Elsevier

Computers & Chemical Engineering

Volume 110, 2 February 2018, Pages 106-114
Computers & Chemical Engineering

Review
Challenges and opportunities in biopharmaceutical manufacturing control

https://doi.org/10.1016/j.compchemeng.2017.12.007Get rights and content

Highlights

  • A perspective is provided on the control of biopharmaceutical manufacturing.

  • Microscale and modular plug-and-play unit operations are technologies of interest.

  • The uncertain nonlinear systems involve mixed continuous-discrete operations.

  • Process data analytics and adaptive and hybrid control strategies are discussed.

  • Novel protein separation methods are discussed that may significantly reduce costs.

Abstract

This article provides a perspective on control and operations for biopharmaceutical manufacturing. Challenges and opportunities are described for (1) microscale technologies for high-speed continuous processing, (2) plug-and-play modular unit operations with integrated monitoring and control systems, (3) dynamic modeling of unit operations and entire biopharmaceutical manufacturing plants to support process development and plant-wide control, and (4) model-based control technologies for optimizing startup, changeover, and shutdown. A challenge is the ability to simultaneously address the uncertainties, nonlinearities, time delays, non-minimum phase behavior, constraints, spatial distributions, and mixed continuous-discrete operations that arise in biopharmaceutical operations. The design of adaptive and hybrid control strategies is discussed. Process data analytics and grey-box modeling methods are needed to deal with the heterogeneity and tensorial dimensionality of biopharmaceutical data. Novel bioseparations as discussed as a potential cost-effective unit operation, with a discussion of challenges for the widespread application of crystallization to therapeutic proteins.

Introduction

Biopharmaceuticals, which are also widely known as biologics or biologic drugs, are products derived from biological organisms for treating or preventing diseases. The global sales of biopharmaceuticals, which have continually increased for many years, were ∼$300 billion in 2014, and are projected to reach ∼$450 billion by 2019 (Deloitte, 2016). Over 30% of the drugs in the drug pipeline are biopharmaceuticals (Informa, 2016), with hundreds of approved products on the market and over 7000 medicines in development (PhRMA, 2016). The rate of biopharmaceuticals approval has remained relatively steady, with monoclonal antibodies (mAbs) accounting for an increasing proportion of the approvals (Walsh, 2014).

Monoclonal antibodies are the highest selling class of biopharmaceuticals with ∼1500 drugs in the product pipeline in 2016 (Informa, 2016). This class is of particular interest due to their specific action and reduced immunogenicity. With continued growth in sales of existing mAb products and a growing pipeline of mAb product candidates being developed, the total sales of mAb products and all biopharmaceuticals will continue to increase in the coming years (Ecker et al., 2015). The development of mAbs is expected to grow further as more diseases are understood at molecular and cellular levels.

Traditional biopharmaceutical manufacturing consists of a similar sequence of unit operations that are divided into two main parts: upstream and downstream. The upstream unit operations typically include cell culture and harvest steps, and the downstream consists of purification with multiple steps of chromatography, filtration, and diafiltration. For example, Fig. 1 shows a process flow diagram for a typical platform used for the production of mAbs (see Kelley, 2009, Shukla and Thömmes, 2010; and citations therein for the figure and more details than the summary provided here). The upstream process starts with cell culture, which is fed inoculum prepared and expanded from a cell bank to a series of batch bioreactors of successfully larger volume for expansion of the cells and finally to the production bioreactor for protein expression. Then cells and cell debris are removed by centrifugation followed by depth and membrane filtration.

The downstream process for mAbs begins with capturing mAb by protein A affinity chromatography. The protein A affinity chromatography captures mAb by specific binding interactions between mAb and protein A ligand. Because protein A ligand has high binding affinity and specificity to mAbs, protein A affinity chromatography provides >98% purity in a single step. The products bind to the stationary phase while impurities, such as host cell proteins and DNA, pass through at neutral pH. The products are then eluted from the adsorbents at low pH, which inactivates viruses. Next, two polishing chromatographic steps are typically used for further removal of impurities. Protein A affinity chromatography is unable to remove aggregates and product variants due to their chemical similarity with the derived protein, and introduces leached protein A as a new impurity. The most commonly used steps are cation exchange (CEX) chromatography and anion exchange (AEX) chromatography. CEX chromatography uses resin with negatively charged groups to bind the products during the loading step, then elutes the products by increasing pH or conductivity. AEX chromatography uses resin with positively charged groups and typically run in flow-through mode due to the high pI of mAbs, which is often >8. The operating conditions are chosen to allow the products to flow through while the impurities bind to the resin. A viral filtration step is then used to ensure viral safety by a sized-based virus removal. A subsequent step of ultrafiltration/diafiltration (UF/DF) formulates and concentrates the product for last step of the process.

Process control engineers have an important role to play in biopharmaceutical manufacturing, and this article describes associated opportunities, needs, and challenges in process control and operations. The next section introduces three trends in biopharmaceutical manufacturing that are described in more detail in subsequent sections. A section on process data analytics discusses challenges in dealing with high-dimensional and heterogeneous data, and the potential for “grey-box” models that supplement first-principles models with data-based models. Another section discusses a recent trend towards continuous-flow operations, and how this development opens many opportunities for mathematical modeling and process simulation and model-based design, control, and optimization. A section on novel bioseparations primarily discusses crystallization as a non-chromatographic method for the purification of proteins from a large number of other components in solution. The challenges of crystallizing mAbs and other large-molecule therapeutic proteins are described along with potential control approaches for biopharmaceutical crystallization. The article ends with a summary and some closing comments.

Section snippets

Trends in biopharmaceutical manufacturing

The continued growth of biologics motivates developing a deeper understanding and advancement of biopharmaceutical manufacturing operations. This growth has increased interest in the application of process analytical technology (PAT), which is “a system for designing, analyzing, and controlling manufacturing through timely measurements (i.e., during processing) of critical quality and performance attributes of raw and in-process materials and processes, with the goal of ensuring final product

Using high-dimensional and heterogeneous data

Biopharmaceutical manufacturing data are often heterogeneous in both time scale and data type (Charaniya et al., 2008). For example, some measurements in a bioreactor can be performed continuously on-line such as dissolved oxygen, optical density, and pH. Other measurements are performed off-line, such as the distribution of oligosaccharides (glycans) attached to the proteins, usually at non-periodic, asynchronous intervals. Many process monitoring and control algorithms such as differential

Continuous operations

Classically, chemically and biologically derived pharmaceuticals were manufactured in a series of batch processes. Increased attention has been towards transitioning to continuous operations, where “continuous” can refer to (1) a unit operation with the capability of operating under continuous flow and minimal holdup volume, or (2) a manufacturing plant with integrated continuous-flow unit operations with minimal hold volume in between (Konstantinov and Cooney, 2015; and citations therein).

Novel bioseparations

The currently dominant method for bioseparation is packed-bed chromatography, due to its high resolution. Chromatography for high-dose biopharmaceuticals is expensive, however, even when using operational improvements such as periodic countercurrent continuous chromatography (Bryntesson et al., 2011). When scaling manufacturing processes from bench-scale to production, it is generally desirable for the operating cost to scale sub-linearly with the material throughput. The method of purification

Conclusion

This article describes opportunities and challenges for the manufacturing of biopharmaceuticals from the perspective of chemical process control. One of the trends in the field is towards increased PAT, especially with the goal towards online sensor technologies. Characteristics of biopharmaceutical data such as heterogeneity in time scale and data type and tensorial dimensionality provide challenges for developing effective process data analytics methods. Existing proposed methods for

Acknowledgments

This research was developed with funding from the Defense Advanced Research Projects Agency (DARPA); U.S. Army Medical Research and Material Command (MRMC); and the Army Research Office (ARO) [grant number N66001-13-C-4025]. The views, opinions, and/or findings contained in this article are those of the authors and should not be interpreted as representing the official views or policies of the Department of Defense or the U.S. Government.

References (76)

  • M. Jiang et al.

    Effect of jet velocity on crystal size distribution from antisolvent and cooling crystallizations in a dual impinging jet mixer

    Chem. Eng. Process.

    (2015)
  • A. Jungbauer

    Continuous downstream processing of biopharmaceuticals

    Trends Biotechnol.

    (2013)
  • K.B Konstantinov et al.

    White paper on continuous bioprocessing may 20-21, 2014 continuous manufacturing symposium.

    J. Pharm. Sci.

    (2015)
  • H. Le et al.

    Multivariate analysis of cell culture bioprocess data- lactate consumption as process indicator

    J. Biotechnol.

    (2012)
  • R. Luttmann et al.

    Sequential/parallel production of potential malaria vaccines – a direct way from single batch to quasi-continuous integrated production

    J. Biotechnol.

    (2015)
  • T. McGlone et al.

    Oscillatory flow reactors (OFRs) for continuous manufacturing and crystallization

    Org. Process Res. Dev.

    (2015)
  • A.S. Myerson et al.

    Control systems engineering in continuous pharmaceutical manufacturing

    J. Pharm. Sci.

    (2015)
  • Z.K. Nagy et al.

    Distributional uncertainty analysis using power series and polynomial chaos expansions

    J. Process Control

    (2007)
  • R.K Pearson et al.

    Gray-box identification of block-oriented nonlinear models

    J. Process Control

    (2000)
  • M.L. Rasche et al.

    Mathematical modeling and optimal design of multi-stage slug-flow crystallization

    Comput. Chem. Eng.

    (2016)
  • A.S. Rathore

    Roadmap for Implementation of quality by design (QbD) for biotechnology products

    Trends Biotechnol.

    (2009)
  • P.A.J. Rosa et al.

    Aqueous two-phase systems: a viable platform in the manufacturing of biopharmaceuticals

    J. Chromatogr. A

    (2010)
  • K. Severson et al.

    Elastic net with monte carlo sampling for data-based modeling in biopharmaceutical manufacturing facilities

    Comput. Chem. Eng.

    (2015)
  • A.A. Shukla et al.

    Recent advances in large-scale production of monoclonal antibodies and related proteins

    Trends Biotechnol.

    (2010)
  • S. Streif et al.

    Robustness analysis, prediction, and estimation for uncertain biochemical networks: an overview

    J. Process Control

    (2016)
  • A.P. Teixeira et al.

    Advances in on-line monitoring and control of mammalian cell cultures: supporting the PAT initiative

    Biotechnol. Adv.

    (2009)
  • C. Ündey et al.

    Monitoring of batch pharmaceutical fermentations: data synchronization, landmark alignment, and real-time monitoring

  • P. Wechselberger et al.

    PAT method to gather bioprocess parameters in real-time using simple input variables and first principle relationships

    Chem. Eng. Sci.

    (2010)
  • S. Wold et al.

    The chemometric analysis of point and dynamic data in pharmaceutical and biotech production (PAT) – some objectives and approaches

    Chemom. Intell. Lab. Syst.

    (2006)
  • Baker, J.C., Roberts, B.M. (1997). Preparation of Stable Insulin Analog Crystals. U.S. Patent...
  • Bryntesson, M., Hall, M., Lacki, K.M. (2011). Chromatography Method. U.S. Patent...
  • M.S. Croughan et al.

    The future of industrial bioprocessing: batch or continuous

    Biotechnol. Bioeng.

    (2015)
  • I.J. del Val et al.

    A dynamic mathematical model for monoclonal antibody N-linked glycosylation and nucleotide sugar donor transport within a maturing golgi apparatus

    Biotechnol. Prog.

    (2011)
  • 2016 Global Life Sciences Outlook: Moving Forward with Cautious Optimism

    (2016)
  • D.M. Ecker et al.

    The therapeutic monoclonal antibody market

    MAbs

    (2015)
  • J.K. Eggersgluess et al.

    Multi-stage aqueous two-phase extraction for the purification of monoclonal antibodies

    Chem. Eng. Technol.

    (2014)
  • Guidance for Industry: PAT – A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance

    (2004)
  • Guidance for Industry: Q8(R2) Pharmaceutical Development

    (2009)
  • Cited by (81)

    View all citing articles on Scopus
    View full text