PAT enables process development, manufacturing

Most biopharmaceutical facilities today are based on batch processes and unit operations. The process is run through one unit operation until completed, based on the application to the FDA for the product/drug being manufactured. Once the unit operation is complete, typically a sample of the batch is taken to a lab for analysis to determine whether it has met the criteria for the specific unit ...

By Dennis McKinley, ABB Inc. and Alex Brindle, NNE PharmaPlan November 1, 2009

Most biopharmaceutical facilities today are based on batch processes and unit operations. The process is run through one unit operation until completed, based on the application to the FDA for the product/drug being manufactured. Once the unit operation is complete, typically a sample of the batch is taken to a lab for analysis to determine whether it has met the criteria for the specific unit operation.

If it passes, the product is deemed good, and is allowed to transfer to the next unit operation in the process. If it fails, a determination is required to define what, if any, additional steps must be taken to get it to pass, or if the batch must be scrapped or recycled. This cycle is repeated until the campaign has completed, which could be anywhere from hours to weeks, depending on the process.

Inherent inefficiencies

Today’s procedures make the biopharmaceutical manufacturing process very inefficient as there are a number of points during the process where there is no value-add or work being done to the product. Downtime occurs at the end of every batch. Depending on the testing requirements, it could take hours or days to get the “go/no-go” answer on the product. When you add up the number of stops and downtimes associated with the batch processes in a typical, large scale manufacturing facility, you have the potential for months of time wasted while waiting for the determination of the quality and yield of a given campaign.

No other industry runs like this. Industries such as chemical, petrochemical, oil and gas, pulp and paper, minerals, food and beverage, etc., are typically based on a continuous model, where testing is done on line or at line, continuously. Outputs of the different unit operations feed directly into the next step in the process. The unit operations are highly instrumented and automated. They use advanced modeling and control techniques to keep the process in control. This ensures proper quality and appropriate yields of the end product — in other words, minimal waste. When a line is down in these industries, there is a lot of emergency attention paid, as they are losing valuable time, money, products and profits.

What is PAT?

One way to help pharmaceutical manufacturing “catch up” with other industries’ use of modernized operations and technology is through implementation of Process Analytical Technology (PAT). PAT has been defined by the U.S. Food and Drug Administration as “a system for designing, analyzing and controlling manufacturing through timely measurements (i.e., during processing) of critical quality and performance attributes of raw and in-process materials and processes, with the goal of ensuring final product quality.”

PAT is a leading-edge quality control concept in the biopharmaceutical industries that promotes integration of analytical measurements into the manufacturing process to streamline and improve new product development. It improves the quality and ensures the safety of the end product. It also facilitates Quality by Design (QbD) and real-time product release.

Simply put, PAT is a system that combines analytical process data with traditional process control data such as temperature, pressure and flow to allow better understanding of the process. This system can be used in development, drug discovery, pilot manufacturing and full-scale manufacturing.

Once all of the data are collected and combined, then with the use of advanced modeling techniques, you can derive the variables that have the most effect on the desired outcome of a specific process or unit operation. Once a better understanding of the process is achieved, the operating bands or design space can be defined. As long as the process stays within the design space, the process is considered by the FDA to be under control.

When the process deviates outside the predefined design space, corrective actions can be taken — either manually or automatically, based on modeling — to get the process back under control, i.e., within the design space. Thus PAT allows for closed-loop feedback control, based on the results of advanced process control and modeling.

Quality by Design drives PAT

It’s easier to understand the benefits of PAT if you first understand the need for Quality by Design. The FDA has challenged pharmaceutical manufacturers to achieve QbD by increasing their level of process understanding so that they can control process variability and assure product quality in real time — while a batch is being manufactured.

This is a huge improvement over relying on post-production product testing, wherein you don’t know whether you’re producing a quality batch until it’s too late — it’s already produced. The FDA’s message is clear: “Lower your costs and increase your quality by enabling the ability to achieve the appropriate quality outcome during (not after), the manufacturing cycle.”

QbD can be implemented throughout the entire product lifecycle, achieving cost reductions and process optimization while improving time-to-market. The goal is to develop a design space that defines a range of variability within which the right product quality outcomes are consistently produced.

What is a PAT system?

There are systems available today that provide simplified engineering and streamlined integration with enterprise systems, as well as broad connectivity to a host of analyzers to help achieve continuous process improvement, real-time product release and QbD. A good, complete PAT solution provides the level of process understanding required to enable QbD. The system should have the capability of providing the following functionality in order to be effective across the enterprise as well as the life cycle of the products manufactured:

Seamless scalability from development to manufacturing — The solution for PAT should allow integration of multiple analytical and automation products in a flexible and scalable platform. It should provide a single configuration point for multi-vendor analytical instrumentation, data acquisition and central storage of all analyzer and process-related data.

Data collection, advanced analyzers — There is a large variety of analyzers, provided by multiple vendors, that provides many different types of critical data. The best way to accomplish the level of analyzer integration required for PAT is through broadly-adopted open standards, versus “one off” interfaces. The recently announced OPC-ADI (Analyzer Device Integration) standard will help facilitate open communication standards with instrument and control suppliers.

The OPC-ADI standard is based on the OPC Unified Architecture specification. It facilitates the building of complex systems composed of products from multiple vendors. It also provides the infrastructure for responding to both simple and complex information integration opportunities. It facilitates open communication standards among instrument and control suppliers. The PAT solution provider should fully support the OPC-ADI standard and actively add OPC-ADI-compatible products to its portfolio.

Data management, visualization, advanced control — The PAT data management and control platform should allow users to manage analyzer control parameters; store analyzer data and traditional process data; and perform multivariate data analysis for process monitoring and process control.

Enterprise connectivity, productivity tools — A good PAT system allows for connectivity to a variety of systems in the labs and manufacturing facilities such as LIMS, ERP, batch management and EBR. Such connectivity delivers significant opportunities to increase productivity by allowing for data-informed decisions and actions to be taken. With the increased amount of data available in one system, consolidating and using the information becomes much easier to do. In addition, the information can be accessed by additional users within the company environment in a timelier manner.

Benefits of PAT

Gaining better process understanding translates to lower costs and increased quality. Cycle times can be reduced by an average of 40%, while costs are reduced as much as 30%. Using a PAT system helps pharmaceutical manufacturers to reduce product quality variations such as rejects, scrap and reprocessing; reduce inventory, lab testing and associated paperwork, regulatory overhead and compliance costs; and increase automation to improve operator safety and reduce human errors.

PAT can be used in the development of new products, and later scaled up for the manufacturing process. It can be used to further understand and refine the design space within a process to enable better controls while reducing the amount of after process testing required.

With an effective system in place from Research and Development through manufacturing, PAT can enable the bidirectional flow of process-related information that can be used to compare desired process versus actual process related parameters, and correct process related upsets in the manufacturing space.

In order to be effective at deploying a PAT strategy, companies must involve a number of different working groups into a team. Successful teams typically include representatives from the analytical, automation, IT and quality departments.

PAT has tremendous potential to help the life science industries improve their product quality, safety, time to market and overall efficiency. Today’s leading edge technologies provide these manufacturers with the tools they need to dramatically improve their overall process. The implementation of Process Analytical Technology and the resulting improvements will benefit the pharmaceutical industry as a whole, as well as the consumers of these medicines.

Using PAT, pharmaceutical manufacturers can control process variability and quality while a batch is being manufactured.

A univariate control model is based on statistical process control concepts, and is typically based on a singular input.

A multivariate control model uses multiple elements, thereby yielding improved outcomes.

Quality by Design enables biopharmaceutical manufacturers to increase their level of process understanding in order to control process variability and assure product quality in real time.

Author Information
Dennis McKinley is Life Sciences sales manager for North America at ABB Inc. Alex Brindle is a PAT consultant at NNE PharmaPlan.