Genotoxicity: a neglected but potentially critical aspect of adenoviral COVID-19 vaccines

A prophylactic immunization against the disease COVID-19, caused by the SARS-CoV-2 virus, can be achieved through vaccination [1]. Vaccination programs have had a remarkable impact on the control of the COVID-19 burden with reports of greater than 90% reductions in mortality rate observed in vaccinated individuals of all ages and reductions in hospitalization by 71–87% [2], though it should be noted that the efficacies of different vaccines against variants of SARS-CoV-2 are variable [1]. However, the long-term safety of COVID-19 vaccines may have been overlooked as a result of the immediacy of the pandemic. Some short-term side effects of the vaccines have gradually become evident; apart from general side effects such as fatigue, headaches, muscle pain and chills, COVID-19 vaccination has been shown to trigger the development of various autoantibodies associated with autoimmune diseases including IgA nephropathy and autoimmune vasculitis [3,4]. Adenoviral-based vaccines are one type of vaccine platform that has been developed for COVID-19 vaccination and warrants further investigation for their potential long-term side effects. These vaccines use a viral vector to deliver a genetic sequence encoding an immunogenic antigen into host cells to elicit transient antigen expression and a prophylactic immune response, but some reports suggest that they may have the potential for genome integration.

Viruses have been harnessed as gene-delivery tools in genetic engineering for their biological functions and, while there are also non-viral gene-delivery approaches, viral vectors are known as the most efficient tools for gene delivery in research and translational medicine. The earliest use of recombinant vector viruses dates to the use of simian virus 40 (SV40) in the 1970s [5]. Adeno-associated viruses, lentiviruses, adenoviruses and herpes viruses have since been used in the development of recombinant vector viruses and through several generations, their safety and efficiency have been improved. These viruses are usually packaged and produced in the laboratory by simultaneous transfection of gene transfer plasmids and helper plasmids that encode the structural components of viral particles in packaging cells. The best choice of viral vector depends on the target cells and the purpose of gene delivery [6]; adenoviral vectors have been explored as suitable candidates for use in DNA vaccine platforms due to their ability to cause transient gene expression [7]. These vectors are usually based on replication- and packaging-deficient generations of modified Ad2 and Ad5 serotypes of human adenovirus C (HAdV C), a non-enveloped double-stranded DNA virus with a capacity of packaging up to 7.5 Kb of foreign DNA. Upon the binding of an adenoviral vector to susceptible host cells, the viral material is internalized through endocytosis, transported into the nucleus and transcription of the transgene antigen is initiated. The low immunogenicity of viral particles, the ability to produce high titer viruses in the laboratory and the supposedly non-oncogenic properties of these vectors have made them suitable candidates for their use in vaccine platforms.

Although adenoviral vectors are known for transient gene expression, their expression has been documented for up to 7 years and there is evidence that adenoviral vectors are capable of genome integration in a random manner [6]. Random integration of foreign DNA into human chromatin is a mutagenic phenomenon known as insertional mutagenesis. Mutations in the regions of the genome encoding tumor suppressor genes or proto-oncogenes, which play a key role in the regulation of the cell cycle and thus oncogenesis, could result in malignancies [8]. Though adenoviral vectors are known to remain extrachromosomal as an episome with a low propensity for integration into the host's genome [9], a minor integration of exogenous DNA fragments existing within the nucleus is inevitable [10]. The non-homologous end-joining (NHEJ) DNA repair system of cells can drive the heterologous recombination of the vector with the genome even though they lack an integration mechanism. Any delivery of exogenous DNA into the nucleus is associated with heterologous recombination [10]. In rodent models, genome integration within intergenic regions of hepatocytes has been observed following administration of recombinant adenoviruses [11–13]. There have also been similar reports of tumorigenesis following adenoviral vaccination in rodents [14,15]. In vitro studies in human and animal cells have also demonstrated the integration of recombinant adenoviruses into genomic DNA, primarily through heterologous recombination [11,12,15]. The conventional assays used to examine vaccine safety are typically based on biocompatibility properties [16]. However, these assays may fail to assess genetic toxicity and other long-term concerns. Insertional mutagenesis can lead to silent effects in the short term, and even in cases of carcinogenesis, it can take a considerable amount of time to become diagnosable [17]. Therefore, complementary genome integration evaluations are critical for assessing the risk of insertional mutagenesis in adenoviral vaccines.

Even in cases where vector integration rates seem negligible, a benefit-risk assessment is crucial to ensure the long-term safety and efficacy of this class of vaccines. Various countries have policies regarding the clinical use of gene therapies that discuss and recommend safety issues for each gene therapy approach. For example, the US FDA provides recommendations for the use of vector-based biologicals. According to this instruction, the probability of replication-competent virus production should be evaluated. When assessing the risk of genome integration, the FDA notes that some viral backbones are capable of genome integration. Therefore, long-term follow-up of subjects receiving gene therapy is necessary after ruling out short-term toxicities [18]. Despite the reported evidence of the integration of some adenoviral vectors, the FDA classifies adenoviruses as non-integrating vectors that do not require long-term evaluations. Yet, according to this classification, herpesviruses, gammaretroviruses, lentiviruses, transposon elements and genome editing tools are all capable of genome modification and require long-term monitoring [17]. According to the evidence discussed, there may yet be a need to evaluate adenoviral vaccines in more comprehensive genetic studies to assess their potential mutagenic properties and reassess their classification as non-integrating vectors. Although adenoviruses have a lower risk of integration, the potential for genome integration needs to be further assessed in human cell lines and vaccinated individuals. Genetic studies would aid in the risk evaluation of adenoviral vaccines and, together with immunologic studies, inform reliable risk-benefit assessments for this category of COVID-19 vaccines.