As small-molecule APIs become more potent, the amount of drug substance needed in a process has decreased while manufacturing complexity and demands on chemical manufacture has increased.
Continuous manufacturing facilities are essential to address these challenges, ensuring efficient development and small-scale production. In this article, we talk to Gareth Jenkins, VP Science & Technology, about FlowInova, a platform is used for continuous drug substance manufacturing developed in collaboration with Quotient Sciences and the University of Nottingham. It offers a more sustainable manufacturing approach that can reduce the time and cost associated with new drug development.
Why was the FlowInova platform created?
Small-molecule active pharmaceutical ingredient (API) candidates are becoming more potent as the trend towards developing more targeted, precise drugs continues. The amount of drug substance required, both through the clinical development phase and at commercial launch, has reduced significantly although the molecular complexity and demands on chemical manufacture have only increased.
Developers must reconsider the technology required for manufacturing drug substance--and continuous manufacturing facilities that can meet development and small-scale manufacturing requirements will be critical to solving this problem.
The FlowInova platform started in 2018 as a research project when we set out to develop a data-driven methodology for process development and continuous manufacturing that enables efficient and intensive scale-up for early-phase development of API candidates. In collaboration with the Department of Chemistry at the University of Nottingham, the project was the first of its kind to be implemented by a UK-based CDMO.
The novel approach combines high-throughput experimentation with process modeling and uses this to rapidly complete a number of plan-do-review cycles of process development. This leads to highly efficient, scalable processes that can produce kilograms of material within a laboratory setting, reducing scale-up time and quantities of material required for development.
The FlowInova project was the first to demonstrate that using automated reaction equipment and building models early in the process enables better decision-making for the next round of experiments. As more data is acquired and knowledge of the process increases, the process models become more predictive and allow for virtual design of experiments to be carried out. This permits a greater focus on confirming and optimizing the predicted process parameters, leading to more robust and reliable scale-up.
What was it like collaborating with the University of Nottingham on this project?
Professor Michael George and Professor Martyn Poliakoff at the University of Nottingham have considerable research experience in fundamental understanding of reactions and applications of this understanding to build prototype reactors with optimal process conditions. With access to an engineering team that could make bespoke prototype reactors, our collaborators at the University of Nottingham were able to look at some challenging chemical processes and develop novel reactors that would be suitable for transfer and scale-up at our Alnwick site.
Some of the reactions we were interested in were poorly understood mechanistically and required the handling of corrosive or low-boiling-point reagents. While these reactions should be suitable for continuous processing, the reaction engineering skills at the University of Nottingham were critical to make this a reality. The Arcinova and Quotient Sciences teams were involved in the review of the prototype reactors, providing guidance on the throughputs required to make them commercially viable and to ensure that we could transfer the knowledge to the Alnwick site to build a larger, kilo-scale system ourselves.
As a drug substance contract manufacturer, all the chemistry we work on is on behalf of our clients. This means that we are constrained by the amount of time any process development project can run before we have to scale up and supply the product, which is often on the critical path during pre-clinical and Phase I clinical development. This insight helped our collaborators at the University of Nottingham focus on the most promising early prototypes, ensuring that both groups learned and developed skills from the other.
How has the FlowInova platform been used?
We developed a continuous-flow approach to the kilo-scale manufacture of (2R,6R)-hydroxynorketamine based on the original literature route described in the 1960s. Initially, the University of Nottingham team looked at the bromination and ammonia reactions, which had no significant precedence and would need considerable reaction engineering.
In parallel, our teams focused on the thermal rearrangement and setting up our labs to be able to carry out continuous-flow chemistry using pumps, mass flow meters, membrane separators, and a commercially available plug flow reactor. This meant that we could quickly demonstrate some significant improvements in the overall process at our Alnwick site.
Based on the literature, to produce 1 kg of the final product via a batch process would have required 18 kg of starting material to be processed through 24 batches (across all stages of the chemistry) at our 20 L laboratory scale. By running the thermal rearrangement in continuous flow, we improved the yield of that step from 60% to 95%—producing the first kilogram of the final product using only 10 kg of the starting material. And we did that in less than 4 weeks.
Following this, we transferred the work from the University of Nottingham team and set up continuous-flow systems to carry out the bromination, amination, solvent swap, and thermal rearrangement stages. This was done using a combination of continuously stirred tank reactors (CSTRs) and plug flow reactors, daisy-chained together so that the output from one provided the input to the next. We ran a full demonstration, bringing all of the reactors up to steady state and then running for over 24 hours before sequentially stopping each system. Over 1.2 kg of material was collected in 24 hours.
How has this project changed the company's approach to drug substance manufacturing?
Continuous-flow chemistry is now embedded as a key tool in our process research and development offering for drug substance manufacturing. To be able to develop a continuous-flow chemistry process, there is a need to generate more information about the chemistry we take on as a drug development and manufacturing accelerator.
This collaboration helped our evaluation of chemistry automation platforms, such as the EasyMax systems from Mettler Toledo and process modeling using Dynochem. With online process analytical technologies (PAT) giving time-course data on each experiment we run, we can develop a greater understanding of the process and operating windows faster than before. Whether the process is better suited to continuous or batch reactors, all of our drug substance projects can benefit from this approach to accelerate process optimization and de-risks scale-up.
This has created 4 production suites, which can house a range of reactors, both continuous and batch, with the latter going up to 150 L in scale. Our data-rich approach to process research and development allows us to be much more material-sparing during the process development phase. We can get more information out of fewer experiments, while improving process robustness and our ability to deliver the 10s of kilograms our clients need on time and in full.
What are the benefits of the FlowInova platform?
FlowInova supports faster, reliable delivery of the first kilograms of a new drug candidate. Potential new medicines can accelerate through pre-clinical and early-phase clinical development, and API supply can be kept off the critical path.
Additional benefits arise from our ability to efficiently evaluate and compare alternative chemistries to make the same new medicine. This can be applied to a single transformation in a synthetic sequence–by evaluating different methodologies and catalysts used for a hydrogenation step, for example.
Does the FlowInova platform offer any sustainability benefits?
Yes, as well as optimizing for yield, we have successfully developed a conjugate reduction that scored 77 out of 100 using the Green Motion™ assessment tool, which considers all 12 Principles of Green Chemistry.
As FlowInova helps capture and develop process understanding, we can embed efficient, scalable, and sustainable chemistry into the synthetic routes we develop. Demonstrating that continuous, process-intensive unit operations work at the kilo-scale allows us to supply API material to support critical Phase I and Phase II clinical studies without the need to re-develop the process. The transfer to Phase III and commercial manufacture can be based on a comprehensive data package and removes the barriers often associated with introducing new technologies such as continuous processing in late-stage API manufacture.
Finally, what does the future look like for the FlowInova platform?
With the emphasis on science and data-led experimentation, we are now starting to consider multiple criteria when optimizing any given process. As well as chemical yield, we include process mass intensity (PMI) and other green metrics as a measure of process development outcomes.
Continuous processing lends itself to more automated ways of working, and we are now looking to bring automation into the earlier stages of process research and development. The success of this first collaborative innovation project led to an invitation to join an EU Horizon collaboration, Project ETERNAL, to develop future pharmaceutical products to be more sustainable by design and reduce their environmental impact throughout their lifecycle.
As a result, there is less overall waste, and the processes used to make the material can be optimized for sustainability. Building on our data-rich approach to process development, we plan to use digital manufacturing technologies for efficient chemical process transfer from lab to kilo scale and from one site to another. Other collaborators are looking at solvent reduction and recovery, removal of carcinogenic impurities, and intrinsically less environmentally harmful products.