Left to right: Patricia Seymour, Richard Snyder, Paulo Carvalho, Holger Wesche, and Marc Better (PHOTO BY LEAH J. ROSIN)

At noon on Tuesday, 20 June 2017, BioProcess International presented a panel discussion as part of the “Emerging Therapies” session of its BPI Theater at the Biotechnology Innovation Organization’s annual convention in San Diego.

Moderated by Patricia Seymour (senior consultant at BioProcess Technology Consultants), this panel comprised Holger Wesche (vice president of research at Harpoon Therapeutics), Richard Snyder (chief scientific officer of Brammer Bio), Marc Better (vice president of product sciences at Kite Pharma), and Paulo Carvalho (associate director of manufacturing at Alnylam Pharmaceuticals).

Emerging therapies — using cell and gene therapy, ribonucleic acid interference (RNAi), T-cell engagers, and others — present new and unique manufacturing challenges for biopharmaceutical companies. These include process development and manufacturability, access to manufacturing facilities (whether through outsourcing or building), and predicting future demand.

Discussion

After introducing the panel, Seymour gave each member time to describe his company and its programs. Then she facilitated discussion by posing a series of questions.

Paulo, not many companies can make your unique product (RNAi), yet you have a combination of in-house and outsourcing. How did your company decide to go with that strategy? Carvalho said that there aren’t many options for manufacturing RNAi. “There might be five contract manufacturing organizations (CMOs) that can robustly generate the drug substance repeatedly, and Alnylam has been using as many of those as possible while continually looking to expand that network.” The biggest challenge has been supply chain robustness. The company has brought some production in house because of challenges with manufacturing and partnership agreements. Bulk drug-product formulation for its lead program is why Alnylam got into internal manufacturing. Success so far has brought more manufacturing in-house. But the existing CMO network remains a requirement for the company to expand its pipeline and move programs into the clinic. The amount of material that can be produced is a challenge. Alnylam is mostly competing with itself for CMO space; however, a few other competitors have recently appeared.

All the panelists’ companies are producing first-in-class innovations. CMOs haven’t manufactured such products yet in compliance with good manufacturing practices (GMPs). So convincing them that GMP is a requirement is important and can take several inspections. Raw materials must be GMP sourced, which is also a challenge for CMO partners.

With novel products it can be difficult to find someone who has enough experience. How have you found people to hire or CMOs with the expertise to produce these products? Carvalho said that people with the necessary expertise typically work at the CMOs with which his company contracts. It is difficult to hire people away from them to work in-house without shutting down processes at the CMOs. Academia provides support, and retention is a big focus for keeping innovators who develop the products. The area of Cambridge, MA, is home to many experienced biotech workers, but people need to learn new ways of using process equipment.

Holger, will you talk about your different challenges? Wesche said that his company was founded just two years ago. It decided not to build a manufacturing facility, but rather to use CMOs aggressively. To its advantage, biologics production is “almost commoditized these days.” Doing so enables Harpoon (a small company with fewer than 20 people) to develop a new platform focused on engineering biology. While designing that platform, the company works with consultants and CMOs early on to gain knowledge and keep manufacturing in mind. For stable molecules, the same Chinese hamster ovary (CHO) cell line can be used from laboratory bench to full-scale bioreactor.

Wesche added that T-cell engagers involve a host of unique issues. For one, they are incredibly potent with doses measured in micrograms. Yield is not a major concern; however, these products are a challenge for analytics. “You have to be able to assess viability and measure aggregates and contaminants at low concentrations.” What gets into the solid-tumor targets will be at higher concentrations, so contaminants could cause more harm.

Are they designing new analytics or making existing analytics more sensitive? The challenge becomes even harder with 100-fold dilutions for in-use studies. Wesche said Harpoon goes back to bioassays, which have better sensitivities than other analytical methods.

Marc, you also had some interesting challenges. Better explained that Kite’s product is a single dose of engineered autologous T cells. A patient enrolls in a clinical trial and donates white blood cells, which go to Kite’s central manufacturing facility. There, they are purified and stimulated to get T-cell production started. The T cells are modified by introducing a recombinant gene to express the chimeric antigen receptor (CAR) protein. Cells are cultured and expanded until there are enough for a dose. Then they are washed, frozen, and tested to make sure that they meet specifications before getting sent back to the clinical center, where they are readministered to the same patient. Better said that half those people are getting a measurable long-term benefit that will prolong their lives substantially.

The developers were in academic laboratories five years ago. They had to figure out how to make the therapy commercially and how to work with vendors, build laboratories, and convert to a GMP environment. Challenges included finding cell culture media free of human serum and closing the system to prevent contamination. The first equipment used was the sort found in hospitals. Vendors didn’t support cell therapy manufacturing early on, but now it’s becoming an attractive market space for companies such as GE Healthcare and Thermo Fisher Scientific.

A further challenge was scaling the process. For each patient treated, the dose is ~900 mL. To treat 1,000 patients, the challenge is how to make product for individual patients side-by-side in small bags. Such closed systems are not difficult to manipulate, but the process is very different from typical biomanufacturing. There are no downstream processing steps in making this product. Living cells cannot be filtered and must be kept sterile at all times. That requires care and forethought regarding liquid handling.

At first, Kite depended completely on external suppliers for everything. It had wonderful experiences with CMOs and their ability to manufacture cell therapies in sufficient quantity for clinical trials (>300 patients have been treated). About two years ago, the company built clinical and commercial manufacturing facilities near LAX airport. Transportation logistics are critical for autologous therapies. The patients are very sick, so a couple days can make a difference. Turnaround must be as fast as possible (usually two weeks).

Kite has been mostly successful in solving those issues. It is taking the product to Europe and has partnerships in China and Japan.

Could you talk a little more about rapid analytics? Sterility tests are a big issue, but what other kind of tests are needed, and how do you turn them around quickly? What kind of staffing do you need? Better responded that FDA guidance allows for treatment of patients before all the analytical tests are complete (in clinical trials). A 14-day sterility assay takes too long, for example. So products are released based on preliminary sterility testing and a gram stain for gross contamination. But for commercial products, “you would probably want a complete sterility package before you start treating patients.” Kite has initiated a rapid in-house sterility method: a complete and fully validated test in seven days. That allows the company to make and completely release the product, then ship it. Another favored rapid analytical method is polymerase chain-reaction (PCR)–based mycoplasma screening.

“In this case, “right-first-time” is literally a matter of life or death. If it doesn’t go right, and you can’t process the dose for that patient, you might not get another opportunity.” Better said that Kite has back-ups built into its systems to ensure a 99% success rate. Although a single dose is produced, often enough cells are expanded to make two doses. Only once during the trial has a second dose been needed. Starting materials can be frozen as well, so in case of a catastrophic failure, the process can be run again.

What is Brammer Bio doing to help solve some of these issues? Snyder replied that his company’s main challenge is the diversity of products it manufactures. The CMO has established platforms for meeting necessary requirements. For viral vectors, it uses five different platforms, is adept at scale-out processes, and has adopted a number of analytic technologies (e.g., digital PCR and analytical ultracentrifugation).

Can you describe any examples of challenging projects? A benefit of working with Brammer, said Snyder, is its deep knowledge of viruses, vectors, and related advanced therapies. Clients have brought in vector configurations with inappropriate viral-genome elements. The CMO encouraged them to engineer those out and then it did so. To advance products into commercial stage, the company not only acquired two facilities from Biogen (now being outfitted for viral vector production), but it also hired 100 Biogen employees who bring 20 years of commercial experience.

Audience Questions
The first audience question was about new analytical methods and tools used to characterize the products. An example was given on measuring potency for understanding whether a product has bioactivity and how relevant that bioactivity is to a given individual (and across products). The classic method for autologous T-cell products is to put them together with cells expressing relevant antigens on their surfaces. Typically, gamma interferon is measured, but recently other types of cytokines have been as well. Analysts can look at what a target cell does instead of just looking at what T cells do: Does it die (apoptosis), and what can be measured to indicate potency?

For personalized treatments, what kind of data system are you using to keep track of batches from beginning to end? Kite uses paper batch records at this point but plans to become more sophisticated in the future. It does have an electronic system for bar-coding and uses an Oracle database to reconcile raw materials. Keeping track of all the data generated and correlations between batches and clinical results is handled by a sophisticated analysis system.

A Final Question
Seymour’s final question to the panel was “What was the biggest challenge you overcame recently?” Snyder’s answer was rooted in capability and capacity. Brammer is doubling its cleanrooms in Florida, brought on facilities in Massachusetts, added 115,000 ft² to its facilities in January, and has added a capable commercial team. This will enable the company to assist clients with market applications and commercial launches.

Carvalho said that Alnylam’s biggest challenge is leading up to its new drug application (NDA), coordinating with collaborators, and ensuring that the chemistry, manufacturing, and controls (CMC) sections are robust for a harmonized filing across countries. The manufacturing process is well defined, and its success rate is high. So the company is spending time on characterization, gap closing for NDA filing, and commercial readiness.

From Harpoon’s point of view, Wesche is most proud of the quality and speed of developing a new platform without in-licensing anything. The second thing was getting two product candidates ready for INDs in about two years. What will be challenging is designing a manufacturing process and meeting regulatory requirements. Growth brings more oversight and levels of bureaucracy, so another challenge will be to keep moving fast and delivering the same quality without forgetting that patients cannot wait.

Better said that Kite is building a commercial organization to market its products. The company has grown and is putting a great deal of effort into sustaining the growth. Critical to success is working with vendors to ensure supply because many of them are small companies or new to the market without their own infrastructure. Kite is working closely with them to make sure that the supply chain doesn’t get interrupted.

Alison Center is editorial assistant (acenter@bioprocessintl.com) for BioProcess International, PO Box 70, Dexter, OR 97431.