COVID-19 Vaccine Platforms: Challenges and Safety Contemplations

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has become a pandemic as of March 2020, creating a global crisis and claiming millions of lives. To halt the pandemic and alleviate its impact on society, economy, and public health, the development of vaccines and antiviral agents against SARS-CoV-2 was a dire need. To date, various platforms have been utilized for SARS-CoV-2 vaccine development, and over 200 vaccine candidates have been produced, many of which have obtained the United States Food and Drug Administration (FDA) approval for emergency use. Despite this successful development and licensure, concerns regarding the safety and efficacy of these vaccines have arisen, given the unprecedented speed of vaccine development and the newly emerging SARS-CoV-2 strains and variants. In this review, we summarize the different platforms used for Coronavirus Disease 2019 (COVID-19) vaccine development, discuss their strengths and limitations, and highlight the major safety concerns and potential risks associated with each vaccine type.


Introduction
The coronavirus disease 2019 , caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), first reported in late 2019 in Wuhan, the capital of Hubei province in Central China, has become a global pandemic with devastating effects worldwide [1]. Since then, and until 29 June 2021, this newly emerging disease caused by the enveloped SARS-CoV-2 virus, which belongs to the Coronaviridae family and the lineage B of the betacoronavirus (β-CoV) genera, has brought over 181 million confirmed cases and claimed the lives of about 4 million people worldwide [1]. SARS-CoV-2 has a positive-sense, single-stranded genome that encodes a large non-structural polyprotein (ORF1a/b) proteolytically cleaved to generate proteins, four of which are structural proteins including spike (S), envelope (E), membrane (M), and nucleocapsid (N) (Figure 1a) [2,3]. Among these proteins, the S surface glycoprotein plays a critical role in receptor recognition and attachment to host cells [4]. The S protein also induces T-cell responses and is the main target of highly potent neutralizing antibodies (nAbs) against the virus, presenting it as the major antigenic pick out for vaccine design [5]. The structure of SARS-CoV-2 is similar to other β-CoVs, including the severe acute respiratory syndrome coronavirus (SARS-CoV) and the Middle Eastern respiratory syndrome-related coronavirus (MERS-CoV), the causative agents of SARS and MERS, two previously reported viral pneumonia disease outbreaks, respectively [6]. Compared to SARS-CoV and MERS-CoV; however, SARS-CoV-2 has higher infectivity and transmissibility due to its high-affinity binding to the host cell receptors and high viral shedding levels during the early stage of infection, contributing Therefore, despite the taken measures and as a consequence of not implementing immediate lockdown, the COVID-19 death toll increased [13,14]. This necessitated the development of an effective and safe vaccine as an imperative solution to control the pandemic and prevent future outbreaks [13,15]. As such, and since the release of the SARS-CoV-2 genome sequence in January 2020, all efforts have been directed towards COVID-19 vaccines development [16,17]. The hope and hype placed on vaccines to prevail over the disease stand up from the success of previously developed vaccines to control other infectious diseases [13]. The route for vaccine development; however, was not always paved, and several historical attempts of vaccines production were doomed with defeats [18]. Until today, and despite all the knowledge and technology at one's disposal, scientists are still unable to conclude the safest and most effective vaccine platform [18]. Back in time, particularly following the outbreak of SARS-CoV in 2002, vaccines against the emerging virus were also developed, a few of which reached phase I clinical trials; yet, did not achieve the final stages and obtain the United States Food and Drug Administration (FDA) approval as the virus was eradicated from the human population in 2004 [16,[19][20][21]. Similarly, several vaccines against MERS-CoV were under development, none of which have obtained FDA approved thus far [21]. Within the same notion, and in relay for safe and effective COVID-19 vaccine production, censorious steps are currently followed in all phases of COVID-19 vaccine development, including manufacturing, dispersal, and vaccination [22]. For the time being, many of the newly developed COVID-19 vaccines are undergoing clinical evaluation and have reached phase III of clinical trials. A few of which have been approved for emergency use [13] (Figure 2a, Table 1), with the research and discovery phase being skipped [21,23]. Several approaches, including traditional platforms (inactivated and live attenuated virus vaccines), and newly established ones (replicating and non-replicating viral vector vaccines, nucleic acid (DNA and RNA) vaccines, recombinant subunit vaccines, and peptide-based/virus-like particles vaccines), have been adopted for COVID-19 vaccine development (Figure 1b-h) [16,24,25]. As of 29 June 2021, and according to the World Health Organization (WHO), out of the 293 total COVID-19 vaccine candidates, 105 are currently in the clinical phase of development and 184 are still in the pre-clinical phase (Figure 2a) [26].Presently, and besides the FDA consideration of the possibility of booster vaccine shots, several standpoints are now advocating the notion that "hybrid immunity" and "the mix and match of different vaccines strategy" could provide an even stronger immune boost, presenting such approaches, if supported by data, as plausible pandemic game-changers. In this review, we detail the different COVID-19 vaccine platforms and highlight their strengths, limitations, and major risks and safety concerns associated with each type, particularly those relevant to the fast-track pace taken for their production. We also summarize all candidate COVID-19 vaccines currently in the clinical phase of development and categorize them according to the platform used for their development.

Inactivated Vaccine
Purified inactivated viruses have been widely used for over a century in vaccine development against various emerging infectious diseases, including influenza, polio, rabies, and hepatitis A [27][28][29][30][31]. Today, inactivated vaccines are typically produced by propagating the virus in cell culture systems, followed by purification, concentration, and chemical and/or physical inactivation to demolish infectivity while retaining immunogenicity ( Figure 1b) [32,33]. This type of vaccine is notably featured by its highly efficient proliferation and genetic stability [34]; yet, limited by the viral yield in a cell culture setting, the requirement of a biosafety level 3 facility, and the short duration of the elicited immune response, possibly making the vaccines less effective in preventing viral entry [33,35]. Up to date, 16 inactivated SARS-CoV-2 vaccines have been developed and are currently in clinical trial phases (Figure 2b) [26]. One of which, for example, is the Sinovac's CoronaVac vaccine candidate which has demonstrated sufficient safety and efficacy in phase III of clinical trials in Brazil, Turkey, and Indonesia and is currently in phase IV of clinical trials (Table 1) [26,34,[36][37][38][39]. Another is the BBIBP-CorV vaccine candidate, which showed adequate humoral immune responses in adults aged 18 years and above

Inactivated Vaccine
Purified inactivated viruses have been widely used for over a century in vaccine development against various emerging infectious diseases, including influenza, polio, rabies, and hepatitis A [27][28][29][30][31]. Today, inactivated vaccines are typically produced by propagating the virus in cell culture systems, followed by purification, concentration, and chemical and/or physical inactivation to demolish infectivity while retaining immunogenicity ( Figure 1b) [32,33]. This type of vaccine is notably featured by its highly efficient proliferation and genetic stability [34]; yet, limited by the viral yield in a cell culture setting, the requirement of a biosafety level 3 facility, and the short duration of the elicited immune response, possibly making the vaccines less effective in preventing viral entry [33,35]. Up to date, 16 inactivated SARS-CoV-2 vaccines have been developed and are currently in clinical trial phases (Figure 2b) [26]. One of which, for example, is the Sinovac's CoronaVac vaccine candidate which has demonstrated sufficient safety and efficacy in phase III of clinical trials in Brazil, Turkey, and Indonesia and is currently in phase IV of clinical trials (Table 1) [26,34,[36][37][38][39]. Another is the BBIBP-CorV vaccine candidate, which showed adequate humoral immune responses in adults aged 18 years and above and currently stands Vaccines 2021, 9, 1196 5 of 38 in phase IV of clinical trials (Table 1) [39][40][41]. Both vaccines have been listed by the WHO for COVID-19 Emergency Use (EUL) and are presently being adopted by several countries worldwide. Despite these promising data, concerns of using inactivated virus vaccines platforms against COVID-19 still reside, some of which relate to the difficulty of confirming a complete virus inactivation status, a risk that could translate into a scenario similar to the 1955 Cutter incident where children receiving the polio vaccine were infected with the inactivated poliovirus [33,42]. Into the bargain, although several developed inactivated SARS-CoV vaccines have been reported to induce nAbs, vaccinated animals still display significant disease upon challenge, which could explain why no vaccines are currently licensed for SARS-CoV [43]. Further, previous studies on animal models have shown that immunizations with inactivated SARS-CoV and MERS-CoV vaccines are associated with hypersensitive-type lung pathology post-challenges with the infectious virus [32,[44][45][46]. Similarly, respiratory syncytial virus (RSV) formalin-inactivated vaccine has been reported to cause enhanced pulmonary disease after live RSV infection [47,48]. In addition, it was suggested that treating the vaccine with formalin could have altered the epitopes, inducing functional antibodies, causing the immune system to produce antibodies against nonprotective epitopes [33,49]. It is worth noting here that none of these concerns and/or complications of using inactivated virus vaccines have been thus far reported from the use of recently developed COVID-19 inactivated vaccines.

Live Attenuated Vaccine
Live attenuated vaccines, which embody a weakened version of the live virus with reduced virulence, are considered one of the oldest and most effective immunization approaches to elicit life-long immune responses ( Figure 1c) [32,50]. A remarkable advantage of such a vaccine type is its relatively low production and delivery costs, given that the attenuated virus can replicate and propagate within the host. As such, a relatively small dose of the virus can be enough to induce immunity [51]. Moreover, live attenuated vaccines can be given intranasally, allowing the attenuated virus to replicate in the mucosal tissue of the upper respiratory tract, a major portal for coronaviruses entry into the host [52]. For the time being, only six SARS-CoV-2 live attenuated virus vaccines have been developed, four of which are in the pre-clinical phase, and two are in phase I of clinical trials ( Figure 2b, Table 1) [26]. Both COVI-VAC and MV-014-212 vaccines are attenuated via codon pair deoptimization, a strategy that involves synthetic recoding of the viral genome by amending the positions of synonymous codons, thereby raising the number of suboptimal codon pairs and cytosine phosphoguanine (CpG) dinucleotides in the recoded genome [25,[53][54][55]. In parallel to live attenuated SARS-CoV-2 vaccine studies, ongoing studies on other live attenuated virus vaccines such as the RSV vaccine have shown success in using the codon pair deoptimization strategy in vaccine production evidenced by the robust humoral and cellular immune responses triggered in non-human primates [56].
Despite the aforementioned advantages and the pulled off accomplishments of using live attenuated virus vaccine in combating different infectious diseases, the overt risk of using such a type of vaccine still resides in the use of a live replicating virus, which can revert under any condition to its pathologic phenotype, causing disease after vaccination, especially in immunocompromised individuals [57,58]. Although this anticipated scenario is considered relatively rare, the degree of unpredictability regarding the virus stability and the arising safety considerations after that should never be ruled out [59]. Further, live attenuated vaccines could result in viral shedding into the environment, posing a potential risk to the unvaccinated community [60]. It also goes without saying that these highlighted disadvantages are acquainted with time consumption and technical difficulties associated with the virus modification approaches if such a vaccine platform is to be implemented [16].

Viral Vector Vaccine
Viral vector vaccines, in both replicating and non-replicating forms, utilize modified viruses such as adenoviruses or poxviruses as the vector to deliver the genetic material coding for a viral antigen of interest into the host cell ( Figure 1d) [57,61]. In self-replicating (replication-competent) viral vector-based vaccines, and through the host cell machinery used by the virus vector, new viral particles are produced in infected cells, which then infect other new cells, resulting in additional vaccine antigen production [62]. On the contrary, non-replicating (replication-incompetent or deficient) viral vector-based vaccines cannot produce new viral particles, and the host cell machinery is used to produce the vaccine antigens, after which the viral vector gets cleared [61,62]. Both viral vector vaccine forms do not cause infection from neither the loaded virus nor the viral vector as the delivered genetic material does not become integrated into the host genome [61,63]. Typically, the advantage of this type of vaccine lies in promoting the expression of viral antigens within infected host cells for efficient major histocompatibility complex (MHC) class I and class II presentation [61]. Moreover, viral vectors are characterized by their high gene transduction efficiency, high specificity of genes delivered to target cells, and the immune response they elicit with increased cellular response [64]. Further, although viral vector vaccines are generally considered less robust than traditional vaccine types, the fact that they persist as genetic material in the host, directly infect antigen-presenting cells, and possess a strong inherent adjuvant activity triggering innate and adaptive immune responses and generating high titers of nAbs, could suffice a single vaccine dose for adequate immunization as in the case of the vesicular-stomatitis virus -(VSV)-based Ervebo vaccine against Ebola virus [62,63,65]. In COVID-19 vector-based vaccine production, replicating and non-replicating vectors have been utilized to deliver genes encoding for either the SARS-CoV-2 S glycoprotein or the receptor-binding domain (RBD) [16,26]. Thus far, vaccinia and adenovirus are the predominantly used virus vectors for vectored vaccines development [64]. The adenovirus, for example, has been previously utilized in developing SARS-CoV vaccines expressing the S and N proteins [32,43,66]. Currently, it is also being used for developing COVID-19 vectorbased vaccines. Up to date, 4 replicating and 17 non-replicating COVID-19 vector-based vaccines have been developed, of which 2 have reached phase III clinical trials, and 3 are currently in phase IV (Table 1, Figure 2b) [26]. All five vaccines are adenovirus-based nonreplicating vaccines containing the gene encoding for SARS-CoV-2 S glycoprotein [67][68][69][70]. Among these vaccines, Janssen's (Ad26.COV2.S) vaccine has recently received the FDA EUA for use in in 18 years old and elder individuals after showing good efficacy data in phase III of clinical trials [71]. Although the Ad26.COV2.S vaccine showed around 65-66% efficacy in moderate to severe/critical and around 76-83% in severe/critical COVID-19 patients, its efficacy dropped to 52 and 64% against the Beta (B.1.351) variant in moderate to severe/critical disease conditions, respectively [69] (Table 1). Low efficacy data were also reported for AstraZeneca vaccine against the Beta variant, with an efficiency of 10.4% only reported in South Africa and 48% in Canada [72,73], contrarily to the 70.4% retained efficacy against the Alpha (B.1.1.7) variant as reported in a study conducted in the UK [74]. The other three viral vector vaccines at stages II/III-IV of clinical development are CanSino's adenovirus type-5 (Ad5) vectored vaccine, Gamaleya Research Institute's Gam-COVID-Vac vaccine, and ReiThera's GRAd-COV2 (Table 1). Although clinical trials have revealed that these vaccines are tolerable and immunogenic, age and the presence of high pre-existing anti-adenovirus immunity were shown to partly diminish vaccination-induced specific antibody and T-cell responses [68]. To overcome pre-existing immunity to the adenovirus in vaccinated individuals, a plausible approach could be using a heterologous recombinant vector as in the Gam-COVID-Vac (Sputnik V) vaccine, the only heterologous COVID-19 vaccine that uses both adenovirus 26 (Ad26) and adenovirus 5 (Ad5) as vectors to express the SARS-CoV-2 S protein [70,75]. Of note, the general principle of prime-boost with two distinct vectors was not exclusively used in recent COVID-19 vaccine platforms but has been largely implemented experimentally and was also previously used in developing the GamEvac-Combi Ebola virus vaccine [76].

Nucleic Acid (DNA and RNA)-Based Vaccine
In nucleic acid-based vaccines, only the genetic material (DNA or RNA), but not the recombinant/live virus, is taken up by host cells and translated into the protein to elicit an immune response (Figure 1e,f) [77]. Although various messenger RNA (mRNA) vaccines, including those against influenza, Zika, and rabies viruses, have been thus far developed, this vaccine development platform is still considered relatively new [78]. The pronounced advantage of some types of nucleic acid vaccines generally lies in the largescale production pace and cost [16]. DNA vaccines, for example, are based on the use of highly stable plasmid DNA that can be easily propagated at a large scale in bacteria, as the plasmid DNA typically encloses mammalian expression promoters and the gene encoding the protein of interest [16]. On the other hand, presenting mRNA vaccines as promising alternatives for conventional vaccines mainly lies in the ability to produce the vaccine completely in vivo, along with their high potency, cost-effectiveness, rapid development, and safe delivery [16,78,79]. Currently, lipid nanoparticles (LNPs) are among the most commonly used in vivo RNA delivery vectors, protecting the mRNA from enzymatic degradation and facilitating endocytosis and endosomal escape [80]. Contrarily to the highlighted recognition of mRNA vaccines, the physiochemical properties of the mRNA that may impact its cellular and organ dispersal, the questioned safety and efficacy of mRNA vaccine use in humans, them being unlikely to induce strong mucosal immunity due to their intramuscular administration, and the uncertainty from what could arise with large-scale production, storage, and stability are among the alarming concerns tailored to mRNA vaccines production [16,57,80]. Likewise, potential disadvantages also relate to DNA vaccines, particularly those relevant to their low immunogenicity and to the need of DNA molecules to traverse the nuclear membrane to be transcribed, necessitating complicated delivery systems such as electroporators for better efficiency [16,57]. In addition, introducing mutation and dysregulated gene expression by the plausible stable integration of transfected DNA into the somatic or germline host cells genome is another arising concern [81] though unconventional as per relevant follow-up studies [82][83][84][85]. Up to date, 28 nucleic acids (10 DNA and 18 mRNA)-based COVID-19 vaccines have been developed and are currently in the clinical stages, and 24 mRNA vaccines are in the pre-clinical stage ( Figure 2b, Table 1) [26]. Two mRNA-based vaccines, developed by Pfizer/BioNTech and Moderna, are currently in phase IV clinical trials and have received the FDA EUA for protection against COVID-19 [26,86,87]. Preliminary results showed astoundingly 94-95% efficacy for both vaccines [88,89]. Though promising, a major concern relevant to mRNA vaccines resides in their rapid pace of development and the uncertainty of potential longterm adverse effects associated with them, particularly because these are the first approved mRNA vaccines with no other FDA-approved mRNA vaccines to date [90]. Another concern is the efficacy of these vaccines against the newly emerging SARS-CoV-2 variants with mutations in the S protein, the main target in COVID-19 vaccines development [91]. As of yet, Pfizer/BioNTech COVID-19 vaccine was reported to protect against four variants of concern (VOCs), including Alpha, Beta, Gamma, and Delta (Table 1) [91][92][93][94]. Interestingly, a recent study by Zakhartchouk et al. reported that combining DNA vaccine and whole killed virus vaccines augments immune responses to SARS-CoV [95], a propitious tactic worth considering in ongoing COVID-19 vaccine development approaches [95].

Protein Subunit and Virus-Like Particles Vaccine
As compared to the whole-pathogen vaccine platform, a protein subunit vaccine is composed of in vitro harvested and highly purified viral protein antigens carefully chosen for their ability to elicit an immune response (Figure 1g) [96]. Being incapable of causing disease, the protein subunit vaccine platform is considered safer than the whole-virus (live attenuated and inactivated) platforms [97]. Not displaying the full antigenic complexity of the virus and enclosing small antigens deficient of pathogen-associated molecular patterns (PAMPs); however, it may promote skewed immune responses, bringing the immunogenicity potential and protective efficacy of protein subunit vaccines into question [57,97]. Subunit vaccine design and production could be also costly and might necessitate specific adjuvants to boost the immune response [98], in addition to the potential occurrence of antigen denaturation, which could lead to non-specific binding [99]. Examples of developed subunit vaccines include the recombinant RBD subunit vaccine, which was reported to elicit partial protective immunity in rhesus macaques against MERS-CoV challenge [100], and S protein-based subunit vaccines against SARS-CoV infection with potency to induce nAbs and protect against SARS-CoV intranasal infection in mice [32,101]. Up to date, 33 COVID-19 protein subunit vaccines based on the S protein or the RBD have been developed and are in the clinical stages. Of which, 10 vaccines, including Novavax's (NVX-CoV2373) are in phase III [26,102]. Recent reports showed that a two-dose regimen of the NVX-CoV2373 vaccine exhibited 89.7% efficacy against SARS-CoV-2 infection, with high efficacy against the Alpha, Beta, and other VOCs [102,103] (Table 1). Virus-like particles (VLPs) vaccine is another type of protein-based vaccine composed of proteins from the viral capsid only with no viral genetic material (Figure 1h) [57,104]. In addition to being safe, VLPs elicit potent immune responses due to their repetitive structures [104]. VLP vaccines against many viruses, including Hepatitis B virus, Human papillomaviruses, and Influenza A virus, do exist [104][105][106][107]. Likewise, VLP vaccines against MERS-CoV and SARS-CoV infection have been also developed, with eosinophilic pulmonary immunopathology detected after viral challenge in some cases [21,46,108]. For the COVID-19 status quo particularly, five VLPs vaccines in different phases of clinical trials are thus far available ( Figure 2b, Table 1) [26]. NR Not yet approved in any country [26,133] Vaccines 2021, 9,1196   NR Not yet approved in any country [26,180] Vaccines 2021, 9,1196 [26,73,88,92,94,110,137,145,[194][195][196][197][198] [26,73,89,92,93,110,141,142,145,196,197,[199][200][201][202][203][204][205][206][207][208][209]  NR Not yet approved in any country [26,227] Vaccines 2021, 9,1196  NR Not yet approved in any country [26,278] Vaccines 2021, 9,1196   NR Not yet approved in any country [26,296]