Current Trends in Raman Spectroscopy for Upstream Bioprocess Monitoring and Control: Capability and Efficiency

IntroductionMore than ten years have passed since the creation of the Food and Drug Administration’s (FDA) Process Analytical Technology (PAT) framework [1], during which the bioprocess has continuously evolved resulting in new challenges that have been addressed with a multitude of innovative measurement and control solutions.The FDA has identified a number of hurdles that will need to be addressed if these innovations are to scale-up from research, into the process development suite, and ultimately the GMP manufacturing suite [2,3]. However, as evidenced by the growing number of success stories being shared by the industry, this is well underway [4-17].This application note covers some new trends in bioprocess monitoring and control. It focuses on the increasing use of Kaiser Optical Systems Inc. (Kaiser) Raman spectrometers as multi-measurement bioprocess sensors to enable quality by design, process optimization, process control and continuous processing. Links to mature manufacturing methodologies such as Six Sigma and Lean Manufacturing are also made throughout this note to highlight the value of Raman-based solutions for improvement of capability and efficiency.

Bioprocess Monitoring Using Raman SpectroscopyBioreactor monitoring and control are particularly challenging given the complex nature of biological matrices. Furthermore, in the absence of a design space, conservative nutrient and by-product control ranges are usually established during development in order to de-risk the process and achieve a consistent process capability. In order to keep the process working within tight ranges, the process needs frequent cumbersome in-process testing with multiple analytical techniques, many of them not suited for in situ and/or real-time measurements. These factors can add significant non-value-added cost and activities (inefficiencies) to the manufacturing process which goes against advocated lean practices.

In order to improve upstream bioprocess probing and to facilitate control, Kaiser has developed a series of solutions to measure multiple components at once in an in situ and real-time manner, Figure 1. These solutions enable frequent spectral measurements from biological matrices easily and safely without the typical water interference encountered when using other vibrational spectroscopy analyzers such as NIR. Raman measurements can be used to make process adjustments in real time when evolving away from optimal operation conditions, guaranteeing a consistent and reliable process. This approach is well aligned with Right First Time and Zero Defect initiatives.

Mammalian Cell Culture

Figure 2 (a). Raman spectral data of mammalian cell culture media with different glucose concentrations (b) Raman glucose prediction for in-line obtained data (black trace) against off-line reference measurements (red circles).

(a) (b)

Mammalian Cell Culture MonitoringOne of the most popular Kaiser Raman bioprocess applications is the monitoring of mammalian cell cultures. Kaiser Raman spectrometers have been successfully used during mammalian cell cultures to monitor multiple critical constituents and metabolic indicators such as glucose, glutamine, glutamate, lactate, ammonium, osmolality, viability, viable cell density (VCD), and total cell density (TCD) [7-15]. Naturally, glucose has been the most critical monitored constituent due to its role as the primary source of nutrients for the cells.Figure 2 (a) shows clear systematic spectral variation of glucose in typical mammalian cell culture media, whereas Figure 2 (b) shows off-line reference measurements (red circles) superimposed to in-line Raman glucose predictions (solid line) for a mammalian cell culture process.

Figure 1 Example of Kaiser’s bioprocess solution

Figure 3. PLS model prediction results for major Critical Performance Perameters (CPPs). Calibration data from batches at three scales (3,200 and 2000 L) were used to predict results from a batch at the 2000 L manufacturing scale. Reprinted with permission from Ref. 14. © 2014 American Institute of Chemical Engineers.

Figure 3 shows a comparison between off-line reference results and in-line Raman predictions for additional critical nutrients such as glutamate, by-products such as ammonia and lactate, as well as VCD, TCD, and osmolality. In addition, Figure 3 shows the application & scalability of Raman to three different process scales from PD, Pilot, and GMP Manufacturing. Figure 4 shows a comparison for critical culture parameters such as VCD and titer. Comparisons depicted in Figures 2 to 4 were obtained from different media types and cell lines, demonstrating Raman’s versatility throughout the bioprocess pipeline. In addition, Raman-based methods that are applicable across different cell lines have been successfully developed using spectral data pooled from multiple media types and multiple cell culture stages [12, 13].

Microbial FermentationsMicrobial Fermentation is a very rapid process, progressing in days not weeks, and as such has proven challenging to most analytical instruments but not Raman. This is attributed in no small part to Ramans real time in situ measurement of the fermentation process [9-11]. Figure 5 (a) shows the in-line−obtained Raman data of a simultaneous saccharification and fermentation of corn mash. Figure 5 (b) shows good agreement between Raman ethanol predictions and High Performance Liquid Chromatography (HPLC) off-line method results.

Figure 4. VCD and titer trajectories measured in situ using a Raman PLS model (blue trace) superimposed with off-line reference values (red circles).

Quality by DesignRaman spectroscopy’s real-time and in situ measurement capabilities can play a fundamental role not only in production, in order to ensure that the process runs according to the drug license, but also during process development. In the latter Raman monitoring allows a more scientific manner of establishing critical parameter ranges by the use of representative sampling frequency strategies. These strategies enable the process’s reaction to disturbances during DoEs to be characterized in a more comprehensive and dynamic manner. Also, given the non-destructive nature of Raman, it’s now possible to characterize mini as well as micro scale bioprocesses without diminishing

Figure 5 (a). In-line obtained spectral data of a microbial fermentation. (b) In situ Raman ethanol prediction against reference measurement (HPLC)

the working volume, and thus compromising the run. The above-mentioned benefits make Raman spectroscopy a very attractive tool for developing processes under Quality by Design (QbD) principles.

Continuous Bioprocessing Continuous manufacturing is a stated goal of the FDA for the purpose of improving process efficiencies and reducing costs. One of the requirements of continuous processes is the use of robust process measurement techniques. These are needed in order to ensure stable operations as well as to support the high degree of automation needed to enable continuous processing. Kaiser bioprocess monitoring solutions enable monitoring multiple critical-to-quality parameters in real time for extended periods without degradation of the measurement source. Robust bioprocess analyzers such as Raman, that enable truly continuous bioprocessing open a new realm of opportunities for process developers, and enable a wide array of novel solutions to traditional bioprocessing challenges and emerging challenges alike.

Bio-Process ControlImproved process control, achieved through a better understanding of that which is occurring in the process, is the ultimate goal of the PAT initiative. With Raman’s proven ability to output its multitude of measurements to any bioprocess control platform the resulting control strategies can be as simple or complex as is required. Figure 6 shows an example of a model predictive controller for a bioprocess application [18].

References1. PAT — A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance

2. Christine M. V. Moore, Ph.D., Guidance for Industry Multivariate Tools for Modern Pharmaceutical Control – FDA Perspective, IFPAC, Annual Meeting, 2013

3. Sharmista Chatterjee, Ph.D., FDA Perspective on Continuous Manufacturing, IFPAC, Annual Meeting, 2012

4. Ronald a Rader and Eric Lander, 30 Years of Upstream Productivity Improvements, Bioprocess International

5. C. Undey, B. Looze, S. Oruklu, T. Wang and R. Woolfenden “Process Analytics Experiences in Biopharmaceutical Manufacturing,” European Pharmaceutical Review, v7 issue 3 2012, 22-29

6. R Johnson, O. Yu, A O Kirdar, A Annamalai, S. Ahuja, K Ram and A Rathore, “Applications of Multivariate Data Analysis in Biotech Processing,” BioPharm, Oct 2007, vol 20 num 10

7. N. R. Abu-Absi, B. M. Kenty, M. E. Cuellar, M. C. Borys, S. Sakhamuri, D. J. Strachan, M. C. Hausladen, and Z.-J. Li, “Real Time Monitoring of Multiple Parameters in Mammalian Cell Culture Bioreactors Using an In-line Raman Spectroscopy Probe,” Biotechnology and Bioengineering, May 2011, pp. 1215-1221.

8. J. Moretto, J.P. Smelko, M. Cuellar, B. Berry, A. Doane, T. Ryll and K. Witberger, “Process Raman Spectrosccopy for In-line CHO Cell Culture Monitoring,” American Pharmaceutical Review, April 2011

9 M. E. Cuellar, “New PAT Tool for In-line Fermentation Monitoring,” presented at IFPAC, January 19, 2011, Baltimore, MD

10. S. Romero-Torres, “Raman as a Multipurpose Sensor for Biopharmaceutical Processes,” presented at Eastern Analytical Symposium, Nov 13, 2012, Somerset, NJ

ConclusionProcess Raman can play a fundamental role in biopharmaceutical production and bioprocess development due to its real-time, in situ capabilities and low water sensitivity. During bioprocess development, Raman spectroscopy can be pivotal when adopting QbD principles aiming to define manufacturing design spaces and demonstrate holistic process and quality understanding. During production, a single Raman sensor can be used to monitor and enable control of multiple critical parameters. This can help to improve bioprocess reliability and promote leaner manufacturing by reducing the cost and risk associated with contamination consumables, equipment maintenance, and operator overburden. Also, a bioprocess control strategy which utilizes process Raman as a real-time PAT method is well suited for continuous processes, 60, 90, 120 days or more, thereby maximizing output while reducing costs. The breadth and depth of insight into the bioprocess facilitated by process Raman, and the associated control strategies made possible, give those utilizing this technology a competitive advantage at the process development, pilot and GMP manufacturing scale.

Figure 6. Schematic diagram of a model predictive controller system and illustration of its incorporation into a bioprocess.

11. S. Romero-Torres, “Use of Raman Spectroscopy to Monitor Fermentation & Key Constituents in Small & Large Molecule Fermentations to Enable Intelligent Bioprocesses,” presented at IFPAC, January 24, 2013, Baltimore, MD

12. D. Didion, A. L. Hantson, and E. Lecomte Norrant, “PAT on Mammalian Cell Culture – Comparison of Two Technologies: NIR and Raman,” presented at 3rd Annual Cell Culture & Bioprocessing Congress, November 18th 2014

13. H. Mehdizadeh, “Generic Raman-based Calibration Models Enabling Real-time Monitoring of Cell Culture Bioreactors,” Biotechnology Progress, 2015, 10.1002/btpr.2079

14. B. Berry, J. Moretto, T. Matthews, J. Smelko, and K. Wiltberger, “Cross-Scale Predictive Modeling of CHO Cell Culture Growth and Metabolites Using Raman Spectroscopy and Multivariate Analysis,” Biotechnology Progress, 2014, 10.1002/btpr.2035

15. J. Whelan, S. Craven, and B. Glennon, “In Situ Raman Spectroscopy for Simultaneous Monitoring of Multiple Process Parameters in Mammalian Cell Culture Bioreactors,” Biotechnology Progress, 2012, 10.1002/btpr.1590

16. I. R. Lewis “Process Raman Spectroscopy,” Handbook of Raman Spectroscopy: From the Research Laboratory to the Process Line, edited by I. R. Lewis and H. G. M. Edwards, New York: Marcel Dekker 2001, pp. 919-974.

17. K. L. Davis, M. S. Kemper, and I. R. Lewis “Raman Spectroscopy for Monitoring Real-Time Processes in the Pharmaceutical Industry,” Pharmaceutical Applications of Raman Spectroscopy, edited by S.Sasic, Hoboken: Wiley-Interscience 2008, pp. 117-162.

18. S. Craven, J. Whelan, and B. Glennon, “Glucose Concentration Control of a Fed-batch Mammalian Cellbioprocess Using a Nonlinear Model Predictive Controller,” Journal of Process Control, 2014, 10.1016/j.jprocont.2014.02.007