June 1, 2016 (Vol. 36, No. 11)
Meghaan M. Ferreira Ph.D. Contributor GEN and Clinical OMICs
Protein A Is Adapting To New Manufacturing Challenges and Becoming More Competitive All the Time
Protein A continues to be the method of choice for the purification of monoclonal antibodies (mAb), a sector of the biopharmaceutical industry that has grown substantially since the first FDA-approved therapy in 1986.
On average, four new mAb products are commercialized each year, and three new therapies have already received FDA approval in 2016.
“Taking into account continuously increasing titers and the high doses required for mAb therapy, it is obvious that downstream processing forms the bottleneck in manufacturing operations,” says Eric Langer, president and managing partner at BioPlan Associates. Despite concerns about ligand leaching, durability, low capacity, and high cost (typically over $10,000/L), companies remain reluctant to abandon Protein A for mAb purification.
According to BioPlan Associates’ 12th Annual Report and Survey of Biopharmaceutical Manufacturing Capacity and Production, the percentage of biomanufacturers planning to move away from Protein A purification for existing commercial production units in the next 12 months has decreased from 14.0% in 2013 to 6.9% in 2015. “The current near total dominance of Protein A products for initial monoclonal antibody capture can be expected to continue,” reports Langer. “Industry is open to considering Protein A alternatives, but with few available, or at least, few yet proven and documented cost-effective at large scale, most have not yet formed strong opinions, and are sticking with Protein A resin products for lack of any better alternatives.”
In fact, according to Jonathan Royce, product manager for antibody affinity media, GE Healthcare Life Sciences, companies developing biosimilar versions of the legacy antibodies losing patent protection are electing to purify their products with Protein A, even in cases where the original versions used alternative methods.
“There’s absolutely no technical reason that you can’t replace Protein A,” explains Royce, “but the loss of simplicity and the loss of purity and yield do not make the switch one that’s either cost-effective or effective from a manufacturability standpoint when you start to think about doing regular commercial manufacturing.”
Its broad applicability as an initial capture step across multiple mAb products makes Protein A one of the key processes that enables the production of multiple products in the same facility. Leveraging their existing infrastructure in this way, allows companies to reduce both cost and the likelihood of manufacturing deviations.
Additionally, the purity and yield obtained with Protein A remains unmatched; the high specificity and affinity of Protein A can provide removal of more than 98% of impurities in a single step. “Protein A is a fascinating result of millions of years of evolution inside a bacteria that created its own immune system to fight our immune system,” explains Royce. “It’s sort of hard to engineer something that’s better,” he adds.
Mother nature is not the only one who had a hand in the evolution of modern Protein A resins. User demands have required Protein A suppliers to continually evolve their products as the industry has grown. Royce observes that, since its initial release of a recombinant Protein A product in 1996, GE Healthcare has brought a new Protein A resin to market approximately every five years. Each new product generation addressed the latest challenges facing Protein A chromatography, generating resins capable of handling larger volumes with longer lifetimes, greater alkali stability, and higher capacities at lower residence times.
Lengthening the lifetime of Protein A resins and improving their ability to withstand cleaning with harsh chemicals has indirectly reduced costs by increasing the re-usability of columns, but the cost of manufacturing the resin remains one of the most formidable adversaries to Protein A chromatography. Drawing from the corporation’s vast experience in making polymer beads for industrial applications, Purolite Life Sciences plans to reduce production costs by using a patented continuous process, called “jetting,” to manufacture the agarose beads used in their Protein A affinity resins. Purolite already uses this process to generate beads for the majority of its polymer resins.
“Today, the vast majority of agarose resins are produced using traditional batch emulsification, resulting in a very broad particle distribution that requires extensive sieving to narrow the distribution,” explains Hans Johansson, application manager at Purolite.
Jetting generates more uniform beads with tighter size distributions than traditional batch emulsification, and the increased bead uniformity can lend itself to more robust chromatography methods by improving pressure and flow properties. If successful, the new process will also decrease solvent consumption and raw material waste, which will reduce both manufacturing costs and environmental impact.
While Johansson does not view Protein A chromatography as a bottleneck for regular manufacturing, he does see concern from customers about the cost of manufacturing therapeutics for early-phase clinical trials. “Once you have a product on the market the price of Protein A actually doesn’t make a big difference, because it’s such a small part of the overall cost anyway. But most of the products going though clinical trials will eventually fail, and then the cost is quite substantial,” he says.
To address these concerns, Purolite brought Praesto™ AC to market. The cost-effective Protein A resin is designed for clinical manufacturing, where the resin is used only for a few cycles, and it uses a recombinant nonalkaline stable version of the Protein A ligand no longer under IP protection to reduce manufacturing costs, Johansson explained. The resin exhibits a shorter lifetime than others on the market, but Purolite has shown that it retains 90% of its capacity after 20 cycles using NaOH to clean the column. For regular manufacturing, Purolite recently introduced a new, high-capacity, alkaline-stable Protein A resin named Praesto AP.
Perhaps the biggest contribution to resin cost is the Protein A ligand itself. MicroProtein Technologies founder and president Vinod Pandiripally, Ph.D., claims the company has developed a new way to produce recombinant biologics, including Protein A, on an industrial scale with lower manufacturing costs than many competitors.
To enable an inexpensive and reliable production of various proteins, enzymes, and plasmid DNA, MicroProtein Technologies developed a solid-phase manufacturing platform, called the MPTxpress, that bypasses many of the complexities faced in standard liquid fermentations. The MPTexpress system consists of an incubator holding approximately 800 square 12-inch trays filled with gelled agar culture media and covered by a semi-permeable membrane. Bacteria are harvested by scraping the thick cell paste off the membrane after inoculating the trays with transformed E. coli and culturing overnight. The system’s modular design makes it highly adaptable and easy to scale, according to Dr. Pandiripally.
In contrast to conventional liquid fermentations, the MPTxpress system does not have the same requirements for energetic mixing, heating, and cooling, which minimizes the need for complex control systems and simplifies the infrastructure required for a manufacturing facility. However, according to Dr. Pandiripally, “the real advantage [of the MPTxpress platform] is very high specific yields. If I take one gram of cell paste, we can get up to 300 mg titers per gram. That’s unheard of for any process in fermentation.”
In April, MicroProtein Technologies launched bulk distribution of a Protein L ligand for the purification of antibody fragments. Like other companies in the business, MicroProtein Technologies is expanding its portfolio to meet the new demands of biopharma. Biopharmaceutical companies are developing therapies based on antibody fragments and bi-specific antibodies; these therapies have interesting clinical properties, but they can not be purified using Protein A.
“Where Protein A can be used, I think by and large people are sticking with Protein A. Where I think people are investigating alternatives are cases where they’re exploring a diversification of their antibody pipeline and Protein A is no longer an option because of the lack of that Fc region,” explains Royce from GE Healthcare whose portfolio already includes Protein L- and camelid-based alternatives Capto L and LambdaFabSelect.
Despite the reluctance of biomanufacturers to adopt alternatives, there is no shortage of buzz around new and upcoming downstream processing technologies. According to Renaud Jacquemart, Ph.D., principal scientist and director, vaccines process sciences, Natrix Separations, the solution is not to move away from Protein A, but rather to move Protein A into a new format: “utilizing a faster, flexible, and more affordable technology still leveraging the power of Protein A is a better option than trying to walk away from this industry standard.”
Natrix Separations specializes in membrane technology, and while they currently offer a salt-tolerant cation exchange membrane (HD-Sb) as an alternative to Protein A for mAb purification, their customers have pressed for the development of a Protein A membrane. As a result, the R&D team at Natrix Separations has entered a collaboration with Merck & Co., as well as other key players in the industry, to develop a high-productivity Protein A membrane.
“Many people in the industry see membrane chromatography only as a polishing option,” comments Dr. Jacquemart, “The first generations of membranes had low binding capacity and were better suited at removing trace impurities than capturing large quantities of the target mAb. The productivity was significantly hindered when compared to traditional Protein A resins despite faster flow rates.”
While skeptics may question the ability of a membrane to compete with current chromatography resins due to restrictions in surface area, Natrix reports that its Protein A membrane prototypes exhibit similar or higher binding capacities than existing resins with process speeds more than two orders of magnitude faster. To achieve this performance, the membrane technology uses a three-dimensional macroporous hydrogel structure with a high density of binding sites and rapid mass transfer.
Natrix believes its membrane technology is one example of how new technologies are reformatting Protein A. For now, biomanufacturers remain reliant on Protein A and find that its advantages make it well worth the price tag, and the main driving force behind the adoption of alternatives relies on the expansion of the pharmaceutical pipeline as antibody fragments and bi-functional antibodies come to market.
Over the last 40 years the biotechnology industry has seen a clear shift in processing bottlenecks from cell culture to downstream unit operations. This trend is, in part, the result of substantial increases in bioreactor titers driven by innovation in cell-line engineering and media and upstream process development. The ubiquitous discussions on downstream process bottlenecks in the literature provide clear evidence of this trend. With that said, increased resin loadings, larger columns, increased in-process vessel volumes, and bigger skids are not going to solve all the problems.
“Continuous downstream processing using a variety of different techniques, most incorporating simplified versions of simulated moving bed, are being evaluated at most large biotechnology companies as well as some smaller ones,” explains Paul Jorjorian, director global technology transfer at Patheon.
“Another more advanced downstream bottlenecking technique involves the coupling of downstream unit operations, which can incorporate the use of inline buffer and product adjustments. In this case the product may run continuously through the post-capture polishing and viral filtration steps, and even be coupled to a continuous capture step.
“As a contract and development manufacturer operating multiple sites globally our ability to select for technologies such as continuous processing is more limited due to a mixture of both in-house and customer-developed processes,” notes John Ward, vp, engineering.
“In our experience, intelligently balancing the utilization of our suites can both generate and alleviate bottlenecks depending on the product mix we are running. In general, we find putting as many reactors (or batches) as possible through the downstream suites, in a balanced manner, is our critical lever to maximize throughput.
“Executing on this methodology presents unique challenges when a diverse product mix and demands for multiplexing (i.e., running of multiple bioreactors into a single downstream) are considered.”
Both Jorjorian and Ward say the company’s approach to debottlenecking in a stepwise manner by maximizing equipment utilization and then incrementally adding additional equipment, support functions (i.e., buffer prep or WFI generation), and automation. It is worth noting that at times this means a transition from single-use to stainless steel systems or via versa,” points out Jorjorian. “This approach is by no means cheap, but is cost-efficient with respect to capital investments having the largest impact on throughput.”
“Although we often talk about debottlenecking in terms of technical demands and capital requirements, the softer side of the business should also be considered,” continues Ward. “The rapid growth in the biotechnology sector over the last 20 years has resulted in a shortage of experienced people. The inability to, or delay in hiring qualified individuals for technical positions may end up bottlenecking certain departments and ultimately product delivery. The other bottleneck is time, specifically with respect to accelerating the time between drug discovery and first in human trials.”
The Patheon officials explain that these softer bottlenecks are unrelated to the ability of manufacturing equipment to deliver product. When considering bottlenecks it is important to account for equipment capabilities, upstream and downstream line balancing, and the softer side of product realization, they add.