There has been a recent surge in interest for continuous processes in the pharma industry as the benefits have become more widely known. This is due to the availability of more expertise in the area of flow chemistry over the last decade, in combination with the need for the industry to develop safer, faster and more sustainable processes, with higher quality and less expensive products. But the first thing we need to do is define what we mean by continuous processing and flow chemistry. The industry is running two broad types of continuous processing, in finished dosage form and API manufacturing – often commentary in the media has made little effort to separate these, and they invariably get confused. Whilst continuous processing in finished formulations with the potential of on demand dosing is extremely exciting, for the purpose of this article we are instead going to specifically look at the improvements flow chemistry can bring to process development and manufacturing for API’s.
It is time the wider pharma community set aside the decade-old views of continuous manufacturing as a ‘luxury but impractical tool’ and looked at the technology as a practical and valuable approach that can resolve our every day chemical processing issues. Flow chemistry offers a more streamlined and continuous synthesis process as well as a variety of advantages compared to a batch operation. Incorporating a flow operation results in increased production with decreased capital. In terms of safety, flow reactors for pharmaceutical reactions are normally run in much smaller volumes than those of batch reactions, therefore lowering the risk of hazardous reactions. Dealing with toxic chemicals is also safer – cytotoxic APIs can be produced in inexpensive, dedicated, and disposable equipment sets for production of low volumes of these compounds in the laboratory fume hood.
Not only is flow chemistry safer than batch, it is also more efficient – better heat and mass transfer alongside less back mixing contribute to enhancing the purity profile and product recovery. The small size of the microreactors, either PFR or CSTR, allows for much higher reaction temperatures and pressures as compared to batch reactors, allowing for safe reactions that were previously unstable in batch.
Reactions can also undergo superheating, enabling them to be heated above their boiling point, further resulting in faster reaction rates. The flow operation reduces the break time between consecutive steps and can significantly reduce the manufacturing time An important advantage of flow chemistry is the ability to fully control many of the parameters, such as mixing, temperature, and reaction time. By having the capability to add or remove heat almost instantaneously, one could remove the heat generated from a reaction; for exothermic reactions or a reaction requiring hazardous materials, this is an especially important benefit. Flow reactors also allow control over residence time, which is the time that the reaction is exposed to a set temperature, allowing for far more precise reaction times. This is immensely beneficial, particularly if a reaction creates more than one product.
There is also continuous monitoring of the quality – such as purity – by online or offline sensors, so parameters can still be fine-tuned during the operation in order to obtain the best product quality. During a batch process, you would need to wait until a reaction is completed, by which point it may be too late to make any adjustment. The intense mixing in flow chemistry is provided by microreactors, which enables scientists to use multiple phase systems, fewer solvents, and produce purer material – reducing unit operation and work up steps. The high- temperature, high-pressure flow reactors reduce reaction time and provide better conversion whilst using starting material more efficiently. This requires tightly controlled Process Analytical Technologies (PAT), and resolution of any Quality Assurance issues related to acceptability of the intermediates. Microreactors can be designed to fit the requirements needed for the reaction, therefore providing customisation opportunities. In addition, microreactors have low maintenance and operational costs without abandoning productivity and efficiency, which provides an economic incentive. These technological advancements are valuable and vital assets in flow chemistry and have expanded the versatility in its use. Although the advantages are clear, before a flow process can be developed a working small-scale batch process should still exist since, in general, developing a flow step may take much longer than its batch equivalent. However, once a flow step has been developed, its scale up is far easier and encounters fewer issues than in batch. The reason for this is that the sizes of the reactors in the scale up version are normally less than 20 times the size of the lab version. For example, the diameter of lab scale PFR tubing is normally around 1/16”- 1/4”, and its pilot plant version is around 3/8” - 1/2” – these are not very different in size. The scale up in the batch process could be 100 to 1000 times bigger than the original lab scale process – this is impractical as mixing and other engineering aspects can complicate such large scale up operations.
The total cost of producing a final product depends on the cost of the process R&D, starting materials and the operational costs – of these, the latter two have the greatest impact on the overall cost. Using the flow operation, cost incentives include the reduction of energy costs and reduction of impurities and waste products. Awareness by chemists of the capabilities of flow chemistry as an enabling technology gives them the power to design shorter synthetic routes, and therefore also reduce the cost once operational. Of course, reducing waste promotes efficiency, enhances purity, and is beneficial to the environment. Companies may realize too late that their drug has an excessive, multi-step process that could have been shortened and since their cost would be unnecessarily high due to a longer synthetic route, this could result in losing substantial amounts of profit.
If we assume that flow facilities provide major benefits at larger scales, we could see that later phase and commercial products are more amenable to continuous processing and will see the greatest benefits. Big pharma such as Lilly, GSK, and Novartis are already preparing for launch of their pilot or commercial plant facilities and, at this time, these plants are built within their own companies. However, over time these companies may decide to outsource such operations to CMOs or CDMOs – we have one such flow chemistry partnership with big pharma, but we’re very much in the minority. Our belief is that it’s only a matter of time until much more flow work is outsourced, and we are building up capacity in anticipation of this.
During development, flow steps seem to be more appropriate for early steps of the synthetic route where less expensive raw material are available for process development and the volume of the material to be processed is larger. Perhaps the majority of the API’s currently produced at a commercial stage have the required volume to be turned into flow. However, due to regulatory issues, limited changes can be applied to the existing commercial processes but it can still potentially be achieved with some investment and time.
The transition from batch to flow operation is generally thought of as both costly and inconvenient, but implementing this change in early development is simple and beneficial. Comparing Phase I and Phase III, it is much easier to manage changes in development and regulation if switched at Phase I, but to switch at Phase III could result in a delay in market release and a loss in both time and money.
Yet, the number of flow steps during development currently remains below 5%. The message here is clear: for flow chemistry to deliver on its huge promise, pharma and CDMOs need to build the platform into the phase I process R&D of innovative API programmes. This requires commitment from the beginning of a project, and a wider commitment to running in flow whenever possible.
Flow chemistry has suffered slow implementation into the industry – especially as compared to some other industries such as oil and gas – even though more are beginning to recognize its benefits. This is largely due to the increase in demand for flow chemists while there remains a lack of experience and education in the field. With an entirely new manufacturing process people may be reluctant to adopt it as they think it may slow down the manufacturing process. The safety, efficiency, and flexibility of flow chemistry are what drives its high interest and is why it should become an essential component in research, development, and for the future of the industry.
Q&A on the future of flow chemistry and its industry adoption
Q) What are the major drivers for the recent surge in efforts within the pharma industry in choosing to use flow processes and investing in greener chemistries?
The need of the industry to develop safer, faster, and more sustainable processes with higher quality and less expensive products, combined with the availability of more expertise in the area of flow chemistry built up in the last decade are behind the recent increase in the industry using more continuous flow processes. Parallel to major pharma companies, we actively initiated our activities in the area of continuous processing and in the last two years we have had around 40 steps processed in flow in our Jinshan plant (in Shanghai).
Q) What uptake do we predict over the next 5 years – how and why?
Currently, many pharmaceutical companies believe that there should be a reaction specific driver to using flow technologies in plant operations, although this might not be the case in the future. At STA, for example, we have great batch plant facilities in Shanghai and Changzhou in addition to many experienced batch scientists to support the PRD and plant operations. It is more economical and meaningful to utilize the existing knowledge and investments in batch operations, and do the chemical steps in flow when there is a driver. In the next 5 years however, more advanced flow technologies will be developed and more dedicated flow companies will be developing continuous processes, while the existing scientists will build up more experience in this area. As a result we should expect a wider range of reactions to be processed through flow, and a lower bar to be established for selecting a process to be operated in flow. During this time, we expect to see more multistep flow processes performed.
For companies with a strong flow team, we should also see more work--up operations steps coupled with the reaction steps in flow. We expect them to move from the currently 1-2 steps to 5-6 chemical steps in flow.
This should be coupled with 1-2 unit operation steps per a chemical step including solvent switch, phase separation, extraction, crystallization, filtration, and dissolution bringing the flow steps to 10-15 steps (including unit operations). This brings the system closer to full flow operations, but it is unlikely that within the next 5 years all of the chemical steps on a project will be carried out using flow technologies.
Q) As a company you not only utilize more efficient processes, but you are currently looking to reduce your PMI (Process Mass Intensity), do you think this is something that will set a trend and we will see more companies moving to flow chemistry in the future with waste management and safety regulations getting tighter?
Yes, environmental regulations including control of PMI are one the industry’s big concerns. The smart design of a flow step process could end up using less solvent mass compared to the batch systems. Flow chemistry can not only be applied to chemical steps, but can also be utilized for unit work-up operation steps. In fact, it makes sense to do the work-ups between two consecutive chemical flow steps to be done in continuous mode. For example, a more effective work-up is provided if a multi-step counter current extraction system is used to separate the liquid phases instead of using batch mode phase separation; this means less solvent is used in the process. In addition a flow chemical step might generate purer intermediates and less work-up might be needed. Furthermore because the flow reactors are smaller, less solvent is used to wash these reactors after run completion. Additionally, the high temperature accessibility provided by flow may mean less solvent requirements due to the higher solubility for compounds at high temperatures or even going fully green by performing the reactions at melt conditions.
Q) Does using continuous flow techniques allow you produce drugs that you were perhaps unable to make using batch processes?
Experience has shown that the batch chemists can ultimately find routes to develop the final API. Therefore it may not necessarily enable us to produce new drugs, but it allows us to use shorter routes to the final product. The reason for this is that flow chemistry enables us to adopt a reaction path that might not be possible in batch mode due to the safety or quality concerns. As an example, performing azide or ozonolysis reactions usually reduces the length of synthetic steps – these reactions are much safer in flow than in batch mode. Around 30% of the flow reactions we have performed at STA are related to the safety; without flow capabilities we may not have been able to do these steps or may have had to take different and perhaps longer synthetic routes.
Q) How crucial is support from the authorities in ensuring flow chemistry secures a place in the future of the pharma industry?
This is very crucial. Fortunately, the FDA is a strong supporter of converting from batch to continuous. Continuous processing has less manual operations and logistics and disturbances that caused process review concerns with the FDA. However, last year the FDA gave the green light to Johnson & Johnson to produce its HIV drug Prezista in flow mode and then invited other companies to take the same approach. Johnson & Johnson has now said that it may want to produce up to 70% of its highest volume drugs using flow manufacturing. Vertex has also built a continuous manufacturing plant in Boston for one of its drugs.
Q) Why has pharma and the wider industry been slow to adopt flow chemistry and continuous processing?
Bulk chemicals industries such as fertilizer, sugar, and oil are at least 50 years ahead of pharmaceuticals in using flow technologies. There are at least two reasons for this: the first and most important reason in this difference is the volume of the commodities they are dealing with. The other reason is hidden in the structure of pharmaceutical industry. Prior to the 1990’s, chemists were almost the only lab scientists developing chemical processes, and management of process R&D centres was also done by chemists. By end of the same decade, chemical engineers had been hired as research staff to deal with chemical reaction kinetics and unit operation steps including crystallization and distillation. The same engineers later expanded their scope of work into flow chemistry. By around 2004, several major companies had flow teams. By 2010, almost all major companies and many intermediate size companies had either expanded into or at least touched the area.
Q) How much money across the entire industry do you think could be saved using flow chemistry and/or how much money do you think will be invested into it over the next 5 years?
Perhaps it would be difficult to put a number for saving across the entire industry, but MIT scientists estimate a saving of 15-50% by switching to flow. The wide range of these numbers indicate uncertainty in the estimation, which perhaps depends on what items are counted in the estimation.
Q) Do you believe that technological advancements over the next few years will allow flow chemistry processes to become more accessible on a global scale?
More technologies will certainly be invented and more processes will be amenable to flow systems. The MIT flow system sponsored by Novartis inspired Novartis and other companies to perform multi-steps or full steps including formulation in flow. We have heard that Novartis renewed the sponsorship for the second time for another 5 years term of the MIT setup, which is good news for flow chemistry.
Q) What challenges does the industry as whole face over the next 5-10 years when it comes to using flow chemistries?
The existence of experienced batch scientists combined with lack of experts in flow chemistry makes traditional process development very tempting. In addition, flow chemistry has put forward some new questions for quality of the products. For example, we can clean and validate a multi-purpose batch reactor using established methods, but those are not valid for a plug flow reactors as the access to the internal walls of the tubing is restricted. These are not very difficult obstacles to overcome, however there are very limited opportunities for drug development companies to learn from each other due to the restricted access of information, the reality is that most companies have to face and resolve these challenges on their own.
Dr. Minzhang Chen is CEO at STA Pharmaceutical and Dr. Sam Tadayon is Executive Director at STA Pharmaceutical, a WuXi AppTec Company.
(Courtesy: CPhI Annual Industry Report-2017)