By: Frank Corden, Senior Director of Enterprise Solutions, New England Controls
Let me start off with a little background. As an analytical chemist, I was attracted to measurement science as a means to understand the world around us. In my mind, you can’t understand something if you can’t measure the stuff that impacts the thing you’re trying to understand. That seemed pretty straightforward when trying to determine environmental compliance and the impact of pollutants on an ecosystem. It wasn’t until later in my career when I got involved with process improvement efforts that I realized the concept of measuring the inputs, process attributes and outputs of a process weren’t universally acknowledged as “necessary” or even well understood. So it’s against that backdrop that I start my story of the happy CHO cell.
As I’ve pointed out in prior posts, I don’t view the concrete and stainless steel, or now single use containers, as a biotech factory. Rather, I see a biotech plant as the sum of millions of very small factories, each unique in its own right. In other words, the factory is really all those CHO cells in the bioreactors. And being a measurement scientist, I’m always interested in what we monitor in a bioreactor, and specifically in the cells.
Historically, we’ve focused on keeping the cells alive, give them oxygen, sparge CO2 and feed them once or twice a day. And in addition to keeping the cells warm (but not too warm) we sought to optimize these factors. But that’s like trying to figure out what is going on in a factory by monitoring the warehouse. It’s indicative of the performance of the factory, but it doesn’t give you much detail. So I’ve been more encouraged of late with a focus on really keeping cells “happy”.
In the August 2016 issue of Pharmaceutical Technology, there were several articles focused on measurement and manufacturing trends. In the article “Evaluating Technology and Innovation in Biopharmaceutical Manufacturing” Ranjit Thakur et al. highlighted the increased use of perfusion technologies for continuous manufacturing and for the accelerated uptake of Process Analytical Technologies in our field. Both of these trends seem to me to contribute to the concept of the “happy” cell. And “happy” cells are generally more productive cells.
Though the initial objective of perfusion was to support continuous harvesting as a means to level out downstream processing, published data from Genzyme (UPSTREAM DEVELOPMENT OF HIGH CELL DENSITY, PERFUSION PROCESSES FOR CONTINUOUS MANUFACTURING, Tim Johnson, Oct. 2013) http://www.engconf.org/staging/wp-content/uploads/2013/12/Johnson13AQ-ICB-Monday.pdf and others has indicated substantially improved yields on the basis of harvest volume. The improved titers seemed to correlate well with stability (reduced variation) in viable cell density. Steady state viable cell density leads to better product quality and increased number of passages before a culture is “spent”. We know cells in organisms perform ideally when in specific unique environments (often created by the cells in the environment), and the rapid advances in tissue culture technology and growing artificial organs have focused on identifying these factors. And in many cases, having the right cells in the micro-environment is key. So it would seem obvious that maintaining a consistent cell density in a bioreactor would keep the cells “happy” and productive.
Process analytical technology plays a key role in maintaining those “happy” cells. Updated sensors for in-process viable cell density measurement using RF Impedence/Capacitance can provide real-time data for determining viable cell density. Just as use of optical DO measurements greatly improved DO monitoring over other sampling techniques, sensors from Hamilton and ABER have changed the paradigm compared to at-line or off-line measures. Feed timing, harvest timing and frequency along with sparging and pH control can be modified in concert by looking at the near real-time response of the culture using these on-line cell monitoring techniques and help maintain that steady state cell density.
As noted above, glucose feeding of cell cultures plays a role in improved yields. The use of Raman probes from Brucker and Kaiser along with PAT models like synTQ (Optimal Technologies) and SIPAT (Siemens Automation) enable the real-time measurement of glucose (and other bio molecules as well) to control feed rates on an as-needed basis as opposed to sticking to scheduled feeding times. A well feed cell is a “happy” cell.
The upcoming agenda at BPI this week includes a number of sessions that address Process Analytical Technologies, control approaches, and other novel developments in increasing throughput and yield. Specifically sessions in the Analytical Formulation and Quality Track along with sessions in Upstream Processing and Manufacturing Strategy tracks (URL: https://lifesciences.knect365.com/bioprocessinternational/agenda/1) touch on these topics as well as others. So I look forward to meeting those of you interested in improved measurement and understanding of those factors to create “happy” cells. See you soon.