Clinical aspects

Description

	• The broad scope of gene-based therapies covers gene transfer, gene editing, and gene silencing.
	• Antisense oligonucleotides and RNA interference are included in the gene-based therapies category.
	• Cell-based therapies belong to a stand-alone therapeutic category, which is not completely covered under the scope of this article; genetically engineered cells for therapeutics purposes will be discussed briefly.
	• Gene-based therapies can be performed ex vivo or in vivo, and several techniques are used.
	• Viral and nonviral vectors are used to deliver gene-based therapies. • Approved gene-based therapies for neurologic disorders are limited to monogenic neurogenetic disorders that encompass neurodegenerative, neuromuscular, and neurometabolic conditions.
	• Potential applications of gene-based therapies for non-genetic neurologic disorders are under pre-clinical and clinical development.

A simplified classification of various gene therapy methods is shown in Table 2.

Table 2. A Simplified Classification of Gene-Based Therapy Delivery Techniques

Chemical or physical methods
	• chemical: calcium phosphate transfection • physical • electroporation • gene gun • focused ultrasound
Viral vectors
	• retroviruses, eg, Moloney murine leukemia virus • adenoviruses • adeno-associated virus • herpes simplex viruses • lentiviruses • vaccinia virus • recombinant viruses • other viruses
Nonviral vectors
	• liposomes • ligand-polylysine-DNA complexes • nanoparticles, eg, dendrimers • synthetic peptide complexes • artificial viral vectors • artificial chromosomes
Genetically modified microorganisms as oncolytic agents
	• genetically modified viruses • genetically modified bacteria
Cell or gene therapy
	• administration of cells modified ex vivo to secrete therapeutic proteins in vivo • implantation of genetically engineered cells to replace dysfunctional cells • genetic modification of stem cells--stem cell therapy
Gene or DNA administration
	• direct injection of naked DNA or genes: systemic or at target site • receptor-mediated endocytosis • use of refined methods of drug delivery: eg, microspheres
Gene regulation
	• regulation of expression of delivered genes in target cells by locus control region technology • molecular switch to control expression of genes in vivo • promoter element-triggered gene therapy
Repair or editing of genes
	• correction of the defective gene in situ
	• transcription activator-like effector nucleases (TALENs): restriction enzymes that can be engineered to cut specific sequences of DNA for gene editing
	• zinc finger nucleases (ZFNs): engineered DNA-binding proteins for targeted editing of the genome by creating double-strand breaks in DNA at user-specified locations
	• gene editing, ie, altering the genomes of living cells by adding, changing, or deleting nucleotides
	• clustered regularly interspaced short palindromic repeats (CRISPR)
Gene replacement
	• excision and replacement of the defective gene by a normal gene
RNA gene therapy
	• RNA trans-splicing • RNA editing
Inhibition of gene expression
	• antisense oligonucleotides • RNA interference • ribozymes • tRNA modification

Gene transfer to human patients may be ex vivo or in vivo.

Ex vivo gene therapy. Ex vivo gene transfer techniques usually involve the genetic alterations of cells (cell lines or human cells), mostly by use of viral vectors, prior to implanting these into the tissues of the living body.

Advantages: • Usually more efficient than in vivo methods. • Allows for extensive testing of the transduced cells to mitigate safety concerns. Disadvantages: • Reimplantation of genetically engineered cells grown in culture may not result in long-term survival of a large portion of the cells. • Limited diffusion capacity for large organs (eg, skeletal muscle). • Potential for immunogenicity.

In vivo gene therapy. In vivo gene therapy means direct introduction of genetic material into the cells or tissues within the patient’s body. It can be accomplished by viral or nonviral vectors that are modified to enhance delivery efficiency to target locations in the CNS or PNS. In vivo gene delivery may be local (in situ) or systemic:

	• In situ gene therapy administration means the introduction of genetic material directly into a localized area in the human body, for example, intraocular subretinal injection of AAV2-PRE65, voretigene neparvovec (Luxturna®) to treat Leber congenital amaurosis.
	• In systemic gene therapy, the therapeutic molecule is administered intravenously and can reach its final site of action, for example, the systemic administration of AAV9-SMN1, onasemnogene abeparvovec (Zolgensma®) to target the anterior horn cells in the spinal cord to treat spinal muscular atrophy.

In vivo gene delivery usually combines specific targeting systems (eg, viral vectors with specific tropism) with transcription switches (eg, tissue-specific promoters for the transgene). Depending on the disease, combinations of in vivo gene therapy may be necessary to target the CNS (eg, via intrathecal delivery) and systemic tissues (eg, via intravenous administration).

Various methods of gene transfer as applied to the nervous system are shown in Table 3.

Table 3. Methods of Gene Transfer as Applied to Neurologic Disorders

Direct methods of gene transfer
Viral vectors
	• non-neurotropic viruses: adeno-associated viruses, adenoviruses, lentiviruses
	• neurotropic viruses: herpes simplex and rabies
Nonviral vectors
	• naked DNA
	• liposomes
	• polymer-based vectors
	• nanoparticles
	• extracellular vesicles
	• ultrasound
	• human artificial chromosomes
Indirect methods of gene transfer
Transplantation of genetically engineered cells
	• neuronal cells
	• non-neuronal cells

COMMONLY USED VECTORS FOR GENE-BASED THERAPIES

DNA and RNA are biological macromolecules with large negative charges. This quality leads to considerable extracellular and intracellular barriers during cell uptake and intracellular release that can dramatically influence their efficiency as therapeutic agents (125). Naked DNA and RNA are easily and rapidly degraded by various nucleases, macrophages, and the reticuloendothelial system in the blood and can activate the innate immune system (eg, by activating TLR receptors). The half-life of DNA and RNA in vivo is typically in the range of minutes (76). Thus, they need to be protected to be able to reach their intended target cells. The delivery vehicle or carrier (ie, vector) can be viral (22) or nonviral (125). Chemical modifications of DNA and RNA molecules can also provide nuclease resistance and improve their stability. In addition, conjugation of peptides, lipids, or antibodies to these molecules can improve their selective cellular uptake and targeting.

Viral vectors

Viral transduction involves infection of the cell with a modified virus and introduction of a desired cargo. In gene transfer using viral vectors, the genes required for replication are removed and replaced by the desired promoter, therapeutic genetic sequence, and stabilizing sequences, such as a poly-A tail. The main classes of viral vectors that have been tested for clinical applications include adenoviruses, adeno-associated viruses, retroviruses, lentiviruses, and herpes simplex viruses (45).

Adeno-associated viruses. The development of novel therapeutic strategies for neurologic disorders using AAV vectors has had a striking impact on gene therapy research. AAVs are non-enveloped, single-stranded DNA nonpathogenic parvoviruses. These viruses are not commonly associated with any human diseases. An exception is an outbreak of acute hepatitis of unknown etiology in children in 2022 that was associated with AAV2 infection (56). AAVs are usually used for targeting nondividing cells in tissues such as the liver, nervous system, eye, and skeletal muscle (135). They are subclassified based on amino acid homology and serologic reactions, with 11 naturally occurring serotypes and more than 100 variants. Several artificial serotypes have also been engineered to enhance their tropism for specific tissue and cell types. For example, AAV9, which is a naturally occurring serotype, shows efficient transduction in the nervous system with intraparenchymal injection and more widespread expression in the brain and spinal cord after systemic or intrathecal delivery. Another notable example is MyoAAVs, which are variant AAVs identified by directed evolution in animal models that markedly enhance the skeletal muscle tropism of the vector (120). These variant AAVs hold promise as a safe, efficient, and specific delivery method for treating skeletal muscle diseases (80). Although 80% of humans have pre-existing neutralizing antibodies against AAV strains, AAV has been shown to be less immunogenic than other viruses (54). Both the capsid proteins and the transgene protein product can trigger the innate or adaptive immune systems, including the production of neutralizing antibodies or a T cell-mediated response that could diminish the clinical efficacy of gene therapy (93).

Advantages
	• neuronal gene expression efficient • long-lived • stable • nontoxic and less immunogenic (compared to other viral vectors) • the transgene can exist in a free form within the cells for a prolonged time • does not usually lead to insertion mutations in the host genome (135).
Challenges
	• complicated and expensive production process • immune response to the capsid and the transgene • significant limitation is the packaging capacity (less than 5 kb). However, cell transduction with multiple AAV vectors applying protein trans-splicing mediated by split-inteins is a promising method to overcome this challenge and enable the reconstitution of large full-length proteins (123).

There are several FDA-approved recombinant AAV (rAAV) gene transfer products in clinical use: etranacogene dezaparvovec (Hemgenix®, rAAV5, to treat adults with hemophilia B), voretigene neparvovec (Luxturna®, rAAV2, to treat patients with biallelic RPE65-associated retinal degeneration), valoctocogene roxaparvovec (Roctavian™, rAAV5, to treat adults with severe hemophilia A), onasemnogene abeparvovec Zolgensma®, rAAV9, to treat spinal muscular atrophy), delandistrogene moxeparvovec (Elevidys, to treat children 4 to 6 years of age diagnosed with Duchenne muscular dystrophy).

Adenoviruses. Recombinant adenoviruses were studied extensively for gene transfer. They are typically genetically modified by introducing deletions in the viral genome to create space for the desired transgene to be inserted (up to 48 kb) and to create a replication-defective virus. Due to the challenges of their use, they have yet to be extensively pursued in later stages of clinical development.

Advantages
	• can be used to transduce both nondividing and dividing cells • integrate at very low rates into the host cell chromosomes • suitable for directing transient expression • broad tropism • large space (up to 48 kb) for carrying foreign genetic material
Challenges
	• widely pre-existing viral immunity • adenoviruses may trigger strong and life-threatening inflammatory responses • pre-existing neutralizing antibodies could also hamper their efficiency • potential direct cytotoxicity.

Retroviruses. Retroviral vectors, specifically Moloney murine leukemia virus and its related viruses, have one of the longest pedigrees in studies of gene transfer therapy, particularly for ex vivo gene therapy. They demonstrate high efficiency of target-cell transduction and are prevented from replication by deletion of necessary viral genes. The ability to integrate selectively into dividing cells with persistent gene transfer has made them excellent potential vectors for the delivery of "killer" genes to brain tumors.

Advantages
	• high efficiency of retrovirus-mediated gene transfer • ability to integrate into dividing cells • persistent gene expression.
Challenges
	• the unpredictable site of chromosomal DNA insertion imposes a risk for insertional mutagenesis and oncogene activation • infect only mitotic cells, which precludes transferring genes to postmitotic cells, including neurons • size limit of 8 to 10kb.

Lentiviral vectors. Lentivirus is a type of retrovirus that contains a single-stranded positive-sense RNA that is transcribed into DNA in the cell (81). Lentiviral vectors have emerged as one of the preferred tools for ex vivo transgene delivery for gene therapy, and new modifications may make them suitable for in vivo therapy as well. They provide an advantage over AAVs with higher capacity for the gene of interest (approximately 10 kb).

Advantages
	• transduce postmitotic cells (whereas other retrovirus-based vectors require active cell division for infection) • strong neural tropism • integrating vectors that permit long-term transgene expression (when desired) • ability to express multiple genes from a single vector • elicit relatively weak immune responses • safer profile and reduced genotoxicity compared to other retroviruses.
Challenges
	• complex large-scale good manufacturing practice • risk of insertional mutagenesis • may activate or inactivate neighboring genes • may cause aberrant splicing of cellular transcripts.

Betibeglogene autotemcel (Zynteglo^TM) is an FDA-approved product that consists of an autologous CD34+ cell-enriched population containing the patient’s own hematopoietic stem cells transduced ex vivo with a lentiviral vector encoding the beta-globin gene to treat adult and pediatric patients with beta-thalassemia. In September 2022, the FDA also granted approval for elivaldogene autotemcel (Skysona®) for the treatment of cerebral adrenoleukodystrophy, which adds functional copies of the ABCD1 gene into patients’ hematopoietic stem cells through transduction of autologous CD34+ cells with a lentiviral vector. Atidarsagene autotemcel (Libmeldy™), which is approved in certain European countries to treat metachromatic leukodystrophy, is another gene therapy construct based on autologous hematopoietic stem cells transduced with a lentiviral vector.

Herpes simplex virus. Herpes simplex virus-1 is a neurotropic virus. Following infection at the periphery, the virus spreads preferentially in the nervous system. Herpes simplex virus-1 amplicon vector is a plasmid-based vector that has been successfully used to transduce and express various genes in somatic cells, particularly postmitotic neurons. Such vectors have potential for gene therapy especially with local delivery approaches.

Advantages
	• neurotropic virus • vector engineering allows long-lasting transgene expression (when desired) • remains extrachromosomal (reduces the risk of insertional mutagenesis) • excellent capacity of more than 30 kb foreign DNA (engineered herpes simplex viruses virions can carry 150 kb of foreign DNA).
Challenges
	• human pathogen, conveying the risk for fatal meningoencephalitis, especially in immunocompromised persons (83) • strong cytopathogenicity.

Topical use of this vector mitigates the risks associated with systemic administration. This is exemplified by the safety profile of beremagene geperpavec (Vyjuvek^TM), an FDA-approved, replication-deficient HSV-1-based vector therapy that has been genetically modified to express the human type VII collagen (COL7) protein to treat wounds in patients with dystrophic epidermolysis bullosa.

Other viral vectors. Other viral vectors with different tropisms and qualities to circumvent the disadvantages of current vectors are under active study. Examples include nonpathogenic retroviruses with broad tropism (eg, foamy viruses) (104) and Semliki Forest viruses. These studies may uncover previously unappreciated naturally occurring or artificial serotypes that can help improve the efficiency and safety of viral gene delivery.

Nonviral vectors

Alternative methods of gene delivery using nonviral vectors have been actively pursued to circumvent the limitations and challenges associated with viral vectors. These include cost, manufacturing capacity limitations, regulatory and safety concerns, cytotoxicity, immunogenicity, and related mutagenesis. Nonviral vector approaches enable the delivery of various molecules other than DNA, such as RNA and protein, without strict size limitations. Examples include packaging and delivery of messenger RNA, antisense oligonucleotides, small interfering RNA, and CRISPR-Cas ribonucleoprotein complex (130). They can also be engineered and selected with specific cell types and tissue tropisms.

Nonviral vectors primarily include liposomes and lipid nanoparticles, cationic polymers, inorganic nanoparticles, exosomes, polymer hydrogels, and virus-like particles (125). Novel compounds, chemical modifications, and conjugation strategies are constantly proposed and investigated to improve nonviral vector efficiency and safety. The most commonly used and promising nonviral vector delivery strategies are discussed in this section.

Lipid-based vectors (liposomes) for gene therapy. Liposomes protect their cargo from external degradation, and their similarity to biological membranes enables the delivery of their cargo into the cells or their intended sub-cellular compartments (43). The successful encapsulation of whole virus or DNA into liposomes opened exciting possibilities by enhancing DNA introduction into mammalian cells. Systemic administration of a synthetic liposomal vector that targets the CNS cells could, in principle, provide an efficient tool for gene therapy of neurologic disorders. The performance and tropism of liposomes can be improved by modifying their physical and chemical characteristics, including their charge, size, presence of ester bonds, chain length, and nature of ligand complexation (137).

One example is cationic liposomes. Cationic liposome-DNA complexes are positively charged and bind easily to the negatively charged cell surface with highly efficient transfection. The nucleic acid cargo molecule is not encapsulated but is simply covered with small unilateral vesicles by electrostatic interactions and, following the transfection, is released in the cytoplasm. Another example is the Trojan horse liposome technology, using conjugates of monoclonal antibodies that bind to specific endogenous receptors, allowing the vector to target specific areas (eg, cross the blood-brain barrier) (16). This facilitates receptor-mediated transcytosis of the conjugated vector through the blood-brain barrier and endocytosis into specific cells in the CNS. Trojan horse liposomes may be targeted to multiple receptors in the brain to enhance the precision of delivery to subpopulations of cells implicated in different diseases (96).

Polymer-based vectors. Cationic polymers have been an important nonviral gene therapy vector system. Their versatile chemical structure and potential high-loading capacity make them an appealing nonviral vector system for drug delivery (136). Poly-L-Lysine and polyethyleneimine are cationic polymers that were tested for in vivo CNS gene therapy delivery (92; 112). Poly-L-Lysine has the advantage that different ligands can be coupled with its primary amino groups, and it is biodegradable. Polyethyleneimine has been shown to provide exceptionally high levels of transgene expression in the mouse brain. However, it suffers from insufficient target specificity. In addition, it is a nonbiodegradable polymer that accumulates around the cell and can trigger cytotoxicity (136). Poly-L-Lysine and polyethyleneimine cytotoxicity is directly related to their molecular weight and pKa, with higher molecular weight and more cationic materials being more toxic (89).

Dendrimers. Dendrimers are branched, three-dimensional macromolecules creating spherical, highly branched polymers (63). Polyamidoamine is the most common form of a dendrimer in biological applications and is the first to be used for gene delivery. High cytotoxicity results from dendrimer surface charge and chemical structure and can be alleviated by polyethylene glycol modification (PEGylation) (58).

Nanoparticles. Nanoparticles, organic or nonorganic, are tiny particles in the nanometer size range (1 to 100 nanometers) that can be engineered to carry and deliver isolated genetic material, such as DNA and RNA, as well as nucleotide and protein complexes into target cells. Nanoparticles can be used as nonviral vectors for efficient in vivo gene delivery without significant toxic effects and with efficacy equaling that of viral vectors. Nanocarriers such as liposomes, metallic and polymeric nanoparticles, dendrimers, gelatins, and quantum dots or rods have been developed, and each shows distinct characteristics (131). The versatility of nanoparticles allows the modification of their surface during production to facilitate their interaction with different cell components and facilitate their entry into the blood-brain barrier to release drugs in the CNS (23). Pharmacokinetics and neurotoxicity of a nanocarrier should be assessed when selecting an appropriate vector for gene therapy. Lipid-based nanoparticles are an emerging drug delivery system approved by the FDA for liver siRNA delivery and mRNA vaccine delivery (COVID-19). Developments are underway to improve their cellular tropism and to safely target nonhepatocyte cells (99). Liposomes and lipid nanoparticles that facilitate the delivery of CRISPR-based genome editing machinery are already in clinical trials (82).

Other nonviral systems.

Ultrasound-mediated gene delivery. Various proteins and genes can be directly delivered using sonoporation. This technique utilizes ultrasound waves to induce temporary pores in the cell membrane, allowing for the delivery of genetic material into cells (118).

Human artificial chromosomes. Long synthetic arrays of alpha satellite DNA are combined with genomic DNA to generate artificial chromosomes. The resulting linear chromosomes contain exogenous alpha satellite DNA and contain all the sequence elements required for stable mitotic chromosome segregation and maintenance. A functional human artificial chromosome could serve as a valuable nonviral gene transfer vector.

The ideal vector for gene therapy of neurologic disorders. The vector has a crucial impact on efficacy, durability, and treatment safety. Viral vectors continue to play an important role in genetic manipulations of neuronal and non-neuronal cells with several nonviral delivery approaches also in development. With four decades of accumulated knowledge and experience with viral vectors, the list of promising viral vector-based strategies to treat genetic diseases is growing (22). Drawbacks include high price, large-scale manufacturing challenges, severe side effects (eg, immunogenicity, which precludes re-dosing), potential carcinogenicity, poor target cell specificity, and their inability to transfer large-sized genes because of limited capacity.

The relative inaccessibility of the brain also presents a problem with the systemic administration of a vector that may not cross the blood-brain barrier. This is also important for nonviral vectors. This limitation can be overcome by nonsystemic administration routes, such as intrathecal, intraocular, intranasal, or direct intraparenchymal administration. The ideal neuronal gene transfer vector should have the following primary characteristics:

	• It should be capable of infecting post-mitotic neurons or disease-relevant supportive cells with a high degree of efficiency.
	• It should carry an expression cassette capable of mediating the expression of foreign genes of various sizes without impairing its cellular physiology.
	• It should be relatively easy to construct and purify in a high titer.
	• It should not present a safety hazard for the patient.

Targeted gene therapy. The ability to target gene transfer, gene editing machinery, or nucleic acid therapeutics to a specific subpopulation of cells is an important aspect of improving the efficacy of gene-based therapies and limiting undesirable side effects. Ideal gene therapy vectors would be delivered intravenously but only transduce specific disease-relevant target cells (specific tropism). When using viral vectors, this can be achieved by developing special viral serotypes (eg, MyoAAV) or by conjugation of specific proteins to the viral capsid that enables specific receptor-mediated cell uptake and escape from processing by the liver or other organs. When using nonviral vectors, chemical modifications and an array of conjugation strategies are often proposed to enable targeting specificity.

Controlled induction of gene expression. Several inducible gene expression systems have been developed over the past decade to meet the need for regulated gene expression therapies. These methods use elements within the vector design to enable an external drug or pro-drug to alter ongoing protein expression after in vivo gene transfer. Various methods of controlling gene or protein expression are shown in Table 4.

Table 4. Approaches to Controlling in Vivo Gene Expression in Gene Therapy

• Pharmacological control of gene expression
	- antibiotics: tetracycline - pharmacological rheostat (gene switch): dumbbell shaped molecules
• Pharmacological control of signal transduction pathways
• Use of transcriptional control in expression systems that can be regulated
	- binary system for toxin gene therapy - alteration of splicing of the transgene
• Manipulation of gene regulation by:
	- cytokines - heat shock proteins - radiation - ultrasound
• Gene switch system to control in vivo expression

Transgene promoter. An ideal promoter of a therapeutic gene would mimic its normal regulation; it should also be tissue-specific and even cell-specific. The ability to maintain gene expression in specific cell types within the brain is a fundamental requirement for CNS gene therapy. Ideally, the transgene expression should replicate the dynamic endogenous expression of the wild-type gene in healthy human tissue. Too much gene expression or overexpression in the wrong cells can be deleterious and lead to cell- or immune-related toxicities. The goal is to achieve the right level of expression in enough target cells to provide maximal clinical benefit without the risk of overexpression-associated toxicities (91). Long-term expression of genes that are normally downregulated during neurodevelopment and aging is another concern, with limited long-term studies conducted so far. Thus, in situations where the regulation of the quantity of a transgene protein product is crucial, using an externally regulated (inducible) promoter or enhancer unit might be necessary. Additional genomic elements, such as enhancers, introns, polyadenylation sequences, and transcript-stabilizing elements, may be added to tune the transgene expression to enable a regulatable gene therapy approach, though their use is limited by the packaging capacity of commonly used viral vectors. As an example, an engineered microRNA-based regulatory element (miRARE) can be used to control dose-sensitive genes. This has been demonstrated in mouse models of Rett syndrome through the insertion of miRARE into the miniMECP2 gene expression cassette of AAV9/MECP2 vector (117).

Gene therapy monitoring. Gene therapy monitoring is important to ensure target engagement and inform future similar studies to improve safety and efficacy. Limited data are available regarding the long-term safety and efficacy of gene therapy. Thus, in addition to long-term clinical monitoring, when feasible, autopsy studies of gene therapy-treated patients and research participants can provide important information to help improve the design, safety, and efficacy of future therapies.

PET imaging enables imaging of target molecules in vivo and numerous tracers are available for imaging of reporter gene activity. The selection of radiolabeled substrates that interact with specific transgene proteins has identified several reporter genes that can be used for imaging vector-mediated gene delivery and expression in preclinical and clinical settings (27). Alternative approaches include measuring the levels of the transgene protein product (eg, by mass spectrometry in CSF). Measurement of gene transcription by RNA-sequencing can be pursued in some target tissues (eg, skeletal muscle) but is more challenging in CNS-related disorders due to the inaccessibility of CNS tissue and low levels of RNA in CSF.

Cell-mediated ex vivo gene therapy. This technique involves the genetic manipulation of cells followed by their in vitro amplification and subsequent injection into target tissues. For human gene therapy, the success of cell-mediated methods depends on the capacity of cells to proliferate during the in vitro amplification step. One of the limitations is the phenomenon of senescence of diploid cells after serial divisions. Several types of cells have been explored for cell-mediated gene therapy: hematopoietic cells, fibroblasts, myoblasts, keratinocytes, hepatocytes, neural cells, etc. Genetic modification of stem cells has been proposed as a treatment strategy for a variety of diseases. Autologous pluripotent stem cells, mesenchymal stem cells, and neural stem cells are most often used. The synergy between the regenerative effects of stem cells and the introduction of other desired properties by genetic engineering can provide significant benefits for neurodegenerative changes in the brain. Autologous transplantation of corrected cells after transduction with gene therapy vectors is an appealing option for decreasing the procedure risks and enhancing engraftment efficacy due to immune system activation (17). In addition, genetically engineered cells may be implanted for systemic delivery of recombinant proteins. Encapsulation techniques that usually involve surrounding the cells with protective and selectively permeable membranes can facilitate the delivery of non-autologous genetically engineered cells to deliver recombinant gene products, including neurotrophic factors. The pores of the membranes should be small to block entry of immune mediators but large enough to allow inward diffusion of oxygen and nutrients required for cell survival and for outward diffusion of neuroactive molecules produced by the cells.

Stem cells. Neural stem cells may present an ideal route for gene therapy in addition to offering new possibilities for the replacement of neurons lost to injury or disease. Elivaldogene autotemcel (Skysona®) is an FDA-approved autologous hematopoietic stem cell-based gene therapy to treat active cerebral adrenoleukodystrophy. The product is made of autologous CD34+ hematopoietic cells transduced with the Lenti-D lentiviral vector with functional copies of the ABCD1 gene. Following intravenous infusion of the product, transduced CD34+ hematopoietic stem cells engraft in the bone marrow and differentiate to produce various cell types, including monocytes (CD14+), capable of producing functional adrenoleukodystrophy protein, which is needed for degradation of very long chain fatty acids.

Neural stem cells. Neural stem cells propagated in culture can be reimplanted into mammalian brain or spinal cord, where they might integrate into the tissue and stably express foreign genes. Neural stem cells might aid in reconstructing maldeveloped or damaged CNS at the molecular and cellular levels. As an example, this approach has been investigated in preclinical and early clinical studies of amyotrophic lateral sclerosis.

Gene editing. Gene editing refers to techniques used to make precise DNA (or RNA) sequence changes, such as modifying, adding, or deleting specific nucleotides. DNA editing machinery such as CRISPR-Cas9, TALENs, and zinc-finger nucleases are commonly used in gene editing to target specific sequences and make alterations in a controlled manner in vitro. These methods can potentially treat monogenic disorders and are in pre-clinical development or different stages of clinical trials. The first-generation nuclease-dependent gene editing techniques (ZFNs, TALENs) were followed by CRISPR/Cas9 and second- and third-generation machinery to enable base editing (cytidine base editors and adenine base editors) and template-based editing (prime editing) in addition to other nuclease-free genome editing techniques and epigenetic editing methods (10). Optimization of the CRISPR machinery efficiency and its delivery can help enable a platform-based approach to gene editing. In this model, the same editing machinery is paired with specifically designed guide RNA or editing templates to target different genetic loci implicated in monogenic or more common neurologic diseases.

CRISPR-mediated gene editing. The CRISPR (clustered regularly interspaced short palindromic repeats)-Cas9 system can be used to precisely target the genome in living cells (28). It was first created by modifying bacterial proteins that normally defend against viral invaders. The naturally occurring bacterial protein-RNA complexes recognize and snip viral DNA. Thus, DNA-editing complexes were created that utilize the CRISPR bound to the Cas9 nuclease with the potential to edit any location in the genome (71). Short RNA sequences (single guide RNA, sgRNA) help target the CRISPR-Cas9 complex to specific locations in the genome and direct the Cas9 enzyme to cut the DNA (66). Next, when a double-strand break is introduced into the genomic DNA, cells repair the break. If no repair template is provided, the cells will use the nonhomologous end-joining DNA repair mechanism, an error-prone approach that commonly introduces insertion or deletion variants at the site of repair. Alternatively, using a DNA molecule with high sequence similarity as a template provides the means to incorporate the desired DNA sequences (homology-directed repair), though this process can only occur in mitotic cells and is not very efficient.

Several engineering efforts have aimed to improve CRISPR-Cas9 specificity and efficiency (01; 37). Promising results from pre-clinical studies in iPSCs or mice utilizing the CRISPR-Cas9 system to treat CNS diseases such as retinitis pigmentosa (49), Alzheimer disease (50; 97), amyotrophic lateral sclerosis (126; 32), Huntington’s disease (34), and more (48) demonstrate the feasibility and enormous potential of CRISPR-Cas9 system in treating neurologic disorders.

Gene editing of neural stem cells enhances their targeting ability, with the potential for cell or gene therapy of neurodegenerative disorders and lysosomal storage disorders (29).

In addition to Cas9, other enzymes can be targeted to specific genomic locations using the CRISPR system. Novel biotechnological and engineering approaches, eg, base editing or prime editing, hold promise in improving the versatility, safety, and specificity of gene-editing approaches. Base editors enable the conversion of one target DNA base into another in a programmable manner, without requiring dsDNA backbone cleavage or a donor template (75; 07). Prime editing uses a catalytically impaired Cas9 endonuclease or a nickase fused to an engineered reverse transcriptase. The prime editing guide RNA (pegRNA) specifies the target site and also provides a template for a desired edit (07). An example is the successful correction of disease-causing variants by delivering prime editing machinery to human myoblasts (46) or humanized mouse models (20).

CRISPR-based therapeutics remain hampered by delivery limitations, specific targeting needs, off-target activity, and adverse immune responses (35). The latter may be associated with pre-existing immune responses to CRISPR effector proteins or the delivery vector as was highlighted by the fatal outcome of a 27-year-old patient with Duchenne muscular dystrophy treated with customized CRISPR gene-edited transgene therapy (78). Both viral and nonviral vectors can be used to deliver the CRISPR machinery and sgRNAs (or pegRNA for prime editing) into cells, though the capacity limitation of smaller viral vectors (eg, AAV) necessitates the use of dual vector systems. Nonviral vectors that provide larger packaging capacity can help circumvent this limitation and also be used to deliver ribonucleoprotein complexes that may mitigate some of the safety issues related to off-target activity or immunogenicity related to long-term expression of CRISPR machinery.

Zinc-finger nucleases Zinc-finger nucleases represent the first developed modular genome editing technique. ZFNs are a group of enzymes that can cleave precise sites within the DNA. ZFNs offer several advantages, making them appealing for therapeutic applications. Their compact size is compatible with AAV vectors, enabling efficient delivery into cells. Their all-protein structure allows access to mitochondrial DNA. Moreover, ZFNs possess a highly adaptable DNA-binding interface and reduced susceptibility to pre-existing immunity (98). Despite their immense potential, challenges persist in refining their specificity and delivery mechanisms to minimize off-target effects and enhance efficiency.

Transcription activator-like effector nucleases. TALENs consist of two main components: transcription activator-like (TAL) proteins, a type of protein produced by a plant pathogen (xanthomonas), and a nuclease domain that induces double-stranded breaks in the DNA. The DNA-binding domain of TALENs can be engineered to target specific DNA sequences by modifying repeat-variable di-residues within the TAL repeats. Once the TALEN complex binds to its target sequence, the nuclease domain induces a break in the DNA helix, triggering the cell's repair machinery, by nonhomologous end-joining and homology-directed repair. This repair process can lead to targeted gene modifications, such as gene knockout, insertion, or correction. Like CRISPR-Cas9 and ZFNs, TALENs need to be delivered into cells using delivery vectors. In vivo, delivery of TALENs is limited by their large size (approximately 3 kb for a single TALEN) but offers good efficiency, simplified design, and enhanced selectivity due to their reliance on single-base recognition rather than triplets in DNA binding. The latter lowers the risk of off-target effects and minimizes cytotoxicity, promising a more precise and safer approach to genome editing.

RNA-directed therapies

Antisense oligonucleotide therapy. This approach involves the use of chemically modified oligonucleotides, short (12 to 25 nucleotides) single-stranded DNA or RNA that modify the processing of genetic information in the cell. The most common mechanisms include recruiting the endogenous cell machinery (most commonly RNAseH) to break down a target RNA molecule or alter splicing of pre-mRNA in the nucleus. They can also be designed to interfere with transcription or translation. Contrary to gene transfer or genome editing, antisense therapies do not alter the genetic repertoire of the target cell, and modifications are not passed down to daughter cells. As antisense oligonucleotides can be degraded by the cell, the treatment is reversible (21); however, depending on their chemical modifications, some antisense oligonucleotides may have very long half-lives compared to small molecules. The backbone chemistry and specific sequence of the antisense oligonucleotide determines its mode of action: degradation of the target RNA by recruiting endogenous enzymes (eg, RNAseH), translation blockade, or pre-mRNA splicing modification (21). Chemical modifications of antisense oligonucleotides are highly diverse, and further purification (eg, stereoisomers) and conjugation of peptides, antibody fragments, or other molecules can affect their specificity, cell targeting, and toxicity (72).

The antisense oligonucleotides are precisely designed to hybridize with specific target sequences, and their backbone chemistry is modified to escape nuclease degradation in the body. The intended use determines the balance between antisense oligonucleotide backbone chemistry modification and the sequences that remain unaltered (72). For example, splice-switching antisense oligonucleotides work by steric hindrance, and their backbones are usually fully chemically modified. They bind the target sequence (pre-mRNA) and prevent the binding of proteins of the splice machinery, enhancers, or silencers. In contrast, antisense oligonucleotides that target RNA molecules for knockdown typically contain a stretch of unmodified or less modified nucleotides (usually with a DNA backbone) flanked by chemically modified nucleotides on either side to prevent nuclease degradation. This design is also broadly referred to as a “gapmer.” After hybridization with their target RNA sequence, the double-stranded RNA/DNA hybrid is recognized by cell machinery (eg, RNaseH), and the RNA molecule is degraded. Antisense oligonucleotides can also be designed to ultimately increase the expression of an endogenous gene, for example, by targeting the regulatory RNA molecules (eg, micro-RNAs) that naturally work to reduce gene expression for knockdown (79).

In December 2016, after early encouraging results from a Phase 3 clinical trial for infants diagnosed with spinal muscular atrophy type 1, Biogen and Ionis obtained accelerated FDA approval for nusinersen (Spinraza®), marking the first example of intrathecal antisense oligonucleotide therapy for a neurodegenerative disease. Nusinersen, which is a chemically modified RNA molecule, operates by altering the splicing mechanism of survival motor neuron 2 (SMN2) pre-mRNA, which leads to increased levels of survival motor neuron protein production.

Since then, several additional antisense therapies have either received FDA or EMA approval or are in clinical trials. These encompass treatments designed to address both CNS and peripheral nervous system disorders. Selected examples include several splice-switching antisense oligonucleotides specifically designed for exon skipping, aiming to restore the reading frame in Duchenne muscular dystrophy; antisense oligonucleotides that target the mRNA of the TTR gene for knockdown, achieving a reduction of abnormal TTR protein production; and antisense oligonucleotide therapies that target the SOD1 gene and decrease the SOD1 protein as a treatment for one of the common familial forms of amyotrophic lateral sclerosis. The landscape of ongoing research and development in antisense therapies for neurologic disorders extends far beyond these examples, now encompassing diverse neurologic conditions.

RNA interference and gene therapy. RNA interference is an endogenous cellular mechanism to regulate the expression of genes and the replication of viruses. RNA interference or gene silencing involves the use of a double-stranded RNA. Once in the cell, the double-stranded RNAs are processed into small interfering RNAs (siRNAs), which are used in a sequence-specific manner to recognize and destroy complementary RNAs by the endogenous enzymes of the (RNA-induced silencing complex).

RNA interference has some similarities to the knockdown by antisense oligonucleotides, but it involves using a different cell machinery (RNA-induced silencing complex) and can only work by targeting mature mRNAs in the cytoplasm. In contrast, antisense oligonucleotides designed for knockdown can be designed to target the pre-mRNA or mRNA in the nucleus or cytoplasm. Because of their high specificity and efficiency, siRNAs represent a promising class of drugs for gene therapy applications. Viral and nonviral vectors can be used for delivery of RNA interference. In addition, chemical modification and conjugation of different molecules (eg, cholesterol or antibodies) to RNA interference molecules can improve their stability and direct their target delivery to specific cell types (95). For example, patisiran (Onpattro®), an FDA-approved medication to treat hereditary transthyretin-mediated amyloidosis, is an siRNA molecule encapsulated in lipid nanoparticles. RNA interference has great potential to become a successful therapeutic strategy for repeat expansion disorders, such as spinocerebellar ataxia, Huntington disease, or myotonic dystrophy, where involved genes have been identified, and the toxicity of the RNA or its protein products have been established and can be specifically silenced for therapeutic effect. Innovative techniques are under investigation for improving targeted delivery of siRNAs.

ROUTES OF GENE THERAPY DELIVERY

The restrictive qualities of the blood-brain barrier impose an obstacle to drug delivery to the CNS. In preclinical studies, transgenes encoding therapeutic proteins, microRNAs, antibodies, or gene editing machinery have been successfully delivered to the central nervous system with natural or engineered viral capsids via various routes of administration (30). The many approaches that have been explored to optimize and successfully deliver gene-based therapies to the central and peripheral nervous systems are beyond the scope of this article. This section highlights some of the approaches in clinical practice or clinical trials.

Direct injection into CNS. Direct access to the parenchyma enables the delivery of doses and concentrations that otherwise, via systemic administration, would require a high vector load with its associated toxicity and off-target delivery. The development of stereotactic techniques in neurosurgery has enabled the injection of genes and vectors into specific locations in the brain in experimental animals and humans. These techniques still carry some risk, and certain areas of the brain remain inaccessible by these approaches. There have been a few human trials employing intraparenchymal injection in both adults (Huntington, Parkinson, and Alzheimer diseases) and children (spinal muscular atrophy, Canavan disease, Batten disease, mucopolysaccharidosis, and metachromatic leukodystrophy). The predominant approach in these trials used AAV vectors administered via injections into the deep gray nuclei or white matter regions (102).

With an intraparenchymal delivery approach, it is crucial to consider the disease's specific anatomy and pathophysiology when determining target sites. The choice of the target region is a delicate balance between efficacy and safety. For example, targeting the white matter in multiple sites may enhance vector spreading, as AAV vectors easily travel along axons, aiding in the distribution of the therapeutic gene (102).

Delivery into the cerebrospinal fluid. Drugs can be introduced directly into the CSF either by lumbar puncture (IT injection), cisternal puncture (at the craniospinal junction), or into the lateral ventricles of the brain. The last route is more complicated because it involves traversing the brain tissue, and repeated punctures can damage the brain. One potential way to reduce this trauma is to insert a cannula connected to a subcutaneous reservoir (Ommaya reservoir). The drugs or other agents can then be introduced by an injection into the reservoir through the skin. However, consistent and uniform diffusion through any CSF route is difficult to attain regardless of the specific route of administration.

For example, delivery of AAVs into the CSF has been shown to efficiently target the CNS and bypass the blood-brain barrier. In addition, viral delivery via CSF can help bypass pre-existing neutralizing systemic antibodies, addressing a limitation of systemic delivery. Lumbar intrathecal delivery of single-stranded AAV9 is a safe and efficient way of targeting the CNS in adult mice (14). Studies in larger animals have been done to find the optimal route for administration and define the biodistribution of the vector. One study has compared intracerebroventricular and intracisternal AAV administration in dogs and found that both routes resulted in similarly efficient transduction throughout the brain and spinal cord, but intracerebroventricular injection produced encephalitis associated with a T-cell response to the transgene product (55). An extension of this study evaluated vector administration via lumbar puncture in nonhuman primates, with some animals placed in the Trendelenburg position after injection to facilitate entry into the brain. Those pre-clinical studies paved the way for a first-in-human clinical trial of intrathecally administered scAAV9/JeT-GAN to treat the neurodegenerative disease, giant axonal neuropathy (NCT02362438) (11).

Antisense oligonucleotides can penetrate the CNS tissue and enter both glial cells and neurons (61). Compared to serum, there is little intrinsic nuclease activity in the CSF, which confers CSF-administered antisense oligonucleotide resistance to degradation. The success of nusinersen has opened the door for customized intrathecal splice-modulating antisense oligonucleotides for orphan, monogenic neurodegenerative diseases, as was demonstrated with milasen (73).

Alternative approaches

Introduction of the genes into the arterial cerebral circulation. Vascular delivery of viral vectors would optimize delivery to the CNS without the need for multiple invasive procedures. Most of the efforts to infect neurons and neuroglia after vascular delivery have been unsuccessful, presumably because of the inability of the viral particles to cross the blood-brain barrier.

Ependymal-leptomeningeal administration. This route can be used to inject non-replicative vectors into the cerebrospinal fluid space. These vectors can transduce the ependymal and leptomeningeal cells consistently and produce the “therapeutic” product of the transgene in the CNS for extended periods.

Spinal-subarachnoid space administration. Implantation of genetically engineered encapsulated cells producing ciliary neurotrophic factor in the spinal subarachnoid space is an example of gene delivery into the CNS via this route. Deposition of transgenic constructs into the subarachnoid space may present an efficient route for transducing not only in the subependymal region but also for disseminating products of gene expression into the brain.

Intranasal delivery. Intranasal administration of plasmid DNA nanoparticles has demonstrated successful transfection and expression of a reporter protein in rat brain (52). Genes may be successfully delivered in a dose-dependent manner via the nose to the cerebral cortex (36). Preclinical studies advance the field toward noninvasive nose-to-brain delivery of gene therapy; this has been studied for Huntington disease and applies to other hereditary brain disorders (39).

Intraocular delivery. Intraocular delivery of gene therapy represents a promising approach to treating neurodegenerative diseases of the eyes safely and effectively. Among the advantages of this delivery route are easy access, an immune-privileged environment, and the reliable and noninvasive measurement of therapeutic efficacy (101). This is exemplified by the FDA-approved drug oretigene neparvovec-rzyl (Luxturna®), and the promising results from the first-in-human phase 1 and 2 trial in treating X-linked retinitis pigmentosa caused by mutations in the RPGR gene through subretinal gene delivery through AAV8-coRPGR (24).

Intravenous administration of vectors. Tissue- and cell-specific gene expression in the brain is possible after intravenous administration of a nonviral vector with the combined use of gene-targeting technology and tissue-specific gene promoters. Intravenous delivery to target the anterior horn cells in the spinal cord (and part of the CNS) has been shown to be efficient in patients with spinal muscular atrophy and, single intravenous infusion of AAV9-SMN1 vector resulted in longer survival, superior achievement of motor milestones, and better motor function (86).

Delivery of gene therapy to the peripheral nervous system. Targeted expression of foreign genes in the peripheral nervous system has many potential applications, including gene therapy of neuromuscular diseases, such as hereditary neuropathies, neuromuscular junction disorders, primary genetic muscle disorders, and nongenetic disorders affecting muscle and nerve.

Peripheral nerves. Direct administration of therapeutics into the dorsal root ganglia by microneurosurgical techniques or into a peripheral nerve (intraneural delivery) may efficiently target the injected peripheral nerves but is technically challenging and can cause tissue damage. As such, indirect administration methods to target Schwann cells and neurons (eg, intrathecal or intravenous delivery) provide alternative, safer options but are, in turn, hampered by the need for specific targeting, limited transduction efficiency, and potential off-target effects (122). The intrathecal route is gaining popularity for testing in preclinical models and for clinical trials (103), and targeting therapies to Schwann cells appears to be a promising strategy to minimize off-target toxicity and unpredictable effects in other cell types (108). For example, intrathecal delivery of a transgene under the control of the myelin protein zero (Mpz) promoter for targeted expression in Schwann cells has been used in mouse models of hereditary neuropathies (109; 70). Similarly, RNA interference conjugation to optimize delivery to Schwann cells holds promise as a therapeutic strategy for hereditary neuropathies due to the gain of function variants.

Muscle. Skeletal muscle is unique in that it is the organ with the largest mass in the human body and has an extensive complex anatomical distribution. Therefore, systemic delivery is required to target all relevant muscles effectively (18). AAV8, AAV9, and AAVrh74 are recognized for their strong affinity for skeletal muscle cells, making them the most utilized serotypes in clinical trials for genetic muscle diseases (87). However, systemic doses necessary to effectively target the muscle tissue in vivo have been linked with severe immune and liver toxicities (114). The highly myotropic capsid, MyoAAV, and nonviral delivery methods bring hope to circumvent this limitation for systemic delivery in gene therapies targeting skeletal muscle.

Indications

Several monogenic neurologic disorders have FDA-approved gene-based therapies. Additional gene therapies for common, rare, and ultra-rare neurologic diseases are at various stages of pre-clinical and clinical development. Examples of FDA-approved drugs are shown in table 5. Separate clinical summaries in MedLink Neurology deal with applications of gene therapy in stroke, glioblastoma, neurogenetic disorders, neurodegenerative disorders, and muscular dystrophy.

Table 5. FDA-Approved Gene Therapies for Neurologic Single Gene Disorders as of January 2024

Product name	Approval date	Indication	Product description
Gene replacement therapies
Voretigene neparvovec-rzyl (Luxturna®)	December 2017	Patients with confirmed biallelic RPE65 pathogenic variants associated with retinal dystrophy	Recombinant adeno-associated virus serotype 2 (rAAV2) vector expressing the gene hRPE65
Onasemnogene abeparvovec-xioi (Zolgensma®)	May 2019	Patients younger than 2 years of age with spinal muscular atrophy and bi-allelic pathogenic variants in SMN1 gene	Adeno-associated viral vector-9 (AAV9)-based gene therapy for intravenous infusion
Elivaldogene autotemcel (Skysona®)	September 2022	Boys 4 to 17 years of age with early, active cerebral adrenoleukodystrophy	Autologous CD34+ cell-enriched population that contains the patient’s own hematopoietic stem cells, transduced ex vivo with the Lenti-D lentiviral vector containing the ABCD1 gene encoding the adrenoleukodystrophy protein
Delandistrogene moxeparvovec-rokl (Elevidys)	June 2023	Ambulatory pediatric patients aged 4 to 5 years with Duchenne muscular dystrophy	Adeno-associated virus vector-based gene therapy, which encodes an engineered protein, micro-dystrophin, that contains selected domains of the dystrophin protein
Transcriptomic (RNA) directed therapies
Eteplirsen (Exondys 51)	September 2016	Patients with Duchenne muscular dystrophy, 7 years and older and adolescents	Antisense oligonucleotide binds to exon 51 of pre-mRNA, resulting in exclusion of this exon, which allows production of an internally deleted dystrophin protein
Nusinersen (Spinraza®)	December 2016	Spinal muscular atrophy	Antisense oligonucleotide, which modulates alternate splicing of SMN2 to produce full-length SMN protein
Patisiran (Onpattro®)	August 2018	Polyneuropathy of hereditary-transthyretin mediated amyloidosis	Double-stranded small interfering ribonucleic acid (siRNA) that induces degradation of mutant and wild-type transthyretin (TTR) mRNA through RNA interference, delivered through lipid nanoparticle
Inotersen (Tegsedi®)	January 2018	Polyneuropathy of hereditary-transthyretin mediated amyloidosis	Antisense oligonucleotide that induces degradation of mutant and wild-type TTR mRNA through binding to the TTR mRNA
Golodirsen (Vyondys 53)	December 2019	Patients with Duchenne muscular dystrophy and genetic mutations that are amenable to exon 53 skipping	Antisense oligonucleotide that binds to exon 53 of dystrophin pre-mRNA, resulting in exclusion of this exon during mRNA processing, which allows for the production of an internally deleted dystrophin protein
Viltolarsen (Viltepso®)	August 2020	Patients with Duchenne muscular dystrophy and genetic mutations that are amenable to exon 53 skipping	Antisense oligonucleotide that binds to exon 53 of dystrophin pre-mRNA, resulting in exclusion of this exon during mRNA processing, which allows for the production of an internally deleted dystrophin protein
Casimersen (Amondys 45™)	February 2021	Patients with Duchenne muscular dystrophy and genetic mutations that are amenable to exon 45 skipping	Antisense oligonucleotide that binds to exon 45 of dystrophin pre-mRNA, resulting in exclusion of this exon during mRNA processing, which allows for the production of an internally deleted dystrophin protein

Gene therapy for traumatic brain injury. Deliberate gene modulation might achieve the goal of upregulation or downregulation of specific genes within a critical period to alleviate the long-term detrimental effects of traumatic brain injury. There are also a few recombinant proteins and neurotrophic factors that have been investigated in various phases of clinical trials (31). For example, injecting male Wistar rats via the cisterna magna with recombinant adenoviral vectors containing the IGF-1 gene cDNA 15 minutes after traumatic brain injury prevented oxidative stress and cognitive deficits induced by traumatic brain injury (90). BDNF and NGF have demonstrated the ability to improve functional recovery in preclinical, and to a lesser extent clinical, studies (116). These approaches hold promise for treating traumatic brain injury in the future.

Gene therapy of epilepsy. Drug-resistant epilepsy remains a clinically difficult problem, and even if seizures are controlled by drugs, long-term exposure to antiepileptic drugs and their adverse effects affects the quality of life of the patients. The ability of viral vectors or other gene-based therapies to deliver therapeutics locally to seizure foci has the potential to overcome several limitations of medical therapy. Gene therapy approaches for epilepsy can be divided into two general categories: those targeting a specific gene associated with a monogenic form of epilepsy and those targeting mechanisms of seizure generation (119). Preclinical research suggests that gene therapy presents a promising avenue for treating epilepsy patients who lack effective treatment options (13).

As an example, for the first category, encouraging results have been obtained for Dravet syndrome, a severe infantile epilepsy syndrome often caused by monogenic loss-of-function variants in the SCN1A gene. ETX101 is a nonreplicating recombinant AAV9 in which a GABAergic regulatory element ensures selective expression in GABAergic neurons of an engineered transcription factor that increases SCN1A gene transcription. In a Dravet syndrome mouse model, it has been shown that a single dose of ETX101 increased SCN1A gene activity and NaV1.1 protein expression in brain GABA neurons, in addition to a considerable decline in spontaneous seizures and improved long-term survival (121). An ETX101 Phase 1/2 study in children with SCN1A-positive Dravet syndrome is expected to start in March 2024.

Focal cortical dysplasia is among the leading causes of drug-resistant focal epilepsy pathology (15). Type II focal cortical dysplasia commonly occurs due to somatic mutations that trigger hyperactivity in the mammalian target of the rapamycin (mTOR) pathway. Surgical resection of the dysplastic focus is not always successful and is often contraindicated because of risks to normal brain function. Therefore, downregulation of the mTOR pathway via a gene-based therapy approach is an appealing method to answer the clinical need for nonsurgical treatment. In recent years, animal models of focal cortical dysplasia have been developed and successfully treated via gene therapy neuromodulation of the pathway. For example, injection of AAV9-CAMK2A-EKC in the dysplastic region in a mouse model of frontal lobe focal cortical dysplasia resulted in a robust decrease (approximately 64%) in the frequency of seizures by inducing the expression of an engineered potassium channel (05).

Gene therapy for peripheral nerve injuries. Gene therapy has the potential to enhance the results of peripheral nerve repair as follows:

	• Accelerating regeneration across nerve suture by overexpression of a neurotrophic factor.
	• Expression of a motoneuron-specific neurotrophic factor to specifically attract regenerating motoneurons toward a nerve branch for facilitating the reinnervation of the denervated muscle, without affecting sensory axons.
	• Preventing target muscle atrophy following proximal nerve injuries by injecting a viral vector encoding a gene that reduces denervation-induced atrophy.

Contraindications

None have been listed for gene therapy as a therapeutic category. Gene therapy is usually indicated for the treatment of diseases for which no cure is available or the conventional treatments are inadequate. Contraindications are product-specific. As with many drugs, a common contraindication is hypersensitivity to the drug product or any component of the formulation. Selected examples for FDA-approved products are given below and demonstrate the complex considerations that may dictate evolving contraindications to gene-based therapy products. Other contraindications are for patients with the specific disease who are not expected to receive benefit from the treatment. For example, only certain types of Duchenne muscular dystrophy variants are amenable to exon-skipping therapies; thus, the treatment is contraindicated in individuals with Duchenne muscular dystrophy who do not have a variant that may benefit from the treatment.

Delandistrogene moxeparvovec (Vyondys 53®) is contraindicated in patients with Duchenne muscular dystrophy and any deletion in exon 8 or exon 9 in the DMD gene. This contraindication is due to a concern about severe immune reactions that may result from a lack of self-tolerance to specific regions encoded by the transgene.

Inotersen (Tegsedi®) may cause an idiosyncratic thrombocytopenia that can be severe and life-threatening. Unfortunately, one clinical trial patient in the inotersen-treated group died from intracranial hemorrhage associated with grade 4 thrombocytopenia (42). In addition, inotersen may result in severe renal failure, requiring dialysis due to glomerulonephritis. Therefore, inotersen is contraindicated in patients with platelet count lower than 100,000/mm³ or patients with a history of acute glomerulonephritis caused by the drug.

Results

Worldwide clinical trials of gene therapy are listed on the NIH website. As of January 2024, the total number of clinical trials of gene therapy on this website was approximately 2,513, including completed and ongoing trials as well as those that have been approved, but in which patient recruitment has not yet started. Among the categories of neurologic disorders, there are many gene therapy trials for nonmonogenic diseases, including brain tumors such as glioblastoma, neurodegenerative disorders such as Alzheimer disease and Parkinson disease, neuro-immune conditions such as multiple sclerosis, and many more. Several trials have established the safety of gene therapy via different modes of delivery, whereas phase 3 trials are in progress to establish efficacy in additional diseases. Per the FDA list of approved cellular and gene therapy products, there are 34 approved cellular and gene therapy products; among them, four are indicated to treat neurologic conditions. Eighteen nucleic acid therapeutics have been approved for the treatment of various diseases in the last 25 years (33); among them, eight are indicated to treat the following neurologic conditions: Duchenne muscular dystrophy (n=4 : eteplirsen, golodirsen, viltolarsen, casimersen), polyneuropathy of hereditary transthyretin-mediated amyloidosis (n=3: patisiran, inotersen, vutrisiran), and spinal muscular atrophy (n=1, nusinersen). In April 2023, tofersen (Qalsody®), an antisense oligonucleotide that targets SOD1 mRNA to reduce the synthesis of SOD1 protein, received an accelerated FDA approval to treat SOD1-related amyotrophic lateral sclerosis (88). Challenges of developing gene therapy for neurologic disorders include targeted delivery to the intended site of action across the blood-brain barrier, control and maintenance of gene expression at optimal levels, and safety of the gene products. A workshop has dealt with the following issues related to gene therapy clinical trials (69):

	• Genetic diagnosis (if applicable) should be identified when selecting the study population to consider any effects on the safety or efficacy of therapy in certain genotypes.
	• Some early trials may be performed in small groups or a single patient, but transparency and rigorous data collection are important in these cases.
	• Close partnerships between patients, families, investigators, and clinicians are important for collecting natural history data to advance the development of gene therapies.
	• Data about the natural history of a disease can be valuable as a control and can guide the development of endpoints as well as the evaluation of the safety and efficacy of therapy.

Adverse effects

Gene therapy has been studied extensively in animal experiments and human clinical trials. Several specific potential adverse effects should be considered and monitored. Innate and adaptive immune responses to vectors and their transgene products constitute a substantial challenge to gene therapy safety (115). The innate immune responses happen early, lack specificity to antigens, and do not establish immunological memory. Adaptive immune responses are triggered by the inflammatory environment created by innate immune signaling, depend on activation and clonal expansion and differentiation of antigen-specific B and T cells, and generate immunological memory. Viral vectors share many commonalities with natural viruses but are distinct in that they are nonreplicative, are delivered in a single high-titer bolus, and are introduced at an unnatural site. Therefore, the unwanted immune response to viral vectors has unique aspects (115). In addition, inappropriate temporal and spatial expression of the transgene or abnormal levels of the protein product (too high or too low) may cause specific adverse effects. Adding to these complexities are the potential interactions of gene therapy with disease-specific alterations (eg, liver disease in MTM1) (114). The following are some of the adverse effects that have been considered or observed.

Adenoviral vectors. One major concern about using adenoviral vectors for repetitive gene delivery is the induction of immune response to the vector that impedes effective gene transduction. Alternate use of adenovirus vectors from different serotypes within the same subgroup can circumvent anti-adenovirus humoral immunity to permit effective gene transfer after repeat administration, although the chronicity of the expression is limited by cellular immune processes directed both against the transgene and the viral gene products expressed by the vector. Significant cellular interactions occurring early after the systemic adenoviral vector delivery involve vascular and hepatic endothelial cells, platelets, Kupffer cells, hepatocytes, splenic macrophages, and dendritic cells (09). The innate responses may occur within minutes to hours, resulting in multisystem effects, such as alterations in blood pressure, thrombocytopenia, inflammation, fever, and other related symptoms (115). Modification of the viral vector to evade the innate immune response is a commonly employed strategy (127). However, the strong immunogenicity of adenoviral vectors has hampered their development for gene therapy.

Adeno-associated virus vector. AAV-delivered therapies can induce immunological toxicity through capsid-triggered and transgene-triggered responses. Early recognition and correct differentiation between these two responses are critical, as each reaction dictates a different approach to monitoring, prevention, and risk mitigation (77). In a meta-analysis of 255 clinical trials using rAAV vectors, there were a total of 11 patient deaths across 8 trials, and 18 out of 30 clinical holds were due to adverse events (113). Cell-mediated immunity directed against the AAV capsid plays an important role in terms of both safety and efficacy of AAV gene transfer in humans (25). Observations from clinical trials with AAVs showed that 4 to 9 weeks after vector administration, activation of capsid-specific CD8+ T cells was associated with increased liver enzymes and concomitant loss of transgene expression (124). Timely initiation of immunosuppression can modify the immune response and is now routinely used after gene therapy. Thrombotic microangiopathy, a potentially life-threatening condition, can occur early after dosing and is characterized by thrombocytopenia, hemolytic anemia, and end-organ damage to the heart, lungs, kidneys, and other organs (85). This has been described using AAV to deliver microdystrophin. Thrombotic microangiopathy in the setting of AAV gene therapy was shown to be anti-capsid antibody-dependent (classical pathway) and amplified by the alternative complement pathway (107). The unpredictable, severe adverse reactions stress the complex and evolving immune biology related to gene-transfer therapies (19).

Strategies to reduce immune responses at various stages during gene therapy include the following (100):

	• Predosing preparation of the immune system for AAV exposure.
	• Directly delivering lower doses to disease-relevant target tissues to minimize toxicity from gene overexpression in nontarget tissues.
	• Using postdose immune management strategies to protect transgene expression and neural tissue and decrease neuroinflammation.

RNA-directed therapies or antisense oligonucleotides. Toxicities and off-target effects are known to be both sequence- and chemistry-dependent. In vivo, foreign DNA may be recognized by toll-like receptors and can lead to activation of the immune system (03). Those undesired effects have been largely mitigated by screening for chemistries and sequences that are better tolerated (110). Systematic administration of antisense oligonucleotides may result in adverse effects, such as liver, hematologic, and renal toxicity (51). Cellular assays of hepatotoxicity and cytotoxicity are routinely used for the preclinical development of such products. Data regarding the immunogenicity of therapeutic oligonucleotides are somewhat limited and will likely grow with post-marketing pharmacovigilance and additional clinical trials.

Measures to improve safety and efficacy of gene therapy. The ability to direct gene transfer vectors to specific target cells is an important task to be tackled, and it will be important not only to achieve a therapeutic effect but also to limit potential adverse effects. Considerable efforts are being made to develop nonviral vectors to avoid some of the adverse effects of viral vectors and to optimize immune suppression regimens for both predicted immune adverse effects and those encountered post-dosing while the patient is monitored for possible side effects.

Special considerations

Pregnancy. Gene therapy is relevant to pregnancy as fetal gene therapy is theoretically feasible. In the case of neurogenetic disorders, therapeutic efforts are usually more efficacious when administered early, including during fetal life, to prevent early disease manifestations and improve neurologic outcomes. Prenatal diagnosis has improved, surgical interventions on the fetus are established, and fetal targeting via maternal administration is an appealing, low-risk delivery option for compounds that can cross the placenta. Fetal gene therapy offers the following advantages:

	• Correction of the abnormality before fetal tissue damage occurs.
	• Targeting of the still expanding stem cell population of organs that is not accessible in later life.
	• Avoidance of immune sensitization against the vector systems or transgene products.
	• It provides a third alternative to termination of pregnancy or acceptance of an affected child in cases where prenatal diagnosis of an inherited disease has been made.

Experimental disease models have validated the potential of intrauterine gene modification therapy (known as in-utero gene therapy) in addressing various hereditary conditions, including neurogenetic disorders (84).

Ethical aspects of gene therapy. A detailed discussion of gene therapy ethics is beyond the scope of this article. In practice, public attitudes are also important considerations. There has been an increasing acceptance of gene therapy in North America during the past quarter of a century since the first gene therapy trials on humans. Results of an opinion survey show that most of the respondents (greater than 90%) accept gene therapy as a treatment for severe or untreatable illnesses such as Alzheimer disease, but this receptivity decreases for conditions perceived as less severe (105). Additional gene therapy clinical trials are becoming available for myriad medical conditions. Given this anticipated growth, patients and their parents or caregivers will need to choose whether to consider gene therapy and participate in clinical trials. A significant area of concern for the application of gene therapy in neurologic disorders is the fear of not receiving sufficient information before undergoing the treatment. Weighing risks and benefits, considering the timing of therapy, and ensuring patients and caregivers understand the safety and efficacy during the informed consent process are all crucial aspects (74). Another theoretical ethical concern is the use of gene therapy methods to enhance certain physical traits or attributes. Although the risk and cost of gene therapies preclude their development for these purposes, this may be an important consideration in the future. Further research and experience are needed to grasp the adverse effects of gene therapy on future generations (67).

Equal access across racial, ethnic, and socioeconomic groups is an important ethical concern with any drug development, but especially with gene therapy programs. Given the high price of such therapies, the exclusive manufacturing process, nonstandard administration, and the high cost of post-treatment monitoring, the limited accessibility of those life-changing therapies may accentuate inequalities and raise the need to facilitate affordable and broad access to all patients in need worldwide. The same rule applies when considering the landscape of gene-based therapy strategies--the development of such therapeutic products should be applicable for patients with rare and ultra-rare genetic conditions for whom no commercial interest exists. Efforts and initiatives to overcome those significant obstacles are rising with partnerships with industry, government, academia, and patient advocacy groups. One example is the AMP Bespoke Gene Therapy Consortium, which aims to develop platforms and standards to speed the development and delivery of customized or “bespoke” gene therapies that could treat millions of people affected by rare diseases.

Future of gene therapy for neurologic disorders. Gene therapies are in development for neurodegenerative disorders, neurogenetic disorders, and neuromuscular disorders, and many of these are in clinical trials with potential for new therapeutic applications. The scope of gene therapy extends beyond genetic disorders. For example, genetically engineered stem cells could be used to repair tissue function for regeneration in neurologic disorders.

Few medical centers and hospitals have sufficiently experienced teams to participate in gene therapy clinical trials for neurologic conditions or to administer approved therapies when available because gene therapy has not been part of training programs for most neurosurgeons and neurologists. Expanded indications and demand for CNS gene therapies will require a worldwide educational effort to supplement the training of clinical neuroscientists, and a few centers of excellence will need to establish relevant educational training requirements and best practices for such therapeutic approaches (38).

Drug delivery is the biggest hurdle for effective gene-based therapies. Intrathecal, direct intracranial, and systemic delivery are associated with specific risks and benefits. Therefore, efforts to improve strategies for safe delivery of gene therapy to treat central and peripheral nervous system disorders are needed. The development of nonviral vectors that could potentially carry larger DNA (or RNA) as well as gene editing machineries will greatly facilitate the treatment of neurologic conditions. Although considerable progress has been made to improve vector design, gene selection, and targeted delivery, there are some limitations, such as immunogenic reactions, nonspecificity of viral vectors, and lack of validated biomarkers to track therapy efficacy (59). Standardized metrics for capturing disease course and progression to deepen the understanding of each disease's natural history and phenotypic subtypes in diverse cohorts of patients are essential for clinical trial readiness (111).

Scientific basis

	• Gene therapy involves the introduction of genetic material into cells to replace or supplement defective genes, edit or suppress a gene, or induce the expression of a novel gene to alleviate a disease.
	• The scientific basis is included in the description of different techniques in the preceding sections of the article.

Gene therapy involves the introduction of genetic material into cells to replace or supplement defective genes, reduce the expression of overactive genes, or induce the expression of a novel gene to alleviate a disease. In addition to gene transfer, modified nucleotide sequence therapeutics, such as antisense oligonucleotides, RNA interference, mRNA-based approaches, and genome editing methods to act specifically on pathogenic target nucleic acids, are in growing clinical use. Both viral (22) and nonviral vectors (132) have been developed to optimize tissue targeting, enhance efficacy, and reduce the risk of toxicities and immune reactions.