Developing drugs and preparing them for commercial markets is a challenging task. But as pharmaceutical companies manage manufacturing footprints and capital budgets, securing experienced manufacturing partners has become a priority.
Drug development is evolving to treat chronic conditions and potent APIs are playing a central role. Many of these new treatments, including a long list of drugs in development, have very sophisticated chemistries. Not only are these compounds tough to formulate in the lab, their chemistries become more challenging to synthesize and scale in commercial volumes.
Early formulation of API chemistry is key for efficient scale-up. Offering insights gleaned from successfully scaling hundreds of API programs, Pfizer CentreOne’s New Business Development and Market Research Manager, Justin Divan discusses API scaling challenges and advises on best practices and offers his insight.
From Good to Best: Begin in the Lab
Going from good to best in commercial API manufacturing begins in the lab with an efficient API synthesis. As you scale-up, the efficiency of your API synthesis will show. Ultimately, the goal is to craft a viable chemistry in the lab along with a process that can successfully scale to commercial volumes.
Not everything you need to know to scale API synthesis can be anticipated in the lab, but there are some practical steps to build efficiency into API synthesis development.
Choosing the Right Path: The Role of Route Selection in Synthesis Efficiency
Route selection plays a vital role in developing a commercially scalable and economically sustainable API synthesis. A poor chemistry route that’s fully optimized is more challenging than a good chemistry route that’s partially optimized.
For example, liquid chromatography in small-molecule chemistry is useful at lab scale to purify a compound. However, when moving to large scale manufacturing, liquid chromatography can quickly become costly. Slow throughput combined with multiple solvents and column packing may lead to increased costs. While it may not always be possible to avoid liquid chromatography to purify API compounds at commercial scale, it’s one example of a lab methodology that should only be integrated at scale when all other avenues have been exhausted.
Sequence Matters: Order Reactions for Best Effect
Ordering reactions properly is another way to achieve better scaling of APIs. At Pfizer CentreOne, we have seen several examples. In one case, a chemist was working on a process that had an oxidation step and a hydrogenation step. At the time, the chemist felt it made sense to perform the oxidation step first. His rationale was based on experience where those reactions tended to create more impurities. The theory was impurities could be dealt with downstream.
The oxidation and hydrogenation steps went well but the process took 14 days. This was fine at small lab scale volumes, but would be time and cost-prohibitive at commercial scale.
Subsequent analysis revealed it was possible to complete the hydrogenation step first in just two days. In this case, the order of the reactions was a critical piece of the puzzle.
Time Matters: Start Optimizing Your Process
Optimizing processes for commercial scale should ideally begin no later than when materials are needed for clinical trial. Although that is the latest when optimization should start, it is a continuous improvement strategy that should never stop -- even after the drug is launched. The “sweet” spot for analyzing the robustness and efficiency of the process relative to commercial scale is usually at the transition from Phase IIb to Phase III clinical trials. Typically, it is at those junctures in the approval process when demand for API sharply grows in response to trial success during critical pre-approval phases.
Reducing Manufacturing Time
Minimizing the time spent in the tank can help with cost control. The longer the reaction takes, the higher the operating expense, especially at a commercial scale.
Consider potential downstream consequences. If a reaction takes 18 hours but the resulting yield takes seven days to dry, that lag has the potential to create an expensive bottleneck. Ideally, the time on the filter should be no longer than the time it takes to finish the next reaction.
Optimizing the time a product spends being filtered is often worth the effort. More experienced process engineers understand the compound’s crystal morphology in solution and can better manipulate and select characteristics to ensure the crystal will filter efficiently.
Evaluating cake washing and drying techniques is another place to look for optimization. The dryness endpoint, for example, can have a direct effect on filter time. Cake washing helps drive impurities out of the yield, but it can also be used to drive out slow-drying solvents. There are many options for cake drying in commercial API manufacturing - from heated recycled nitrogen to single-pass nitrogen or wet nitrogen for hydrates. It’s crucial to understand which is needed and which is best to leverage at a given time.
Finally, a clear endpoint is critical. If the next reaction can handle 5 percent of the solvent used in the previous reaction, why dry to 0.1 percent? Water is the caveat. It is not a good idea to store cake stabilized with water for extended periods as this can start to decompose. Of course, the best way to reduce filter time is to avoid using a filter in the first place.
Side-Stepping Filtration Altogether
Even though analysis may have isolated every intermediate along the development path, and it may be well understood which impurities are created at each step (as well as how they are purged), crystallizations can become quite expensive at large scale.
Best practice dictates looking for opportunities to telescope steps. When telescoping a process, consider which compounds in the reaction sequence are the most effective crystallizers. That exercise will help identify and exclude as many impurities as possible during the crystallization. These experiments can be time-consuming and tedious, but generally yield significant cost efficiencies by reducing crystallizations.
Running a series of experiments to understand the best concentration relative to the reaction is also a good idea. The more concentrated a reaction, the faster tank throughput should be. A higher throughput tank can help speed processing time and save money.
Other ways to concentrate the reaction could be valid as well. For example, another option might be employing a slurry-to-slurry reaction where only a portion of the API compound is in solution at any given time.
Don’t Forget to Factor Equipment into Process Optimization
High-capacity API manufacturing comes with other challenges, including making sure API chemistry works and is efficient in industrial scale equipment. Considering deviations at large scale are more costly, the robustness of the process needs to be carefully evaluated. Best practice is to do this immediately prior to scale-up.
An emulsion, for example, may be manageable and easy to break in a shake flask but almost impossible to deal with in a 15,000-liter tank. Once fully realized in a cGMP manufacturing environment, the improved process has potential to solve most problems associated with chemistry and manufacturing failures.
A Word on Safety and Large Volume Chemistries
High-energy reactions at small scale are dangerous. At scale, high-energy reactions can become dangerous and deadly.
High-energy reactions can be managed safely at scale, but require stringent engineering controls, operator training and often additional safety requirements. This can create ongoing expense and poor manufacturing economies. With proper analysis, however, safer reactions can be achieved.
Case in Point - Enterin
Our work with biotech company, Enterin, exemplifies how high-energy reactions at scale can be modified and managed safely. We partnered to scale-up the company’s ENT-01 molecule, a synthetic derivative of squalamine used in the treatment of Parkinson’s. Our team made changes to the molecule’s tailpiece structure and synthesis after we learned that it was not safely scalable or storable in large quantities because of a highly energetic functional group. We completely redesigned the molecule and synthesis to make this critical starting material much safer to handle, store and use.
Final Thoughts
By the time a supplier is processing quantities of API for clinical trials, the chemistries, processes and equipment required to produce the compound at commercial scale should be firmly established. Successful commercial API manufacturing strategy calls for thorough analysis of potential chemistry problems before they can derail a program.
For an ongoing commercial process that’s been run more than a dozen times, opportunities for process optimization will often present themselves. Regular and robust data collection will help process engineers identify opportunities to increase processing efficiency and to control costs. In effect, process optimization never ends – and that’s the best practice of all.
Author Biography
Justin Divan is the New Business Development and Market Research Manager at Pfizer CentreOne, a global contract development and manufacturing organization (CDMO) and a leading supplier of specialty APIs embedded within Pfizer. His 20 year career at Pfizer CentreOne's Kalamazoo site in Michigan began as a laboratory professional and scientist before progressing into business development in 2008. In his current role he leads the company's business development efforts for custom projects and works closely with customers to understand and track their project requirements.