GENE THERAPY – Solving the Puzzle: Aligning the Pieces of Gene Therapy & Creating Success for Patients


For a two-word phrase, the term “gene therapy” encom­passes diverse and critical elements that are essential for deliver­ing genetic material to target cells in order to treat or cure disease. Each of these elements presents challenges and oppor­tunities for companies developing gene therapies. Each element can be a point of potential failure due to safety concerns, poor targeting of the affected cells types, or insufficient expression to achieve the desired biologic effects. On the flip side, unique and proprietary technologies that can optimize safety and efficacy may enable differentiated approaches to transgene engineering or vector development that provide clinical, regulatory, and com­petitive benefits. Ultimately, the seamless integration of the un­derstanding of vector biology, genetic engineering, manufacturing, and the unmet patient needs is essential for de­veloping transformative therapies that offer substantial clinical and commercial potential.

When embarking on the development of a novel gene ther­apy, it is important to keep in mind that the most direct route to the commercial marketplace may not be a straight line. While leveraging existing vectors and promoters might appear to pro­vide an opportunity to reduce development time and risk, the re­ality is that effective gene therapies must be built from the ground up on a case-by-case basis, if they are to provide optimal per­formance in the context of a specific disease. This article will re­view the various elements that must come together in developing safe, effective, and commercially viable viral-based gene thera­pies.


Viral vectors are a foundation of gene therapies because the viruses from which they are derived have evolved over millions of years to enter and deliver genetic material into human cells with high efficiency. Viruses, however, vary in the types of cells they infect, the efficiency with which they enter cells, how long they persist within an infected cell, the size of the genetic payload they can deliver, and the extent to which they activate host im­mune responses or even cause illness. The ease of vector manu­facturing is also highly dependent on viral type.

Each vector offers unique features and benefits that must be carefully paired with the specific disease indication and the de­sired biologic activity for which it will be used.

For example, adenoviral vectors offer relatively large pay­load capacity and ease of manufacturing, but also have the po­tential to trigger robust immune responses that can lead to serious, and in one case, fatal inflammatory responses. The ma­jority of adenoviral subtypes have a tropism for cells in the respi­ratory system, while others can efficiently transduce renal, intestinal, or ocular cells. The effect of adenoviral vectors is also short lived, which, while potentially beneficial in certain limited circumstances, would require ongoing treatment in the case of genetic or chronic indications, thus aggravating this tendency of problematic immune responses. Lentiviral vectors offer relatively large payload capacity, the ability to treat cells that rapidly turnover, and can be engineered to have tropism for a broad range of target cells. In contrast to adenovirus and adeno-asso­ciated virus (AAV) vectors, however, they can integrate into host cell DNA, resulting in potential interference with other important genes and unwanted alteration of transduced cells. Lentiviral vec­tors also stimulate innate and adaptive immune responses that can limit efficacy and may cause inflam­mation. AAV vectors have somewhat more limited payload capacity and more com­plex manufacturing requirements, but mul­tiple serotypes, combined with careful selection of promoters and other DNA reg­ulatory elements, allow for targeting vec­tors to specific tissues. Additionally, AAV is not known to cause any human disease, which offers potential safety benefits com­pared with other viral vectors.

Which one, then, to choose? It de­pends on the therapeutic needs of the dis­ease for which a gene therapy is being developed. For example, AGTC utilizes en­gineered AAV-based vectors for its inher­ited retinal disease (IRD) programs, because they can, in most cases, accom­modate the size of the genes that need to be delivered and have reduced potential for inflammation compared with adenovi­ral and lentiviral vectors, which is particu­larly important for the treatment of ocular diseases in which inflammation may cause interference with visual function and long-term damage. There are a variety of AAV serotypes that can be used to target spe­cific cell types with high transduction effi­ciency based on the biological need of a given indication, and AAV capsids can also be engineered both for increased transduction efficiency of cells for which they have native tropism, as well as to transduce additional cells types. As an ex­ample, AGTC utilizes an engineered AAV vector (AAV-TYF), which has been opti­mized for transduction of non-human pri­mate photoreceptor cells, for X-linked retinitis pigmentosa (XLRP) and achro­matopsia (ACHM).


The desired therapeutic effect of a gene therapy also guides promoter selec­tion. Promoters, which are the DNA regu­latory sequences that control in which cells a gene is expressed, can be chosen based on tissue specificity, whether they are con­stitutively active or respond to physiologic signals, and the levels of expression that they drive. Again, the goal is to build a vector that expresses an appropriate level of protein in the target cell and minimizes expression in non-target cells. Given that a single vector may have tropism for mul­tiple tissues, or multiple cell types within an organ system, promoter elements play a key role in targeting expression to specific cell types with greater precision. Bottom line? Capsids and promoters can be com­bined to maximize expression in multiple cell types or to restrict expression to specific cells depending on the biology of the delivered gene and the targeted disease indication.

For example, AGTC is currently devel­oping AAV-based gene therapies for XLRP and ACHM, both of which are IRDs. AAV-TYF efficiently transduces both rod and cone cells within the retina, but only cone cells play a role in the pathology of ACHM, and expression of the gene used for ACHM therapy in rod cells could be detri­mental. Consequently, the ACHM vector utilizes a cone-specific promoter in order to restrict expression to cone cells even if the vector transduces rod cells, which should eliminate safety issues arising from off-target gene expression. In contrast, the XLRP construct uses a promoter that is ac­tive in both rod and cone cells, as both cell types contribute to the XLRP disease pathology (Figure 1).

Promoter selection is critical for targeting gene expression to appropriate tissues, which in turn is essential for improving safety and efficacy. The photoreceptor-specific GRK1 promoter (left) is active in both rods and cone cells, while the PR1.7 promoter is active only in cone cells (right). Based on these expression patterns, the GRK1 and PR1.7 promoters are appropriate for use in gene therapies to treat XLRP or ACHM, respectively.

Click image to enlarge


In addition to optimizing delivery and expression to address the molecular basis of disease, the transgene itself plays a crit­ical role in the safety and efficacy of gene therapies. Here again, understanding the biology of a particular indication is essen­tial for defining the parameters that guide transgene selection and engineering. In some cases, therapeutic activity may re­quire a transgene that produces a protein with absolute fidelity to the wild-type pro­tein. In other cases, transgenes may ben­efit from codon optimization that yields a protein with enhanced expression and/or stability (Figure 2, left). Depending on the therapeutic gene and vector payload ca­pacity, transgenes may also need to be en­gineered to be smaller but still produce proteins that provide therapeutic benefit (Figure 2, center).

For example, current clinical pro­grams evaluating gene therapy for the treatment of Duchenne muscular dystro­phy (DMD) encode smaller versions of the dystrophin protein. This is because the wild-type dystrophin gene, which is one of the largest genes in the human genome, is too large for any of the currently avail­able viral vectors. Consequently, DMD gene therapies use mini- or micro-dys­trophin genes that have been engineered to retain critical functions. Although long-term data on the safety and efficacy of this approach have not yet been generated, preliminary evidence suggests that these smaller genes may provide functional ben­efit to patients with DMD.

Alternatively, dual vector systems may enable delivery of large, multi-domain proteins that cannot be packaged in cur­rent viral vector systems. Such systems could enable delivery of larger genes, such as the therapeutic gene for Stargardt dis­ease, ABCA4. This can be achieved by splitting the DNA into two pieces that share an overlapping region at one end, pack­aging the pieces into separate vectors, and then delivering both vectors to a single cell. Once inside the cell, the overlapping regions in each gene fragment would me­diate DNA recombination to create the full-length gene sequence (Figure 2, right). The efficiency of this recombination is highly influenced by the specific sequences chosen, which need to be tested and proven in appropriate animal models. Here again, vector and promoter selection will also be important for ensuring the complete protein is efficiently expressed in target cells.

Gene cassette modification can enhance expression and stability. Left: in codon optimization, DNA sequences are optimized to enhance stability and expression of full-length protein. Center: transgenes can be engineered to produce a protein that is smaller in size than the native protein while retaining essential biological activity. Right: Dual vectors can be used to split a DNA sequence into two pieces that recombine in the cell to produce a protein with desired biological activity.

Click image to enlarge


Other factors beyond the vector com­ponents themselves may also play a sig­nificant role in the safety, efficacy, and commercial viability of gene therapies. The route of vector administration is one such factor. Gene therapies can be adminis­tered via subretinal or intravitreal injection for the treatment of ocular diseases; through the round window of the middle ear for the treatment of otologic diseases; and via injection into the spinal cord or di­rectly into the brain ventricles for the treat­ment of neurologic diseases. In each of these examples, the route of administra­tion may have an impact on the total dose needed to achieve therapeutic benefit, the ability to target specific cell types or sub­types, and the simplicity versus complexity of administration. Additionally, more com­plex or invasive administration routes may pose a barrier to patients wishing to un­dergo therapy, although this would likely be a greater issue for indications in which other effective therapies are available, rather than for diseases in which a gene therapy would be a significant new ad­vancement. Novel routes of administration may also require availability and testing of devices used to deliver gene therapies. Consequently, the route of administration is another important piece of the gene therapy puzzle and should be considered at the earliest stages of development in order to ensure the route and therapy have been optimized for use together.


While the strategic selection of vec­tors, promoters, and transgenes is essen­tial for the safety and efficacy of gene therapies, robust manufacturing processes are critical for commercial success and broad patient access. Scalable processes that can be performed in bioreactor sys­tems similar to those used for decades to produce a variety of protein and antibody therapies can enable cost-effective manu­facturing to meet small rare disease indi­cations, such as IRDs, while providing ca­pacity to support the development of gene therapies for large markets, such as age-related macular degeneration. Bioreactor-based methods also reduce the need for manual tissue culture processes that carry the risk of operator error or contamina­tion.

A key challenge in gene therapy man­ufacturing is balancing the “need for speed” with the development of processes that meet regulatory requirements and will support commercial use. Many gene ther­apies are initially developed in academic laboratories using small-scale production methods that are appropriate for preclini­cal studies and may even have sufficient yield for small Phase 1 trials. These processes are not, however, sufficiently ro­bust for late-stage clinical development or commercial manufacturing. Commercial-scale gene therapy manufacturing will re­quire significant investment in process development, facility build-out, or contract manufacturing. Companies must strike the right balance between investing early in programs that may not warrant continued development and waiting until product candidates near approval and commer­cialization, which could delay the initiation of late-stage trials.

Gene therapy manufacturing processes must also be designed with high yield for vectors that have the complete gene cassette packaged within them, with limited numbers of empty vectors and with the efficient removal of the residual raw materials, such as host cell proteins and DNA. Since the initiation of its AAV manu­facturing efforts, AGTC has significantly in­creased the productivity of its process, resulting in the current process that offers the potential for a 90-fold reduction in the cost of goods sold. Additionally, the process results in 90% complete capsids and undetectable residual raw material. This is important for enabling optimized dosing by delivering a therapeutic dose with the lowest total viral load possible, and without impurities that could lead to side effects or toxicities.

Vector characterization is another crit­ical element of manufacturing, as it en­sures the safety, efficacy, and potency of the finished product. Regulatory agencies put as much emphasis on ensuring the safety, quality, and reproducibility of gene therapies as they do on their clinical safety and efficacy. As an illustrative example, in September 2020, the US FDA requested that Sarepta utilize an additional potency assay for release of the commercial process material prior to dosing in the company’s planned Phase 3 trial of its gene therapy for DMD. This request under­scores the importance of vector character­ization as an intrinsic part of developing gene therapies.


The safe, effective, and successful assembly of the pieces of the gene ther­apy puzzle requires one other critical in­gredient – disease-specific expertise. As highlighted earlier, the choice of vector, promoter, transgene, and route of ad­ministration must be done with an eye to­ward optimally addressing the biology of a disease that a specific gene therapy is designed to treat. This can only be achieved with insight from those who are truly expert in that disease.

For larger biopharmaceutical com­panies, such expertise may reside in-house and can be integrated with internal gene therapy research and de­velopment efforts. Smaller companies may achieve the same objective through strategic partnerships and collaborations that synergistically utilize complementary expertise and technologies. For AGTC, this has meant establishing collabora­tions with Bionic Sight and with Otonomy, to develop gene therapies for vision loss and otologic disease, respectively. In each relationship, AGTC provides expert­ise in gene therapy development while the partner brings expertise in optogenet­ics (Bionic Sight) or genetic hearing disor­ders. This allows development of an optimized (Otonomy) vector and admin­istration strategy for each indication that is based on the biology of the disease and desired therapeutic outcome.


With diverse factors and options at every step of the continuum, the path for developing gene therapies can be viewed as ridden with multiple hurdles and com­plexities. Yet a diligent, data-driven, and patient-focused approach to navigating this path offers unique opportunities to truly advance and transform the treat­ment of serious diseases that have urgent unmet need. With the potential to opti­mize all aspects of gene therapy devel­opment to optimize safety, efficacy, and patient outcomes, we cannot accept any­thing less. The patients who stand to ben­efit from successful gene therapy development certainly won’t.

Sue Washer is President and CEO of AGTC and brings a decade of experience in pharmaceutical management and research with Abbott Labs and Eli Lilly & Company and more than 26 years of senior management experience with entrepreneurial firms, including three start-ups. At AGTC, she has secured private and public investments of over $290 million for AGTC, negotiated and closed two major collaborations with top Biotech companies resulting in over $150 million of cash in-flows, and led AGTC in efficiently completing critical scientific milestones. She is an appointed member of the Small Business Capital Formation Advisory Committee for the SEC, Associate Vice President of the ECSGB Board of Bio, and currently serves on the Board of Directors of BIO, and BioFlorida. She earned her undergraduate degree in Biochemistry from Michigan State University and her MBA from the University of Florida, where she was one of the first graduates from the Warrington College of Business Entrepreneurship program.