Guest Column | December 12, 2019

3 Keys To Realizing FDA's Vision For CAR-T And Other ATMPs

By Mark F. Witcher, Ph.D.

Janet Woodcock’s recent quote on pharmaceutical development — “If the guy under the bridge cannot get the $2 million CAR-T cell curative therapy, then we have failed” — brings the cost of commercializing new products into sharper focus.1 While much can be done to improve patient access through various social support and payer systems, the fundamental problem remains the time and cost of developing and manufacturing complex biopharmaceutical products, including CAR‑T cell therapies.

Dr. Woodcock continues, “We need the whole system to evolve and change if we’re going to do what we set out to do: help every patient feel better and live longer.”1 This evolution will require the industry to achieve the FDA’s 21st century vision of “a maximally efficient, agile, flexible pharmaceutical manufacturing sector that reliably produces high quality drugs without extensive regulatory oversight.”2,3 This article proposes methods for evolving the industry’s approaches for meeting the many challenges and opportunities of CAR T cell therapies as well as other advanced therapy medicinal products (ATMPs).

In a recent excellent article, Anamika Ghosh and Dana Gheorghe (G&G) reviewed the history, limitations, and opportunities of CAR T cell therapies.4 While cell therapy remains a very promising approach to developing much-needed new immunotherapies, some significant challenges will have to be overcome in order to realize the FDA’s 21st century vision of making complex ATMPs widely and cost effectively available to patients.

The limitations and opportunities presented by G&G can be reviewed to define a path for widely commercializing CAR T cells specifically and ATMPs generally. As described by G&G, autologous CAR‑T therapy challenges include:4

  • A three- to four-week “vein to vein” delivery delay required for processing the autologous therapy,
  • High cost of goods and intrinsic variability associated with processing a single patient’s cells,
  • Risks associated with autologous cell processing that include exposures to contamination and supply chain problems,
  • Significant donor uncertainty from initial cell acquisition issues combined with the impact of the patient’s prior treatment regimens,
  • Developing products for solid tumors and new indications, and
  • Significant supply chain risks associated with both centralized and distributed business models for delivering the therapy.

However, G&G also identify a wide variety of important opportunities for meeting the above as well as other important challenges. These opportunities include:

  • Possible use of a portfolio of allogeneic products selected for the patient based on tumor and patient genetic, epigenetic, or other selection criteria,
  • Appropriate allogenic products can be delivered rapidly to begin treatment as soon as possible,
  • Allogeneic manufacturing will have a significant impact on cost of goods from economies of scale while increasing safety by using standard product release mechanisms,
  • Various combinations of multifunctional CAR T cell options can be made available,
  • Formulations for different delivery methods can be provided, and, very importantly,
  • The portfolio of allogeneic products might also contain or not contain various modulators and inducers that could include imbedded pro-inflammatory or immunosuppressing cytokine genes based on a wide variety of factors to increase the therapy’s potency.

The bottom line for successful widespread commercialization is that a large number of different products will have to be both rapidly developed and manufactured at highly uncertain clinical and commercial scales and capacities. Products will be developed in a high-risk environment that includes risk of product failures, highly uncertain demand, and possible unpredictable disintermediation by follow-on versions, such as gene therapies that prove to be more effective.

The objective of any product development effort leading to commercialization is to push the critical path through clinical testing. Given the use of surrogate endpoints, other expedited review approaches, and increasing experience levels with ATMPs, very sophisticated tools and methods will be required to keep the clinical trials on the critical path. The following discussion includes methods that can be used to rapidly develop these and nearly all new 21st century products.

Meeting these challenges will require three complementary and interacting approaches. The first is a process development path that rapidly and efficiently organizes, develops, and validates the manufacturing process. The second is an approach to build the many control strategies and systems for designing and then operating the manufacturing process. The final approach provides a manufacturing facility that can seamlessly develop a large number of products simultaneously from late-stage process development, clinical manufacturing, and then from initial commercial launch through long-term manufacturing. The facility must be capable of supporting the many process scales, configurations, and campaigns that might be required to support the product’s complex and highly uncertain development manufacturing life cycle.

Product/Process Development Path

Given the old, sometimes forgotten wisdom that “the process defines the product,” the primary task of product development is to develop an effective manufacturing process. Efficient and agile process development methods are critical to rapidly develop any product. Process development follows a four-stage cycle that seeks to answer four basic questions: What, How, Will it work, and Did it work. For any biopharmaceutical manufacturing, especially advanced therapies, all four questions are extremely important. Each question can be viewed as a stage in the product development life cycle.5 Briefly, the four stages are:

0 – Define (What?) – Define the product’s critical quality attributes (CQAs) and associated product and process requirements to meet patients’ needs with respect to capacity and cost, etc.

1 – Design (How?) – Design the process and provide the manufacturing capacity necessary to achieve the goals set in the define stage, then develop the process through the following two execution stages to provide the product for clinical testing and commercial supply.

2 – Qualify (Will it work?) – Before any important process is operated to produce material for therapies, it should be appropriately tested within the commercial manufacturing facility to assure it will reliably produce safe and effective high-quality product. As the product moves through its development life cycle, all product material should be qualified to achieve its necessary quality goals to assure successful development and patient safety and efficacy.

3 – Operate & Verify (Did it work?) – If properly designed, the process should both control and provide proof of control such that the product can be used with a very high degree of assurance that it meets the required quality goals.

The four questions can be readily answered using a lifecycle process development and validation (LPDV) paradigm.5 The primary goal of using an LPDV approach is to build and understand the essential product and process knowledge and structure to identify and build the necessary control strategies and systems.

Control System Development

The shortest definition of good manufacturing practices (GMPs) is to “control everything to minimize product quality risks.” The most effective product development approach is to use appropriate GMPs through every phase of product development.6,7 Two tools can provide valuable methods for efficiently building control strategies. The first tool is a well-structured design space (ws-DS)5 that describes all the process inputs and outputs such that appropriate control strategies can be established for all inputs and outputs.

For biopharmaceuticals, especially advanced therapies, product quality is defined by a combination of known and measurable CQAs as well as a set of complex, but very real, unknown CQAs (u-CQAs) that are neither known, understood, or measurable. The product’s quality (both CQAs and u-CQAs) is controlled by the process’ behavior and performance as the product attributes (CQAs and u-CQAs) are formed by the sequence of unit operations that make up the manufacturing process. The process’ performance for each unit operation is measured by critical process responses (CPRs).5 Examples of CPRs include viability, yield, dCO2, etc. The process’ behavior and performance are determined by controlling the following three classes of input variables or parameters:

  • Equipment Parameters (EPs) – EPs define the equipment and instruments used in executing the process. Most are defined during process development and engineering and provide the basis for user requirement specifications (URS) when the equipment is selected or acquired. Examples of EPs are tank volumes, sizes, capacities, stirrer types, etc. While many remain fixed, a few may change with usage, requiring control systems such as maintenance to assure they remain under control.

  • Operating Parameters (OPs) – The performance and behavior of the process is largely determined by OPs. Many are established during process development, such as buffer concentrations and linear velocity in chromatography columns. Others are actively controlled during operation, such as heat input to control temperature or base addition to control pH.

  • Material Parameters (MPs) – MPs are material attributes that fall into two categories. The first is raw materials, such as media and buffers that may contribute CQAs and u-CQAs, such as contaminates. The second is the in-process materials from the previous unit operation described by the CQAs from the previous unit operation. However, this material may also contain u‑CQAs that eventually contribute to the final product’s u-CQAs.

With the process inputs and outputs defined for the process sequence, the next step is to use the second tool — quality risk management (QRM) — to evaluate, mitigate, and accept all the process risks. Controlling risks is the basis for all control systems. Using a simple threat — process — risk structure, control systems for all the inputs and outputs of the well-structured design space described above can be developed. The best QRM method is to identify and structure the process inputs and outputs using a system risk structure (SRS) approach.8 The SRS can be evaluated using prospective causal risk modeling (PCRM) with additional knowledge of the various elements of the process gathered through testing, especially design of experiments (DOE).8

Flexible Manufacturing Capacity

Product development requires a lot of very challenging manufacturing as the product moves through the preclinical, clinical, and commercial launch sequence. Controlling and assuring the quality of the product requires an integrated manufacturing capability that minimizes or eliminates tech transfer by using the same human, procedural, and facility resources to make the products for each stage. Inadequate manufacturing capability and insufficient resources can threaten the successful scale-up of any product.9 A multipurpose facility (MPF)10,11,12 provides a manufacturing facility that is sufficiently flexible and adaptive to run almost any process format at a scale appropriate to efficiently accomplish the required manufacturing goals for every stage from preclinical to commercial manufacturing. The MPF is designed to quickly install and remove processes ranging from suspension and attachment cell culture systems, media and buffer preparation, to any purification trains that may eventually include continuous manufacturing if appropriate.


The three approaches described provide methods for minimizing product development’s impact on the critical path for developing new products. The tools increase the efficiency and reliability of the development effort to successfully create a wide variety of different products. When Dr. Woodcock’s quotes are combined, her other statement — “It’s not working and it won’t work in the future” — illustrates that significant evolution is critical to the industry’s long-term ability to provide for patients’ needs. While the above methods and tools provide some of the necessary approaches for change, many new methods and approaches need to be developed and refined to provide patients with the rapid delivery of cost-effective therapies to satisfy critical medical needs. If the industry takes good care of the patients, the patients will take good care of the industry.


  1. Baumgaertner, E., “‘It’s not working’: An FDA insider’s view of where medical innovation falls short,” Los Angeles Times, Sept. 7, 2019.

  2. FDA – Pharmaceutical CGMPs for the 21st Century – a Risk Based Approach (September 2004)

  3. Yu, L.X. and M. Kopcha; The future of pharmaceutical quality and the path to get there, Int. J. of Pharmaceutics, 528 (2017) 354-359

  4. Ghosh, A. and D. Gheorghe; CAR T-Cell Therapies: Current Limitations & Future Opportunities; Cell & Gene, September 26, 2019

  5. Witcher MF. Integrating development tools into the process validation lifecycle to achieve six sigma pharmaceutical quality, BioProcess J, 2018; 17.

  6. Witcher, M. F.; “Achieving Excellence in Biopharmaceutical Development and Manufacturing by using Appropriate Manufacturing Practices (AMPs),” BioProcess J, Vol. 14, No. 4, Winter 2015.

  7. Witcher, M. F., “Using a Patient Centered Risk-benefit Structure and Appropriate Manufacturing Practices (AMPs) for Successfully Developing and Manufacturing Effective Cell Therapy Products,” BioProcess J, Vol. 15, No. 2, Summer 2016.

  8. Witcher MF. Estimating the uncertainty of structured pharmaceutical development and manufacturing process execution risks using a prospective causal risk model (PCRM), BioProcess J, 2019; 18.

  9. Witcher, M. F., “Phase III Clinical Trials – Ever Wonder Why Some Products Unexpectedly Fail?” ISPE iSpeak Blog Pharmaceutical Engineering, Aug. 7, 2019.

  10. Witcher, M. F. “The Facility Challenges of Developing Continuous Process based Biopharmaceutical Products,” March 2019.

  11. Witcher, M. F. “How FDA’s 21st Century Goals can be realized by using a Multi-purpose Manufacturing Facility,” Pharmaceutical Technology, 2018.

  12. Witcher, M. and H. Silver, "Multi-Purpose Biopharmaceutical Manufacturing Facilities Part 1: Product Pipeline Manufacturing," Pharmaceutical Technology 42 (9) 2018.

About The Author:

Mark F. Witcher, Ph.D., has over 35 years of experience in biopharmaceuticals. He currently consults with a few select companies. Previously, he worked for several engineering companies on feasibility and conceptual design studies for advanced biopharmaceutical manufacturing facilities. Witcher was an independent consultant in the biopharmaceutical industry for 15 years on operational issues related to: product and process development, strategic business development, clinical and commercial manufacturing, tech transfer, and facility design. He also taught courses on process validation for ISPE. He was previously the SVP of manufacturing operations for Covance Biotechnology Services, where he was responsible for the design, construction, start-up, and operation of their $50-million contract manufacturing facility. Prior to joining Covance, Witcher was VP of manufacturing at Amgen. You can reach him at or on LinkedIn.