As we entered the 21st century, there was a host of new technologies on the horizon that could significantly change the way we do things in the pharmaceutical industry. The FDA clearly recognized the need to innovate. This, however, could require that we do things differently. The FDA agreed and wanted to ensure they were not in the way of innovation, which could significantly improve the overall economics of delivering new medicines to the general population. At the 2010 AAPS Annual Meeting, Christine Moore, Deputy Director for Science and Policy for CDER Office of New Drug Assessment, commented that the Agency saw definite economic and quality advantages to continuous manufacturing and that the science exists [1]. Furthermore she said that there are “no regulatory hurdles” for industry to implement continuous practices. At the 2011 AAPS meeting, Janet Woodstock, Director of the Center for Drug Evaluation and Research, stated, “Right now manufacturing experts from the 1950s would easily recognize processes today. In 25 years these same processes will be obsolete.” Adoption of continuous manufacturing will cause this change according to Woodstock [2].
Many of the “new” innovations are not technically new and have been used in other industries. These “new technologies” have been around for many years but were never really embraced by pharmaceutical manufacturers. In “Pharma Time”, it normally takes 40-50 years to truly implement a new technology. There are many reasons for this and one cannot always blame regulatory bodies.
The use of continuous reactions has been around for over half a century. Some of the original patents for continuous chromatography and flow chemistry can date back to the early 1950s. Other industries such as the petro-chemicals and specialty chemicals have used these technologies to manufacture millions of kilos of products over the year.
Continuous technologies will be a game changer for the pharmaceutical industry. While this article will focus more on the manufacturing of Active Pharmaceutical Ingredients (API), similar concepts can be used for the manufacturing of the final dosage. If this is the case, then the only real issue will be to develop these technologies and to implement them under the current regulations. For this paper, we will concentrate on API production and look more at continuous reactions rather than continuous processes since full continuous processes remain farther out of reach [3].
Continuous Chromatography was one of the earliest technologies that the FDA reviewed. It was and is successfully used to produce several APIs. Today, it has become a very useful tool for the manufacturing chemist. The use of various continuous reactor designs continue to be used today and the trend seems to be growing. The key issue is that the FDA has seen these technologies and had the chance to inspect plants that have implemented continuous systems.
To start, the industry is not operating under a single set of rules and guidelines. Few people operate on a real regional base but look to be more global. This means that manufacturers will need to comply with multiple agencies such as the FDA (US), EMA (Europe), PMDA (Japan), sFDA (China), TGA (Australia) and DCG (India). In addition, compliance with local environmental and safety rules could result in additional regulatory burdens.
It has been well-documented that continuous flow chemistry can bring added value to the manufacturing process. This can relate to safer chemistry, pure products and more economic process as well. Reduction of waste along with potentially smaller footprints and lower investment also create the need to implement new and better technology [1].
It is important to note that today; individual continuous steps are run, not full processes. This means that we are, in principle, manufacturing our products using both batch and continuous in the process. In addition, there are many systems and approaches that are already in place that can be adapted to new technologies. Also, the human factor to implementing changes in a facility or process must not be underestimated.
Currently, there are no guidelines or rules covering the use of continuous systems in a pharmaceutical plant. However, under 21 CFR 210, there are a series of definitions that should be very helpful in implementing any technology. A “batch” is defined as a specific quantity of drug or other material that is intended to have uniform character and quality, within specific limits and is produced according to a single manufacturing order during the same cycle of manufacturing. Also the “acceptance criteria” means the product specifications as acceptable quality level and unacceptable quality level, with an associated sampling plan, that is necessary for making a decision to accept or reject a lot or batch (or any other convenient subgroups of manufactured units) [4].
By definition, “batch” refers to some quantity of material produced and does not reflect the method that was used to manufacture the product. “Acceptance criteria” also does not relate to how we got to this point, it relates to getting there consistently using the same process. Therefore, if these two definitions are followed, we should be able to develop a system using any new technology and get it into a regulated environment.
Our original ideas about “batch” and scale up will need to be re-thought to accommodate the new processes. In fact, the question of batch could be the most important issue. In a batch reactor, a batch in principle becomes what comes out of the reactor for a given charge. However, by definition, a continuous unit would not have a specific charge but could go on for months or even years if desired. The first question to ask would be, “Why it is so important to have a well-understood and a welldefined batch?” In addition, the idea of scale also changes from moving to a larger reactor to potentially running longer or the idea of “numbering up” which simply adds equipment of the same size. Thus, doubling the capacity means adding a second unit.
By definition, the “batch” needs to produce a uniform character and quality within certain limits. It implies that if the quality is not matched, one will have the opportunity to trace back to the specific batch in question and understand the manufacturing process as well as the starting materials and reagents used to produce the final product. This becomes an issue of traceability and reliability.
For a continuous system, there are two common methods to determine the “batch” and they rely on either time or charge. Like a batch reactor, one can charge a known amount of material through the reactor [5]. This implies that one makes up a batch of starting material in solution and uses some type of tank to feed the reactor. Most likely, one would process the first tank and then work it up after the “continuous” portion. It is important to remember that the stability of the feedstock or the resultant product in solution should be well-understood when running over a long period of time [5].
The second potential way to produce a batch would be to run a fixed rate of flow over a fixed time (any time desired). There are several critical parameters to remember especially when running for long periods of time. Product stability is still an issue unless it is worked up immediately; this could mean going into some type of continuous distillation and drying. Precise control of the flow rates is also very critical.
In either case, it is important to be able to trace back to specific lots or batches of starting material and reagents. The starting material could dictate the final batch size due to either stability or lot size. However, the definition of the batch size is up to the manufacturer and not the regulating body. The ability to practice good science is crucial for this.
There are several other considerations for determining the actual batch size and one can refer back to the 21 CFR for help here as well. Under 21 CFR 211.165(a), each batch needs to have appropriate testing and 21 CFR 211.165 (c) states that a sampling plan should be in place to assure that batches of drug product meet each appropriate specification. Section 211.192 clearly states that unexplained discrepancies need to be extended and Section 211.150 (b) covers procedures for a recall of a lot [4].
The point is that one should clearly understand what constitutes a batch, be able to identify the batch, and test it appropriately. Ask, “If there were a recall, would I be able to identify the specific batch or batches in question?” And, “If there was a discrepancy to investigate, can I relate back to a starting material, in-process control or other variable?” Defining the batch is extremely important and this does not need to relate to the process whether it is continuous or not.
Once the batch has been defined, one needs to turn their attention to other aspects which are required for compliance. Based on the definitions, the “batch” should yield uniform character and quality, one need to control what happens between batches. Do you want to break down the equipment and do a full clean? Or do we need to add some cleaning and additional testing?
There are a number of ways to handle this and one must look at this during the original set up and design. The simplest system would be to add a rinse step followed by some additional testing which could include some type of FTIR, or another method which could give you access to the correct information.
This assumes that the material is still in solution. For solid material, which could be present in the drying step or milling step or during crystallization,one must consider more. We are already aware of appropriate methods to clean mills and sieves and these could apply here. This would result in cleaning between batches and visual inspections. The same would apply to crystallizers as well. In principle, we already know these methods and need to apply them, taking advantage of your batch definition and building the cleaning time into the process.
Finding what areas result in entrainment of product and developing methods to assure these are clean are extremely important issues to cover. This could mean using multiple sets of pumps, tubing, fittings and even the reactor itself.
The basic concepts around validation do not change going from batch to continuous. You still need to comprehend your process and be assured that you understand the critical operating parameters for your process. There are several additional considerations that must be addressed. Depending on the “batch size”, the raw materials will spend more time before being converted to final product and the product that was produced may be in solution longer than expected.
In addition, there are other operating parameters to consider. For continuous reactions, the feed rates of the solution are critical and need to be well-defined. The resonance time is another important factor. Also because the reactors are often smaller, in process probes for analytical measuring could be more difficult and the probe itself could affect the flow patterns as well as the mixing.
Based on the most recent FDA Guidance documents for Validation (Guidance for the Industry-Process Validation: General Principles and Practices), process validation is divided into three phases and this could easily be applied to continuous processes. Stage 1 requires the process design which should be based on the process knowledge and understanding. In addition, the strategy for process control should be prepared. Stage 2 covers the process qualification and should cover the design of the facility including facility and equipment as well as the process itself. The final Stage (Stage 3) should be continuous process verification. Under Section 501 (a) (2) (B) of the Act (21 U.S.C. 351 (a) (2) (B), process validation for drugs is a legally enforceable requirement3.
Based on the basic principles of Process Validation, quality, safety and efficacy should be designed or built into the process. You cannot control the quality of the product with in-process and finished product testing. The basic principles of validation should not change going from batch to continuous flow chemistry. The key issue to recall is that cGMP requires that manufacturing processes be designed and controlled to assure that in-process materials and the finished product meet predetermined quality requirements.
Following conventional wisdom, we validate the process using three consecutive batches. For continuous reactions, the batch could take many days and you need to monitor it across that time. The stability has the potential to affect your batch size by causing the time to be either increased or decreased. In any case, remember that by definition, the batch should be of “uniform character” and “quality” so for continuous reactions, the uniform character could occur over multiple days.
Based on 21 CFR 314.50 (d) (1) and 21 CFR 314.94 (a) (9) (1), analytical procedures are necessary to assure identity, strength, quality, purity and potency. Under the Guidance for Industry-Analytical Procedures and Methods Validation, there are general guidelines covering the analytical requirements and validation. There is no differentiation for the guidance for batch or continuous processes. The general principles are valid no matter how you get there.
The last point to consider is the where of continuous flow. The chemistry that is being discussed can bring advanced technologies to a wide range of countries and manufacturing plants. Inconsistent electrical flow could affect the process and create additional validation parameters to be considered. Even in the most advanced countries, power losses can occur and issues develop. Consider the potential length that some of these reactions take, one needs to ensure that they understand their process and prepare for any potential issues [6].
Continuous flow technology is here to stay and we will see more of this in the future. Major companies have invested in this and we are already producing drugs using flow chemistry. With regard to the regulatory issues, there is nothing stopping us from bringing this technology forward. What we need is good science and a true understanding of our process. Add our basic understanding of the regulatory environment to this and we can start experiencing the economic and safety advantages of adopting flow [1].
References
- Continuous Manufacturing in Pharma: Beginning to Snowball?, PharmaQbD, April, 19, 2013
- Taylor, Nick, “Continuous manufacturing will make the current methods “obsolete,” FDA says,” In-Pharmactechnologist, October 11, 2011.
- Crosby, Tim, “Designing For the Future of Continuous Processing”, Pharmpro, January, 2006
- Code of the Federal Regulations, Title 21 Volume 4
- Pellek, Alex and Patricia Van Amun, “Continuous Processing: Moving with or against the Manufacturing Flow”, Pharmtech September 2008
- Mollan, Matthew Jr., and Mayur Lodaya, “Continuous Processing in Pharmaceutical Manufacturing”, White Paper
- Cahill, Jim, “Continuous Process Verification per FDA Process Validation Guidance,” Emerson Process Experts, January, 2012
- Moore, Christine, “Continuous Manufacturing-FDA Perspective on Submissions and Implementation”, 3rd Symposium on continuous Flow Reactor Technology for Industrial Applications, October 3, 2011
- Guidance for Industry-Process Validation- General Principles and Practices
- Guidance for Industry-Analytical Procedures and Methods Validation
Author Biography
James Bruno has over 40 years of industrial experience manufacturing APIs. He has been consulting for over 10 years servicing the merging pharma companies focused on manufacturing APIs in a regulated environment, including the technical and regulatory aspects of the product. More strategically, Jim helps CMOs to focus their work and review their assets to function in these markets globally. In order to expedite the production of new compounds, Jim also works to bring technologies like continuous flow chemistry including chromatography as well as catalyst and membrane technology to solve chemical problems. Jim is a DCAT Past President and Scientific Advisory Board Chairman at Rider University.