Development of temperature-controlled batch and 3-column counter-current protein a system for improved therapeutic purification.

Maximizing product quality attributes by optimizing process parameters and performance attributes is a crucial aspect of bioprocess chromatography process design. Process parameters include but are not limited to bed height, eluate cut points, and elution pH. An under-characterized chromatography process parameter for protein A chromatography is process temperature. Here, we present a mechanistic understanding of the effects of temperature on the protein A purification of a monoclonal antibody (mAb) using a commercial chromatography resin for batch and continuous counter-current systems. A self-designed 3D-printed heating jacket controlled the 1 mL chromatography process temperature during the loading, wash, elution, and cleaning-in-place (CIP) steps. Batch loading experiments at 10, 20, and 30 °C demonstrated increased dynamic binding capacity (DBC) with temperature. The experimental data were fit to mechanistic and correlation-based models that predicted the optimal operating conditions over a range of temperatures. These model-based predictions optimized the development of a 3-column temperature-controlled periodic counter-current chromatography (TCPCC) and were validated experimentally. Operating a 3-column TCPCC at 30 °C led to a 47% increase in DBC relative to 20 °C batch chromatography. The DBC increase resulted in a two-fold increase in productivity relative to 20 °C batch. Increasing the number of columns to the TCPCC to optimize for increasing feed concentration resulted in further improvements to productivity. The feed-optimized TCPCC showed a respective two, three, and four-fold increase in productivity at feed concentrations of 1, 5, and 15 mg/mL mAb, respectively. The derived and experimentally validated temperature-dependent models offer a valuable tool for optimizing both batch and continuous chromatography systems under various operating conditions.

Structural and practical identifiability analysis in bioengineering: a beginner's guide.

Advancements in digital technology have brought modelling to the forefront in many disciplines from healthcare to architecture. Mathematical models, often represented using parametrised sets of ordinary differential equations, can be used to characterise different processes. To infer possible estimates for the unknown parameters, these models are usually calibrated using associated experimental data. Structural and practical identifiability analyses are a key component that should be assessed prior to parameter estimation. This is because identifiability analyses can provide insights as to whether or not a parameter can take on single, multiple, or even infinitely or countably many values which will ultimately have an impact on the reliability of the parameter estimates. Also, identifiability analyses can help to determine whether the data collected are sufficient or of good enough quality to truly estimate the parameters or if more data or even reparameterization of the model is necessary to proceed with the parameter estimation process. Thus, such analyses also provide an important role in terms of model design (structural identifiability analysis) and the collection of experimental data (practical identifiability analysis). Despite the popularity of using data to estimate the values of unknown parameters, structural and practical identifiability analyses of these models are often overlooked. Possible reasons for non-consideration of application of such analyses may be lack of awareness, accessibility, and usability issues, especially for more complicated models and methods of analysis. The aim of this study is to introduce and perform both structural and practical identifiability analyses in an accessible and informative manner via application to well established and commonly accepted bioengineering models. This will help to improve awareness of the importance of this stage of the modelling process and provide bioengineering researchers with an understanding of how to utilise the insights gained from such analyses in future model development.

Smart Sensor Control and Monitoring of an Automated Cell Expansion Process.

Immune therapy for cancer patients is a new and promising area that in the future may complement traditional chemotherapy. The cell expansion phase is a critical part of the process chain to produce a large number of high-quality, genetically modified immune cells from an initial sample from the patient. Smart sensors augment the ability of the control and monitoring system of the process to react in real-time to key control parameter variations, adapt to different patient profiles, and optimize the process. The aim of the current work is to develop and calibrate smart sensors for their deployment in a real bioreactor platform, with adaptive control and monitoring for diverse patient/donor cell profiles. A set of contrasting smart sensors has been implemented and tested on automated cell expansion batch runs, which incorporate advanced data-driven machine learning and statistical techniques to detect variations and disturbances of the key system features. Furthermore, a 'consensus' approach is applied to the six smart sensor alerts as a confidence factor which helps the human operator identify significant events that require attention. Initial results show that the smart sensors can effectively model and track the data generated by the Aglaris FACER bioreactor, anticipate events within a 30 min time window, and mitigate perturbations in order to optimize the key performance indicators of cell quantity and quality. In quantitative terms for event detection, the consensus for sensors across batch runs demonstrated good stability: the AI-based smart sensors (Fuzzy and Weighted Aggregation) gave 88% and 86% consensus, respectively, whereas the statistically based (Stability Detector and Bollinger) gave 25% and 42% consensus, respectively, the average consensus for all six being 65%. The different results reflect the different theoretical approaches. Finally, the consensus of batch runs across sensors gave even higher stability, ranging from 57% to 98% with an average consensus of 80%.

Enhanced harvest performance predictability through advanced multivariate data analysis of mammalian cell culture particle size distribution.

The industry's pursuit for higher antibody production has led to increased cell density cultures that impact the performance of subsequent product recovery steps. This increase in cell concentration has highlighted the critical role of solids concentration in centrifugation yield, while recent product degradation cases have shed light on the impact of cell lysis on product quality. Current methods for measuring solids concentration and cell lysis are not suited for early-stage high-throughput experimentation, which means that these cell culture outputs are not well characterized in early process development. This article describes a novel approach that leveraged the data from a widely-used automated cell counter (Vi-CELL™ XR) to accurately predict solids concentration and a common cell lysis indicator represented as lactate dehydrogenase (LDH) release. For this purpose, partial least squares (PLS) models were derived with k-fold cross-validation from the particle size distribution data generated by the cell counter. The PLS models showed good predictive potential for both LDH release and solids concentration. This novel approach reduced the time required for evaluating the solids concentration and LDH for a typical high-throughput cell culture system (with 48 bioreactors in parallel) from around 7 h down to a few minutes.

Next-generation cell line selection methodology leveraging data lakes, natural language generation and advanced data analytics.

Cell line development is an essential stage in biopharmaceutical development that often lies on the critical path. Failure to fully characterise the lead clone during initial screening can lead to lengthy project delays during scale-up, which can potentially compromise commercial manufacturing success. In this study, we propose a novel cell line development methodology, referenced as CLD 4, which involves four steps enabling autonomous data-driven selection of the lead clone. The first step involves the digitalisation of the process and storage of all available information within a structured data lake. The second step calculates a new metric referenced as the cell line manufacturability index (MI CL) quantifying the performance of each clone by considering the selection criteria relevant to productivity, growth and product quality. The third step implements machine learning (ML) to identify any potential risks associated with process operation and relevant critical quality attributes (CQAs). The final step of CLD 4 takes into account the available metadata and summaries all relevant statistics generated in steps 1-3 in an automated report utilising a natural language generation (NLG) algorithm. The CLD 4 methodology was implemented to select the lead clone of a recombinant Chinese hamster ovary (CHO) cell line producing high levels of an antibody-peptide fusion with a known product quality issue related to end-point trisulfide bond (TSB) concentration. CLD 4 identified sub-optimal process conditions leading to increased levels of trisulfide bond that would not be identified through conventional cell line development methodologies. CLD 4 embodies the core principles of Industry 4.0 and demonstrates the benefits of increased digitalisation, data lake integration, predictive analytics and autonomous report generation to enable more informed decision making.

Directed evolution for soluble and active periplasmic expression of bovine enterokinase in Escherichia coli.

Bovine enterokinase light chain (EKL) is an industrially useful protease for accurate removal of affinity-purification tags from high-value biopharmaceuticals. However, recombinant expression in Escherichia coli produces insoluble inclusion bodies, requiring solubilisation, refolding, and autocatalytic activation to recover functional enzyme. Error-prone PCR and DNA shuffling of the EKL gene, T7 promoter, lac operon, ribosome binding site, and pelB leader sequence, yielded 321 unique variants after screening ~ 6500 colonies. The best variants had > 11,000-fold increased total activity in lysates, producing soluble enzyme that no longer needed refolding. Further characterisation identified the factors that improved total activity from an inactive and insoluble starting point. Stability was a major factor, whereby melting temperatures > 48.4 °C enabled good expression at 37 °C. Variants generally did not alter catalytic efficiency as measured by kcat/Km, which improved for only one variant. Codon optimisation improved the total activity in lysates produced at 37 °C. However, non-optimised codons and expression at 30 °C gave the highest activity through improved protein quality, with increased kcat and Tm values. The 321 variants were statistically analysed and mapped to protein structure. Mutations detrimental to total activity and stability clustered around the active site. By contrast, variants with increased total activity tended to combine stabilising mutations that did not disrupt the active site.

Data integrity within the biopharmaceutical sector in the era of Industry 4.0.

Data Integrity (DI) in the highly regulated biopharmaceutical sector is of paramount importance to ensure decisions on meeting product specifications are accurate and hence assure patient safety and product quality. The challenge of ensuring DI within this sector is becoming more complex with the growing amount of data generated given increasing adoption of process analytical technology (PAT), advanced automation, high throughput microscale studies, and managing data models created by machine learning (ML) tools. This paper aims to identify DI risks and mitigation strategies in biopharmaceutical manufacturing facilities as the sector moves towards Industry 4.0. To achieve this, the paper examines common DI violations and links them to the ALCOA+ principles used across the FDA, EMA, and MHRA. The relevant DI guidelines from the ISPE's GAMP5 and ISA-95 standards are also discussed with a focus on the role of validated computerised and automated manufacturing systems to avoid DI risks and generate compliant data. The paper also highlights the importance of DI whilst using data analytics to ensure the developed models meet the required regulatory standards for process monitoring and control. This includes a discussion on possible mitigation strategies and methodologies to ensure data integrity is maintained for smart manufacturing operations such as the use of cloud platforms to facilitate the storage and transfer of manufacturing data, and migrate away from paper-based records.

A decade in review: use of data analytics within the biopharmaceutical sector.

There are large amounts of data generated within the biopharmaceutical sector. Traditionally, data analysis methods labelled as multivariate data analysis have been the standard statistical technique applied to interrogate these complex data sets. However, more recently there has been a surge in the utilisation of a broader set of machine learning algorithms to further exploit these data. In this article, the adoption of data analysis techniques within the biopharmaceutical sector is evaluated through a review of journal articles and patents published within the last ten years. The papers objectives are to identify the most dominant algorithms applied across different applications areas within the biopharmaceutical sector and to explore whether there is a trend between the size of the data set and the algorithm adopted.

Advanced control strategies for bioprocess chromatography: Challenges and opportunities for intensified processes and next generation products.

Recent advances in process analytical technologies and modelling techniques present opportunities to improve industrial chromatography control strategies to enhance process robustness, increase productivity and move towards real-time release testing. This paper provides a critical overview of batch and continuous industrial chromatography control systems for therapeutic protein purification. Firstly, the limitations of conventional industrial fractionation control strategies using in-line UV spectroscopy and on-line HPLC are outlined. Following this, an evaluation of monitoring and control techniques showing promise within research, process development and manufacturing is provided. These novel control strategies combine rapid in-line data capture (e.g. NIR, MALS and variable pathlength UV) with enhanced process understanding obtained from mechanistic and empirical modelling techniques. Finally, a summary of the future states of industrial chromatography control systems is proposed, including strategies to control buffer formulation, product fractionation, column switching and column fouling. The implementation of these control systems improves process capabilities to fulfil product quality criteria as processes are scaled, transferred and operated, thus fast tracking the delivery of new medicines to market.

A rational approach to improving titer in Escherichia coli-based cell-free protein synthesis reactions.

Cell-free protein synthesis (CFPS) is an established method for rapid recombinant protein production. Advantages like short synthesis times and an open reaction environment make CFPS a desirable platform for new and difficult-to-express products. Most recently, interest has grown in using the technology to make larger amounts of material. This has been driven through a variety of reasons from making site specific antibody drug conjugates, to emergency response, to the safe manufacture of toxic biological products. We therefore need robust methods to determine the appropriate reaction conditions for product expression in CFPS. Here we propose a process development strategy for Escherichia coli lysate-based CFPS reactions that can be completed in as little as 48 hr. We observed the most dramatic increases in titer were due to the E. coli strain for the cell extract. Therefore, we recommend identifying a high-producing cell extract for the product of interest as a first step. Next, we manipulated the plasmid concentration, amount of extract, temperature, concentrated reaction mix pH levels, and length of reaction. The influence of these process parameters on titer was evaluated through multivariate data analysis. The process parameters with the highest impact on titer were subsequently included in a design of experiments to determine the conditions that increased titer the most in the design space. This proposed process development strategy resulted in superfolder green fluorescent protein titers of 0.686 g/L, a 38% improvement on the standard operating conditions, and hepatitis B core antigen titers of 0.386 g/L, a 190% improvement.

Multivariate Data Analysis Methodology to Solve Data Challenges Related to Scale-Up Model Validation and Missing Data on a Micro-Bioreactor System.

Multivariate data analysis (MVDA) is a highly valuable and significantly underutilized resource in biomanufacturing. It offers the opportunity to enhance understanding and leverage useful information from complex high-dimensional data sets, recorded throughout all stages of therapeutic drug manufacture. To help standardize the application and promote this resource within the biopharmaceutical industry, this paper outlines a novel MVDA methodology describing the necessary steps for efficient and effective data analysis. The MVDA methodology is followed to solve two case studies: a "small data" and a "big data" challenge. In the "small data" example, a large-scale data set is compared to data from a scale-down model. This methodology enables a new quantitative metric for equivalence to be established by combining a two one-sided test with principal component analysis. In the "big data" example, this methodology enables accurate predictions of critical missing data essential to a cloning study performed in the ambr15 system. These predictions are generated by exploiting the underlying relationship between the off-line missing values and the on-line measurements through the generation of a partial least squares model. In summary, the proposed MVDA methodology highlights the importance of data pre-processing, restructuring, and visualization during data analytics to solve complex biopharmaceutical challenges.

High throughput process development workflow with advanced decision-support for antibody purification.

Chromatography remains the workhorse in antibody purification; however process development and characterisation still require significant resources. The high number of operating parameters involved requires extensive experimentation, traditionally performed at small- and pilot-scale, leading to demands in terms of materials and time that can be a challenge. The main objective of this research was the establishment of a novel High Throughput Process Development (HTPD) workflow combining scale-down chromatography experimentation with advanced decision-support techniques in order to minimise the consumption of resources and accelerate the development timeframe. Additionally, the HTPD workflow provides a framework to rapidly manipulate large datasets in an automated fashion. The central component of the HTPD workflow is the systematic integration of a microscale chromatography experimentation strategy with an advanced chromatogram evaluation method, design of experiments (DoE) and multivariate data analysis. The outputs of this are leveraged into the screening and optimisation components of the workflow. For the screening component, a decision-support tool was developed combining different multi-criteria decision-making techniques to enable a fair comparison of a number of CEX resin candidates and determine those that demonstrate superior purification performance. This provided a rational methodology for screening chromatography resins and process parameters. For the optimisation component, the workflow leverages insights provided through screening experimentation to guide subsequent DoE experiments so as to tune significant process parameters for the selected resin. The resulting empirical correlations are linked to a stochastic modelling technique so as to predict the optimal and most robust chromatographic process parameters to achieve the desired performance criteria.

Influence of Incident Wavelength and Detector Material Selection on Fluorescence in the Application of Raman Spectroscopy to a Fungal Fermentation Process.

Raman spectroscopy is a novel tool used in the on-line monitoring and control of bioprocesses, offering both quantitative and qualitative determination of key process variables through spectroscopic analysis. However, the wide-spread application of Raman spectroscopy analysers to industrial fermentation processes has been hindered by problems related to the high background fluorescence signal associated with the analysis of biological samples. To address this issue, we investigated the influence of fluorescence on the spectra collected from two Raman spectroscopic devices with different wavelengths and detectors in the analysis of the critical process parameters (CPPs) and critical quality attributes (CQAs) of a fungal fermentation process. The spectra collected using a Raman analyser with the shorter wavelength (903 nm) and a charged coupled device detector (CCD) was corrupted by high fluorescence and was therefore unusable in the prediction of these CPPs and CQAs. In contrast, the spectra collected using a Raman analyser with the longer wavelength (993 nm) and an indium gallium arsenide (InGaAs) detector was only moderately affected by fluorescence and enabled the generation of accurate estimates of the fermentation's critical variables. This novel work is the first direct comparison of two different Raman spectroscopy probes on the same process highlighting the significant detrimental effect caused by high fluorescence on spectra recorded throughout fermentation runs. Furthermore, this paper demonstrates the importance of correctly selecting both the incident wavelength and detector material type of the Raman spectroscopy devices to ensure corrupting fluorescence is minimised during bioprocess monitoring applications.

On-Line Control of Glucose Concentration in High-Yielding Mammalian Cell Cultures Enabled Through Oxygen Transfer Rate Measurements.

Glucose control is vital to ensure consistent growth and protein production in mammalian cell cultures. The typical fed-batch glucose control strategy involving bolus glucose additions based on infrequent off-line daily samples results in cells experiencing significant glucose concentration fluctuations that can influence product quality and growth. This study proposes an on-line method to control and manipulate glucose utilizing readily available process measurements. The method generates a correlation between the cumulative oxygen transfer rate and the cumulative glucose consumed. This correlation generates an on-line prediction of glucose that has been successfully incorporated into a control algorithm manipulating the glucose feed-rate. This advanced process control (APC) strategy enables the glucose concentration to be maintained at an adjustable set-point and has been found to significantly reduce the deviation in glucose concentration in comparison to conventional operation. This method has been validated to produce various therapeutic proteins across cell lines with different glucose consumption demands and is successfully demonstrated on micro (15 mL), laboratory (7 L), and pilot (50 L) scale systems. This novel APC strategy is simple to implement and offers the potential to significantly enhance the glucose control strategy for scales spanning micro-scale systems through to full scale industrial bioreactors.

Advanced multivariate data analysis to determine the root cause of trisulfide bond formation in a novel antibody-peptide fusion.

Product quality heterogeneities, such as a trisulfide bond (TSB) formation, can be influenced by multiple interacting process parameters. Identifying their root cause is a major challenge in biopharmaceutical production. To address this issue, this paper describes the novel application of advanced multivariate data analysis (MVDA) techniques to identify the process parameters influencing TSB formation in a novel recombinant antibody-peptide fusion expressed in mammalian cell culture. The screening dataset was generated with a high-throughput (HT) micro-bioreactor system (AmbrTM 15) using a design of experiments (DoE) approach. The complex dataset was firstly analyzed through the development of a multiple linear regression model focusing solely on the DoE inputs and identified the temperature, pH and initial nutrient feed day as important process parameters influencing this quality attribute. To further scrutinize the dataset, a partial least squares model was subsequently built incorporating both on-line and off-line process parameters and enabled accurate predictions of the TSB concentration at harvest. Process parameters identified by the models to promote and suppress TSB formation were implemented on five 7 L bioreactors and the resultant TSB concentrations were comparable to the model predictions. This study demonstrates the ability of MVDA to enable predictions of the key performance drivers influencing TSB formation that are valid also upon scale-up. Biotechnol. Bioeng. 2017;114: 2222-2234. © 2017 The Authors. Biotechnology and Bioengineering Published by Wiley Periodicals, Inc.

An automated laboratory-scale methodology for the generation of sheared mammalian cell culture samples.

Continuous disk-stack centrifugation is typically used for the removal of cells and cellular debris from mammalian cell culture broths at manufacturing-scale. The use of scale-down methods to characterise disk-stack centrifugation performance enables substantial reductions in material requirements and allows a much wider design space to be tested than is currently possible at pilot-scale. The process of scaling down centrifugation has historically been challenging due to the difficulties in mimicking the Energy Dissipation Rates (EDRs) in typical machines. This paper describes an alternative and easy-to-assemble automated capillary-based methodology to generate levels of EDRs consistent with those found in a continuous disk-stack centrifuge. Variations in EDR were achieved through changes in capillary internal diameter and the flow rate of operation through the capillary. The EDRs found to match the levels of shear in the feed zone of a pilot-scale centrifuge using the experimental method developed in this paper (2.4×105 W/Kg) are consistent with those obtained through previously published computational fluid dynamic (CFD) studies (2.0×105 W/Kg). Furthermore, this methodology can be incorporated into existing scale-down methods to model the process performance of continuous disk-stack centrifuges. This was demonstrated through the characterisation of culture hold time, culture temperature and EDRs on centrate quality.

A scale-down mimic for mapping the process performance of centrifugation, depth and sterile filtration.

In the production of biopharmaceuticals disk-stack centrifugation is widely used as a harvest step for the removal of cells and cellular debris. Depth filters followed by sterile filters are often then employed to remove residual solids remaining in the centrate. Process development of centrifugation is usually conducted at pilot-scale so as to mimic the commercial scale equipment but this method requires large quantities of cell culture and significant levels of effort for successful characterization. A scale-down approach based upon the use of a shear device and a bench-top centrifuge has been extended in this work towards a preparative methodology that successfully predicts the performance of the continuous centrifuge and polishing filters. The use of this methodology allows the effects of cell culture conditions and large-scale centrifugal process parameters on subsequent filtration performance to be assessed at an early stage of process development where material availability is limited. Biotechnol. Bioeng. 2016;113: 1934-1941. © 2016 The Authors. Biotechnology and Bioengineering Published by Wiley Periodicals, Inc.

The development of an industrial-scale fed-batch fermentation simulation.

This paper describes a simulation of an industrial-scale fed-batch fermentation that can be used as a benchmark in process systems analysis and control studies. The simulation was developed using a mechanistic model and validated using historical data collected from an industrial-scale penicillin fermentation process. Each batch was carried out in a 100,000 L bioreactor that used an industrial strain of Penicillium chrysogenum. The manipulated variables recorded during each batch were used as inputs to the simulator and the predicted outputs were then compared with the on-line and off-line measurements recorded in the real process. The simulator adapted a previously published structured model to describe the penicillin fermentation and extended it to include the main environmental effects of dissolved oxygen, viscosity, temperature, pH and dissolved carbon dioxide. In addition the effects of nitrogen and phenylacetic acid concentrations on the biomass and penicillin production rates were also included. The simulated model predictions of all the on-line and off-line process measurements, including the off-gas analysis, were in good agreement with the batch records. The simulator and industrial process data are available to download at www.industrialpenicillinsimulation.com and can be used to evaluate, study and improve on the current control strategy implemented on this facility.