As an associate research fellow at Pfizer, Eric Cordi has developed a unique set of skills within chemical engineering. His challenge is to take molecules discovered at gram scale and build the right recipe and technology to produce the same molecule in kilograms or tons. 

We spoke with Eric to explore the nuances of chemical process development, and learn how continuous processes are changing the pharmaceutical processing landscape. 

 

What challenges do you face implementing continuous pharmaceutical processes?

Continuous processing of active pharmaceutical ingredients is an exciting area of development and becoming more common in the industry. It offers several compelling advantages including the integrated processing of multiple steps, wider processing conditions, increased speed of manufacturing, lower product costs overall and reduced in-process inventory. Beyond the business impacts, probably the greatest benefit to the customer will be the consistency of the product quality delivered through continuous processing.

Given that the pharmaceutical industry is highly regulated, changes in manufacturing technology will go through multiple reviews with regulators such as the Food and Drug Administration (FDA) before being approved as a new drug. These regulators are concerned with understanding the processing parameters that are critical to product quality and how they will be controlled as part of the process.  We expect that improved product quality and consistency will result in strong support from these agencies for further investments in continuous processing.

In a continuous process, a process upset may result in a period of short-lived impurity formation that is inside or outside established limits for accumulated product quality control. To judge the overall impact to quality calls for a clear definition of what is considered a product lot for a continuous process. For a batch process, the definition is straightforward since it is typically the amount of product manufactured simultaneously in a plant vessel. 

However, the lot definition for a continuous process could be any amount of accumulated product or time period of manufacture. An agreed-upon definition of lot quantity is one element that allows manufacturers to demonstrate that we have robust processes for producing material consistently, have a good control strategy and define the quality of the material in a way that makes sense.

 

How do you use technology in process development?

One of the biggest advancements I have seen in the last 10 years is the use of online analytical technology in pharmaceutical process development and manufacturing. Unlike in refineries and other continuous manufacturing industries, online analytical sensors have not been commonly used in batch processing of pharmaceuticals or specialty chemicals. 

In continuous processing, for instance, it is advantageous for us to know how a reaction is progressing and whether any changes in process parameters affect product quality. This is especially important in the process-development phase, where scientists are mapping and defining a robust operating space that meets quality requirements and provides operating efficiency.

Traditionally, kinetics and product quality are determined by taking samples for offline analysis. More often, in-situ sensors such as infrared (IR) spectrometers and inline particle size analyzers are being employed for process development and monitoring of several types of unit operations, including reactions and crystallizations in real time. We can also regress sensor data in modeling software to determine important kinetic parameters, allowing us to simulate the process accurately and refine experiment plans. 

In-situ process sensing allows us to observe the effects of process changes or upsets that are important to product quality without needing to wait for offline sample analysis. This, in turn, allows us to create robust manufacturing recipes quickly for all types of operations.

 

What separates successful projects from failures?

An important measure of project success is process-development efficiency, and that can be improved significantly by incorporating process modeling and simulation in the laboratory workflow.

It is important for scientists to streamline the number of time-consuming experiments by combining computational studies with a smaller set of lab tests. Scientists are often good at iterating through a set of conditions and materials they are familiar with, but process simulation allows them to explore a wider set of materials in addition to extended temperature and pressure ranges. 

With insights gained from simulations, scientists can return to the lab with more confidence in the feasibility of recipes under consideration and proceed toward further refinement through experimentation. Process simulation also exposes the sensitivity of process outcomes to various changes in inputs, and this information provides a much richer understanding of the robust operating space appropriate for manufacturing.

 

In your opinion, what should process-development specialists (chemists, engineers, analysts) focus on during process development and/or scale-up?

Three key focus areas for process developers are: 

• Process optimization for efficiency and product quality
• The effect of increasing scale on equipment selection and recipe specifications
• A thorough understanding of the robust operating space around the optimum conditions that meet product quality requirements 

We use a powerful set of computational and laboratory technologies together to provide answers in each of those areas. 

Process development specialists should concentrate on acquiring the benefits derived from merging simulation, lab automation and online analytics in their research efforts for improved efficiency and confidence in manufacturing recipes.

Eric Cordi hosted a live webinar with AspenTech on October 24th 2017. To learn more about how Pfizer accelerates new process development, click HERE to view the event on-demand.