Real Time Process Monitoring in Biologics Development

Introduction

Real time monitoring of biopharmaceutical process development has evolved significantly in recent years with the advancement of analytical technologies, cyber-physical systems and advanced-data interrogation tools.¹ Robust platforms for ensuring consistent product quality through Quality by Design (QbD) approaches are becoming a necessity as biologic drugs gained popularity over the last decade.^2,3 As defined by the International Conference of Harmonization, QbD is a systematic approach to development that begins with predefined objectives and emphasizes product and process understanding and process control, based on sound science and quality risk management.⁴ A holistic understanding of the development process is a key prerequisite for designing a QbD roadmap where appropriate analytical technologies and cyber-physical systems must be deployed for real time monitoring of critical product quality attributes (CQAs), critical process parameters (CPPs) at critical control points (CCPs).² Per ICH definition, a CQA is a physical, chemical, biological, or microbiological property or characteristic that should be within an appropriate limit, range, or distribution to ensure the desired product quality. Whereas a CPP is a process parameter that affects CQAs and the critical unit operations at which product quality attributes should be monitored are defined as CCPs.^4,5 Real time monitoring of bioprocesses involves integration of process analytical sensors, data management, visualization, and advanced data analytics such as machine learning and deep learning (Figure 1).

Here we review the key components of real time monitoring in biopharmaceutical development with strategic insights into designing a technology roadmap for process monitoring and advance process control. Real Time Release (RTR) and QbD of biopharmaceuticals in relation to Industry 4.0 are emphasized.

Real Time Monitoring – In the Context of Bioprocessing

Integration of physical sensors into the bioprocess stream is one of the main components which allows real time data acquisition. The process Analytical Technology (PAT) framework document published by the US Food and Drug Administration explains the guidance on developing analytical sensors and different modes of integration at the CCPs of the bioprocess.⁶ In-line sensors, where the sensors are placed within the bioprocess stream and data acquisition is performed without taking samples out from the unit operation, are common mode of choice for real time data acquisition. Vibrational spectroscopic probes such as Raman and Fourier Transform Infra-Red (FT-IR) are well-known for real time data acquisition from production bioreactors,^7,8 and other unit operations of the bioprocess.^9,10 Flow cell sensors allow the flow of process fluid in-line for real time data acquisition. In-line light scattering and UV flow cells are some of the common applications in bioprocess monitoring of therapeutic protein concentration and product-related impurities such as high-molecular weight species.¹¹ On-line PAT tools, in which a sample from the process stream is extracted for analysis, require automation capabilities.⁶ Automated-samplers, sample distributors and robotics provide cell-free sampling from bioreactors, piping to sample preparation and analysis systems and pre-treatment procedures. Generally, chromatographic and mass spectrometric PAT platforms require automation capabilities.^12-14 Automated data analysis and visualization tools are an integral part of real time monitoring infrastructure (Figure 2). In-line and on-line real time sensors generate signifycant amounts of data which opens up the need of robust data aggregation, management and processing platforms (Figure 2). Automated multivariate data analysis and visualization technologies permit reporting, analysis, visual summaries, and passage of data between enterprise-level systems.¹⁵ Soft sensors such as multivariate chemometric models or advanced machine learning models can be used for instant prediction of product quality and process behavior. For example, Enterprise Manufacturing Intelligence systems enable complete integration of data from physical sensors, distribution control systems and data historians.¹⁶ Upon analysis the data can be further utilized to maintain target product profile by either feedback or feedforward control activities. Data interrogation techniques play a crucial role in unravelling information for better process understanding. Advanced data-mining techniques, mechanistic modeling, machine learning, and deep learning provide a multi-dimensional perspective of the process, predictions and understanding.^1,17

Biologic drugs are mainly macro-molecules, hence unlike small molecular chemical drugs, biologics are more complex in structure and manufacturing processes. Thus, the implementation of real time process monitoring in biopharmaceuticals is challenging in comparison to the traditional chemical industries.¹⁸ The development of a meaningful process monitoring platform where analytical outcomes can be used for precise process control and understanding is essential. This involves the clear identification of CQAs, CPPs and CCPs of the process to achieve the target product profile.² The selection of CQAs and CPPs are dependent upon multiple factors such as the type of the biological molecule, manufacturing process and the historical knowledge base. As an example, for typical antibody therapeutic molecules glycosylation, charge variant profile and purity are often CQAs during the cell culture process, while monitoring of product and process related impurities are crucial during downstream purification unit operations. Analytical technologies are ever evolving, and characteristic of their own capabilities and limitations. For instance, vibrational spectroscopic techniques such as FT-IR and Raman are capable of generating data within a fraction of a minute time scale, however, they may lack the sensitivity and specificity as chromatographic or mass spectrometric techniques. On the other hand, chromatographic techniques are often on-line PAT tools where sample automation and a few minutes-to-hours of assay time are required. CCP of the process may also influence the appropriate selection of PAT tools for real time monitoring. Some attributes such as the titer of a therapeutic molecule during typical fed-batch culture process often changes every several hours, thus deployment of an on-line chromatographic technology which can monitor titer once or twice per day is sufficient for real time understanding of the titer changes during the process. In contrast, concentration of a therapeutic molecule of interest during the elution step of a bind-&-elute unit operation changes in seconds-to-minutes time scale. Online chromatography is incapable of providing real time concentration measurements within the time scale of the bind-&-elute stage, where in-line PAT tools such as FT-IR chemometric or UV sensors with faster response time should be employed.^9,10

Real time monitoring of bioprocesses require precise identification of CPPs, CQAs and CPPs based on historical information followed by the employment of appropriate PAT tools. On-line PAT sensors typically require automated-sampling infrastructure in place. A systematic data management and repository framework should be a part of the real time monitoring strategy for analysis, visualization and data interrogation.

Advanced Process Monitoring and Control

In the era of digital transformation of the biopharmaceutical industry, the utility of advanced data analytics and modeling are becoming increasing-popular for process monitoring applications.^1,19 Data from physical sensors and soft sensors from various unit operations of the process can be used for mechanistic, statistical or hybrid modeling for more holistic monitoring of unit operations and prediction of CQAs and process performance (Figure 3).^20,21 As modeling based process monitoring and control platforms of bioprocess are gaining momentum in recent years, regulatory expectations on model predictive product quality is essential to be considered. The ICH Q8,9,10 Q&A document classifies model-based approaches as low-impact, medium-impact and high-impact, where low-impact models are information-only models that do not affect the process or the product quality assurance strategy, while high-impact models are the sole control/assurance of product quality.²² Medium-impact models can affect the product quality but other engineering controls are in place within the process to mitigate any redundancies due to false predictions. One of the key features of all the regulatory guidance which are available to date, is that they all clearly emphasize the fact that multivariate models should demonstrate the mechanistic, scientific and statistical understanding with supporting data.^22,23

Advance Process Control (APC) is one of the main advantages of using real time monitoring of processes, which can be achieved through fault detection, classification, or soft sensors.²⁴ Real time detection of any atypical behavior of the process can be recovered via feedback or feedforward control routes to ensure final product quality. APC through fault detection involves qualitative or quantitative monitoring of any abrupt performance that resides outside of the typical process experience, which triggers corrective feedback or feedforward controllers.²⁰ For instance, real time control of process variables such as glucose and lactate in bioreactors can be achieved through multivariate spectroscopic models.²⁵ Process monitoring and APC allows superior understanding and consistent maintenance of product quality.

Future Perspective

Despite significant advances in technologies for process monitoring, and control, complete process understanding is still challenging in biologic drug development. While monoclonal antibody based biopharmaceuticals continue to dominate in the space, novel modalities such as bi-specific, tri-specific, fusion proteins, nucleic acid and cell based therapies are emerging through development pipelines.³ Thus, a single platform of manufacturing process is typically not suitable for different types of molecules and CQAs, and CPPs quite often change from molecule to molecule. Owing to these complexities, complete process understanding and implementation of RTR of biopharmaceuticals are challenging in comparison to small molecular drugs. International Conference on Harmonization Q8(R2) guidance defines RTR as “the ability to evaluate and ensure the quality of in-process and/or final product based on process data, which typically include a valid combination of measured material attributes and process controls”.26 Even though RTR is an attractive alternative to current manufacturing paradigms as it enables improved-product quality, improve productivity, and reduce cycle times, a failed RTR batch of pharmaceuticals may not be able to be replaced by a successful end-product test.⁵ Therefore, in addition to PAT tools, advanced process monitoring and APC, other cyber physical systems related to “Industry 4.0” should be in place. The industry 4.0 concept includes the utility of internet of things, sensors, global communication and control.²⁷ As the concept of Industry 4.0 heavily engages data-driven cyber physical systems, it allows real time snapshot of the process and facilitate complete control.²⁷

Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special offers
from American Pharmaceutical Review – all delivered right to your inbox! Sign up now!

Conclusion

Biologic drugs continue to dominate the pharmaceutical industry by exerting unprecedented mechanisms of action and therapeutic end points. Thus, the manufacturing process needs to adopt technical modernization to improve production capabilities, efficiency and consistent quality. Real time monitoring of bioprocess is a paramount necessity to achieve RTR and the concepts of Industry 4.0. Analytical sensors such as spectroscopic or chromatographic PAT tools, and sample automation capabilities are essential for real time acquisition of data from various unit operations of the process. Tools for multivariate and univariate data analysis followed by visualization allow real time monitoring of quality attributes, and processed-analytical outcome can be used for feedback/feedforward control of the process. Data interrogation techniques such as machine learning and deep learning enable in-depth understanding of the process which should be a part of real monitoring framework. A successful roadmap for a real time process monitoring platform entails precise identification of CQAs, CPPs and deployment of appropriate PAT tools at CCPs. It’s essential to employ the right technologies with the understanding of their inherent analytical capabilities and limitations to gain accurate information of the process in real time fashion. Advanced data-driven modeling can also be used for complete monitoring and prediction of the complete process (not by a unit operation basis) followed by APC. Real time monitoring of bioprocesses facilitate RTR and Industry 4.0 initiatives for manufacturing biopharmaceuticals with consistent quality and efficacy for patients.

References

1. Ündey C, Ertun S, Mistretta T, Looze B, Applied advanced process analytics in biopharmaceutical manufacturing: Challenges and prospects in real-time monitoring and control. J Process Control. 2010; 20(9):1009-1018. doi:10.1016/j.jprocont.2010.05.008.
2. Rathore AS, Roadmap for implementation of quality by design (QbD) for biotechnology products. Trends Biotechnol. 2009;27(9):546-553. doi:10.1016/j.tibtech.2009.06.006.
3. Walsh G, Biopharmaceutical benchmarks 2018. Nat Biotechnol. 2018;36(12):1136-1145. doi:10.1038/nbt.4305.
4. International Conference of Harmonization (ICH). Pharmaceutical development Q8(R2) Guideline. https://database.ich.org/sites/default/files/Q8_R2_Guideline.pdf. Accessed 01 February 2020.
5. Jiang M, Severson KA, Love JC, Madden H, Swann P, Zang L, Braatz RD. Opportunities and challenges of real-time release testing in biopharmaceutical manufacturing. Biotechnol Bioeng. 2017;114(11):2445-2456. doi:10.1002/bit.26383.
6. Food and Drug Administration (FDA). Guidance for Industry PAT - A framework for innovative pharmaceutical manufacturing and quality assurance. http://www.fda.gov/cder/OPS/PAT.htm 969/. Aaccessed 01 February 2020.
7. Abu-Absi NR, Kenty BM, Cuellar ME, Borys MC, Sakhamuri S, Strachan DJ, Hausladen MC, Li ZJ. Real time monitoring of multiple parameters in mammalian cell culture bioreactors using an in-line Raman spectroscopy probe. Biotechnol Bioeng. 2011;108(5):1215-1221. doi:10.1002/bit.23023.
8. Chavez B, Brorson K, Agarabi C, Khan M. Real time monitoring of bioreactor mAb IgG3 cell culture process dynamics via Fourier transform infrared spectroscopy: Implications for enabling cell culture process analytical technology. Front Chem Sci Eng. 2015;9(3):386-406. doi:10.1007/s11705-015-1533-3.
9. Wasalathanthri DP, Tewari JC, Kang X, Hincapie M, Barrett S. Multivariate Spectral Analysis and Monitoring for Biomanufacturing. Patent No. US 2019 / 0272894 A1. United States. https://patents.google.com/patent/US20190272894A1/en. Accessed 01 February 2020.
10. Großhans S, Rüdt M, Sanden A, Brestrich N, Morgenstern J, Heissler S, Hubbuch J. In-line Fourier-transform infrared spectroscopy as a versatile process analytical technology for preparative protein chromatography. J Chromatogr A. 2018;1547:37-44. doi:10.1016/j. chroma.2018.03.005.
11. Patel BA, Gospodarek A, Larkin M, Kenrick SA, Haverick MA, Tugcu N, Brower MA, Richardson DD. Multi-angle light scattering as a process analytical technology measuring real-time molecular weight for downstream process control. MAbs. 2018;10(7):945-950. doi:10.1080/19420862.2018.1505178.
12. Patel BA, Pinto NDS, Gospodarek A, Kilgore B, Goswami K, Napoli WN, Desai J, Heo JH, Panzera D, Pollard D, Richardson D. Brower M, Richardson, DD. On-Line Ion Exchange Liquid Chromatography as a Process Analytical Technology for Monoclonal Antibody Characterization in Continuous Bioprocessing. Anal Chem. 2017;89(21):11357-11365. doi:10.1021/acs.analchem.7b02228.
13. Dong J, Migliore N, Mehrman SJ, Cunningham J, Lewis MJ, Hu P. High-Throughput, Automated Protein A Purification Platform with Multiattribute LC-MS Analysis for Advanced Cell Culture Process Monitoring. Anal Chem. 2016;88(17):8673-8679. doi:10.1021/acs. analchem.6b01956.
14. Rogstad S, Yan H, Wang X, Powers D, Brorson K, Damdinsuren B, Lee S. Multi-Attribute Method for Quality Control of Therapeutic Proteins. Anal Chem. 2019. doi:10.1021/acs. analchem.9b03808.
15. Barenji RV, Akdag Y, Yet B, Oner L. Cyber-physical-based PAT (CPbPAT) framework for Pharma 4.0. Int J Pharm. 2019;567(June):118445. doi:10.1016/j.ijpharm.2019.06.036.
16. Parker K, 2013. Enterprise manufacturing intelligence software supports processing. https://www.processingmagazine.com/maintenance-safety/asset-management/article/15580663/enterprise-manufacturing-intelligence-software-supports-processing. Accessed 01 February 2020.
17. Bayrak ES, Wang T, Tulsyan A, Coufal M, Undey C. Product Attribute Forecast: Adaptive Model Selection Using Real-Time Machine Learning. IFAC-PapersOnLine. 2018;51(18):121-125. doi:10.1016/j.ifacol.2018.09.286.
18. Rathore AS, Bhambure R, Ghare V. Process analytical technology (PAT) for biopharmaceutical products. Anal Bioanal Chem. 2010;398(1):137-154. doi:10.1007/s00216-010-3781-x.
19. Sokolov M, Ritscher J, MacKinnon N, Souquet J. Broly H. Morbidelli M. Butté A. Enhanced process understanding and multivariate prediction of the relationship between cell culture process and monoclonal antibody quality. Biotechnol Prog. 2017;33(5):1368-1380. doi:10.1002/btpr.2502.
20. Romero-Torres S, Wolfram K, Armando J, Armando J, Ahmed SK, Ren J, Shi C, Hill D, Guenard R. Biopharmaceutical process model evolution-enabling process knowledge continuum from an advanced process control perspective. Am Pharm Rev. 2018;21(4):1-10.

Author Biographies

Dhanuka Wasalathanthri, PhD is a Senior Scientist at Bristol Myers Squibb where he leads the Process Analytical Technology (PAT) initiatives and strategies at Biologic development in Devens, MA. His broad spectrum of expertise in the PAT space features the utility of multi-dimensional chromatography, spectroscopy and chemometrics, microfluidic sensors, and mass spectrometry for real time monitoring of product quality attributes and process parameters. He represents several academic, and industry consortia. Dhanuka received his Ph.D. in Analytical Chemistry from University of Connecticut in 2014, and his work is published in peer-reviewed scientific journals.

Julia Ding, PhD is a Director of Analytical Development at Bristol Myers Squibb (BMS) where she is leading the BMS global process analytical network. Julia Ding plays an important role in analytical strategy advancement at BMS where she serves an active role in BMS biologic specification committee and biologic comparability council. Prior to this, Julia Ding was a Director of Commercial Method Development and Process Analytical Development for late phase programs at BMS. Julia Ding was also a manager leading a multifunctional analytical team at PPD before joining BMS in 2016. Julia Ding obtained her Ph.D. in physical organic chemistry from Emory University and postdoc research from University of California at Berkeley.

Zheng Jian Li, PhD is an Executive Director at Bristol Myers Squibb (BMS) where he is leading the Biologics Analytical Development and Attribute Sciences. Before this role, he was the Executive Director at BMS leading the Late Stage Biologics Development. He has extensive experience in manufacturing process development for biological molecules including various modalities and expression systems. His teams have extensive publications in various well known journals, which covers all aspects of issues related to biologics commercialization. Dr. Li received his Ph.D. in Chemical and Biochemical Engineering from University of Maryland Baltimore County in 2000.