Every biopharmaceutical manufacturing process is unique. While these processes take common approaches to unit operations, they combine unit operations differently for each drug that is produced.
The same is true for how each biopharmaceutical production process is monitored, the parameters measured, and the systems used to gather data. Certainly, some variables, such as pH, carbon dioxide, and concentration, are almost always measured. However, the technologies used tend to differ process to process. Equally, there are some parameters, such as online pressure, viscosity, conductivity, and optical density, that are measured less often, but which also impact the biosensor and monitoring technologies that are used.
For commercial-scale manufacturing usually it is the data needs that determine which parameters are measured, says Julia Brück, marketing specialist at Infors. Some developers want biosensors to provide a process overview, she adds.
Raman and near-infrared (NIR) spectroscopy measure many of the key biochemicals in culture media, supporting the creation of models for process analytics technology (PAT). These models are not in widespread use, but they show the trend toward processing more biochemical data to create predictive models.
Advanced monitoring technologies are part of a paradigm shift, according to Brück: “Rather than parameter control, this moves things closer to the realm of whole bioprocess control. This is the direction of travel for the future.”
The change is being driven by evolving regulations and the rising demand for safer, higher quality medicines. Growing regulatory preference for Quality by Design (QbD) is a key factor driving biopharmaceutical industry interest in process data and shaping demand for biosensors and monitoring technologies, notes Brück.
“QbD must be looked at as a part of a wider emphasis on getting key data quickly and having the potential to use it in real time,” she explains. “If a process is not going to meet defined safety, productivity, or regulatory standards, it needs to be stopped as quickly as possible. This saves resources and time, and generates useful data by analyzing what has gone wrong.”
This QbD focus is reshaping all aspects of process development, from the very initial stages through to final design. For example, it is increasingly common for manufacturers to determine key parameters using multiple small-scale experiments, the so-called Design-of-Experiments (DoE) approach.
The method allows manufacturers to define how interactions between parameters influence productivity or quality. “Excursion limits,” Brück points out, “can be defined that support some flexibility in how much a key parameter can vary before the process is considered to have gone outside acceptable limits.”
The regulatory quality push is also moving the industry closer to statistical process control, whereby archived data is analyzed and modeled to identify potential areas where improvements can be made.
“The net result is that it encourages a manufacturer to do a lot of testing at the laboratory scale and feed that information into control of the larger process—a major aim of the FDA’s PAT initiative,” says Brück. “It is even possible to reverse this process and take a sample from a larger bioreactor and run small-scale experiments in real time to adjust a running production process.
“QbD is very much concerned with the nuts and bolts of the bioprocess. The emphasis is on defining the key parameters and critical quality attributes (CQA) in order to improve them. This is a practical subject and feeds back into optimizing both the process and its validation. The values of concern are mostly biochemical and physiological in nature.”
Per Lidén, product strategy manager from GE Healthcare Life Sciences, also says regulatory pressure is impacting how drug makers approach process monitoring, particularly with respect to the use of data.
“The expectations from regulators on process monitoring keep growing,” he asserts. “Independent of that, biomanufacturers have strong incentives to step up process monitoring for risk mitigation by detection.
“While there is a strong interest in new sensors to detect species that cannot currently be detected, the industry is still ramping up on consistency monitoring. Techniques that fall into this category include multivariate statistical process monitoring or other more sophisticated soft-sensor technologies and the use of spectral sensors such as Raman and NIR as a fingerprint for process consistency.”
Demand for increased consistency is also shaping how the industry approaches process monitoring. The focus is on the parameters that have the greatest impact on biopharmaceutical product quality.
“A few values are so critical they cannot be allowed to change by much,” Brück says. “For example, a temperature of 38°C or above will kill animal cells.” The accurate measurement of temperature, then, is critical for success.
“Parameters in the critical category can have implications for production environments,” she continues, explaining that manufacturers often employ multiple sensors to ensure they are measured accurately and continuously.
Cost and competition pressures
Regulatory demand for quality is impacting process monitoring and the use of biosensors in other ways, in particular, with regard to capital expenditure and technology investment. According to Lidén, “Manufacturers are now thinking much more holistically about the factors that contribute to cost than in the past, and process variability is a key driver cost of many sorts, including cost of quality and capacity utilization.” One driver Lidén sees is greater focus on generating more data during process characterization to increase process understanding ahead of market approval.
Cost pressures have also impacted drug industry spending on process monitoring technology. “New monitoring technology is being introduced in the control strategies for new processes,” Lidén says. “If I were to mention one such technology where I see a lot of interest today, it would be MAM— multiattribute methods.” Brück is of a similar opinion: “The more critical a process, the more will be spent on the best measurement and control possible.”
Beyond this, quality is also being used by innovative drug companies to protect market share. “A factor to consider, is the use of high quality standards to differentiate products from a large pharma company from smaller companies producing biosimilars,” Brück points out. “The large pharma company has resources to invest in the best sensors, devices, models, and quality control to a level not attainable by the smaller manufacturer.
“This has nothing to do with product safety or basic regulatory compliance. It could, however, could reflect differences in scale, experience, and innovation.”
The biopharmaceutical industry’s growing interest in cell therapy development has also changed process monitoring.
From a conceptual standpoint, cell therapy production is similar to “traditional” biopharmaceutical manufacturing. Cells are cultured in strictly controlled and monitored conditions and then processed into therapies.
The difference is that processes have yet to become “standardized” in the cell therapy industry, says Erik Kakes, international sales and marketing director at Applikon Biotechnology.
“Many cell manufacturers are still in the starting phase of their process development and are slowly entering production,” he notes. “The robustness of the processes is not at the level of regular biopharma production processes.”
“Volumes are smaller,” Kakes continues. “This also limits the process monitoring options on this smaller scale.” Smaller volume is also a key differentiator for Brück, who says, “Process monitoring systems—which exist for primary cell culture of individual patient’s cells—will usually be smaller, single use, and tied to very tight quality procedures.”
The approach to process monitoring later in the cell therapy production process also differs from traditional biopharmaceutical manufacturing. “There is no ‘downstream processing’ step to consider,” Brück says. “However, freedom from unwanted biochemicals, cell fragments, etc., from the growth steps is just as important. Measurement of these elements is almost always part of offline analytical processes.”
In addition, regulators tend to impose stricter safety requirements on cell-based products, often requiring that their efficacy and integrity has to be demonstrated to a higher standard of proof than biopharmaceuticals. For example, engineered cells such as chimeric antigen receptor T cells undergo rigorous testing before use to reduce the risk of common complications such as inflammation and neurotoxicity. Part of this testing process involves the assessment of data gathered during production.
The realization that cell therapy process monitoring differs from traditional biopharmaceutical manufacturing is an opportunity for technology developers. “At Applikon,” asserts Kakes, “we are developing miniaturized and preferably noninvasive sensor technologies that can be applied in these cell and gene therapy processes.
“Our process control systems and process management and analysis software is prepared to accept these novel technologies in combination with our small-scale single-use Appliflex ST bioreactors to offer a simple to operate but functionally advanced turnkey solution for the cell therapy market.”