.6 (95% CI, 8328.1-9669.2) for the AstraZeneca full dose group. Immunogenicity according to Pseudovirus Neutralizing Antibody Titer (PVNT50) against delta variant at day 28 were 468.3 (95% CI, 409.8-535.2) for AstraZeneca half dose group, 530.6 (95% CI, 470.4-598.5) for AstraZeneca full dose group [Suvimol Niyomnaitham, 2022 ]

Thiruvengadam R et al was a case control study conducted in India that included data from 2,766 cases of confirmed SARS-CoV-2 infection. The aim was to was to assess ChAdOx1 nCoV-19 vaccine effectiveness during the massive surge from April 1st to May 31st 2021 predominantly due to the more infectious B.1.617.2 (Delta) variant. Full vaccination prevented moderate-severe COVID-19 in 81.5% (95% CI, 9.9% to 99.0%). The effectiveness of single-dose vaccine was 46.2% (95% CI, 31.6% to 57.7%) against infection but 79.2% (95% CI, 46.1% to 94.0%) in preventing moderate-severe COVID-19. [Ramachandran Thiruvengadam, 2021 ]

Gamma (P.1)
Behrens et al. conducted a longitudinal monitoring of ChAd/ChAd (n = 41) and ChAd/BNT (n = 88) vaccinated individuals and the impact of a third vaccination with BNT. Whilst the majority of participants had neutralizing antibodies against the Wuhan strain in plasma before the third vaccination, neutralizing antibodies against Gamma variant were particularly in the ChAd/ChAd group less frequent or virtually absent. In the ChAd/ChAd group, the third immunization profoundly increased neutralization of Gamma variant, which was virtually absent after prime/boost. All BNT-boosted ChAd/BNT vaccines had efficient neutralizing activity against Gamma variant and titers were mostly above those identified after the second vaccination [Behrens GMN, 2022 ].

Omicron (B.1.1.529)
Suntronwong N et al. included 458 fully vaccinated participants (2 doses Sinovac=248; 2 doses Sinovac plus AstraZeneca booster=210). Data were collected 34 days after infection. The study showed that binding antibody levels in sera from patients with breakthrough infection were significantly higher than those in individuals who had received AstraZeneca as a booster dose. However, neutralizing activities against wild-type and variants including alpha, beta, and delta were similar between patients with breakthrough infections and those who received a booster vaccination with AstraZeneca. Omicron was neutralized less effectively by serum from breakthrough infection patients, with a 6.3-fold reduction compared to delta variants [Nungruthai Suntronwong, 2022 ].

Niyomnaitham et al. was a comparative study conducted in Thailand. The study recruited 210 participants equally divided (n=30) to receive Sinovac, AstraZeneca and Pfizer-BioNTech vaccines either as a first or second dose. Participants who received only Sinovac or AstraZeneca first and second doses also received a booster dose with Pfizer-BioNTech. The study found antibody levels were highest among the groups that received Pfizer-BioNTech as a second dose, levels were similar between homologous and heterologous regimens. These levels were significantly higher compared with the groups who received AstraZeneca or Sinovac as a second dose.  The groups who were given Pfizer-BioNTech as second dose had significantly higher neutralizing titers against Delta and Beta than the groups that received AstraZeneca or Sinovac as the second dose. Neutralizing titers against the Beta variant were reduced by 2 to 5-fold compared to the Delta variant. Overall, neutralization against Omicron was low across the groups and were 28- to 229-fold lower than Delta, depending on the vaccine schedules. Neutralization against both the Delta and Omicron variants were significantly lower among groups who received Sinovac as a second dose compared to the other groups. Seropositivity rate for the Omicron variant was : 80% (45/56), 50% (30/60) and 21% (21/58) for the groups that received  Pfizer-BioNTech, AstraZeneca and: Sinovac as a second dose respectively [Suvimol Niyomnaitham, 2022 ].
 

Behrens et al. conducted a longitudinal monitoring of ChAd/ChAd (n = 41) and ChAd/BNT (n = 88) vaccinated individuals and the impact of a third vaccination with BNT. 14 days after the third vaccination Omicron-neutralizing antibodies were present in 28/29 (97%) and 55/58 (95%) of vaccines in the ChAd/ChAd and ChAD/BNT group, respectively [Behrens GMN, 2022 ].

Buathong et al conducted an observational study to determine the levels of neutralizing antibodies against the SARS-CoV-2 ancestral strain, Delta and Omicron variants of concern (VOCs), in 125 healthcare workers who received CoronaVac as their primary vaccination and later received either a single ChAdOx1 or a combination of two consecutive boosters using either two ChAdOx1 doses or a ChAdOx1 or BNT162b2 as the primary and second boosters, respectively, or two doses of BNT162b2. Coronavac Coronavac + ChAdOx1 
- NT positivity rate and titers against Omicron VOC at day 0 post-booster was 0% with an MI of 1.2% (95% CI: 1.0–1.3%), which increased significantly (p < 0.0001) at week 1 to 18.6% with an MI of 21.8% (95% CI: 13.3–30.3%) and to 21.2% with an MI of 22.2% (95% CI: 12.9–31.5%) at week 4. Coronavac Coronavac + ChAdOx1 ChAdOx1.
- 4% were NT positive against the Omicron VOCs at Day 0 post second booster, with an MI of 7.4% (95% CI: 0–16.0%). NT positive rates increased to 16% at week 1 with an MI of 15.1% (95% CI: 2.6–27.7%). Coronavac Coronavac + ChAdOx1 Pfizer
-  NT positive against the Omicron VOCs went from 11.4% with an MI of 10.0% 95% CI: 0.2–19.8%) at Day 0 to 71.4% with an MI of 58.7% (95% CI: 45.4–72.0%) at week 1.
- Subjects in the 2CoronaVac-ChAdOx1/BNT162b2 and 2CoronaVac-2BNT162b2 groups had significant higher MI against Omicron VOC than the CoronaVac(x2)/ChAdOx1(x2) group at week 1 post second booster (p = 0.034 and p < 0.0001, respectively). [Buathong R, 2022 ]

Niyomnaitham et al was a phase 2 randomized clinical trial conducted in Thailand. The study included 1243 participants with CoronaVac primary schedule: 312 in the AstraZeneca booster (half dose) group, 307 in the AstraZeneca booster (full dose) group, 316 in the Pfizer booster (half dose) group and 308 in the Pfizer booster (half dose) group. Vaccination-induced immunogenicity to Ancestral, Delta and Omicron BA.1 strains were evaluated by assessing anti-spike (‘anti-S’), anti-nucleocapsid antibodies, pseudovirus neutralization (‘PVNT’), micro-neutralization titers, and T-cells assays. Within platforms and irrespective of dose or platform, seroconversions were greater than 97%, and greater than 90% for neutralising antibodies against pseudovirus, but similar against SARS-CoV-2 strains. Anti Spike RBD IgG Geometric means concentration (U/mL) at day 28 were 8237.0 (95% CI, 7679.3-8835.2) for the AstraZeneca half dose group and 8973.6 (95% CI, 8328.1-9669.2) for the AstraZeneca full dose group. Pseudovirus Neutralizing Antibody Titer (PVNT50) against Omicron (GMT) were 118.9 (95% CI, 97.3-145.2) for the AstraZeneca full dose group. [Suvimol Niyomnaitham, 2022 ]

Nantanee R et al was a randomized clinical trial conducted in Thailand that included 100 adults with a median age of 59.3 years with two doses of AZD1222: 50 received a half-dose BNT162b2 dooster (15 µg/dose) and 50 received a standard-dose BNT162b2 booster (30 µg/dose). The immunogenicity was evaluated by a surrogate virus neutralization test (sVNT) against Omicron variants and anti-spike-receptor-bindin-domain IgG (anti-S-RBD IgG). At day 14, the geometric means (GM) of sVNT against the omicron variant were 74.4% (95% CI, 68.8% to 80.5%)  inhibition post-half-dose BNT162b2 boosters and 67.3% (95% CI, 57.9% to 78.1%) inhibition poststandard-dose boosters. At day 14, anti-S-RBD IgGs were comparably boosted with GMs of 2329.8 BAU/mL (95% CI, 2109.3–2573.4) post half-dose BNT162b2 boosters and 2574.7 BAU/mL (2262.5–2929.9) post standard-dose boosters. At 14 days after boosters, T cell responses were significantly enhanced to 406 SFU/106 PBMCs (IQR 206–718, p < 0.001) in the half-dose and 402 SFU/106 PBMCs (148–678, p < 0.001) in the full-dose groups, without a difference between the dosage groups (p= 0.80). [Nantanee R, 2022 ]

Prasert Assantachai et al was a randomized controlled trial conducted in Thailand that included 210 participants aged ≥65 years with AstraZeneca primary schedule: 35 received intradermal mRNA-1273, 35 intramuscular mRNA-1273, 70 intradermal BNT162b2, and 70 intramuscular BNT162b2. The group that reported the highest geometric mean titers against the ancestral strain and Omicron BA.1 variant was the group that received the intramuscular mRNA-1273 vaccine (GMT 1,717.9 and 617), followed by the group that received the intradermal mRNA-1273 vaccine (GMT 1,212 and 318), the group that received the intramuscular BNT162b2 vaccine (GMT 713 and 230 and the group that received the intradermal BNT162b2 vaccine (GMT 587 and 148), respectively. [Prasert Assantachai, 2022 ]

Efficacy outcomes
Alpha (B.1.1.7)
Emary KRW et al. was a randomized clinical trial conducted in the United Kingdom (COV002 phase 2/3). The study included 8,534 paraticipants ≥18 years (4,244 vaccinated group and 4,290 control group). The study measured neutralizing antibody responses using a live virus microneutralization assay against lineage B.1.1.7 and a non-canonical lineage B.1.1.7 (Victoria). The results measured at least after 14 days after applying the second dose showed an efficacy of 70.4% (95% CI: 43.6 to 84.5) against primary symptomatic COVID-19 and an efficacy of 28.9% (95% CI −77.1 to 71.4) against asymptomatic or unknown infection [Emary KRW, 2021 ].

Beta (B.1.351)
Madhi SA et al. was a multicenter, randomized and controlled trial conducted in South Africa (COV005 phase 1/2). The study included 2,021 participants ages 18 to 65 (1,011 in the vaccine group and 1,010 in the placebo group). The study measured neutralizing antibody titers and anti-SARS-CoV-2 spike-binding antibody (bAb) levels 28 days after two standard doses (5 × 1010 viral particles). The results measured at least after 14 days after applying the second dose showed that among the 42 participants with COVID-19, 39 cases (92.9%) were caused by variant B.1.351, obtaining an efficacy of the vaccine of 10.4% (95% CI, −76.8 to 54.8) IR 73.1 (94,881) [Madhi SA, 2021 ].

Effectiveness outcomes
Alpha (B.1.1.7)
López Bernal J. et al. carried out a case-control study (test-negative) in England, which included 19,109 COVID-19 cases (the delta variant was detected in 4,272 samples). The study sought to estimate the effectiveness of vaccination against symptomatic disease caused by the delta variant or the predominant strain (B.1.1.7, or alpha variant) during the period when the delta variant began to circulate. The results showed an effectiveness of vaccination against symptomatic disease of 67.0% (95% CI, 61.3 to 71.8) [Lopez Bernal J, 2021 ].

Gamma (P.1)
Costa Clemens SA was a case-control study (test-negative design) conducted in Brazil. The study included data from 7,958 participants: 2,109 cases and 5,849 controls. Odds of being vaccinated among cases vs controls were calculated to estimate vaccine effectiveness. All individuals aged 18–60 who received a PCR test after May 16th and were unvaccinated prior to this date were included in the analysis. Vaccine effectiveness against any COVID-19 illness was 74.5% (95% CI, 23.7% to 51.5%). Vaccine effectiveness against any COVID-19 illness was 76.0% (95% CI, 49.5% to 87.8%) against Delta, and 81.3% (95% CI, −298.0% to 97.9%) against Gamma [Costa Clemens SA, 2022 ].

Delta (B.1.617.2)
Andrews et al conducted a test-negative case-control design to estimate vaccine effectiveness (VE) against symptomatic COVID-19 and related hospitalization and death in England. 6,056,673 SARS-CoV-2 PCR test with a sample within 10 days after symptom onset were successfully linked to the NIMS database for vaccination status. VE against symptomatic Covid-19 with the delta variant peaked in the early weeks after receipt of the second dose and then decreased by 20 weeks to 44.3% (95% CI 43.2 to 45.4) with the ChAdOx1-S vaccine. Waning of VE was greater in persons 65 years of age or older than in those 40 to 64 years of age. At 20 weeks or more after vaccination, VE decreased less against both hospitalization to 80.0% (95% CI 76.8 to 82.7) with the ChAdOx1-S vaccine, and death to 84.8% (95% CI 76.2 to 90.3). Greater waning in VE against hospitalization was observed in persons 65 years of age or older in a clinically extremely vulnerable group and in persons 40 to 64 years of age with underlying medical conditions than in healthy adults. Results from the sensitivity analysis with the use of only hospitalizations that had been coded as respiratory admissions were similar to those of the primary analysis, showing VE of 82.6% (95% CI, 79.1 to 85.4) for the ChAdOx1-S vaccine against the delta variant at 20 weeks or more after vaccination. [Andrews N, 2022 ].

Costa Clemens SA was a case-control study (test-negative design) conducted in Brazil. The study included data from 7,958 participants: 2,109 cases and 5,849 controls. Odds of being vaccinated among cases vs controls were calculated to estimate vaccine effectiveness. All individuals aged 18–60 who received a PCR test after May 16th and were unvaccinated prior to this date were included in the analysis. Vaccine effectiveness against any COVID-19 illness was 74.5% (95% CI, 23.7% to 51.5%). Vaccine effectiveness against any COVID-19 illness was 76.0% (95% CI, 49.5% to 87.8%) against Delta, and 81.3% (95% CI, −298.0% to 97.9%) against Gamma [Costa Clemens SA, 2022 ].

Omicron (B.1.1.529)
Soledad Gonzalez et al was a test-negative, case-control study conducted in Argentina. The study included 422,144 individuals ≥50 years: 221,933 with a positive test and 200,211 with a negative test. The aim of the study was to estimate the protection against laboratory-confirmed SARS-CoV-2 infection, hospitalizations, and death after homologous or heterologous third-dose (booster) in individuals with primary vaccination schemes with rAd26-rAd5, ChAdOx1nCoV-19, BBIBP-CorV or heterologous combinations, during the period of Omicron BA.1 predominance. The odds ratio of booster dose against confirmed SARS-CoV-2 infections were 0.64 (95% CI 0.61 to 0.67) for ChAdOx1 primary schedule and mRNA booster (≤ 60 days) and 0.97 (95% CI 0.93 to 1.01) for ChAdOx1 primary schedule and vectored booster (≤ 60 days). The odds ratio of booster dose against hospitalizations were 0.14 (95% CI 0.03 to 0.77) for ChAdOx1 primary schedule and mRNA booster (≤ 60 days) and 0.28 (95% CI 0.22 to 0.37) for ChAdOx1 primary schedule and vectored booster (≤ 60 days) [Soledad Gonzalez, 2022 ].

Monge S et al,  was a nationwide Cohort study conducted in Spain. This study included 7,036,433 participants older than 40 years: 3,111,159 in the booster group and 3,111,159 in the no booster group. The aim of this study was to  estimate the effectiveness of mRNA-based vaccine boosters against infection during the period of the predominance of the omicron variant in Spain. Vaccine effectiveness of AstraZeneca primary schedule and ARNm booster was 58.6% (95% CI 55.5–61.6) (Moderna or Pfizer booster dose). [Monge S, 2022 ]

Stowe et al. conducted a test-negative case-control study in UK, to estimate Vaccine Effectiveness (VE) against hospitalisation with the Omicron and Delta variants using PCR testing linked hospital records (Emergency Care Data Set; ECDS). The total number of tests in the study period was 409,985 of which 115,720 were cases and 294,265 controls.

VE against hospital admissions from ECDS within 14 days of the test date by the Omicron variant in symptomatic individuals 18 to 64 years of age:

Primary Schedule ChAd0x1-S 
After 14-174 days: 38.5% (95% CI 20.2 to 52.6)
After 175+ days: 31.6% (95% CI 22.3 to 39.7)

VE against hospital admissions from ECDS within 14 days of the test date by the Omicron variant in symptomatic individuals 65 years of age and older:

Primary Schedule ChAd0x1-S 
After 14-174 days: 55.3% (95% CI -20.1 to 83.3)
After 175+ days: 56.3% (95% CI 39.6 to 68.3)

[Stowe J, 2022 ]

Agrawal et al was a retrospective cohort study conducted in the United Kingdom that included 16,208,600 participants, 7,589,080 who received a Pfizer-BioNTech primary schedule and 8,619,520 participants who received a ChAdOx1 primary schedule. The study was based on data from the Oxford-Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) database, Vaccine Management System, Early Pandemic Evaluation and Enhanced Surveillance of COVID-19 (EAVE II), and Secure Anonymised Information Linkage Databank platform, during the Omicron period between December 2021 and February 2022. The adjusted Rate Ratio (aRR) 3-5 weeks after booster vaccination against hospitalization was 0.43 (95% CI, 0.37-0.50) for the ChAdOx1 primary schedule and Pfizer-BioNTech booster group and 0.41 (95% CI, 0.33-0.50) for the ChAdOx1 primary schedule and Moderna booster group. [Agrawal U, 2022 ]

Baum U et al was a cohort study conducted in Finland including 896,220 participants aged 70 years and older. The study was based on a nationwide register-based cohort, starting on December 27th 2020 and ending on March 31st 2022. Vaccine effectiveness against hospital admission during the Omicron period was 100% for the primary schedule, 94% (95% CI, 81% to 98%) for the AstraZeneca schedule + Moderna booster, and 99% (95% CI, 91% to 100%) for the AstraZeneca schedule + Pfizer booster. [Baum U, 2022 ]

Intawong K et al was a test-negative case-control study conducted in Thailand. The study included 36,170 participants, 14,682 cases and 21,488 controls. Based on data from the Epid-CM platform, including participants aged 18 years or older, between October 2021 and April 2022.The aim was to assess the effectiveness of different booster doses. The adjusted vaccine effectiveness against infection during the omicron period after the third dose was 31% (95% CI, 15% to 44%) for the Pfizer-BioNTech booster, 26% (95% CI, 8% to 40%) for the AstraZeneca booster and 31% (95% CI, 13% to 45%) for the Moderna booster. The adjusted VE after the fourth dose was 71% (95% CI, 60% to 79%) for the Pfizer-BioNTech booster, 73% (95% CI, 48% to 89%) for the AstraZeneca booster, and 71% (95% CI, 59% to 79%) for the Moderna booster. [Intawong K, 2022 ]

Kirsebom FCM et al was a cohort study conducted in England including participants who received an AstraZeneca primary schedule: 43,171 received an AstraZeneca booster and 13,038,908 received a Pfizer booster dose. Based on data from the National Immunisation Management System on all adults aged 18 years and older in England who had received a ChAdOx1-S primary schedule followed by a ChAdOx1-S or BNT162b2 booster, until March 14th 2022. The vaccine effectiveness against symptomatic infection was 51.7% (95% CI, 38.9% to 61.8%) after the AstraZeneca booster, the VE against hospitalization was 61% (95% CI, 49.8% to 69.7%) for the primary schedule, and 82.3% (95% CI, 64.2% to 91.3%) for the AstraZeneca booster. [Kirsebom FCM, 2022 ]

Huilberts A et al. was a prospective cohort study conducted in Netherlands that included 27,646 participants from the Omicron period: 3,802 received primary vaccination schedule, 23,352 received first booster dose and 492 were unvaccinated. The primary outcome was contracting SARS-CoV-2 during July 12th 2021 and June 6th 2022. Vaccine effectiveness against contracting COVID-19 during  Omicron BA.1-2 period was 58.4% (95% CI, 48.8% to 66.1%) and 67.6% (95% CI, 60.1% to 73.7%) for AstraZeneca primary schedule with Pfizer booster and AstraZeneca primary schedule and Moderna booster, respectively. [Anne J. Huiberts, 2023 ]

Cerqueira-Silva T et al was a case-control study, test-negative design, conducted in Brazil and Scotland. The study included 5,832,210 participants: 5,276,385 from Brazil and 555,825 from Scotland. This study assessed vaccine effectiveness of a mRNA booster after AstraZeneca or Pfizer primary schedule during the period of Omicron dominance. Vaccine effectiveness against severe outcomes was 93.5% (95% CI, 93% to 94%), 94.4% (95% CI, 87.7% to 97.5%) and 92.7% (95% CI, 91% to 94%) with AstraZeneca primary schedule and Pfizer booster, AstraZeneca primary schedule and Moderna booster and Pfizer homologous booster, respectively. [Cerqueira-Silva T, 2023 ]

Vaccine efficacy and effectiveness for booster dose

Immunogenicity outcomes
COV002 is a phase 2/3 trial sponsored by the University of Oxford and conducted in the United Kingdom that started in 28 May 2020. The study included 552 healthy adults Older than 18 years: 206  Low-dose group (2.2 × 1010 viral particles); 208 Standard-dose group (3.5–6.5 × 1010 virus particles). Anti-SARS-CoV-2-spike binding antibody (bAb) levels 28 days following the booster dose was measured. Result showed neutralizing antibody titres after a boost dose were similar across all age groups. [Ramasamy, Maheshi N, 2021 ]

COV001 phase 1/2, randomized trial conducted in the United Kingdom. The study included 52 healthy adults 18 to 55 years: 20 Low-dose group (2.2 × 1010 viral particles); 32 Standard-dose group (5 × 1010 viral particles). Result showed total antigen-specific T cell responses peaked 28 days after a seconde 5 × 1010 viral particles dose of ChAdOx1 nCoV-19. There was no significant difference in the magnitude of spike-specific T cell responses at 28 days following boost vaccination between the 28- or 56 days interval groups. [Barrett JR, 2020 ]

Bruminhent J was a randomized trial conducted in Thailand, the study included 85 kidney transplant recipients, amongst which 43 were mRNA vaccine recipients and 42 were AstraZeneca recipients.  At 2 weeks post-vaccination, anti-RBD (IgG) levels were measured, results showed titers of 28.3 ( 2.4 to 87.1)(Median, IQR). [Bruminhent J, 2022 ]

The MnM study was a phase 4 randomized trial conducted in India including 404 participants, 200 received an AstraZeneca primary schedule and 204 received a Covaxin primary schedule. The participants were randomized to receive an AstraZeneca or Covaxin booster dose. The study assessed the imunnogenicity 28 days after the booster dose. The geometric mean concentration of anti SARS-CoV-2 Spike IgG was 97,445.09 (95% CI, 82,626.97-114,920.7) with a geometric mean fold ratio of 5.06 (95% CI, 3.88-6.61) [Rose W, 2023 ].

Effectiveness outcomes
Kirsebom FCM et al was a cohort study conducted in England including participants who received an AstraZeneca primary schedule: 43,171 received an AstraZeneca booster and 13,038,908 received a Pfizer booster dose. Based on data from the National Immunisation Management System on all adults aged 18 years and older in England who had received a ChAdOx1-S primary schedule followed by a ChAdOx1-S or BNT162b2 booster, until March 14, 2022. The vaccine effectiveness against symptomatic infection was 51.7% (95% CI, 38.9% to 61.8%) after the AstraZeneca booster, the VE against hospitalization was 82.3% (95% CI, 64.2% to 91.3%) for the AstraZeneca booster. [Kirsebom FCM, 2022 ]

dos Santos et al was a comparative cohort study conducted in Brazil that included data from 108,625,066 booster-vaccinated participants. The aim was to assess the effectiveness of the primary series of COVID-19 vaccination and booster shots in protecting against severe cases. Vaccine effectiveness (VE) against severe outcomes were 66.5% (95% CI, 62.8% to 70.0%), 73.0% (95% CI, 67.8% to 77.6%), 39.9% (95% CI, −15.3% to 77.2%) and 82.4% (95% CI, 63.2% to 93.7%), for the AstraZeneca primary schedule and Pfizer booster, the CoronaVac primary schedule and Pfizer booster, the Pfizer primary schedule and AstraZeneca booster and the Janssen primary schedule and Pfizer booster, respectively. VE against severe outcomes were 81.1% (95% CI , 80.3% to 81.9%), 84.7% (95% CI, 83.7% to 85.5%), 70.7% (95% CI, 66.9% to 74.2%) and 90.3% (95% CI, 89.5% to 91.0%), for AstraZeneca, CoronaVac, Janssen and Pfizer, respectively. [Santos CVBD, 2023 ]

Vaccine efficacy and effectiveness for heterologous schedule

Immunogenicity outcomes
Com-COV was a multi-center single-blind phase II randomized parallel study. The trial included 830 healthy adults 50 years and older: 115 Pfizer/ AstraZeneca vaccine; 114 ​​AstraZeneca/ Pfizer vaccine group; 234 Control group (Two doses of AstraZeneca or Pfizer COVID-19 vaccine). Immunogenicity was measured by anti-SARS-CoV-2-spike binding antibody (bAb) levels 28 days and 6 months following the second dose. Results showed SARS-CoV-2 anti-spike IgG concentrations of both heterologous schedules were higher versus homologous vaccine schedule after 28 days, all groups showed a marked decline in IgG levels after 6 months. [Liu X, 2021 ]
    
Com-Cov was an open-label randomized controlled trial that included 664 participants. Participants were divided into two groups: one group received the primary schedule with a 14-week interval between doses (83 participants received ChAd-ChAd schedule, 83 ChAd-BNT schedule, 84 BNT-BNT schedule and 83 BNT-ChAd schedule), and the other group received the primary schedule with a 12-week interval between doses (89 received ChAd-ChAd schedule, 77 ChAd-BNT schedule, 81 BN-BNT schedule and 78 BNT-ChAd schedule). The aim of the study was to assess the effect of priming interval on humoral response 28 days and 6 months post-second dose. Anti-spike IgG (ELU/mL) at 28 days after second dose for the 4-week interval study groups were 1,444 (95% CI, 1,205−1,732), 12,979 (95% CI, 11,217−15,018), 14,349 (95% CI, 12,470−16,511) and  7,530 (95% CI, 6,811−8,325) for ChAd-ChAd schedule, ChAd-BNT schedule, BNT-BNT schedule and BNT-ChAd schedule, respectively. Anti-spike IgG (ELU/mL) at 28 days after second dose for the 12-week interval study groups were 2,622 (95% CI, 2,152−3,195), 13,465 (95% CI, 11,391−15,917), 19,011 (95% CI, 16,468−21,947) and  10,642 (95% CI, 8,936−12,673) for ChAd-ChAd schedule, ChAd-BNT schedule, BNT-BNT schedule and BNT-ChAd schedule, respectively. [Shaw RH, 2022 ].
    
CombiVacS was a phase 2, open-label, randomized, controlled trial. The trial included 676 adults aged 18-60 years: 450 Intervention group (ChAd/BNT ); 226 Control group. Immunogenicity was measured 14 days following the second dose. BNT162b2 given as a second dose in individuals prime vaccinated with ChAdOx1-S induced a robust immune response  [Borobia AM, 2021 ].

ComFluCOV was a phase 4, randomized controlled trial. The trial included 679 volunteers aged ≥ 18 years: 129 AstraZeneca /QIVc group; 146 AstraZeneca /aTIV group; 128 AstraZeneca /QIVr group; 139 Pfizer /QIVc group; 79 Pfizer/ aTIV group; 58 Pfizer/ QIVr group. Immunogenicity was measured 21 days after receiving Astrazeneca. Results showed concomitant vaccination with AstraZeneca COVID-19 vaccine plus an age-appropriate influenza vaccine raises preserves antibody responses [Lazarus R, 2021 ].
        
Niyomnaitham S et al was a randomized clinical trial conducted in Thailand. This study evaluated the immunogenicity and reactogenicity of heterologous COVID-19 primary schedules in healthy adults, as well as booster response to BNT162b2 following heterologous CoronaVac and ChAdOx1 nCoV-19 regimens. 210 adults participated in the study: 30 with CoronaVac/AstraZeneca primary schedule, 30 with CoronaVac/Pfizer, 30 with AstraZeneca/CoronaVac, 30 with  AstraZeneca/Pfizer, 30 with Pfizer/CoronaVac, 30 with Pfizer/AstraZeneca and 30 with Pfizer/Pfizer. At 2 weeks after the second dose, the anti-RBD IgG levels were 2132.7 BAU/mL (95%CI 1696.1 to 2,681.7) for AstraZeneca/Pfizer schedule, 851.4 BAU/mL (95% CI 649.5 to 1116.1) for CoronaVac/AstraZeneca schedule, 1201.2 BAU/mL (95%CI 947.9 to 1522.1) for Pfizer/AstraZeneca schedule and 137.04 BAU/mL (95% CI 103.6 to 186.4) for AstraZeneca/CoronaVac schedule [Niyomnaitham S, 2022 ]

Fernández-Ciriza et al. conducted a prospective study enrolling 709 healthcare workers receiving two doses of mRNA-1273, BNT162b2, ChAdOx1, ChAdOx1/BNT162b2 or ChAdOx1 single dose to compare humoral and cellular immunogenicity across 9 months. The mean age was 44 ± 11 years old and 85.9% were women. After 21 and 90 days after the second dose, mRNA-1273/mRNA-1273 and ChAdOx1/BNT162b2 were the most highly immunogenic schedules among COVID-19 naïve individuals, and ChAdOx1/ChAdOx1 yielded the lowest level of Anti-S-RBD up to 6 months after the second dose. The half-life of anti-S-RBS antibody concentration in those participants who received ChAdOx1/BNT162b2 was 36 days. For those who received ChAdOx1/BNT162b2 combinations, cellular immunity was found in 92.3%, after 6-months, of COVID-19 naïve participants. All tests to assess cellular immunity among individuals with a previous infection were reactive. [Fernández-Ciriza L, 2022 ]

Alfouzan W et al was a cross sectional study conducted in Kuwai that included data from 19,363 participants in the pre-vaccination period and 4,973 participants in the post-vaccination period. The aim was to assess pre-vaccination and post-vaccination seroprevalences of anti-SARS-CoV-2 antibodies. SARS-CoV-2 seroprevalence in the post-vaccination period in general population with AstraZeneca/Pfizer primary schedule was 95.86% (95% CI, 76.39% to 99.11%). [Alfouzan W, 2023 ]

Effectiveness outcomes
Chung et al. conducted a case-control study with a test-negative design to estimate vaccine effectiveness (VE) against SARS-CoV-2 infection after the primary schedule of any combination of BNT162b2, mRNA-1273, and ChAdOx1 between January 11th and November 21st 2021 in Ontario, Canada. It included 261,360 test-positive cases (of any SARS-CoV-2 lineage) and 2,783,699 individuals as test-negative controls. VE for ChAdOx1/mRNA-1273 7-59 days after second dose was 91% (95% CI, 89% to 93%) against any infection, 96% (95% CI, 93% to 97%) against symptomatic infection and 99% (95% CI, 95% to 100) against severe outcomes. VE for ChAdOx1/mRNA-1273 60-119 days after second dose was 87% (95% CI, 85% to 89%) against any infection, 93% (95% CI, 91% to 94%) against symptomatic infection, and 99% (95% CI, 98% to 100%) against severe outcomes. VE for ChAdOx1/BNT162B2 7-59 days after second dose was 89% (95% CI, 87% to 91%) against any infection, 95% (95% CI, 92% to 97%) against symptomatic infection, and 98% (95% CI, 94% to 99%) against severe outcomes (hospitalization/death). VE for ChAdOx1/BNT162B2 120-179 days after second dose was 77% (95% CI, 73% to 80%) against any infection, 89% (95% CI, 86% to 91%) against symptomatic infection, and 98% (95% CI, 96% to 99%) against severe outcomes (hospitalization/death). [Chung H, 2022 ]

Monge S et al. was a retrospective cohort study conducted in Spain that included 59 laboratory-confirmed cases of SARS-CoV-2 participants vaccinated with AstraZeneca+mRNA heterologous schedule and 7,712 controls. The study was based on data from RENAVE and REGVACU including participants aged 50-59 years old during August 2021. The vaccine effectiveness of AstraZeneca+mRNA schedule against the confirmed infection was 90% (95% CI, 88% to 93%), against symptomatic infection was 86% (95% CI, 80% to 90%), and against hospitalization was 98% (95% CI, 88% to 100%). [Monge, Susana, 2022 ]

Nittayasoot et al was a test negative case-control study to examine the effectiveness of COVID-19 vaccines during January to April 2022 in Thailand. They analyzed secondary data from four main national health data bases: Co-Lab, Co-Ward, COVID-10 Death and MOPH-IC, using the national identification numbers of each individual as a unique identifier to link the same person across databases. They obtained a total of 3,059,616 records including: 1,015 cases of COVID-19 pneumonia requiring invasive ventilation from 652,854 cases with SARS-CoV-2 detection and 2,046,762 controls or non-SARS-CoV-2 detection. Vaccine Effectiveness against pneumonia requiring invasive ventilation for schedule: Coronavac + ChAdOx1 was 71.70% (95% CI, 63.22% to 78.22%) and ChAdOx1 + Pfizer was 83.18% (95% CI, 62.37% to 92.48%). [Nittayasoot N, 2022 ] 

Vaccine efficacy and effectiveness for heterologous booster schedule

Immunogenicity outcomes
Tobudic S et al. was a clinical trial that evaluated the efficacy and safety of a booster dose in patients in whom serconversion did not occur after the second dose. The additional booster dose was delivered with the AstraZeneca or mRNA vaccines against COVID-19. Efficacy was measured by the difference in the SARS-CoV-2 antibody seroconversion rate between patients vaccinated with the AstraZeneca vaccine (heterologous) and the mRNA vaccines (homologous) at the fourth week. The results demonstrated that seroconversion rates at week four were comparable between patients who received the AstraZenaca vaccine (6/27 patients, 22%) versus the mRNA vaccines (9/28, 32%) (p = 0, 6). Overall, 27% of the patients seroconverted; furthermore, no serious adverse events related to immunization were observed [Michael Bonelli, 2021 ].

COV-BOOST et al. was a clinical trial that evaluated the immunogenicity of seven different COVID-19 vaccines as a third dose after two doses of ChAdOx1 nCov-19 (Oxford–AstraZeneca; hereafter referred to as ChAd) or BNT162b2 (Pfizer–BioNtech, hearafter referred to as BNT). Efficacy was measured by neutralizing antibody titers at 28 days post-boost dose. The results demonstrated that all study vaccines boosted antibody and neutralising responses after AstraZeneca/AstraZeneca initial course and all except one after Pfizer/Pfizer, with no safety concerns [Munro, Alasdair P S, 2021 ].
        
Andrews N et al. was a case-control study (Test-negative) conducted in United Kingdom. The study included 893,845 eligible tests in those aged 18 years and over. The objective was to estimate the effectiveness of the Pfizer and Moderna booster vaccines against symptomatic disease, hospitalization, and death in adults in England. The study results showed that the booster dose was associated with an absolute vaccine efficacy from 14-34 days after a Pfizer booster of 94.4% (95% CI 94.1 to 94.7) following either an AstraZeneca or Pfizer primary scheme in individuals 50 years and older. With a Moderna booster, absolute vaccine effectiveness was 97.0 (95% CI 96.0 to 97.8) after an AstraZeneca primary scheme and 94.8% (95% CI 92.7 to 96.3%) Pfizer primary scheme. [Andrews N, 2022 ].

Fadlyana et al was a randomized controlled trial conducted in Indonesia that included 949 participants primed with CoronaVac that received one booster dose: 193 half-dose ChAdOx1, 192 full-dose ChAdOx1, 190 half-dose BNT162b2, 193 full-dose BNT162b2 and 192 full-dose parameter CoronaVac. The primary outcome was to evaluate the seroconversion rate and geometric mean titres (GMTs) of IgG anti S-RBD 28 days after the booster in the per-protocol population. Seroconversion rates were highest for BNT162b2 (97.8% and 92.0% for full and half-dose), followed by ChAdOx1-S (87.9% and 81.5% for full and half dose) and CoronaVac (41.3% to 66.3%). For participants primed within 6–9 months before booster, GMT values 28 days post-booster were highest for BNT162b2 (19999.84 and 17017.62 for full and half-dose), followed by ChAdOx1-S (11258 and 7853.04 for full and half-dose) and CoronaVac (1440.55). [Fadlyana E, 2023 ]

Chuang et al was a randomized trial conducted in China that evaluated the safety, reactogenicity, and immunogenicity of heterologous booster vaccination in health care workers (HCW) who had received two doses of AstraZeneca. The study enrolled 340 participants, 83 allocated to receive a Pfizer booster, 85 received a Moderna booster, 85 received half dose Moderna booster and 85 received a MVC-COV1901 booster. Neutralizing antibody titers and Anti-SARS-CoV-2-spike binding antibody levels 28 days following after booster dose vaccination were assessed. Compared with pre-boost, all study vaccines elicited significantly higher anti-Spike IgG at 28 days post-boost. The fold-rise ranged from 8.4 in MCVCOV1901 group to 63.2 in mRNA1273 group. [Chuang CH, 2022 ]

Behrens et al. conducted a longitudinal monitoring of ChAd/ChAd (n = 41) and ChAd/BNT (n = 88) vaccinated individuals and the impact of a third vaccination with BNT. Anti-SARS-CoV-2 spike IgG (anti-S IgG) levels were significantly higher in the ChAd/BNT group short after prime-boost vaccination when compared to the ChAd/ChAd group but declined significantly over time in both groups, with lower anti-S IgG after homologous vaccination prior to the third immunization. Following a third immunization, they found greatly increased anti-S IgG responses in both groups. Additional immunization of the homologous ChAd/ChAd immunized group led to a significant 46.9 -fold increase in anti-S IgG (p<0.0001). In both groups, anti-S IgG levels were considerably higher when compared to the situation observed 14 days after the second vaccination (the third vaccination diminished previous differences between the heterologous ChAd/BNT and homologous ChAd/ChAd prime-boost vaccination groups, since anti-S IgG were comparable in both groups after the third vaccination). The numbers of spike-specific memory B cells generated after prime-boost vaccination gradually increased during the following months with no significant difference between the ChAd/ChAd and the ChAd/BNT group, in line with increased amounts of spike-specific antibodies, highlighting the impact of the third vaccination for better protection from SARS-CoV-2 infection. 2 weeks after a third immunization with BNT, spike-specific memory B cells were significantly higher in the ChAd/ChAd as compared to the ChAd/BNT prime-boost group. The frequencies of spike-specific CD4+ and CD8+ T cells in blood samples collected after the second vaccination were significantly higher in the ChAd/BNT than in the ChAd/ChAd group. Both cell populations declined over time after heterologous immunization, while they remained at low levels after homologous vaccination. Whilst spike-specific CD4+ T cells declined to frequencies similar to individuals after homologous ChAd/ChAd vaccination, spike-specific CD8+ T cells remained above the frequencies of the heterologous vaccinated group. A third immunization with BNT in the ChAd/ChAd group significantly raised numbers of spike-specific CD4+ T cells above levels observed after the second vaccination. A third vaccination with with BNT did result in an expansion of spike-specific CD8+ T cells above levels observed after second vaccination in ChAd/ChAd vaccinated individuals Like for spike-specific CD4+ T cells, raised numbers in spike-specific IFN-γ-producing T cells in the ChAd/ChAd was confirmed by cytokine measurement in supernatants after SARS-CoV-2 spike peptide stimulation [Behrens GMN, 2022 ].

Niyomnaitham S et al was a randomized clinical trial conducted in Thailand. This study evaluated the immunogenicity and reactogenicity of heterologous COVID-19 primary schedules in healthy adults, as well as booster response to BNT162b2 following heterologous CoronaVac and ChAdOx1 nCoV-19 regimens. 210 adults participated in the study: 30 with CoronaVac/AstraZeneca primary schedule, 30 with CoronaVac/Pfizer, 30 with AstraZeneca/CoronaVac, 30 with  AstraZeneca/Pfizer, 30 with Pfizer/CoronaVac, 30 with Pfizer/AstraZeneca and 30 with Pfizer/Pfizer. At two weeks after the booster dose with Pfizer, the anti-RBD IgG levels against the ancestral strain were 2518.8 BAU/mL ( 95%CI 1960.4 to 3236.4) for participants primed with CoronaVac/AstraZeneca and 2610.6 BAU/mL (95%CI 2037.7–3344.5) for participants primed with AstraZeneca/CoronaVac. GMR were 2.5 and 18.9 for CoronaVac/AstraZeneca/Pfizer  and AstraZeneca/CoronaVac/Pfizer groups, respectively. [Niyomnaitham S, 2022 ]

Buathong et al conducted an observational study to determine the levels of neutralizing antibodies against the SARS-CoV-2 ancestral strain, Delta and Omicron variants of concern (VOCs), in 125 healthcare workers who received CoronaVac as their primary vaccination and later received either a single ChAdOx1 or a combination of two consecutive boosters using either two ChAdOx1 doses or a ChAdOx1 or BNT162b2 as the primary and second boosters, respectively, or two doses of BNT162b2. Coronavac Coronavac + ChAdOx1
- When measured against the SARS-CoV-2 ancestral strain, 13.6% of subjects were NT (Neutralizing Antibodys) positive at the time of booster dose and a mean percentage inhibition (MI) of 14.3% (95% CI: 7.1–21.5%). 98.5% of subjects were NT positive at week 1 (p < 0.0001) with an MI of 81.9% (95% CI: 77.8–86.0%) and a 100% at week 4 post-booster, with an MI of 81.3% (95% CI: 78.2–84.4%)
Coronavac Coronavac + ChAdOx1 ChAdOx1 
- 96% of the subjects who received a second ChAdx1 booster dose were NT positive at the time of their second booster (day 0) with an MI of 88.1% (95% CI: 82.2–92.9%). All (100%) were NT positive at week 1 post second booster and an MI of 91.6% (95% CI: 87.5–95.7%).  
Coronavac Coronavac + ChAdOx1 Pfizer
- 100% of the subjects were NT positive against the ancestral strain at day 0 (MI of 84.5%, 95% CI: 80.5–88.6%; and MI of 84.6%, 95% CI: 73.2–95.9%, respectively) and at 1 week (MI of 85.2%, 95% CI: 81.8–88.7%; and MI of 93.3%, 95% CI: 86.3–100.3%, respectively) after their second booster dose. [Buathong R, 2022 ]

Sira et al. (TCTR20210722003) conducted a randomized clinical trial in Thailand, with 422 adults with a mean age of 44 years old. The aim of the study was to evaluate the immunogenicity and reactogenicity of standard-dose and low-dose boosters of Oxford-AstraZeneca vaccine, after the full CoronaVac schedule in healthy adults. 206 individuals received low dose and 208 individuals standard dose. At baseline, geometric means (GM) of sVNT against delta variant and anti-S-RBD IgG were 18.1% inhibition (95% CI 16.4-20.0) and 111.5 (105, 1-118.3) BAU/ml. GM of sVNT against delta variant and anti-S-RBD IgG in DS were 95.6% inhibition (95% CI 94.3-97.0) and 1975.1 (1841.7-2118.2) BAU /ml at day 14, and 89.4% inhibition (86.4-92.4) and 938.6 (859.9-1024.4) BAU/ml at day 90, respectively. The GMRs of sVNT against delta variant and anti-S-RBD IgG in LD compared to SD were 1.00 (95% CI: 0.98-1.02) and 0.84 (0.76-0, 93) at day 14, and 0.98 (0.94-1.03) and 0.89 (0.79-1.00) at day 90, respectively. [Sira Nanthapisal, 2022 ].

Mrak et al conducted a randomized controlled trial to determine the best vaccination strategy for immunosuppressed patients. Patients who failed to seroconvert upon two mRNA vaccinations (BNT162b2 or mRNA-1273) were randomized to receive either a third dose of the same mRNA or the vector vaccine ChAdOx1nCoV-19. 22 patients were vaccinated with the vector vaccine, and 24 received an mRNA vaccine. All patients subsequently presented at follow-up visits and completed the trial at week 4 after vaccination. Seroconversion rates at week 4 after the third vaccination were significantly lower in the vector group (4/22, 18%) as compared to the mRNA group (15/24, 63%; p=0.006). In a multivariable logistic regression model, boost with vector vaccine and age over 65 favored the likelihood of non-seroconversion (OR 0.05 95% CI 0.01–0.28 and OR 0.14 95% CI 0.02–0.73, respectively. [Mrak D, 2022 ].

Niyomnaitham et al was a phase 2 randomized clinical trial conducted in Thailand. The study included 1243 participants with CoronaVac primary schedule: 312 in the AstraZeneca booster (half dose) group, 307 in the AstraZeneca booster (full dose) group, 316 in the Pfizer booster (half dose) group and 308 in the Pfizer booster (half dose) group. Vaccination-induced immunogenicity to Ancestral, Delta and Omicron BA.1 strains were evaluated by assessing anti-spike (‘anti-S’), anti-nucleocapsid antibodies, pseudovirus neutralization (‘PVNT’), micro-neutralization titers, and T-cells assays. Within platforms and irrespective of dose or platform, seroconversions were greater than 97%, and greater than 90% for neutralising antibodies against pseudovirus, but similar against SARS-CoV-2 strains. Anti Spike RBD IgG Geometric means concentration (U/mL) at day 28 were 8237.0 (95% CI, 7679.3-8835.2) for the AstraZeneca half dose group and 8973.6 (95% CI, 8328.1-9669.2) for the AstraZeneca full dose group. [Suvimol Niyomnaitham, 2022 ]

Tawinprai K et al was a randomized controlled trial conducted in Thailand. The study included 125 participants with 2 doses of CoronaVac: 41 received AstraZeneca intramuscular standard dose, 41 received AstraZeneca intradermal fractional dose and 43 received AstraZeneca intradermal fractional dose (20%). The primary endpoint was the geometric mean ratio of anti-receptor binding domain antibody in the ID1/ID2 vs. the IM groups 14 days post-vaccination. Geometric mean concentration of anti-SARS-CoV-2 spike protein RBD antibodies 2 weeks after ChAdOx1/AZD1222 vaccination were 6,414.62 (95% CI, 5,107.56–8,056.17), 5,669.49 (95% CI, 4,560.30–7,048.45) and 8,230.37 (95% CI, 6,697.04-10,114.8) for the intramuscular standard dose group, intradermal fractional dose group and intradermal fractional dose (20%) group respectively. [Tawinprai K, 2022 ]

Bánki Z et al was a randomized clinical trial conducted in Austria. The study included 234 participants between 18 and 65 years with no prior history of SARS-CoV-2 infection and a first dose of AZ or BNT were included. The AZ/AZ and the AZ/BNT arms were randomized, the third arm (BNT/BNT) was observational. 116 participants received AstraZeneca as second dose and 118 Pfizer as second dose. This study compared the reactogenicity between the study groups and immunogenicity for the heterologous AZ/BNT compared to the homologous AZ/AZ regimen using neutralizing antibody titers as primary endpoint. The antibody levels after the boost vaccination were statistically significant higher in the heterologous AZ/BNT group than in the homologous AZ/AZ group (p<0.0001). Median IgG titers for AZ/BNT were 2538 BAU/ml (95% CI 2217 - 3045) at day 10 and 1478 BAU/ml (95% CI 1234 - 1826) at day 30. [Bánki Z, 2022 ]

The MnM study was a phase 4 randomized trial conducted in India including 404 participants, 200 received an AstraZeneca primary schedule and 204 received a Covaxin primary schedule. The participants were randomized to receive an AstraZeneca or Covaxin booster dose. The study assessed the immunogenicity 28 days after the booster dose. The geometric mean concentration (GMC) of anti-SARS-CoV-2 Spike IgG was 36,190.78 (95% CI, 30,526.64-42,905.88) with a geometric mean fold ratio (GMFR) of 2.81 (95% CI, 2.21-3.57) in the AstraZeneca+Covaxin group and the GMC was 241,681.6 (95% CI, 201,380.2-290,048.3) with a GMFR of 31.19 (95% CI, 20.38-47.74) in the Covaxin+AstraZeneca group [Rose W, 2023 ].

REFUERZO was a phase 3 randomized trial conducted in Chile that included 523 participants with 2 doses of CoronaVac: 127 with AstraZeneca booster, 134 with CoronaVac booster, 91 with Pfizer booster, and with 95 placebo. This study reported results of the immunogenicity arm, which determined neutralizing antibody titers in plasma, measured at baseline as well as 14, 30, 60 and 90 days post-booster inoculation. Mean of the fold-increase, with respect to the baseline value, of the geometric mean of ID50 at day 30 was 129.4 (95% CI, 98.9-166.4) for AstraZeneca booster, 12.1 (95% CI, 9.4-15.4) for CoronaVac Booster, 214.2 (95% CI, 160.7-280.5) for Pfizer booster and 0.8 (95% CI, 0.6-1.1) for placebo. [Acevedo J, 2022 ]

Nantanee R et al was a randomized clinical trial conducted in Thailand thatincluded 100 adults with a median age of 59.3 years with two doses of AZD1222: 50 received a half-dose BNT162b2 booster (15 µg/dose) and 50 received a standard-dose BNT162b2 booster (30 µg/dose). The immunogenicity was evaluated by a surrogate virus neutralization test (sVNT) against Omicron variants and anti-spike-receptor-bindin-domain IgG (anti-S-RBD IgG). At day 14, the geometric means (GM) of sVNT against the Omicron variant were 74.4% (95% CI, 68.8% to 80.5%) inhibition post-half-dose BNT162b2 boosters, and 67.3% (95% CI, 57.9% to 78.1%) inhibition post standard-dose boosters. At day 14, anti-S-RBD IgGs were comparably boosted with GMs of 2329.8 BAU/mL (95% CI, 2109.3–2573.4) post half-dose BNT162b2 boosters and 2574.7 BAU/mL (95% CI, 2262.5–2929.9) post standard-dose boosters. At 14 days after boosters, T cell responses were significantly enhanced to 406 SFU/106 PBMCs (IQR 206–718, p < 0.001) in the half-dose and 402 SFU/106 PBMCs (148–678, p < 0.001) in the full-dose groups, without a difference between the dosage groups (p= 0.80). [Nantanee R, 2022 ]

Prasert Assantachai et al was a randomized controlled trial conducted in Thailand that included 210 participants aged ≥65 years with AstraZeneca primary schedule: 35 received intradermal mRNA-1273, 35 intramuscular mRNA-1273, 70 intradermal BNT162b2, and 70 intramuscular BNT162b2. The group that reported the highest geometric mean titers against the ancestral strain and Omicron BA.1 variant was the group that received the intramuscular mRNA-1273 vaccine (GMT 1,717.9 and 617), followed by the group that received the intradermal mRNA-1273 vaccine (GMT 1,212 and 318), the group that received the intramuscular BNT162b2 vaccine (GMT 713 and 230 and the group that received the intradermal BNT162b2 vaccine (GMT 587 and 148), respectively. [Prasert Assantachai, 2022 ]

Sauré D et al was a cross-sectional study conducted in Chile. The study included 101,070 participants: 65,902 with CoronaVac primary schedule, 10,095 with Pfizer booster, 5,435 with AstraZeneca booster and 608 with CoronaVac booster. The aim was to assess IgG seropositivity dynamics after primary and booster vaccination. After homologous booster with CoronaVac the proportion of participants with IgG positivity was 95.7% (95% CI, 89.8-100). After heterologous booster with Pfizer the proportion of participants with IgG positivity was 99.7% (95% CI, 99.3-100.0). After heterologous booster with AstraZeneca the proportion of participants with IgG positivity was 99.5% (95% CI, 98.8–100.0). [Sauré D, 2023 ]

Jäger M et al was a comparative cohort study conducted in Austria that included 137 participants, of which 79 were vaccinated with AstraZeneca primary schedule and mRNA vaccine booster, and 40 with 3 doses of mRNA vaccine. The aim was to assess humoral and cellular response. Median IgG titer were 1655.0 (95% CI, 1163.0-1901.0), 2673.0 (95% CI, 2189.0-3517.0) and 2891.0 (95% CI, 1582.0-4458.0) for particpants with 2 AstraZeneca doses and Pfizer booster group, 2 AstraZeneca doses and Moderna booster and 3 Pfizer doses, respectively. [Jäger M, 2023 ]

Dedroogh S et al was a prospective cohort study conducted in Germany that included 1,338 participants with Pfizer booster dose: 565 Pfizer primary schedule and 773 AstraZeneca primary schedule. The aim was to assess SARS-CoV-2 antibodies booster immunization with Pfizer and AstraZeneca vaccines. Mean of anti-S RBD antibody levels were 25,000 U/mL (95% CI, 18,474 − 25,000) for homologous Pfizer booster and 16,488 U/mL (95% CI, 11,192-24,785) for Pfizer primary schedule and AstraZeneca booster, respectively. [Dedroogh S, 2023 ]

Effectiveness outcomes
Vivaldi et al conducted a population-based cohort study (COVIDENCE UK) to identify risk factors for SARS-CoV-2 infection after primary and booster vaccinations. This UK study in adults (≥16 years) vaccinated against SARS-CoV-2, assessed the risk of breakthrough SARS-CoV-2 infection up to February 2022, for participants who completed a primary vaccination course (ChAdOx1 nCoV-19 or BNT162b2) and those who received a booster dose (BNT162b2 or mRNA-1273). Primary vaccination with ChAdOx1 (vs BNT162b2) was associated with higher risk of infection in both post-primary analysis (adjusted hazard ratio 1,63, 95% CI 1,41–1,88) and after an mRNA-1273 booster (1,26 [1,00–1,57] vs primary and booster BNT162b2 schedules) [Vivaldi G, 2022 ].

Soledad Gonzalez et al was a test-negative, case-control study conducted in Argentina. The study included 422,144 individuals ≥50 years: 221,933 with positive test and 200,211 with negative test. The aim of the study was to estimate the protection against laboratory-confirmed SARS-CoV-2 infection, hospitalisations, and death after homologous or heterologous third-dose (booster) in individuals with primary vaccination schemes with rAd26-rAd5, ChAdOx1nCoV-19, BBIBP-CorV or heterologous combinations, during the period of Omicron BA.1 predominance. Odds ratio of booster dose against confirmed SARS-CoV-2 infections were 0.64 (95% CI 0.61 to 0.67) for ChAdOx1 primary schedule and mRNA booster (≤ 60 days) and 0.97 (95% CI 0.93 to 1.01) for ChAdOx1 primary schedule and vectored booster (≤ 60 days). Odds ratio of booster dose against hospitalizations were 0.14 (95% CI 0.03 to 0.77) for ChAdOx1 primary schedule and mRNA booster (≤ 60 days) and 0.28 (95% CI 0.22 to 0.37) for ChAdOx1 primary schedule and vectored booster (≤ 60 days). [Soledad Gonzalez, 2022 ]

Monge S et al,  was a nationwide Cohort study conducted in Spain. This study included 7,036,433 participants older than 40 years: 3,111,159 in the booster group and 3,111,159 in the no booster group. The aim of this study was to  estimate the effectiveness of mRNA-based vaccine boosters against infection during the period of the predominance of the omicron variant in Spain. Vaccine effectiveness of AstraZeneca primary schedule and ARNm booster was 58.6% (95% CI 55.5–61.6) (Moderna or Pfizer booster dose). [Monge S, 2022 ]

Stowe et al. conducted a test-negative case-control study in UK, to estimate Vaccine Effectiveness (VE) against hospitalisation with the Omicron and Delta variants using PCR testing linked hospital records (Emergency Care Data Set; ECDS). The total number of tests in the study period was 409,985 of which 115,720 were cases and 294,265 controls.

VE against hospital admissions from ECDS within 14 days of the test date by the Omicron variant in symptomatic individuals 18 to 64 years of age:

Primary Schedule ChAd0x1-S Booster BNT162b2
After 14-34 days: 84.1% (95% CI 80.5 to 87.0)
After 105+ days: 38.2% (95% CI 16.0 to 54.5)

Primary Schedule ChAd0x1-S Booster mRNA-1273
After 14-34 days: 88.3% (95% CI 84.3 to 91.3)
After 70+ days: 74.1% (95% CI 56.3 to 84.7)

VE against hospital admissions from ECDS within 14 days of the test date by the Omicron variant in symptomatic individuals 65 years of age and older:

Primary Schedule ChAd0x1-S Booster BNT162b2
After 14-34 days: 93.5% (95% CI 89.3 to 96.0)
After 105+ days: 82.9% (95% CI 75.4 to 88.1)

Primary Schedule ChAd0x1-S Booster mRNA-1273
After 14-34 days: 98.1% (95% CI 92.1 to 99.5)
After 70+ days: 89.6% (95% CI 73.8 to 95.9)

[Stowe J, 2022 ]

Agrawal et al was a retrospective cohort study conducted in the United Kingdom that included 16,208,600 participants, 7,589,080 who received a Pfizer-BioNTech primary schedule and 8,619,520 participants who received a ChAdOx1 primary schedule. The study was based on data from the Oxford-Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) database, Vaccine Management System, Early Pandemic Evaluation and Enhanced Surveillance of COVID-19 (EAVE II), and Secure Anonymised Information Linkage Databank platform, during the Omicron period between December 2021 and February 2022. The adjusted Rate Ratio (aRR) 3-5 weeks after booster vaccination against hospitalization was 0.43 (95% CI, 0.37-0.50) for the ChAdOx1 primary schedule and Pfizer-BioNTech booster group and 0.41 (95% CI, 0.33-0.50) for the ChAdOx1 primary schedule and Moderna booster group. [Agrawal U, 2022 ]

Baum U et al was a cohort study conducted in Finland including 896,220 participants aged 70 years and older. The study was based on a nationwide register-based cohort,  starting on December 27th 2020 and ending on March 31st 2022. Vaccine effectiveness against hospital admission was 98% (95% CI, 91% to 99%) for the AstraZeneca schedule + Pfizer booster, and 94% (95% CI, 92% to 98%) for the AstraZeneca schedule + Moderna booster. [Baum U, 2022 ]

Intawong K et al was a test-negative case-control study conducted in Thailand that included 36,170 participants, 14,682 cases and 21,488 controls. Based on data from the Epid-CM platform, including participants aged 18 years or older, between October 2021 and April 2022. The aim was to assess the effectiveness of different booster doses. The adjusted vaccine effectiveness (VE) against infection during the Omicron predominance period after the third dose was 31% (95% CI, 15% to 44%) for the Pfizer-BioNTech booster, 26% (95% CI, 8% to 40%) for the AstraZeneca booster and 31% (95% CI, 13% to 45%) for the Moderna booster. The adjusted VE after the fourth dose was 71% (95% CI, 60% to 79%) for the Pfizer-BioNTech booster, 73% (95% CI, 48% to 89%) for the AstraZeneca booster, and 71% (95% CI, 59% to 79%) for the Moderna booster. [Intawong K, 2022 ]

Nittayasoot et al was a test negative case-control study to examine the effectiveness of COVID-19 vaccines during January to April 2022 in Thailand. They analyzed secondary data from four main national health data bases: Co-Lab, Co-Ward, COVID-10 Death and MOPH-IC, using the national identification numbers of each individual as a unique identifier to link the same person across databases. They obtained a total of 3,059,616 records including: 1,015 cases of COVID-19 pneumonia requiring invasive ventilation from 652,854 cases with SARS-CoV-2 detection and 2,046,762 controls or non-SARS-CoV-2 detection. Vaccine Effectiveness against pneumonia requiring invasive ventilation for schedule: Coronavac + Coronavac + ChAdOx1 was 94.92% (95% CI, 61.91% to 99.32%), ChAdOx1 + ChAdOx1 + Pfizer was 85.84% (95% CI, 74.00% to 92.29%), ChAdOx1 + ChAdOx1 + Moderna was 87.69% (95% CI, 46.60% to 97.16%), Coronavac + ChAdOx1 + ChAdOx1 was 88.24% (95% CI, 66.80% to 95.83%), Coronavac + ChAdOx1 + Pfizer was 86.05% (95% CI, 70.67% to 93.37%), ChAdOx1 + Pfizer + Moderna was 100.00% (95% CI, 99.99% to 100.00%), Coronavac + Coronavac + ChAdOx1 + ChAdOx1 was 100.00% (95% CI, 99.99% to 100.00%) and Coronavac + Coronavac + ChAdOx1 + Pfizer was 100.00% (95% CI, 99.99% to 100.00%). [Nittayasoot N, 2022 ] 

Huilberts A et al. was a prospective cohort study conducted in Netherlands that included 27,646 participants from the Omicron period: 3,802 received primary vaccination schedule, 23,352 received first booster dose and 492 were unvaccinated. The primary outcome was contracting SARS-CoV-2 between July 12th 2021 and June 6th 2022. Vaccine effectiveness against contracting COVID-19 during  the Omicron BA.1-2 period was 58.4% (95% CI, 48.8% to 66.1%) and 67.6% (95% CI, 60.1% to 73.7%) for AstraZeneca primary schedule with Pfizer booster and AstraZeneca primary schedule and Moderna booster, respectively. [Anne J. Huiberts, 2023 ]

Cerqueira-Silva T et al was a case-control study, test-negative design, conducted in Brazil and Scotland. The study included 5,832,210 participants: 5,276,385 from Brazil and 555,825 from Scotland. This study assessed vaccine effectiveness of a mRNA booster after AstraZeneca or Pfizer primary schedule during the period of Omicron dominance. Vaccine effectiveness against severe outcomes was 93.5% (95% CI, 93% to 94%), 94.4% (95% CI, 87.7% to 97.5%) and 92.7% (95% CI, 91% to 94%) with AstraZeneca primary schedule and Pfizer booster, AstraZeneca primary schedule and Moderna booster and Pfizer homologous booster, respectively. [Cerqueira-Silva T, 2023 ]

Sritipsukho P et al was a case-control study, test-negative design conducted in Thailand that included data from 7,971 participants: 3,104 cases and 4,867 controls. The aim was to assess the vaccine effectiveness (VE) after receiving 2, 3 and 4 doses. VE was 76% (95% CI, 52% to 88%), 58% (95% CI, 28% to 75%), 69% (95% CI, 36% to 85%), 57% (95% CI, 7% to 80%) and 82% (95% CI, 59% to 92%) after CoronaVac primary schedule and Pfizer booster, AstraZeneca primary schedule and Pfizer booster, CoronaVac primary schedule and AstraZeneca booster, CoronaVac primary schedule and 2 doses of Pfizer booster and CoronaVac primary schedule, AstraZeneca booster and Pfizer second booster, respectively. [Sritipsukho P, 2023 ]

Kamal SM et al was a comparative cohort study conducted in Saudi Arabia that included data from 1500 vaccinated participants: 503 AstraZeneca, 521 Pfizer, and 476 Moderna, and 1500 in the control group. The aim was to assess the effectiveness of the AstraZeneca, Pfizer, and Moderna vaccines, and two Pfizer boosters. Relative risk of infection after the first booster was 0.024 (95% CI, 0.009-0.057). [Kamal SM, 2023 ]

dos Santos et al was a comparative cohort study conducted in Brazil that included data from 108,625,066 booster vaccinated participants. The aim was to assess the effectiveness of the primary series of COVID-19 vaccination and booster shots in protecting against severe cases. Vaccine effectiveness (VE) against severe outcomes were 66.5% (95% CI, 62.8% to 70.0%), 73.0% (95% CI, 67.8% to 77.6%), 39.9% (95% CI, −15.3% to 77.2%) and 82.4% (95% CI, 63.2% to 93.7%) for the AstraZeneca primary schedule and Pfizer booster, the CoronaVac primary schedule and Pfizer booster, the Pfizer primary schedule and AstraZeneca booster and the Janssen primary schedule and Pfizer booster, respectively. VE against severe outcomes were 81.1% (95% CI , 80.3% to 81.9%), 84.7% (95% CI, 83.7% to 85.5%), 70.7% (95% CI, 66.9% to 74.2%) and 90.3% (95% CI, 89.5% to 91.0%) for AstraZeneca, CoronaVac, Janssen and Pfizer, respectively. [Santos CVBD, 2023 ]

Safety of the vaccine in preclinical studies

ChAdOx1 nCoV-19 was tested in mice and non-human primates. No cases of death or obvious clinical signs in the vaccinated groups were observed. In rhesus macaques, there were no pathological changes in the lungs [van Doremalen N, 2020 ].

Safety of the vaccine in clinical trials

Key messages

AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine did not increase the risk of adverse events.

AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine did not increase the risk of serious adverse events.

Main safety outcomes of AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine

Any adverse event (1 to 3.4 months)

The relative risk of any adverse event in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 1.05 (95% CI 0.54 to 2.03). This means AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine increased the risk of any adverse event in 5%, compared with placebo vaccine.

Figure - Forest plot of risk ratio meta-analysis. Outcome: any adverse event. Comparison: AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus control vaccine

In the trial identified in this review, 4297 people not receiving AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine out of 22517 presented this outcome (191 per 1000) versus 12157 out of 33608 in the group that did receive it (200 per 1000). In other words, 9 more people per 1000 did not develop the outcome because of the vaccine. This is the same as saying that the intervention led to an absolute risk increase of 0.9%, or that the intervention increased the risk of any adverse event by 0.9 percentage points. Another way of presenting the same information about the absolute effects is the number needed to treat for an additional beneficial/harmful outcome (NNTB/H), the number of participants who need to receive the intervention for one of them to experience the outcome. In this case, the NNTH is 111. Which means that 111 people need to receive the vaccine for one of them to experienced any adverse events.

Applying the GRADE approach, we assessed the certainty of the evidence for this outcome as high.

Serious adverse events (1 to 3.4 months)

The relative risk of serious adverse events in the group that received AstraZeneca-Oxford/SK Bioscience/Serum Institute of India COVID-19 vaccine versus the group that received placebo vaccine was 0.87 (95% CI 0.71 to 1.07). No statistically significant differences between groups were found for serious adverse events.

Applying the GRADE approach, we assessed the certainty of the evidence for this outcome as high.

Summary of findings (iSoF)

Safety of the vaccine in subgroups

Sex
Randomized trials
The proportion of females in the combined analysis of COV002 and COV003 was 60.5% (7045 out of 11636 participants) [Voysey, Merryn, 2021 ].
The subgroup effect in safety outcomes has not been reported in the trials.

Age
Randomized trials
The proportion of patients >55 years of age in the combined analysis of COV002 and COV003 was 12.2% (1418 out of 11636 participants) [Voysey, Merryn, 2021 ].
Combined safety estimates for different age have not been yet reported. The researchers plan to present them in future analyses when a larger dataset is available [Voysey, Merryn, 2021 ]. However, in individual trials it has been reported that the vaccine seems to be better tolerated in older adults than in younger adults [Ramasamy, Maheshi N, 2021 ]

Children and adolescents
Randomized trials
Children were excluded from the COV001, COV002, COV003 and COV0005 trials, so no data are available for this subgroup [Voysey, Merryn, 2021 ]. 
COV006 was a phase 1/2 participant-blinded, randomized controlled trial evaluating the safety in children 6-17 years old. Participants were randomly assigned in a 4:1:4:1 ratio to receive two doses of AstraZeneca COVID-19 vaccine or control. Pain and tenderness were the most common local solicited adverse events for all the Astrazeneca COVID-19 vaccine and capsular group B meningococcal arms following both first and second doses. For systemic solicited adverse events, fatigue and headache were the most commonly reported with higher proportions reported in the Astrazeneca COVID-19 vaccine arms compared with capsular group B meningococcal arms. The increase in systemic reactogenicity was less obvious after 48 hours post-immunisation. For participants receiving Astrazeneca COVID-19 vaccine, there were fewer reported solicited adverse events after the second dose than the first dose [Li G, 2022 ].

Other studies
COVA is an ongoing non-randomized study [The University of Hong Kong, 2021 ] evaluating the safety in children 11-16 years old.

Morciano C et al was  a self-controlled case series study that included 15,986,009 participants who received at least one dose of a COVID-19 vaccine: 10,833,284 Pfizer, 1,706,979 Moderna, 2,863,950 AstraZeneca and 581,796 Janssen. This study investigated the risk of Guillain Barré Syndrome after vaccination with anti-COVID-19 vaccines. Relative incidence of Guillain Barré Syndrome for the second dose of AstraZeneca vaccine was 6.52 (95% CI, 2.88-14.77). [Cristina Morciano, 2023 ]

Pregnancy
Randomized trials
Pregnant females have been excluded from the COV001, COV002, COV003 y COV005 trials, so no data are available for this subgroup. [Voysey, Merryn, 2021 ]

Comparative studies
Calvert C et al was a comparative cohort study conducted in Scotland that included data from the COPS study database: 93,900 pregnant women, 18,780 vaccinated and 75,120 controls. The study assessed the incidence of adverse pregnancy outcomes after COVID-19 vaccination. The adjusted odds ratio after ChAdOx1 vaccination for miscarriage was 0.92 (95% CI, 0.76–1.11), and 1.15 (95% CI, 0.69–1.91) for ectopic pregnancy. [Calvert C, 2022 ].

Covas D et al was a cross-sectional study conducted in Brazil that included 2,486 pregnant and postpartum women vaccinated with: 187 Sinovac, 572 Pfizer, 1,712 AstraZeneca and 15 Janssen. The aim was to describe the adverse events of COVID-19 vaccines in pregnant and postpartum women in the early stage of the vaccination campaign in Brazil. Incidence of adverse events notified for pregnant women was 92.5 (95% CI, 78.95-106.04), 150.14 (95% CI, 137.72-162.57), 4463.3 (95% CI, 4,255.55-4,671.04) and 4,087.19 (95% CI, 2061.54-6112.85) for Sinovac, Pfizer, AstraZeneca and Janssen vaccine, respectively. Incidence of adverse events notified for postpartum women was 15.28 (95% CI, 5.3-25.26), 10.78 (95% CI, 4.68-16.88), 65.5 (95% CI, 35.25-95.75 and 0 (95% CI, 0-0) for Sinovac, Pfizer, AstraZeneca and Janssen vaccine, respectively [Covas DT, 2023 ].

Woestenberg PJ et al was a cohort study conducted in the Netherlands, including 4,348 pregnant women who received at least one dose of Pfizer, Moderna or AstraZeneca. After the first dose, 59% of the women reported 1 or more adverse reactions: 68% for Moderna, 87% for AstraZeneca and 56% for Pfizer. After the first dose, injection site reaction was most commonly reported (32% of women), followed by myalgia (26%) and fatigue (20%). After the second dose, fatigue was most commonly reported (28% of women), followed by myalgia (24%) and injection site reaction (23%).
[Woestenberg PJ, 2022 ].

Breast-feeding
Randomized trials
Breastfeeding females were excluded from the COV001, COV002, COV003 and COV005 trials, so no data are available for this subgroup [Voysey, Merryn, 2021 ].

Persons/ individuals with comorbidities
Randomized trials
Available data are currently insufficient to assess safety in individuals with comorbidities.

Comparative studies
The non-randomized study ILBS-COVID-05 is currently evaluating the effectiveness/safety of the vaccine in liver cirrhosis and non liver cirrhosis groups [Institute of Liver and Biliary Sciences, India, 2021 ].

Immunocompromised persons
Randomized trials
Available data are currently insufficient to assess safety in individuals with comorbidities.

Comparative studies
The cohort study COVATRANS is currently evaluating the effectiveness/safety of the vaccine in kidney transplant recipients aged 15 years and older who receive Pfizer, Moderna and Astra-Zeneca vaccines [University Hospital, Strasbourg, France, 2021 ].

The non-randomized study COVAC-IC is currently evaluating the effectiveness/safety of the vaccine in immunocompromised and immunocompetent hematology patients aged 18 years and older [University Hospitals of North Midlands NHS Trust, 2021 ].

The prospective, cohort, non-interventional, single-center clinical study ImmunoHaema-COVID-VAX-21 is currently evaluating the effectiveness/safety of the vaccine in patients with hematological malignancies 18 years of age and older [Ospedale di Circolo - Fondazione Macchi, 2021 ].

Available data are currently insufficient to assess safety in individuals with comorbidities. 

Persons living with HIV
Randomized trials
Available data are currently insufficient to assess safety in individuals with comorbidities.