There are over 6,000 genetic disorders and the list grows continually. Gene therapy works to treat or prevent disease of monogenic cause by modifying an individual's genes. In 2012, alipogene tiparvovec was the first to receive European regulatory approval for treating lipoprotein lipase deficiency by delivering a functional copy of the enzyme to muscle cells using an adeno-associated viral (AAV) vector. This approval (now withdrawn due to low demand) was followed closely in 2017, in the US, by voretigene neparvovec, for treating Leber congenital amaurosis, an inherited eye disease of progressive blindness, in biallelic carriers of mutations to the RPE65 gene.
Since these landmark approvals, there has been several additional approved gene therapies. Aside from treating monogenic disorders, AAV-based gene therapies have also been approved for treating certain cancers, through engineered chimeric antigen receptor T cells, and they are being explored against infectious diseases via delivered recombinant antibodies.
The gene therapy landscape is dynamic and constantly shifting. The technology holds the promise of potential disease modification, providing life-altering benefits for patients that otherwise would have faced dire outcomes. In the midst of this shifting landscape, the gene therapy field has celebrated victories and faced obstacles.
Herein, we will review major advances in disease modification for illnesses that presently still lack effective treatments, as well as some challenges and approaches to overcome them.
Gene therapy: Panacea for neuromuscular and neurodegenerative disease?
Neuromuscular diseases (NMDs) and neurodegenerative diseases (NDDs) are generally progressive illnesses that often shorten lifespan. Unfortunately, many also lack disease-modifying therapies. This has made NMDs and NDDs of genetic etiology prime candidates for gene therapy, by addressing the root cause of disease, i.e., the defective gene.
The 2019 approval by the US Food and Drug Administration for onasemnogene abeparvovec, an AAV therapy for spinal muscle atrophy (SMA), marked a turning point in the treatment of NMDs.
“SMA is a NMD caused by mutations to the gene SMN1, which codes for survival of motor neuron, a protein essential to neuronal health,” explained Kathrin Meyer, assistant professor at the Ohio State University and principal investigator at the Center for Gene Therapy at Nationwide’s Children Hospital, who was involved in the preclinical and clinical studies that led to the approval of onasemnogene abeparvovec. SMA patients exhibit gradual loss of motor neurons, leading to progressive muscle weakness, although the rate of decline varies across individuals. In the most severe cases, respiratory failure occurs. “The gene therapy treatment for SMA involves delivering a functional copy of SMN1 using an AAV9 carrier by intravenous injection. In the clinical trial, recipients experienced improved functional and survival outcomes compared to natural history data of the disease progression,” Meyer elaborated. Meyer is presently assessing the feasibility of gene therapy for additional NMDs and NDDs.
One focus is Batten disease and the family of neuronal ceroid lipofuscinoses, which are NDDs caused by mutations to lysosome storage genes called ceroid-lipofuscinosis neuronal (CLN). “We found that delivery of AAVs carrying a functional CLN copy improves lifespan and brain and behavioral pathology in CLN6 and CLN8 animal models of Batten disease,” Meyer summarized of this line of work.
Guangping Gao, professor at the Horae Gene Therapy Center, at the University of Massachusetts Medical School, is also actively researching gene therapy as a potential treatment avenue for NMDs and NDDs. “We are especially interested in Canavan disease, a fatal rare recessive NDD caused by loss-of-function mutations to aspartoacylase (ASPA),” Gao described of his research. Loss of aspartoacylase activity prevents breakdown of N-acetylaspartate, which accumulates in the central nervous system and interferes with neuron myelin sheaths. Disease onset occurs in infancy and manifests with poor motor function with reduced head control, lack of visual fixation and muscular hypotonia, which develops to include spasticity, ataxia and seizure. “We have shown that AAV-mediated delivery of a functional ASPA copy to ASPA–/– mice drastically improves neuropathology, myelination and motor function. Moreover, using our Canavan disease model and gene therapy, we have been able to shed insight into disease pathophysiology, such as peripheral involvement, would could help further refine our therapeutic approach,” Gao explained.
Gene therapy can replace genes with defective loss-of-function mutations with a functional copy of the gene. Alternatively, the AAV vehicle can be used to deliver short hairpin (shRNA) or interfering RNAs (RNAi) to knock down expression of a gene with toxic gain-of-function mutation. This is a viable approach, for example, for cases of the motor neuron disease amyotrophic lateral sclerosis caused by mutations to the gene superoxide dismutase 1 (SOD1). In animal models of ALS, AAV-mediated delivery of shRNA or RNAi slows disease progression in SOD1G93A mice.
Rather than targeting a specific mutation, gene therapy can also be used to support damaged tissues, e.g., delivering a copy of gene that strengthens muscles to treat NMDs with muscle weakness, or a growth factor to regenerate injured nerves. Ultimately, a gene therapy with a base-editing CRISPR-Cas cargo could correct genes, with either loss- or gain-of-function mutations.
Gene therapy has already celebrated some successes, with approved therapies that tangibly improve outcomes for patients, and potential developments in place for additional diseases. Like all technologies, however, it also faces some obstacles.
Overcoming the immune response
A prominent challenge facing gene therapies is that AAVs and their transgene cargo can elicit an immune response, innate or adaptive, which diminishes transduction efficiency and long-term gene expression and increases the chances of adverse events. “Ideally, gene therapy was envisioned as a one-shot treatment, but in practicality, re-dosing is likely needed depending on the target tissue and age of treatment. Moreover, although AAVs are quite effective, high systemic doses are sometimes needed. Both these factors potentially provoke adverse events associated with AAV and transgene immunogenicity,” Gao discussed.
One route is to change the injection method for AAV administration. “It is possible to reduce a systemic immune response by employing intrathecal injection. Although this delivery route is safer, local immune responses in highly targeted regions, so called “hotspots” can still be of concern,” Meyer explained. “However, this injection method preferably targets central nervous system tissue, such as the brain and spinal cord, which will not be the optimal tissue target for many diseases.”
Another option is to eliminate preexisting AAV antibodies, resulting either from prior AAV gene therapy doses or from natural infection to wild-type AAV, which is endemic to humans. “There are several ways to accomplish this,” Gao outlined. “Pharmacological agents, for instance, rituximab to deplete B cells, or physical methods, such as plasmapheresis and AAV serotype-specific IgG absorption to remove circulating antibodies. We can also block the innate response through complement deactivation with antibodies against complement components, such as anti-C5 eculizumab.”
Moreover, it is possible to engineer the AAV capsid, vector genome and transgene, e.g., by “cloaking”, to lower their immunogenicity. The transgene expression from AAV vector can also be engineered to evade immune cells, such as antigen presenting cells (APCs), an arm of the adaptive immune response. Gao developed a microRNA (miR)-based strategy to evade APCs. In a proof-of-concept study using an AAV expressing the highly immunogenic ovalbumin, Gao demonstrated that incorporating the target sites of the APC-specific miR142 and other miRNAs into the transgene detargeted ovalbumin expression from APCs, i.e., they did not display ovalbumin antigens, blunting their immune response against it. In tandem, the cytotoxic T-cell response against ovalbumin was also blunted. More recently, Gao extended this approach by evaluating combinatorial microRNAs to better block the APC and cytotoxic T-cell response.
Towards tailored tissue targeting and transduction
Another major goal of gene therapy research is improving AAV tissue specificity and augmenting tissue transduction, i.e., that the AAVs target the intended tissue and disseminate widely across it. Wild-type AAVs occur in several serotypes, distinct variations in the capsid, that differ in their tissue specificity and ability to cross the blood–brain barrier, impacting their tissue targeting and transduction efficiency.
“Widespread transgene delivery is a major challenge,” Meyer elaborated. “There are AAVs, such as AAV9, with good ability to cross the blood–brain barrier and access the brain and spinal cord, which are optimal target tissues for NMDs and NDDs.” Using intrathecal AAV9 injection, Meyer successfully showed widespread transgene expression in mouse and nonhuman primate animal models, which achieves up to 80% targeting to spinal motor neurons as well as across various brain areas. Transduction can be further enhanced by placing the animal in the Trendelenburg position – lain flat on the back with the head tilted 15 to 30 degrees below the feet. “Although we’ve developed injection methods that enhance transgene distribution across neural tissue, penetration to the deep layers remains relatively limited, but we are actively researching better methods,” Meyer added.
An alternative to wild-type AAVs is to rationally engineer or evolve novel capsids with distinct or improved tissue targeting and blood–brain barrier and trans-vascular penetrance. “Improved capsid specificity and transduction efficiency will mean lower doses, which will mean fewer possible complications arising from immune responses to high dose AAV gene therapy,” Gao explained. “Further refinement may be achieved by tissue-specific codon-optimized and CpG motif-reduced transgenes, to enhance translation within the target tissue versus other tissues and blunt innate immune response.”
Gene expression in moderation and in context
Another facet of gene therapy research is modulating transgene expression. The overarching goal is stable and durable expression, but to levels commensurate with physiological expression. “You need to be able to physiologically regulate transgene expression because uncontrolled expression, even of endogenous proteins, becomes toxic at excessive levels,” Gao elaborated. “Furthermore, you want to be able to turn some gene therapies off. For example, if using an AAV vehicle to deliver CRISPR-Cas cargo, you don’t want the Cas nuclease expressed in perpetuity because it will increase off-target cleavage and cause unintended genomic damage.”
Meyer suggested that the field needs to understand expression of the target gene in its native environment better. “Animal models are great and have helped us test the efficacy of potential gene therapies for rescuing disease phenotypes. However, oftentimes we don’t understand the disease mechanisms in detail. On paper it looks simple; it seems like you might need to deliver a functional gene copy for diseases with loss-of-function mutations, and deliver a shRNA or RNAi for diseases with gain-of-function mutations. In practice, it isn’t always this straightforward.” In amyotrophic lateral sclerosis, for example, the most common known monogenic cause are repeat expansions to the gene C9orf72, which cause both toxic gain-of-function as well as haploinsufficiency and loss-of-function. “Generating models using patient-derived cells allows us to investigate disease pathophysiology and gene expression patterns of genetic mutations in context, and to dissect contributions from various cellular compartments. Failing to understand the complex mechanisms in context in patient cells could lead to patients not responding to a gene therapy.”
A hefty price tag
Although gene therapies have profoundly improved outcomes for some patients, their hefty price tags limit access. Moreover, some NMDs and NDDs are caused by rare mutations, so patients are not always diagnosed early.
“There are three major hurdles that are currently impacting translation of gene therapies to patients,” Meyer explained. “On the first front, we need to streamline manufacturing, which is currently expensive and variable.
On the second front, regulatory aspects are difficult to navigate because methods to assess gene therapies, such as dosage, are not standardized, making it problematic to compare efficacy between different gene therapy products. This was the scenario with therapeutic antibodies a while back, which has since been streamlined, so we should be able to achieve the same for gene therapies.
On the third front, we need to make sequencing technologies more accessible to patients with rare diseases to identify mutations and diagnose patients earlier. We will need to be able to treat patients earlier, pre-emptively, before too much tissue damage has already occurred to truly modify the disease course.”
Gao echoed similar sentiments regarding manufacturing when asked about future directions for gene therapy. “I think if we look to the future, we have to identify what the current problems are, the challenges we are presently facing,” he replied. “If we are to bring gene therapies to patients more readily down the drug development pipeline, we will need to develop better manufacturing platform technologies and lower costs.”
