AAV-mediated gene transfer has great potential as a therapeutic approach [8]. Most of the currently developed AAV vectors are directed toward monogenic diseases, which belong to the category of rare diseases [25]. The FDA has approved the gene therapy products based on two viral vectors, which are both AAV vectors: LUXTURNA (Spark Therapeutics, Inc.) for the treatment of patients with confirmed biallelic RPE65 mutation-associated retinal dystrophy and ZOLGENSMA for the treatment of pediatric patients below two years of age having spinal muscular atrophy (SMA) with bi-allelic mutations in the survival motor neuron 1 (SMN1) gene. The use of recombinant AAV serotypes with unique tropisms to deliver cytotoxic therapy may also be considered a local antitumor therapy approach. EBV⁺ B-cells exhibit increased susceptibility to rAAV6.2 infection. Therefore, the introduction of a functional suicide gene into the rAAV6.2 genome could serve as a candidate vector for the development of rAAV-based oncolytic therapy targeting focal EBV-bearing B-lymphoproliferative disorders [26]. Intracranial interferon-beta (IFN-β) gene therapy based on the local administration of AAV vectors was reported to have successfully treated non-invasive orthotopic glioblastoma models and was also effective against migrating tumors [27]. The AAV vectors have several critical properties that could be exploited for gene delivery in cancer therapy [28].

AAV was initially discovered as a contaminant in adenovirus preparations [29]. The AAV genome comprises a single-stranded DNA approximately 4.8 kilobases (kb) in length. In addition, the AAV has a small (~25 nm) icosahedral capsid composed of three types of structural proteins, namely, VP1, VP2, and VP3 [30]. AAVs are replication-deficient parvoviruses, which have traditionally required co-infection with a helper adenovirus or herpes virus to achieve efficient infection [31]. Currently, the AAV-Helper-free system is used mostly in clinical research. On their own, AAVs are thought to be non-pathogenic and are yet to be concretely linked to any of the known human diseases. AAVs have at least 12 natural serotypes, each of which exhibits different tissue tropisms (Table 1). This is mainly because of the different affinities of these serotypes to an array of primary cell surface glycoprotein receptors and secondary receptors or coreceptors. For instance, heparan sulfate proteoglycan is thought to act as a primary receptor for AAV-2. The reported co-receptors for AAV include alpha V beta5 integrin, fibroblast growth factor receptor 1 (FGFR-1), and hepatocyte growth factor receptor (c-Met). The attachment of AAV-3 strain H relies on heparin, heparan sulfate, and FGFR-1 [32].

AAV Serotype	Tissue-Specific Tropisms	Key Pipeline	Disease	The Delivered Gene	Sponsor (s)	The Clinical Trial Stage
AAV1	Muscle, heart, skeletal muscle (including cardiac muscle), nerve tissue	Glybera	Lipoprotein lipase deficiency	Lipoprotein lipase	UniQure	Approved #
AAV2	Central nervous system, muscle, liver, brain tissue, eye	BIIB111	Hereditary ophthalmopathy	Rab escort protein 1	NightstaRx Ltd., a Biogen Company	Phase III completed, suspended
AAV3	Muscles, liver, lung, eye	N/A	N/A	N/A	N/A	N/A
AAV4	Central nervous system, muscle, eye, brain	N/A	N/A	N/A	N/A	N/A
AAV5	Lung, eye, central nerve, joint synovium, pancreas	BMN 270	Hemophilia type-A	Coagulation factor VIII	BioMarin Pharmaceutical	Approved
AAV6	Lung, heart	SB-525	Hemophilia type-A	Coagulation factor VIII	Pfizer	Phase III
AAV7	Muscle, liver	N/A	N/A	N/A	N/A	N/A
AAV8	Liver, eye, central nerve, muscle	BIIB112	X-linked retinoschisis	Pigmentosa GTPase regulator	NightstaRx Ltd., a Biogen Company	Phase III, suspended
AAV9	Heart, muscle, lung(alveolar), liver, central nervous system	PF-06939926	Duchenne muscular dystrophy	Truncated dystrophin	Pfizer	Phase III
AAV-DJ	Liver, retina, lung, kidney	N/A	N/A	N/A	N/A	N/A
AAV-DJ/8	Liver, eye, central nervous system, muscle	N/A	N/A	N/A	N/A	N/A
AAV-Rh10	Lung, heart, muscle, central nervous system, liver	LYS-GM101	GM1 gangliosidosis	beta-galactosidase	Lysogene	Phase II
AAV11	Unknown	N/A	N/A	N/A	N/A	N/A
AAV12	Nasal	N/A	N/A	N/A	N/A	N/A
AAV13	Unknown	N/A	N/A	N/A	N/A	N/A

AAV is a non-enveloped virus that may be engineered to deliver DNA to target cells. The virus genome is not integrated into the host cell but rather forms episomal concatemers in the host cell nucleus. These head-to-tail circular concatemers remain intact in non-dividing cells, such as neurons and cardiomyocytes, and are, therefore, capable of expressing transgenes over several months [33]. AAV vectors provide a relatively stable expression in dividing cells as well. The frequency of integration events may increase if an extremely high multiplicity infection is used or if the cell is infected in the presence of an adenoviral replicase. Recently, Dhwanil A et al. indicated that chromosomal integrations occurred at a surprisingly high frequency of 1–3% both in vitro and in vivo [34]. Moreover, according to recent research, high copy numbers of the AAV9 vector led to severe toxicity in animal models [35].

The AAV vectors are usually preferred for in vivo gene therapy due to several advantages, including the ability to transduce both dividing and quiescent cells, robust in vivo transduction efficiency, long-term transgene expression in quiescent cells, tropism for specific tissues and cell types, relatively low immunogenicity, non-pathogenicity, and a history of clinical safety.

Ad is a large and complex, non-enveloped, double-stranded DNA (dsDNA), icosahedral virus, which is 70 to 90 nm in size. Ad possesses an icosahedral protein capsid that accommodates a 26–45-kb linear, double-stranded DNA genome. Over 100 serologically different types of adenoviruses exist, among which 49 types infect humans [57,58]. According to their specific type, these viruses may bind to various cell surface proteins to facilitate their entry into the target cells [59]. As gene therapy tools, the high efficiency of Ads has resulted in over 450 protocols being approved so far for clinical trials [60].

Similar to AAV, Ad does not integrate into the host genome. Ad is the most efficient gene delivery system for a broad range of cell and tissue types. This is because most human cells express the primary adenovirus receptor and the secondary integrin receptors, such as Coxsackie and Adenovirus Receptor (CAR), CD46, and desmoglein-2 (DSG-2), as well as the glycans GD1a and polysialic acid [61,62]. Ad was the first DNA virus to enter rigorous therapeutic development, largely because of its well-defined biology, genetic stability, large transgene capacity (up to 36 kb), and ease of large-scale production. In addition, Ad leads to side effects that are considerably milder compared to chemotherapy [63,64,65]. Adenoviral vectors were initially used for brain cell transduction in the early 1990s [66]. The non-human canine adenovirus type 2 (CAV-2)-based vectors are capable of directing a gene to the neurons in the brain, spinal cord, and peripheral nervous system [67]. Adenovirus vectors may be divided into two groups: (1) replication-deficient viruses and (2) replication-competent, oncolytic viruses (OVs) [64]. The most commonly used adenoviral vector is the human Ad serotype 5, which is a common cold virus that circulates in humans with a seropositivity rate of 40–60% [68]. This virus has been rendered replication-defective through the deletion of the E1 and E3 genes [69]. The other contemporary Ad vectors have been derived from human adenovirus serotype 2 (HAd2) [70].

So far, three generations of adenoviral vectors have been developed. The first generation of Ad vectors was engineered by replacing the E1A/E1B region with transgene cassettes that could be up to 4.5 kb in length. These Ad viral vectors could induce high-level innate inflammatory responses within the first 24 h of transduction [71]. In the second generation of adenoviral vectors, the transgene capacity was enhanced further by additionally deleting the E2/E4 site, although the overall production yield remained low due to the decreased replication ability in producer cell lines [72]. The third-generation adenovirus vectors, also referred to as the helper-dependent or gutless adenovirus, have all of their viral sequences deleted, except for the ITRs and the packaging signal. The associated in vivo immune response in these viral vectors is highly reduced compared to the first- and second-generation adenovirus vectors, although high transduction efficiency and tropism are maintained [73].

Ad vectors are the most commonly used vectors in cancer gene therapy. Ad vectors are also used in vaccines to express foreign antigens [74]. Among all diseases, cancer remains the leading cause of death worldwide and accounted for nearly 10 million deaths in 2020 [75]. The overall risk accumulation is combined with the tendency for cellular repair mechanisms to become less effective as the individual grows older [76]. Cancer may be treated with surgery, radiation therapy, and/or systemic therapy (chemotherapy, hormone therapy, and targeted biological therapy). However, traditional treatments, such as surgery, may lead to side effects, including the inhibition of cellular immunity, reduction in the activity of natural killer cells, and reduction in the levels of anti-angiogenic factors [77,78,79]. Recently, viral vector gene therapy has received much attention as a novel treatment modality for cancer because of the flexibility and effectiveness it offers [80,81].

Most cases of cancer, when detected at an advanced stage, cannot be cured with traditional therapeutic modalities. Therefore, to improve tumor penetration and local amplification of the antitumor effect, oncolytic agents were developed, such as the conditionally replicating adenoviruses (CRAds). Viral infection in tumor cells results in the replication, oncolysis, and subsequent release of the virus progeny. Importantly, this replication cycle allows for a dramatic local amplification of the input dose. In theory, CRAds would replicate until all cancer cells are lysed [82]. On the other hand, similar to the other types of oncolytic virotherapy, oncolytic adenoviruses may, in addition to de-bulking the tumor, elicit powerful antiviral and antitumor immune responses. These viruses may transform a cold immunosuppressive tumor into one which is inflamed [83,84]. In other words, antitumor immunity is more important than direct oncolysis, as the former allows for the generation of tumor-specific memory T cells [65,85]. Consistent with this, the 2018 Nobel Prize in Physiology or Medicine was awarded for the discovery of cancer therapy based on the inhibition of negative immune regulation. Immune checkpoints (ICPs), in addition to controlling autoimmunity, play a key role in host defenses aimed at eradicating pathogenic microbes and microbial strategies, while also regulating the balance among tolerance, autoimmunity, infection, and immunopathology [86]. The antibodies targeting the T cell inhibitory checkpoint proteins, namely, cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), programmed cell death 1 (PD1) protein, and the PD1 ligand (PDL1), have been approved for the treatment of a variety of cancers, including melanoma, non-small-cell lung cancer (NSCLC), head and neck cancer, bladder cancer, renal cell carcinoma (RCC), hepatocellular carcinoma, and several types of tumors [87].

In addition, adenoviral vectors may be used in therapeutic cancer vaccines. These vaccine adenoviral vectors are capable of inducing both innate and adaptive immune responses in mammalian hosts [88]. An example is ETBX-011, which has been developed to treat patients with cancers that express the carcinoembryonic antigen [89]. Another example would be Ad-E6E7, which generates an enhanced immune response against HPV-positive tumors [90]. In particular, great progress has been achieved recently in utilizing the Ad-based vectors as a vaccine platform for HIV and cancer immunotherapy approaches as well as in the vaccination for other infections. The recent pandemic of coronavirus disease 2019 (COVID-19), which was caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has led to an unprecedented development of multiple vaccines. Among these vaccines, Ad-vectored vaccines are also playing important roles in the global vaccine efforts against COVID-19. Certain examples include Ad26.COV2.S, ChAdOx1 nCov-19, Ad5 nCoV, and Gam-COVID-Vac vaccines, all of which have demonstrated efficacy in protecting against symptomatic COVID-19 disease in humans [91]. Despite these successes, the innate and pre-existing immunity against Ad vectors remains a serious challenge in the development and application of these vectors [92]. Moreover, according to clinical records, the administration of an adenovirus serotype 5 (Ad5) vector in a gene therapy trial led to lethal systemic inflammation in the subject [93]. One approach that could be used to overcome this obstacle is to sequentially administer two or more antigenically distinct viruses. This approach would ensure that the specific immunity that arises after the administration of the first virus does not inhibit the therapeutic effects of the second virus [94]. In addition, various non-human Ad vectors have been considered for development. Anurag Sharma et al. reported that the non-human adenovirus (Ad) vectors derived from bovine Ad serotype 3 (BAd3) or porcine Ad serotype 3 (PAd3) could circumvent the pre-existing immunity against human Ad (HAd) [95].

Although adenoviruses are tissue-specific and flexible, an intravenous administration of these viruses may induce acute liver injury, as has been reported in animal models [96]. In comparison to AAV, Ad has a short duration of expression in vivo [97]. Ad vectors have been studied in rodents, primates, and humans, and variable results have been achieved, which highlights the necessity for further detailed investigations on the natural history of Ad infection in humans and for questioning the value of animal models in determining the safety of virus vectors [96]. However, developing an ideal model that mimics the human infection remains the key focus of biomedical research.

LVs belong to the orthoretroviridae subfamily of the genus retroviruses [98]. LVs may be divided into two major classes—primate and non-primate LVs [99]. The morphology and genome organization of all LVs are similar in several aspects: all LVs are pleomorphic spherical-shaped particles with diameters of approximately 100 nm [100], containing a diploid genome comprising two single-stranded positive-sense RNA molecules. LV vectors typically included the following required elements: 5′ long terminal repeat (LTR) through the Ψ packaging signal, central polypurine tract/chain termination sequence (cPPT/CTS), Rev responsive element (RRE), and 3′ LTR, including the poly (A) signal [101]. The classification and the specific structures of different LVs have been detailed in a previous report [9].

Currently, four generations of lentiviral vectors have been developed. The first-generation lentiviral vectors contained a significant portion of the HIV genome and exhibited a high frequency of transfer of genetic material into the host cell [102]. The lentiviral accessory genes vif, vpr, vpu, and nef, and LV regulatory genes tat and rev, were included in the first-generation LV vectors [102]. In order to achieve further safety, the second-generation LV vectors were developed by removing vif, vpr, vpu, and nef, which used to be present in the first-generation of LV vectors, as these are not necessary for the transfer of genetic material to the host cell [103]. The third-generation LV vectors are considered to be replication-incompetent and self-inactivating vectors. In this generation, the viral tat gene, which is essential for the replication of the wild-type human immunodeficiency virus type 1 (HIV-1), has been deleted. In addition, the vector packaging functions have been separated into three separate plasmids rather than two plasmids to reduce the risk of recombination during plasmid amplification and viral vector manufacture. An altered 3′ LTR renders the vector “self-inactivated”, which prevents the integrated genes from being repackaged. A heterologous coat protein [e.g., vesicular stomatitis virus G protein (VSV-G)] is used in place of the native HIV-1 envelope protein, and such vectors allow for the infection of a broad range of host cell types. These reasons render the third generation of LV vectors safer than the second generation LV vectors, which has allowed for the widespread application of the former [104]. The third-generation LV vectors generate virus particles using four plasmids and a producer cell line. The rationale behind including four plasmids is to enhance safety, as separating genetic components reduces the chances of recombination [105]. However, homologous recombination between the constructs remains possible nevertheless since the RRE sequence and a part of the packaging sequence in the gag gene are present in both transfer and structural packaging constructs. In order to resolve these concerns, the RRE sequences were replaced with heterologous sequences that have a similar function and do not require the REV protein. Another approach to resolving the above-stated issue is based on codon optimization. These described solutions led to the emergence of the fourth generation of LV vectors. However, the titers had been affected in the fourth-generation LV vectors, which has limited their extensive application [106].

LVs offer several potentially unique advantages over traditional gene delivery systems. Unlike adenoviral or adeno-associated vectors, neutralizing antibodies are rarely generated against lentiviral vectors [107]. The most important advantage of LV vectors is their ability to provide long-term and stable gene expression, which is crucial for adolescents or pediatric patients; LV vectors are capable of infecting dividing/non-dividing cells, such as neurons [108] and osteocytes [109], and due to their relative low-immunogenic characteristics [110], LV vectors may incorporate constructs up to 9~10 kB in size [107,111].

LV vectors are mainly used in ex vivo gene therapies, such as the one for B-cell acute lymphoblastic leukemia (B-ALL) [112]. B-ALL is a clonal malignant disease that originates in a single cell. B-ALL is characterized by the accumulation of blast cells that are phenotypically reminiscent of the normal stages of B-cell differentiation [113]. B-ALL remains a leading cause of non-traumatic death in children, and most adults diagnosed with it also succumb to the disease [114]. CAR-T therapy has successfully achieved extraordinary clinical outcomes in the treatment of B-ALL [115]. In order to develop CAR T-cells ex vivo, LVs appear particularly appealing due to their ability to stably integrate relatively large DNA inserts [116]. In the CAR T-cell therapy, the patient’s T-cells, either CD4+ or CD8+, are isolated and activated prior to transduction. The CAR transgene is then delivered into the activated T-cells via LV vectors and then expanded. Finally, the produced CAR T-cells are formulated in an adjusted buffer in a defined ratio of CD4⁺:CD8⁺ CAR T-cell [116,117]. A complete explanation of engineered CAR-T cells in cancer immunotherapy would not be provided in the present report and one could refer to other reports for the same [118,119,120]. The emergence of CAR-T cell therapy has paved a new way for cancer treatment. In 2017, Novartis received its first FDA approval for a CAR-T cell therapy, Kymriah (TM) (CTL019), for children and young adults with B-cell ALL that is refractory or has relapsed at least twice. In 2019, EMA approved Zynteglo, a medicine used for the treatment of patients aged 12 years and older with transfusion-dependent β-thalassemia (TDT) who do not have a β0/β0 genotype and for whom hematopoietic stem cell (HSC) transplantation is appropriate although a human leukocyte antigen (HLA)-matched related HSC donor is not available. Zynteglo (betibeglogene autotemcel) is a genetically modified autologous CD34+ cell-enriched population that contains HSCs transduced with the LV vector encoding the βA-T87Q-globin gene. LV vectors have also been demonstrated as efficient gene transfer vehicles for human solid tumor cells, such as ovarian cancer cells [121], prostate cancer [122], and hepatocellular carcinoma [123].

The major concerns associated with LV vectors-based gene therapy include the possible generation of replication-competent LVs during vector production, mobilization of the vector by endogenous retroviruses in the genomes of patients, insertional mutagenesis that may lead to cancer, germline alteration resulting in trans-generational effects, and dissemination of new viruses from the gene therapy patients [108]. LV vectors typically insert into the host DNA as a single non-rearranged copy, and while these vectors exhibit improved stability and durability, the random insertion method nonetheless has the risk of activating the cancerous gene in the genome. Several products of Bluebird that are based on lentivirus vectors have led to such events in the clinical stage. For instance, a patient was detected with myelodysplastic syndrome [124], while another patient had developed acute myeloid leukemia after treatment with LentiGlobin gene therapy [125]. In CAR T-cells therapy for ALL patients, serious although manageable adverse events, including B-cell aplasia, tumor lysis syndrome, and cytokine release syndrome, have been reported. Basically, the use of non-integrating LVs (NILVs) reduces insertional mutagenesis and the risk of malignant cell transformation due to the integration of the lentiviral vectors [126]. As stated earlier, the usage of VSV-G alters the native tropism of lentiviral vectors to allow for the infection of a broad range of host cell types [127,128,129], which implies targeting such viruses to particular cell types is challenging due to non-tissue-specificity [129,130].

However, the observed differences could have been due to the differences in vector design, final formulation, immunomodulatory regimens (transient around vector administration), and surgical approach, among other reasons. With extensive and detailed studies on LV vectors over the past few years, this platform has been used widely in both research and clinical trials. Although certain problems remain to be addressed, the safe and efficient LV vectors are nonetheless considered promising as a tool for human gene therapy.