Guest Column | February 17, 2021

Developing Future Delivery Models For ATMPs: Practical Considerations

By Erich H. Bozenhardt and Herman F. Bozenhardt


Advanced therapy medicinal products (ATMPs) are therapies based on gene editing, cells, and/or derived tissues. In the U.S., the FDA refers to these therapies as human cells, tissues, and cellular and tissue-based product (HCT/P). In the words of the EMA, “They offer groundbreaking new opportunities for the treatment of disease and injury.” One of the types of ATMPs that has gained a lot of attention is chimeric antigen receptor T (CAR T) cell therapy . There are more than 500 individual clinical trials within the CAR T cell therapy/ex-vivo gene therapy space. This type of therapy is estimated to impact a patient population of 900,000 in the next five years and 2 million in the next 10 years.1

There are multiple other cellular-based therapy types in trials that are targeting even broader areas. With all the promise and technology, the production rate of the current manufacturing systems is incredibly slow. Therefore, there is real need for scalability. Today’s CAR-T facilities are planning for treating 2,000 to 10,000 patients a year per facility with individualized treatments. That would imply there will be 90 to 450 facilities to address the potential patient population in the next five years for CAR-T therapies alone. We as an industry will need the facilities to increase their footprints and scale out to process more batches. With this seemingly daunting task, the industry might require hospitals to take up some of the manufacturing. Volume and patient outreach are the now the predominant concern.   

With a small number of approved cell therapies/ex-vivo gene therapies, and with those approved therapies reaching small patient populations, there is not a proven strategy to answer the questions about large-scale commercial manufacturing. The developers of these processes are taking different strategies for commercial manufacturing (e.g., centralized, regionalized, point of care, outsourced) as they weigh several factors.   

Process Types And Key Aspects In The Process

There are many factors to consider in the commercialization of ATMPs. Some of the major ones are:

  • process development approach – plan to continue development after starting trials
  • automation – transition to increasingly greater levels of automation to improve efficiency and reduce cycle time
  • closure – close the process and leverage biopharma’s functionally closed and aseptic technologies and methods
  • logistics – the supply chain of the volume of needed disposables, cryogenic transport of the raw material and product, and the client-specific nature of the autologous chain of custody
  • facility capabilities – flexibility and adaptability, whether in a hospital, clinic, or pharma plant, all accommodating the process development plan

New technologies like allogenic CAR-T will shift which factors and advanced technologies to emphasize and which risks to address but without removing any of the factors above from consideration.

Key Aspects Of A Process Development Plan

The current processes are first-generation products that are jumping from research labs to manufacturing in an expedited fashion. Speed to market is more than being the first mover in a therapeutic class; it is a lifesaving drive to solve an unmet medical need. The FDA and EMA have recognized this and have provided pathways to allow accelerated approval. The challenge for us is to develop these novel therapies in less time. A well -thought-through process development plan is key. This will not only direct the development scientists, but it will drive the facility design and be a discussion point with the regulators. Laying out where the process technology is today and how you intend to evolve it from an engineering viewpoint allows a practical approach to building flexibility into the facility without trying to “design for everything.” The following are the keys to that path:

1. Process

Current approved treatments are for small patient populations, but even at this scale we are seeing that many manual operations are not sustainable and are fraught with risk (deterioration, contamination, etc.). The advantages of automation are convincing, but specific applications like robotic handling are still being developed. That development is a siloed effort where therapy providers are working to come up with solutions themselves. Many are pursuing this path because they believe their processes are unique, but are they really?

At the same time, equipment suppliers are trying to develop innovative systems before customers ask for a solution. These suppliers, who have the advantage of seeing different processes in development, have developed solutions that cover the majority of the CAR-T process. They have taken equipment development at different ends of the process delivery spectrum, to produce solutions that use different equipment for each operation and, in some cases, an “all-in one” machine. This gives the therapy process developer a defined set of tools or a platform to work from, assuming they research the market rather than taking on the equipment burden themselves.

2. Closing the open process using off-the-shelf tools

Process closure is a major aspect of de-risking and scaling out these therapies to provide simplicity and maintain an aseptic environment.  There has been a lot of discussion, specifically, around ways to close the process for CAR-T and for this class of ATMPs:

  • The addition of some materials (e.g., immunomagnetic beads)
  • Initial processing for the apheresis bag requires consideration of functionally closed systems using filtration or aseptic techniques that are typical to medical practice but foreign to the biopharma industry.
  • Looking at other classes of ATMPs that start with a patient biopsy or require bedside processing, the need to evaluate functionally closed systems and conduct a risk assessment become critical.
  • “Portable” assemblies or systems can be plugged into a hospital, clinic, or home. These would be single-use pre-sterilized systems with little skill required for use.

3. Automation

Many of the processing technology solutions being deployed today increase the level of automation in a process, but for only one specific step.  Even with the all-in-one type of machines, the handling of individual patient batches for autologous therapies still needs to be done to maintain the entire supply chain from patient to process and back to the patient.

Currently, organizations are utilizing paper or “paper on glass” batch records for manufacturing in support of clinical trials. Batch record review/release has become a bottleneck. EBRS (electronic batch record systems) could be used to reduce it or at least manage the supply chain data as processes scale.

Chain of custody tracking is another layer in the data management-heavy process. Tying the materials to equipment and operator schedules across thousands of batches per year that are running in parallel requires a manufacturing execution system (MES). This multilayered system of data is critical for operation of the facility and its impact on the process. IT and automation groups need to be brought into the design early to assess the needs of the different networks, servers, and wide area broadband access and to plan the space. Beyond the network closet and server rooms, the devices the operators use to interface with the different systems need space and charging locations.  The operators themselves need a simple, small handheld universal device that can be used everywhere in a plant.

With the rapid pace of process development, robotics is a consideration, although it currently has limited practicality because it typically serves one function. Liquid handling systems can deal with dosing and loading wells but require humans to move them around. The practicality of employing robotics on allogenic therapies at commercial scale is favorable even if it is only at robotic filling and packaging. We do expect many new robotic applications in the near future.

4. Logistics

The first generation of cell therapies are autologous, thus requiring the collection of a donation from a patient and individualized tracking back to the patient. Procurement of the donation from the patient (the first step in the process) is going to be happening in a clinical setting. There may be a need to supply the required consumables or have the process adaptable to variations in containers/quality of material received.

Most therapies are planning for cryogenically frozen donations, which will require specialized shippers and cold chain handling (and documentation). There are a few therapies that are planning for simply refrigerated donations. These require a short (on the order of hours) transport time, limiting the distance from clinic to manufacturing site.

When manufacturing, there is a need to segregate material from different patients and protect operators from blood-borne pathogens or viral vectors. To find the right facility solution, we need to focus on the closure discussion above, because we may have to design the facility for BL-2 operations. Fully closed systems can utilize a common area with controls around material movement. Fully closed systems need to ensure that even disconnected materials to be discarded are not opened, thus creating a biosafety risk. Functionally closed systems need to consider the risk of cross contamination associated with the particular closure method. Most prudent is to conduct the closure in the local protection of a biosafety cabinet. Open operations require rigorous cross contamination controls, typically limiting only one patient’s material to be processed at a time. These operations require facility designs that utilize walls or dividers to segregate flows (air, people, materials). Ultimately, for efficiency and expediency, we may need to design these facilities like a biopharma cell culture facility with uni-directional flows of materials, patient tissue, intermediates, products, waste, and personnel. This will also require HVAC design to provide containment.

Allogenic therapy processes mitigate but don’t remove the all the logistics issues of delivering cell therapies. The shipment and storage of the final product require close coordination with the receiving clinic. Our main challenge is the logistics around the hospital, especially if it is not providing any of the process. We need to determine if the hospital is equipped to store the material and how is it received and handled at the hospital.

The impact of logistics (consumables, patient donation, and distribution) will lead to a large and highly trafficked warehouse. This makes remote warehousing impractical at large scale.

5. Facility requirements

Site selection is always a major consideration for any facility. Proximity to hospitals and available transit systems is more critical than ever. For many of these therapies, time is truly of the essence. Maintenance of safety stocks of single-use assemblies and chemicals is a typical consideration in sizing a building. What manufacturers have been seeing is that certain critical materials, like viral vectors and plasmids, have lead times of over a year at current demands. As the scale of production increases, manufacturers need to consider if it makes sense to invest in outsourced product or to internalize manufacturing. If they internalize the manufacturing, they can radically reduce their risk, control their supply chain, and, importantly, maintain their quality.


Process development approach, automation, process closure, and logistics are the major factors to consider in developing a strategy for commercialization of ATMPs. The large-scale manufacturing of these therapies is transitioning from a fledgling experiment to a calculated business risk as many companies build out facilities that will come on-line in the next few years. Delivering these therapies at large scale will eventually change the outlook for millions of patients.



About The Authors:

HermanHerman Bozenhardt has 44 years of experience in pharmaceutical, biotechnology, and medical device manufacturing, engineering, and compliance. He is a recognized expert in the area of aseptic filling facilities and systems and has extensive experience in the manufacture of therapeutic biologicals and vaccines. His current consulting work focuses on the areas of aseptic systems, biological manufacturing, and automation/computer systems. He has a B.S. in chemical engineering and an M.S. in system engineering, both from the Polytechnic Institute of Brooklyn. He can be reached via email at and on LinkedIn.

ErichErich Bozenhardt, PE, is the lead process engineer for regenerative medicine operations at United Therapeutics. He has 14 years of experience in the biotechnology and aseptic processing business and has led several biological manufacturing projects, including cell therapies, mammalian cell culture, and novel delivery systems. He has a B.S. in chemical engineering and an MBA, both from the University of Delaware