Today’s drug manufacturing processes are only about 35% efficient. “We see tons of rework and scrap,” says Bart Reitter, Global Marketing Manager, Life Sciences, GE Fanuc, “because the scientists have not examined or altered their processes in years, and their understanding of them is insufficient for present needs.”

The reasons behind such dismal performance are many, including an adversarial relationship between the FDA and pharmaceutical companies, particularly, the FDA’s concern for patient safety that led to an insistence on verified, locked-down processes, and the widespread pharmaceutical view that manufacturing is a necessary evil.

Today’s business pressures, however, will not permit such inefficiency to continue. Increasing competition, price and delivery pressures, counterfeit drugs, millions of dollars of stored inventory, and advances in biomedical science and technology are squeezing dividends. To please stockholders, pharmaceutical companies are discovering that manufacturing is the new profit territory, offering the best opportunity to reduce expenses and improve the bottom line.

From locked down to open
The FDA’s voluntary Process Analytical Technology (PAT) initiative is an attempt to help pharmaceutical companies survive and thrive in today’s business climate. “PAT is a representation of a risk-based approach,” says John Blanchard, ARC Analyst. “It’s a system for designing, analyzing, and controlling manufacturing for timely measurement during the processing of critical quality and performance attributes of raw and in process material and processes with the goal of ensuring final product quality. It’s quality assurance rather than quality control.”

Adds Woelbeling, “Quality is to be built in, not tested upon. The pharmaceutical companies have to understand their processes to the point that they know the impact when parameters change.”

Data collection will be paramount and ongoing. Data analysis will be key too, as the FDA requires companies to prove their science. In addition, it is this requirement that makes the needs of the pharmaceutical industry different from other industries, which already use online-inline equipment.

Blanchard mentioned that pharmaceutical companies would likely need real-time and historic data collection, packaging technologies, electronic batch records systems, electronic wireless laboratory information management systems, and compliance systems. Other needed systems include product authentication, tracking, bar code and RFID, anti-counterfeiting measures for packaging and product formulation, as well as common site, corporate and trading partner business processes and associated information technologies.

“It’s more the measurement, sensor, and analytical technologies that will be affected,” adds Woelbeling. “The FDA wants the manufacturing process to be completely controlled and understood, so that you can measure exactly what is going on during the process. For example, in granulation, you want to measure particle sizes. In near infrared spectroscopy, you want to verify the exact amount of active pharmaceutical ingredient in the mixture.”

Encouraging pharmaceutical manufacturers to buy such systems and devices will not be without challenges, however. Their culture is established and often resistant to change. Moreover, PAT is voluntary. To meet the PAT recommendations, scientists must open their process. Except for the large pharmaceutical companies, there is reluctance to spend already small profits on new equipment or devices and undergo the required but laborious validation, re-validation, and documentation.

Managing the risks of progress
The primary way to risk continuous process improvement is for “pharmaceutical companies to identify all sources of variability, manage those sources accurately, and reliably predict the outcome of processes,” says Reitter.

This goal is not easy. Today’s drugs contain live, biological components. “There are many degrees of freedom in biological processes as opposed to a limited number in finite chemistry,” adds Blanchard. “Fermentation, for example, does not always result in the same yield or efficacy of the end product. There are too many variables.”

For instance, one manufacturer found that a batch of pills did not have uniform dispersion of the active pharmaceutical ingredient, rendering that batch useless. The scientists had to find out why the dispersion was not uniform. After months of data collection and analysis, they concluded that humidity levels during the summer that year at their Puerto Rican manufacturing plant where higher than the plant’s temperature control system could handle. The scientists never realized this factor could affect a batch. However, this is exactly the kind of depth of process knowledge the PAT initiative promotes. Instruments and devices that can support such depth will be valuable.

Streamline opportunities
Most drugs go through a blending stage. However, scientists often do not precisely know how much blend time is required; every batch is different despite following the drug recipe. One reason for the difference is in the quality of raw ingredients, which, until now have not been examined before blending.

Processes call for frequent sampling and off-line laboratory analyses. Results typically return several days or a week later. Meanwhile, the batch is waiting. “A goal,” says Woelbeling, “is to bring analytical testing online to skip that delay. By measuring all the values that occur during production, operators can adjust parameters for specific qualities in a mix or make corrections before a batch is ruined. For example, if the particle size is increasing too much, a change in the process could correct it.

“The equipment of the future must be able to measure all the quality attributes on line,” continues Woelbeling. “One challenge that complicates this goal of a real-time control loop, though, is that a few processes need some delay time before you can test for results. Pharmaceutical processes need real-time control loops, but not all processes lend themselves to this, so we need to find new analytical methods and technologies to close these control loops.”

To determine precise blending times, for example, scientists are installing laboratory instruments on the manufacturing line. “In one facility, engineers have installed a spectrometer on a blender to actively check the molecular level of a chemical compound to determine the blend time required to achieve batch homogeneity,” says Reitter. “Their recipe calls for 16 hours of blending even though the process might be finished in less time. Without data stating otherwise, the recipe will be followed precisely. However, through our software, the spectrometer, and following PAT guidelines, we were able to gather data that proved the proper blend was achieved in four hours. That saves 12 hours of running equipment; and lets a drug reach its market more quickly.”

Another opportunity for designers is to design and build laboratory instruments to a particular standard. “Instrument manufacturers often have proprietary ways of programming and communicating with their devices,” notes Blanchard. “This affects how a company meets the FDA’s validation requirements. To validate an instrument, you must validate how it takes a measurement, how it communicates, its method of analyzing data, how it calibrates, samples, and so on. Each instrument must go through this process. But most instruments are closed and these parameters and functions cannot be determined. You only receive results, not how they were obtained. There are not enough manpower and financial resources in the pharmaceutical industry to meet the intention in PAT without certain standards at the instrument level.”  

Turning the Titanic
Presently, management awareness of the PAT initiative is much too low. To encourage change, the FDA will have to play a key role. Even though this agency has limited resources, it will decide which companies to inspect based on their voluntary compliance with the PAT initiative.

Designers can play a role as well. One important step is to join organizations such as the ISPE (www.ispe.org) and ISPE CoP PAT Steering committee. Founded in 1980, ISPE is a not-for-profit association dedicated to educating and advancing pharmaceutical manufacturing professionals and their industry.

While some pharmaceutical scientists and managers are taking such steps as hardening laboratory instruments for use on the manufacturing line, more will need to be done. Your expertise in various technologies, especially measurement devices and systems, can be crucial in moving drug production into the world of modern manufacturing.

Taking the wait out of production
Ahead of the release of the PAT initiative, MG America introduced its NETT Weight Control System in 1999. This system is an option for its automatic capsule filling machines. It measures the weight of every capsule produced on-line and in real time, eliminating the need for someone to collect samples for a lab quality check. The system also automatically rejects any capsule outside the specified limit and automatically sends reports to operators or management at the end of each batch.

The system first weighs the empty capsule. Then it orients and separates the capsule lid from the body, fills the capsule body with the drug, replaces the lid and weighs the full capsule. The system then subtracts the two weight measures to obtain the net weight.

“Even though this is an optional add-on to our filling machines,” notes Claudio Radossi, Director of Sales, “we find our customers are opting for it to work with the PAT initiative.”

To handle validation needs, MG America provides a complete validation package consisting of design and function specifications, and installation and operational qualifications.