Guest Column | October 8, 2019

Why Aren't More CDMOs Using Continuous Manufacturing For API Production?

By Ben Littler, senior director, process chemistry, Vertex Pharmaceuticals

Continuous Much More Than Manufacturing

Continuous manufacturing (CM) of small molecule active pharmaceutical ingredients (APIs) is an emerging technique that has garnered much attention in recent years,1 but many factors need to be overcome before API CM is routinely implemented.2 One barrier to widespread implementation of API CM is that a global network of contract development and manufacturing organizations with API CM expertise and equipment needs to be grown because outsourced manufacturing is a central business strategy of the modern biopharmaceutical industry.3 This is the first part in a two-part article exploring the factors creating a demand from innovator biopharmaceutical companies for continuous manufacturing facilities in the U.S. and EU regions and some of the economic and technical challenges that face any contract development and manufacturing organizations (CDMOs) considering building API CM facilities.

Trends Impacting The Outsourcing Of API

Over the past 25 years there has been a significant shift by innovator biopharmaceutical companies away from in-house manufacture of APIs to predominantly outsourced manufacture at CDMOs. Even in cases where the innovator company might use in-house manufacture to prepare the materials to supply clinical development studies and launch supplies of an asset, having a parallel supply network with CDMOs is an important part of ensuring a robust supply chain of material.

Another significant change in recent decades is the shift away from blockbuster therapies for a wide patient population toward more targeted precision medicines. The emergence of precision medicine has two important effects that dramatically shrink the annual demand for a given API. First, the patient population is smaller. Second, precision medicines often have a low daily dose since the underlying biological target is clearly identified. The combination of these factors means that annual API requirements for a precision medicine can be as low as a few hundred kilograms per year, even for medicines that aren’t classified as highly potent.

A Misalignment In Scale

Conventional semi-batch manufacturing facilities are typically designed and scaled to prepare tens or hundreds of metric tons of material per year. Indeed, many of the facilities owned by U.S. and EU CDMOs were divested by innovator biopharmaceutical companies as they moved away from in-house long-term manufacture of blockbuster therapies. As a result, the existing CDMO infrastructure in the U.S. and EU is poorly aligned with the changes in emerging demand from the innovator biopharmaceutical companies.

To illustrate the mismatch in scale, it is worth considering that a typical facility constructed to support a blockbuster product might consist of 10 m3 reactors. A typical well-optimized pharmaceutical manufacturing step will operate with a maximum volume of 20 L (or less) per kg of product isolated. The 10 m3 reactor will therefore prepare at least 500 kg of product per batch – which in many cases would meet or exceed the entire annual supply requirements.

At first this might seem like an ideal outcome, but for the following reasons, making only a few batches per year is a situation that biopharmaceutical manufacturers try to avoid:

  • The cleaning burden on the equipment becomes onerous. Having the reactors sit idle for months between manufacturing batches is not an option for a CDMO looking to maximize its equipment utilization rate. In many cases, it takes longer to perform the cleaning validation and equipment qualification before and after each batch than it does to perform the chemistry itself, which leads to inefficient plant utilization rates resulting in high unit costs for the API manufactured.
  • Infrequent manufacturing means that large stockpiles of material (both API and intermediates) must be built up to cover the times between manufacturing campaigns. These materials are sunk cost for the biopharmaceutical company until they are distributed to the patients.
  • Isolated API has a limited shelf life, so it is undesirable to store the material months between manufacture and use. Additional testing to justify extending the shelf life may be possible, but this adds cost and complexity to the supply chain.
  • If demand for the product should change, it will be very difficult to adjust the manufacturing schedule to meet the new demand.
  • Making only a few batches each year does not give the operators the opportunity to gain familiarity running a process and, with a limited dataset, it is more difficult to track the performance of a process over time, especially if there is a change in the material inputs, whether intentional or not. This can also create a quality concern because limited data is available to track operating trends that provide understanding of whether any changes observed are fluctuations within the normal, acceptable operating window of the process or a sign of a more serious deviation about to occur.
  • Process qualification and validation studies before commercial launch may require many times the annual demand of material if performed at full scale.

Other factors driving innovation in biopharmaceutical manufacture are the emergence of “breakthrough medicines” and innovative clinical trial design, because they are leading to reduced time from API discovery to commercial launch and delivery of an urgently needed new therapy to patients. Historically, it has taken many years to refine processes and solve problems that occurred upon scale-up from the laboratory to the pilot plant and ultimately to the commercial manufacturing facility. In the new development paradigm, much less time is available to refine processes or work around process upsets. Increasingly, API supply is on the critical path throughout the development of a new breakthrough medicine; therefore, the focus of innovator biopharmaceutical companies during clinical development has to be on ensuring a robust supply of API. If the annual demand of a product requires only a few batches be manufactured each year at peak demand, then it becomes very difficult to gain sufficient experience and data during development to demonstrate that the process will run consistently to deliver a quality product each time for many years.

Could Continuous Manufacturing Be The Answer?

CM is an attractive option to address many of the issues highlighted in the preceding paragraphs; however, most equipment and published literature of API CM focuses on scales that do not meet the material demands of hundreds of kilograms to a few metric tons each year. The petrochemical and fine chemical industries have used CM for decades to improve throughput and reduce costs, but the equipment and techniques are designed to produce orders of magnitude more material than is required for many emerging APIs.

At the other extreme are microfluidic devices that can prepare up to a few grams of material per day. Some excellent academic examples of this approach have been published recently,4 but utilization of such an approach would require a massive numbering-up strategy to prepare a few kilograms of material each day. Also, many pharmaceutical compounds, and their precursors, do not possess the solubility profile required to prevent fouling due to solid formation that plagues microfluidic devices. Overcoming precipitation by operating at high dilution is not economically viable nor environmentally desirable. An alternative approach to improve solubility using dipolar aprotic solvents, such as DMSO and DMF, can be effective in a limited number of cases but, often, multi-step synthesis sequences perform best in different solvents for each transformation. Also, many dipolar aprotic solvents pose environmental and/or safety risks that make them undesirable choices for manufacturing.

Part 2 of this two-part article will explore small-volume continuous (SVC) manufacturing as an approach to bridge the gap between these two classical CM manufacturing approaches.


  1. Lee, S. L., O’Connor, T. F., Yang, X., CruzS, C. N., Chatterjee, h., Madurawe, R. D., . . . Woodcock, J. (2015). Modernizing Pharmaceutical Manufacturing: from Batch to Continuous Production. J. Pharm. Innov., 10, 191-199.
  2. Baxendale, I. R., Braatz, R. D., Hodnett, B. K., Jensen, K. F., Johnson, M. D., Sharratt, P., . . . Florence, A. J. (2015). Achieving Continuous Manufacturing: Technologies and Approaches for Synthesis, Workup, and Isolation of Drug Substance. May 20–21, 2014 Continuous Manufacturing Symposium. J. Pharm. Sci, 104(3), 781-791. doi:10.1002/jps.24252
  3. McWilliams, J. C., Allian, A. D., Opalka, S. M., May, S. A., Journet, M., & Braden, T. M. (2018). The Evolving State of Continuous Processing in Pharmaceutical API Manufacturing: A Survey of Pharmaceutical Companies and Contract Manufacturing Organizations. Org. Process Res. Dev, 22(9), 1143-1166. doi:10.1021/acs.oprd.8b00160
  4. Adamo, A., Beingessner, R. L., Behnam, M., Chen, J., Jamison, T. F., Jensen, K. F., . . . Zhang, P. (2016). On-demand continuous-flow production of pharmaceuticals in a compact, reconfigurable system. Science, 352(6281), 61-67. doi:10.1126/science.aaf1337

About The Author:

Ben Littler received his B.A. in chemistry and his Ph.D. in organic chemistry in the United Kingdom and then completed postdoctoral studies at North Carolina State University before joining the Chemical Development team at AMRI. In 2004, Littler joined Vertex Pharmaceuticals, and he has led the Vertex Process Chemistry team in San Diego since 2005. He has worked on projects from preclinical development to Phase 3 manufacture in disease areas including cystic fibrosis, hepatitis C, oncology, and pain. Littler is currently a senior director in Process Chemistry and the chair of Vertex’s cross-functional team to implement advanced manufacturing techniques for small molecule APIs.