COVID-19 Vaccines: A Review of the Safety and Efficacy of Current Clinical Trials

Various strategies have been designed to contain the COVID-19 pandemic. Among them, vaccine development is high on the agenda in spite of the unknown duration of the protection time. Various vaccines have been under clinical trials with promising results in different countries. The protective efficacy and the short-term and long-term side effects of the vaccines are of major concern. Therefore, comparing the protective efficacy and risks of vaccination is essential for the global control of COVID-19 through herd immunity. This study reviews the most recent data of 12 vaccines to evaluate their efficacy, safety profile and usage in various populations.


Introduction
The COVID-19 pandemic is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Up till February 2021 it had infected more than 110 million patients, causing 2.4 million deaths worldwide, according to data recorded by the World Health Organization (WHO) [1].
The prevention and control of the epidemic in 2020, other than treatment of symptomatic patients, has included monitoring of asymptomatic infections, follow-up and monitoring after cure and discharge, close contact tracking, high-risk population screening, and disinfection of the epidemic source, but the only way for the radical control of COVID-19 infections is by effective vaccination. Vaccines stimulate the body to produce specific antibodies, with anamnestic response when the body is exposed to this pathogen again.
During 2020, there has been extensive research to look into the use of vaccinations to prevent further transmission of SARS-CoV-2. Globally, several prospective vaccines have been produced and used by the public ( Table 1). The protective efficacy and immunogenicity profile of each vaccine is also documented ( Table 2).
There are currently two forms of messenger ribonucleic acid (mRNA) vaccines: nonreplicating mRNA (NRM) vaccines and self-amplifying mRNA (SAM) vaccines. The constructed mRNA is formulated into a carrier-usually lipid nanoparticles-to protect them from degradation and promote cellular uptake [2]. After the carrier particles are ingested into the cell, mRNA is released, which is translated by the ribosome to produce the target protein (recognizable antigen) [3]. After the target protein is secreted by the cell, it is rec-ognized by the immune system and stimulates an immune response.    3 µg: 5.6 (95% CI 3.6-8.7); 6 µg: 7.7 (95% CI 5.2-11.5); Placebo: 2.0 (95% CI 2.0-2.0) The average IFN-γ positive spot-forming cells per 100,000 cells were: 3 µg group: 7.4 (95% CI 3.9-11.1); 6 µg group: 3.9 (95% CI 1.0-6.7); Placebo: DNA vaccines, also known as nucleic acid vaccines or genetic vaccines, have also been studied. DNA vaccines are eukaryotic expression plasmid DNA (sometimes also RNA) that encode immunogens or immu-nogens4. It can enter animals through a certain route, and be transcribed and translated after being taken up by host cells. The antigen protein can stimulate the body to produce two kinds of non-specific and specific immune responses, thereby playing a role in immune protection [33,34]. The production process of mRNA is not complicated. The difficulty lies in the fact that mRNA is prone to folding and failure in the absence of protection [35]. Therefore, there is the shortcoming of extremely poor stability. It is questionable whether unstable mRNA is safe for the human body [36]. The comparison between DNA and RNA vaccines is shown in Figure 1.
As of 10 April 2021, the top five countries with vaccination programs are the United States (6.129 million), China (4.052 million), the European Union (2.66 million), the United Kingdom (1.82 million) and India (1.084 million) [37]. Although the implementation of vaccination is one of the important factors to achieve global herd immunity, there is no consensus concerning the superiority of one vaccine over the others in terms of protective efficacy and safety profile, even thigh previous reviews have commented on some of the vaccines [38,39].
To date, there are 86 vaccines under development in clinical phase trials. They are developed with different methods such as protein subunits, inactivated virus, DNA-based vaccine, RNA-based vaccine, viral vectors, and live-attenuated viruses. (see Table 3) [40]. However, many of them are currently in preclinical or phase 1 trials, or without publishing on academic journals at the time of writing. The inclusion criteria of this review are: (1) vaccines that has at least finished their phase 2 clinical trials; and (2) the clinical data of the trial has been published in academic journals and accessible on databases (PubMed, Embase, MedLine, Cochrane) at the time of writing. Exclusion criteria includes: (1) vaccines that are on preclinical phases at the time of writing. (2) vaccines that have not gone through at least phase 2 trials 3) vaccines that have phase 2 trials but have not published their data in academic journals nor accessible on databases (PubMed, Embase, Medline, Cochrane).
This study reviews 12 vaccines in production to evaluate their protective efficacy, safety profile and usage in high risk populations such as children, elderly and patients with co-morbidities.
Pharmaceuticals 2021, 14, 406 7 of 28 route, and be transcribed and translated after being taken up by host cells. The antigen protein can stimulate the body to produce two kinds of non-specific and specific immune responses, thereby playing a role in immune protection [33,34]. The production process of mRNA is not complicated. The difficulty lies in the fact that mRNA is prone to folding and failure in the absence of protection [35]. Therefore, there is the shortcoming of extremely poor stability. It is questionable whether unstable mRNA is safe for the human body [36]. The comparison between DNA and RNA vaccines is shown in Figure 1. After electroporation, cell membrane permeation will be increased, allowing DNA enter into cytoplasm thereby reaching to the nuclear. Subsequently, DNA will be translated into mRNA, which will be further translated into SARS-CoV-2 spike proteins and express on cell membrane. Nanoparticleencapsulated mRNA encoding SARS-CoV-2 antigen will be integrated into cytoplasm. The spike Figure 1. Schematic graph of the comparison between DNA and mRNA vaccine in terms of mechanisms. DNA vaccine is a circle DNA which contains the spike gene of SARS-CoV-2. After electroporation, cell membrane permeation will be increased, allowing DNA enter into cytoplasm thereby reaching to the nuclear. Subsequently, DNA will be translated into mRNA, which will be further translated into SARS-CoV-2 spike proteins and express on cell membrane. Nanoparticle-encapsulated mRNA encoding SARS-CoV-2 antigen will be integrated into cytoplasm. The spike mRNA utilizes ribosome and bases to translate spike proteins, which express on the cell membrane. The membrane spike protein will be recognized by antigen presenting cell (APC) thereby activating immune reaction.

BiONTech (BNT162b1 and BNT162b2)
The BiONTech trials focus on two candidates: BNT162b1 and BNT162b2. Both vaccines are lipid-based, nucleoside-modified mRNA vaccines that encode the trimerized receptorbinder (RBD) of the spike glycoprotein SARS-CoV-2. The RBD-IgG concentrations and SARS-CoV-2 neutralizing titres were measured after complete course of the vaccines. In the trial of BNT162b112, serum IgG geometric mean concentra-tion (GMC) of the recipient after first dose was comparable to the convalescent sera of COVID-19 patient. The trial showed a strong, dose-dependent vaccine-induced antibody response: the GMC of vaccine recipients is 8 times and 42 times the convalescent sera in the 10 µg and 30 µg group, respectively. A further increase to 100 µg showed no additional elevation of RBD IgG concentration, compared with 10 µg and 30 µg trials [4,5].
BNT162b1 induced functional CD4 + and CD8 + T cell responses in almost all recipients: 95.2% participants mounted RBD-specific CD4 + T cell responses. There is a positive correlation between RBD-binding IgG and SARS-CoV-2 neutralizing antibody titres [6]. Severe adverse events, such as grade 3 decrease of lymphocyte count and grade 2 neutropenia, were manageable. No clinical deteriorations were observed.
The overall serological responses of BNT162b2 and BNT162b1 were similar [7]: Phase 2/3 trial showed they conferred 94.6% (95% CI 89.7-97.3) protection against COVID-19 in persons older than 16 years of age [8]. Double dose vaccination further boosts the immune response in both younger and older adults, while the response was weaker in participants 65 to 85 years old. Exploration of dose elevations of vaccinations in the elderly should be conducted in future research.
Serious adverse events such as death from arteriosclerosis and cardiac arrest, paroxysmal ventricular arrhythmia were recorded. However, cardiovascular events occurred similarly in the placebo group, with two deaths due to haemorrhagic stroke and myocardial infarction, and two with unknown causes. It is uncertain whether the vaccine increases cardiovascular risk.
COVID-19 infections is associated with a higher inflammatory burden that can induce vascular inflammation, myocarditis and cardiac arrhythmias [17]. Vaccinations for other acute respiratory virus infection show the possibility of a transient increase in the risk of vascular events [18]. Some studies showed a 10-fold increase of acute myocardial infarction admission within the seven days for of testing positive for influenza B, and a 5-fold increase of risk with influenza A [41][42][43]. Another study suggests that binding of SARS-CoV-2 to ACE2 can cause acute myocardial and lung injury through the alteration in ACE2 signaling pathways [44]. The effect of vaccinations for patients with pre-existing cardiovascular diseases have to be further elucidated.

Moderna (mRNA1273)
mRNA1273 is manufactured by Moderna. It encodes stabilized prefusion S-2P antigen, consisting of the SARS-CoV-2 glycoprotein with a transmembrane anchor and an intact S1-S2 cleavage site [9]. A preliminary report showed the binding antibody IgG GMT to S-2P increased after vaccinations, with 100% serocon-version rates by day 15. Doseresponse relationship was observed with higher dosage eliciting stronger IgG GMT. Both low dose (25 µg) and medium dose (100 µg) elicited CD4 + T cell responses by expression of Th1 cytokines.
The phase 1 clinical trial showed a dose-response relationship [45]. It also elicited a strong CD4 + cytokine response involving Th1 helper T cells. The higher dosage (100 µg) was chosen for phase 3 clinical trials. Robust neutralizing activity to the 614G variant was observed for the 100 µg dose, regardless of the patients' age.

ChadOx1 nCoV-19 (AZD1222)
ChadOx1 nCoV-19 consists of replication-deficient simian adenovirus vector ChA-dOx1, containing the full-length structural surface glycoprotein of SARS-CoV-2, with a tissue plasminogen activator leader sequence [12]. It expresses a codon-optimised coding sequence for the spike protein. Upon vaccination, antibodies against SARS-CoV-2 spike protein peaked by day 28 and remained elevated up to day 56 in participants receiving 1 dose. The median titre of the booster-dose group was more than five times higher than the single-dose group. Paracetamol was used to reduce local regional side effects such as fever and myalgia. Prophylactic paracetamol was prescribed in certain participants, but serological response was independent of prophylactic paracetamol prescription.
ChAdOx1 nCoV-19 appears to be better tolerated in older adults than in younger adults, and it provides similar immunogenicity across all age groups after a booster dose [13]. Serological response was independent of dosage and age after booster, with the IgG level being consistently higher than those without booster vaccinations. Median IgG titres peaked by day 42 in most groups who received two-dose vaccinations. A higher vaccine efficacy was observed when the participants first received a low-dose followed by a stand-ard-dose (90%, 95% CI 67.4-97.0, p = 0.01), compared with two standard-dose recipients (62.1%, 95% CI 41.0-75.7) [24].
In terms of safety profile, 13 serious adverse events occurred but none was considered related to either study vaccine as assessed by the investigators [13]. There was a reported case of hemolytic anemia and three cases of transverse myelitis. The independent neurological committee considered two of them were unlikely to be related to vaccination, and one of them was an idiopathic, short segment spinal cord demyelination [14].
Phase 3 trials are being performed in the United Kingdom, Brazil and the United States of America to assess the protective efficacy and safety [13].
Various thromboembolic events were reported after participants have received Cha-dOx1 nCoV-19 (AZD122) vaccinations. One of the reasons may be related to post-vaccination immune-mediated thrombo-cytopenia [46]. In a report including 28 patients after receiving AZD122 with thromboembolic events, all of them were tested positive for anti-platelet factor 4(PF4)-heparin antibodies, which clinically mimics auto-immune heparin-induced thrombocytopenia [47]. This was similarly observed in another study where five participants with thromboembolic events (100%) tested positive with high level of IgG anti-PF4polyanion complexes, measured by enzyme linked immunoassay (ELISA) [48]. The adverse reaction may be related to the adenovirus-platelet-leukocyte complexes formed after vaccination, which are taken up by the liver by interaction [28] with membrane-associated heparan sulphate proteoglycan (MAHSP) [49,50]. MAHSP acts as a receptor for viral entry. Heparin can lead to dose-dependent inhibition of this reaction, leading to induction of anti-PF4/heparin antibodies [51]. Subsequently, heparin-induced thrombocytopenia and thrombophilia was observed in patients after receiving AZD122 vaccination.

Convidecia (Adenovirus Type-5 Vectored COVID-19 Vaccine)
Adenovirus type-5 (AD-5) vectored COVID-19 vaccine is a replication of defective Ad5-vectored vaccine expressing the spike glycoprotein SARS-CoV-2 [15]. It clones an optimized full-length spike gene based on Wuhan-Hu-1 with the tissue plasminogen activator signal peptide gene into an E1 and E3 deleted Ad-5 vector, and constructed the Ad-5 vectored COVID-19 vaccines using the Admax system. The vaccine demonstrated a dose-response relationship at day 28 after vaccination: the T-cell responses in the high dose group were significantly higher than that in the low-dose group (p < 0.0010), but not significant compared with that in the middle group. TNF-α expression from CD4 + T cells was significantly lower in the low dose group than in the high dose (p < 0.0001) and middle dose groups (p = 0.0032). TNF-α expression from CD8 + T cells was higher in the high-dose group than that in both the middle dose group (p = 0.016) and the low-dose group (p < 0.0001).
Phase 3 trial are being performed globally, with 40,000 participants. It is expected to be completed by January 2022 [17].

Gam-COVID-Vac (Recombinant Adenovirus Type 26 and Recombinant Adenovirus Type 5 Vaccine)
rAd26-S and rAD5-S are vaccines made by Russian manufacturer which carry the gene for SARS-CoV-2 full-length glycoprotein S. Phase 1/2 studies showed both rAd26-S and rAD5-S formulations were safe and well tolerated [18]. Patients receiving combined rAD26-S and rAD5-S were associated with a higher se-roconversion rate (100%) and neutralising antibody GMT (49.25) on day 28 [19]. Combined regimen was better than individual rAD26-S or rAD5-S injection. Increased CD4 + T cells, CD8 + T cells and IFN-γ secre-tion were observed in all vaccine recipients. No serious adverse events were reported.

Covovax (NVAX-CoV2373)
NVAX-CoV2373 is a recombinant SARS-CoV-2 nanoparticle vaccine composed of trimeric full-length sARS-CoV-2 spike glycoproteins and Matrix-M1 adjuvant. The phase 1 study showed two-dose 5 µg regimen with adjuvant induced IgG GMT and neutralization responses that exceeded convalescent serum from most symptomatic COVID-19 patients [20]. The immunological outcomes in 5 µg and 25 µg vaccination groups were comparable. Second vaccinations with adjuvant resulted in GMT level four times greater than the convalescent plasma in symptomatic patients. Adjuvant regimens induced polyfunctional CD4 + T-cell responses that were reflected in IFN-γ, TNF-α and IL-2 production on spike protein stimulation. No serious adverse events were reported. Interim analysis showed the vaccine achieved protective efficacy of 86% against UK variant and 60% against South Africa variant [21]

WIV04-Strain Inactivated SARS-CoV-2 Vaccine
The WIV-04 strain inactivated SARS-CoV-2 vaccine is designed by the Wuhan Institute of Biological Products Co Ltd. The WIV-04 strain was isolated and cultivated in a Verco cell line for propagation, and the supernatant of the infected cells was inactivated by β-propiolactone. Interim analysis of two randomised controlled trials showed a seroconversion rate of 100% in phase 1 trial and 85.7% in the phase 2 trial [10]. A lower-dosage injection was associated with a higher GMT of neutralizing antibody at day 14 after the third injection, compared with other dosage groups. Injection schedule on day 0 and Pharmaceuticals 2021, 14, 406 23 of 28 21 confer a higher GMT, compared with the schedule of day 0 and 14. Most patients started to generate antibody response after the second injection, and remained at high level 14 days after the third injection. The most common adverse reactions were injection site pain and fever, which were mild and self-limiting. The phase 3 study data was not available at the time of writing.

BBIBP-CorV
BBIBP-CorV is developed by the Beijing Institute of Biological Products. It is an inactivated vaccine developed from the strain 19nCoV-CDC-Tan-HB02 (HB02) [11]. The HB02 strain was purified and passaged in Vero cell lines to generate vaccine production by using a novel carrier in a basket reactor. In the phase 1 trial, a higher dosage (8 µg) was associated with a higher seroconversion rate by day 14, while seroconversion rates reached 100% in all three dosage cohorts on day 28. By day 28, the neutralizing antibody GMT was significantly higher in the high-dose group than the low-dose group (2 µg), with no significant difference between medium-dose (4 µg) and high-dose. Younger adults were associated with higher neutralizing anti-body GMT, compared with older adults (>60 years).
The phase 2 trial showed the immunization schedule of 4 µg on day 0 and 21 was associated with the highest neutralizing antibody GMT (282.7, 95% CI 221.2-361.4), compared with other immunization schedules. One grade 3 or above adverse event was documented due to self-limiting grade 3 fever (>38.5 • C).
A phase 3 study is currently underway in Abu Dhabi with 15,000 participants: 5000 participants receiving placebo, another 5000 receiving BBIBP-CorV, and the remaining 5000 receiving another inactivated vaccine manufacturer by Sinopharm [23].
The phase 2 immunization schedule trial showed receiving vaccination on day 0 and 14 resulted in the most promising outcomes: seroconversion rates were 97%, 100% and 0% in the 3 µg, 6 µg and placebo groups on day 28, respectively. The neutralising antibody GMT were 44.1 (95% CI 37.2-52.2), 65.4 (95% CI 56.4-75.9) and 2.0 (95% CI 2.0-2.1) in the three groups, respectively. One case of serious adverse events related to acute hypersensitivity with presentation of urticaria 48 h after the first dose. It was managed with chlorphenamine and dexamethasone, and recovered within 3 days.
The phase 3 study data has not been published in medical journals. An online search of the phase 3 study in Brazil showed a 50.4% protective efficacy in preventing symptomatic infections, 78% protective efficacy in preventing mild cases requiring treatment and 100% prevention of severe cases [52]. Phase 3 studies in Turkey and Indonesia showed a protective efficacy of 83.5% and 65.3%, respectively [53,54].

Ad26.COV2.S
Ad26.COV2.S is developed by Johnson & Johnson. It is a recombinant, replicationincompetent adenovirus serotype 26 (Ad26) vector encoding a full-length and stabilized SARS-CoV-2 spike protein. Early animal studies showed promising efficacy with low-dose single-shot vaccination [25,26]. In the phase 1 clinical trial, binding and neutralizing antibodies were detected in 100% of vaccine recipients by 57 days after single vaccinations [27]. The geometric mean titres (GMT) of spike-specific binding antibodies and neutralizing antibodies ranged from 2432-5729 and 242-449, respectively. A booster immunization on day 57 increased binding antibody titres and neutralizing antibody titres by a mean of 2.56-fold (range 1.58-3.04) and 4.62-fold (range: 3.56-5.68), respectively. An interim study showed the titres remain stable until at least day 71 [28]. Strong immune responses were recorded as CD4 + T cells were detected in 76 to 83% of the young patients (aged 18-55 years), and 60 to 67% in older patients (aged greater than 65). Phase 3 data showed a 66.9% (95% CI 59.0-73.4) protective efficacy across all participant age groups, and 76.3% (95% CI, 61.6-86.0) in participants older than 60 years old [29]. In preventing severe or critical COVID-19, Ad26.COV2.S was associated with 76.7% efficacy at 14 days, and 85.4% at 28 days. Adverse reactions were recorded such as thromboembolic events (15 in vaccination arm and 10 in placebo arm) and tinnitus ( respectively. This may be related to the difference in the prevalence of mutant strain of SARS-CoV-2 in different regions.

Covaxin (BBV 152)
BBV 152 is a whole-viron inactivated SARS-CoV-2 vaccine formulated with a toll-like receptor 7/8 agonist molecule (IMDG) adsorbed to alum (Algel) [30]. It is developed by Bharat Biotech from an isolated NIV-2020-770 strain of a patient with COVID-19 sequenced in India. Previous animal studies showed acceptable safety profiles, humoral and cellmediated responses [31]. Phase 2 trials showed a good reactogenicity, safety profile, and enhanced humoral and cell-mediated immune responses when participants received a higher dose (6 µg) of Algel-IMDG formulation [32]. In the phase 2 trial, the GMT at day 56 was significantly higher in the 6 µg group (197.0, 95% CI 155.6-249.4) compared with the 3 µg group (100.9, 95% CI 74.7-137.4, p = 0.0041). Seroconversion rates were 92.9% (95% CI 88.2-96.2) in the 3 µg group, and 98.3% (95% CI 95.1-99.6) in the 6 µg group. The Algel-IMDG formulation elicited T-cell responses biased to a Th1 phenotype at day 42, with no significant difference in causing local or systemic adverse reactions between the 3 µg and the 6 µg groups. No serious adverse events were reported in the study. Protective efficacy was not reported.

Challenges
In view of the surging infections and promising efficacy in clinical trials of vaccines (Table 2), many countries have advocated vaccination programs for their citizens. However, questions have been raised concerning the efficacy against new variant strains. Experience in Manaus (Brazil) showed secondary immunity alone was not sufficient to arrest transmission [55], possibly due to new variant strains. The B.1.1.7 of the UK and South African 501Y.V2 variants are shown to cause alterations to the spike protein, which may affect immune recognition of antibodies derived from existing vaccines [56]. Further clinical trials are required to test for the efficacy of existing vaccines against mutant variants.
Another problem is the duration of the protective efficacy. It is likely that at least yearly boosters are necessary. Seasonal modification to annual vaccines to arrest the transmission of previous strains may also be considered. It is also doubtful whether circulating neutralizing antibody is protective against COVID-19 infection as animal studies showed robust viral infective activities in nasal turbinate. Reinfection is still potentially possible [57].
Also with the expansion of the vaccination programs in the general population, the relationship of certain side effects, such as the thrombotic events occurring after receiving ChadOx1 nCoV-19, with the vaccines has to be further determined.
The pathological correlation between incidence of cardiovascular adverse events and vaccination with in-activated or live-attenuated virus has to be elucidated. SARS-CoV-2 infection is associated with systemic inflammatory response causing cytokine releases and cytokine storm, resulting in vasculopathy and its complications [58]. Likewise, influenzae carries similar pathogenesis as SARS-CoV-2. However, the experience of influenzae vaccinations (inactivated virus) shows that vaccinations reduced major cardiovascular events significantly, and has become part of the routine care of patients with chronic cardiovascular conditions [59]. COVID-19 vaccinations do not follow the typical trend of influenzae. In general, attenuated patho-gens have the very rare potential to revert to its pathogenic form [60]. Further studies is required to determine whether vaccines with inactivated SARS-CoV-2 can reduce or induce cardiovascular events.
Diabetic patients are associated with a higher risk of inflammatory response and coagulopathy during an infection episode [61]. Close monitoring of inflammatory markers, tight glycemic controls and lifestyle modifications are recommended for diabetic COVID-19 care [62]. Acute complications after vaccinations can be monitored by measurement of prognostic inflammatory markers, such as serum ferritin, lactate dehydrogenase, C-reactive protein (CRP), erythrocyte sedimentation rate, D-dimer level, cardiac troponin and Nterminal pro-brain-type natriuretic peptide (NT-proBNP) [63][64][65][66]. These markers have close associations with the prognosis of COVID-19 infections. However, the interval and duration of monitoring has to be further studied. The relation between thrombotic events and vaccine using as adenovirus vector has been discussed in a previous section.

Conclusions
The COVID-19 vaccines in clinical trials have all shown promising immunogenicity with varying degree of protective efficacy, and an acceptable safety profile. A second dose immunization gives more robust immune response in all vaccines. The immunological outcome in the elderly is poorer than in the younger recipients. Further exploration on immunization schedule is required, such as more frequent vaccinations or higher dosage in each injection. Grade 3 or above side effects are not common in the clinical trials to date.