At a recent industry event, the 14th Annual Bioprocessing Summit in Boston, scientists discussed a variety of advances in gene therapy manufacturing. To learn more about these advances, GEN talked with four scientists who are developing therapeutics that are delivered by adeno-associated virus (AAV) vectors.

Facing the challenges

Kenneth Yancey
Kenneth Yancey, PhD
UPenn

“Gene therapy has shown the potential to positively impact millions of people, but access to gene therapy has been limited among those most in need,” says Kenneth Yancey, PhD, senior director, downstream process development, University of Pennsylvania. “If the industry is going to expand access, then it is critical that the cost of goods be reduced and that the supply of vector be increased.”

On the issue of vector supply, Yancey remarks, “The per batch productivity has been limited due to current vector manufacturing scales. Systems for manufacturing vectors typically operate at smaller scales than systems for manufacturing more established biologics.

“This is due in part to the complexity of transient transfection when utilizing large adherent or suspension bioreactors, and the limitations of operating in existing gene therapy manufacturing facilities that have been designed for operation at small production scales. Operating at small production scales can result in a failure to achieve economy of scale and result in an increase in complexity due to the need for extreme scale-out manufacturing.”

Yancey notes that AAV manufacturing has been an ongoing challenge, one that the COVID-19 pandemic has only exacerbated, intensifying the need for new approaches. To begin addressing this need, Yancey and colleagues began instituting several short-term improvements. “We better utilized limited resources by incorporating new technologies,” he says. “Better resource utilization allowed us to set new records for our gene therapy program in terms of the scale and productivity achieved while maintaining our commitment to quality.” It resulted in successful Investigational New Drug (IND) applications for several critical diseases: GM1 gangliosidosis, Krabbe disease, metachromatic leukodystrophy, and frontotemporal dementia.

For longer-term improvements, Yancey and his colleagues started developing a new suspension platform to increase productivity. “The result,” he asserts, “has been an increase in per liter productivity by approximately an order of magnitude, and an estimated 30-fold increase in per batch productivity.”

In addition to those successes, Yancey and his colleagues will keep working to improve AAV manufacturing. “In the near future, we will continue to transition from traditional AAV manufacturing technologies to processes such as suspension, which allow for higher productivity and larger-scale production of vectors,” he explains. “This will allow us to greatly increase clinical supply and achieve economy of scale, which will drastically reduce our cost of goods.”

Perfecting purification

Nick DiGioia
Nick DiGioia
Alexion Pharmaceuticals

Nick DiGioia, currently manager of process development at Alexion Pharmaceuticals, is familiar with the challenges of AAV purification. “The biggest challenge in AAV purification is the empty/full capsid separation step, which is usually performed using ion-exchange chromatography,” he says. “We have found that maximizing full-capsid purity often comes at a trade-off to yield, so it can be difficult to find an appropriate balance between high product quality and adequate step recovery.” To address the challenges in purifying AAVs, Alexion has found, in DiGioia’s words, that “nothing beats smart experimental design and time in the laboratory working with a product.”

The interaction between upstream and downstream steps must also be considered. “Any variability in bioreactor production directly impacts the work of the purification group,” DiGioia explains. “We believe that the mAAVRx production platform implemented by the Alexion upstream team has been impactful in improving our downstream development, as we have seen improvements in our overall upstream titers.” When there is more material for downstream steps, it is possible, DiGioia says, to “more easily standardize the earlier purification steps, as we have observed much less titer variability between our programs.”

At the Summit, DiGioia discussed adjusting the purification steps to work with an upstream process that produced a 15-fold increase in AAV vectors with the mAAVRx production platform. That increase in AAV allowed DiGioia and his colleagues to eliminate a concentration step with post-harvest tangential flow filtration and to load clarified harvest directly onto an affinity column. “In our runs,” he recalled, “we observed that this saved us a day of operation and improved overall process recovery.”

The biggest change, DiGioia says, was “the implementation of the ion-exchange step for empty/full capsid separation to replace the cesium chloride ultracentrifugation step we used in our legacy purification process.” Another helpful change was the addition of a chromatography step. “[It] gave us,” he points out, “a process that appeared to be much better suited for GMP production and ready to scale.”

These changes saved significant time. “The old process took about two weeks to perform,” DiGioia notes. “With our changes, we have been able to perform the purification process in one week—all while maintaining process recovery and potency compared to our legacy process.”

Concentrating on CMC

Nripen Singh
Nripen Singh, PhD
Passage Bio

To improve overall AAV manufacturing, Passage Bio focuses on standardizing chemistry, manufacturing, and control (CMC) methods, says Nripen Singh, PhD, head of process and product development at the company. Singh specifies that CMC methods are used to “select the right production system, optimize downstream processing, and develop standardized analytical methods and quality control assays that continue to be the most common challenges faced in AAV vector manufacturing.”

Currently, AAV-based manufacturing varies across the industry. “Companies use different production systems and downstream processes,” Singh explains. “[This variability suggests that] regulatory standards for manufacturing and quality control topics still need attention.” Although Singh expects that some of the manufacturing processes will be standardized over time, he observes that “key upstream processes will continue to act as important sites of differentiation between AAV gene therapy companies.”

According to Singh, the variations in manufacturing AAV-based therapies mean that gene therapy companies need “to overcome manufacturing challenges as they tailor their processes to their assets, while also making critical decisions at the asset and portfolio levels that will allow them to leverage developments in manufacturing to accelerate patient access.”

To address those challenges, Passage Bio built a portfolio strategy by using existing systems that work well with a specific therapy, and by developing proprietary technologies when needed. Through that combination, Singh asserts, “Passage Bio has been able to internalize all the major CMC capabilities that will enable the late-stage development and commercialization of gene therapies.” He adds, “We are also working toward standardization of our manufacturing and analytical platform that can be leveraged across our preclinical and clinical pipelines.”

Passage Bio has been participating in various collaborations to extend its capabilities. For example, the company has created a wide external network of GMP-compliant contract research organizations and contract development and manufacturing organizations. In addition, Passage Bio manages a dedicated suite at Catalent that is being used to carry out four Phase I programs. Finally, the company works with the University of Pennsylvania Gene Therapy Program to transition AAV gene therapies from preclinical development to late-stage development and clinical development stages.

All of this work and collaborative activity will really pay off as Passage Bio extends into new therapeutic areas. “As we start to shift attention from first launches in rare diseases toward development in non-rare indications,” Singh relates, “Passage Bio is also thinking ahead and making key manufacturing decisions early on that will become critical for success.”

Testing approaches to transfection

In one talk at the Summit, Stephanie Doong, PhD, a senior scientist at Vertex Pharmaceuticals, discussed how to improve AAV production in HEK293 cells. She noted, “Key challenges include achieving high productivity in the bioreactor and ensuring that the process is scalable.”

Vertex Pharmaceuticals lab
Vertex Pharmaceuticals, best known for producing medicines to target the underlying cause of cystic fibrosis, has a pipeline of investigational small-molecule drugs and cell and genetic therapies in other serious diseases. With respect to cell and gene therapies, Vertex says that it is building a portfolio of technologies, teams, and manufacturing capabilities. In this image, Vertex scientists (left: Anita Shrestha, PhD; right: Stephanie Doong, PhD) use a shake-flask model to screen for the optimal parameters for delivering DNA into HEK293 cells.

She points out that efficiently delivering DNA into HEK293 cells can be difficult: “There are many combinations of parameters to optimize in the transfection process. Reagent type, reagent-to-DNA ratio, complexation time, and DNA-to-cell ratio are just a few variables that can impact how well the DNA is delivered to the cells. In addition, these parameters need to be feasible in large-scale operations.”

To address some of those challenges, Doong and her colleagues took several approaches. “We identified a non-aggregating cell line and transfection reagents that worked with that cell line,” she details. “As combinations of parameters are important, we always tested several parameters simultaneously.” During these tests, the scientists measured the production with several DNA ratios to find the best reagent. They also tried various formulations of media and additives in search of the best conditions for transfection. “All of the initial screening work was conducted in shake-flask models, and the optimal combination of conditions was evaluated at a bench-scale bioreactor,” Doong recalls. “Our strategy increased the titer by more than seven times with equivalent product quality.”

Making the most of AAV manufacturing

The possible improvements in AAV manufacturing described here promise many benefits to bioprocessors and patients. This view is likely shared by the scientists who were quoted in this article. For example, Yancey says that the end result of his team’s work “could be a new wave of clinical programs that have been previously on hold due to high clinical supply requirements.”