When the Stars Align

The next wave of highly potent and tailored small molecule drugs poses a serious challenge to the current batch manufacturing infrastructure; frequently, oversized equipment is simply unsuitable for lower volumes. And that’s why companies such as Eli Lilly – with a number of potent cancer treatments that target small patient populations in its pipeline – are considering continuous manufacturing.

Although many groups have been able to continuously manufacture APIs, achieving cGMP and strict regulatory compliance can be challenging. In a recent study (1), Eli Lilly report the development of a multistep continuous process that produced 24 kg of prexasertib monolactate monohydrate under cGMP. Eight continuous unit operations were conducted to produce the target at roughly 3 kg per day. Kevin Cole, Principal Research Scientist at Eli Lilly and Company, tells us more.

Why focus on prexasertib? 

The main technical drivers specific to prexasertib were improved process performance and safety relative to the batch process, both for the step using hydrazine and the ability to control formic acid using distillation, which enabled isolation of the monolactate monohydrate. Continuous manufacturing also makes sense for materials that require high containment, such as prexasertib, because the equipment can either be dedicated to an individual product or disposed of after use.

There is a strong desire within the company and the industry at large to modernize pharmaceutical manufacturing so, to some extent, this was also a test case to assess the feasibility of continuous manufacturing. For a long time, we’ve wanted to run a manufacturing process in this way, and the “stars aligned” in terms of an active clinical project that had material needs and could be developed for flow.

How does your process improve upon current methods of manufacturing prexasertib?

It was a completely new process in terms of the chemical route used, as well as the salt form that was isolated. We believe that the monolactate monohydrate salt could be produced from a batch process, but we have not substantially investigated that option. Currently, the continuous process enables removal of formic acid solvent by distillation, which has been shown not to work in batch mode due to stability and mass transfer limitations.

There are several significant benefits derived from the new chemistry route: it is two steps shorter, there was a late stage coupling step in the old route that used undesirable reagents and showed high performance variability, it can be more easily scaled up, uses a much less expensive pyrazine starting material, and offers better control in the later steps in terms of solubility. Specifically, in the previous route, there was a step where the insoluble product crystallized from solution uncontrollably. Some of these benefits are due to inherently better chemistry, while others are benefits of continuous processing.

Are there any limitations to continuous processing?

There are currently several challenging areas that we deal with routinely:

1. The formation of solids in a continuous reactor. We need to understand the solubility of all the reaction components to avoid undesired formation of solids in a continuous reactor (which can cause plugging) and develop high performance reactor types that are more tolerant towards solids. 
2. Challenging implementation of continuous processing in the contract manufacturing network  because of limited capacity/ability in the external network. 
3. Very long reactions that cannot be sped up by heating. As your continuous reactor gets bigger and bigger to accommodate a long reaction, it starts to look more like a batch reactor. 
4. Development timelines and the amount of data needed for a continuous process are improving as we gain experience; however, there is substantial operational complexity to linking flow unit operations with process analytical testing that operate simultaneously – advanced process modeling is required to ensure sufficient understanding of the system.

Does continuous processing have a strong role in the industry’s future?

There are a many different visions out there for what the future holds for pharma manufacturing – especially with regards to continuous processing. Certainly at Lilly, we envision the future for small molecule manufacturing to be different from the 2017 paradigm, which still primarily uses batch methods for production. We have invested aggressively in the development of continuous processing technologies – and we intend to use them! Our new continuous manufacturing building in Kinsale, Ireland, will allow us to pursue continuous processing much more. Batch processing is not going away, and there are certainly many instances where that method will be used in production. We want to use flow chemistry where it makes sense and provides an advantage. It will be interesting to see where some of the academic efforts in flow and additive manufacturing take us over the next few years. Finally, although Lilly isn’t ready for it yet, the concept of a hand-held device that can make individual doses of medicines on-demand using continuous technology already exists – and it will be interesting to see if anything comes of that.