Multivariate PAT solutions for biopharmaceutical cultivation: current progress and limitations

Rethinking Pharmaceutical Industry with Quality by Design: Application in Research, Development, Manufacturing, and Quality Assurance

Quality by Design (QbD) is a transformative and systematic approach to developing top-tier pharmaceutical products, ushering in a departure from traditional trial-and-error methods toward a more science-based, risk-oriented, and holistic strategy. Central to QbD implementation is the meticulous development of formulations and manufacturing processes, consistently fulfilling predefined quality objectives. The core objective of QbD remains unwavering - to guarantee the steadfast alignment of the final pharmaceutical product with predetermined quality attributes, thereby mitigating batch-to-batch variations and potential recalls. This article succinctly explores the multifaceted application of QbD methodology within the pharmaceutical industry. Emphasizing its pivotal role in research and development, manufacturing, quality control, and quality assurance, the discussion navigates through the strategic deployment of QbD elements and tools. Amidst the evident advantages of QbD, challenges persist in its widespread adoption within the pharmaceutical sector and regulatory frameworks. This article sheds light on the regulatory landscape that currently governs the implementation of QbD in these crucial stages of pharmaceutical processes. For that reason, this review article aims to provide researchers, scientists, and industry professionals with a thorough introduction to QbD so they may adopt this methodical approach to developing and producing high-quality pharmaceutical products, always in compliance with the underlying regulations.

Harnessing near-infrared and Raman spectral sensing and artificial intelligence for real-time monitoring and precision control of bioprocess

Effective monitoring and control of bioprocesses are critical for industrial biomanufacturing. This study demonstrates the integration of near-infrared and Raman spectroscopy for real-time monitoring and precise control of gentamicin fermentation. The orthogonal method reduced redundant features and improved spectral model performance by 9.2-100.4 % in terms of the coefficient of determination (R2). The combinatorial spectral model outperformed single-source models in external validation (R2 > 0.99). An AI-based platform, combining dual-sensors data collection, ML-based prediction, and automated feeding control, was developed for fully automated fed-batch fermentation. This platform dynamically adjusted feeding rates, maintained low glucose concentrations (5 g/L) with accuracy and coefficient of variation below 2 %, and increased gentamicin C1a concentration (346.5 mg/L) by 33.0 % compared to traditional intermittent feeding. These findings underscore the transformative potential of combinatorial spectroscopy and machine learning for real-time bioprocess monitoring, offering a scalable solution for enhancing industrial fermentation efficiency and product titer.

An innovative hybrid modeling approach for simultaneous prediction of cell culture process dynamics and product quality

The use of hybrid models is extensively described in the literature to predict the process evolution in cell cultures. These models combine mechanistic and machine learning methods, allowing the prediction of complex process behavior, in the presence of many process variables, without the need to collect a large amount of data. Hybrid models cannot be directly used to predict final product critical quality attributes, or CQAs, because they are usually measured only at the end of the process, and more mechanistic knowledge is needed for many classes of CQAs. The historical models can instead predict the CQAs better; however, they cannot directly relate manipulated process parameters to final CQAs, as they require knowledge of the process evolution. In this work, we propose an innovative modeling approach based on combining a hybrid propagation model with a historical data-driven model, that is, the combined hybrid model, for simultaneous prediction of full process dynamics and CQAs. The performance of the combined hybrid model was evaluated on an industrial dataset and compared to classical black-box models, which directly relate manipulated process parameters to CQAs. The proposed combined hybrid model outperforms the black-box model by 33% on average in predicting the CQAs while requiring only around half of the data for model training to match performance. Thus, in terms of model accuracy and experimental costs, the combined hybrid model in this study provides a promising platform for process optimization applications.

An adaptive modeling approach using spiking-augmentation method to improve chemometric model performance in bioprocess monitoring

Intensified and continuous processes require fast and robust methods and technologies to monitor product titer for faster analytical turnaround time, process monitoring, and process control. The current titer measurements are mostly offline chromatography-based methods which may take hours or even days to get the results back from the analytical labs. Thus, offline methods will not meet the requirement of real time titer measurements for continuous production and capture processes. FTIR and chemometric based multivariate modeling are promising tools for real time titer monitoring in clarified bulk (CB) harvests and perfusate lines. However, empirical models are known to be vulnerable to unseen variability, specifically a FTIR chemometric titer model trained on a given biological molecule and process conditions often fails to provide accurate predictions of titer in another molecule under different process conditions. In this study, we developed an adaptive modeling strategy: the model was initially built using a calibration set of available perfusate and CB samples and then updated by augmenting spiking samples of the new molecules to the calibration set to make the model robust against perfusate or CB harvest of the new molecule. This strategy substantially improved the model performance and significantly reduced the modeling effort for new molecules.

Statistically-aided development of protein A affinity chromatography for enhancing recovery and controlling quality of a monoclonal antibody

Protein A chromatography is widely used for isolation of monoclonal antibodies (mAbs) from cell culture components. In this study, the effect of different process parameters of the Protein A purification namely, binding pH, elution pH, flow rate, neutralization pH and tween concentration, on the concentration and quality of the purified mAb were evaluated. Using design of experiments approach, the critical process parameters of protein A chromatography were identified and experimentally optimized. Their impact on quality attributes, such as size variants and charge variants, of the mAb was studied. Multivariate data analysis was subsequently performed using multiple linear regression and partial least squares regression methods. It was observed that the elution pH primarily governed the concentration of the purified mAb and the content of monomers and aggregates, while the tween concentration primarily influenced the main peak of the charge variants. This is the first study that evaluates the impact of tween concentration in buffers on the protein A chromatography purification step. These studies helped in identifying the design space and defining the target robust and optimal setpoints of the responses, which were subsequently verified experimentally. These setpoints not only passed the target criteria but also resulted in the highest recoveries during the investigation. Through this statistically-aided approach, an optimized and robust protein A chromatography process was rationally developed for purification of mAbs, while achieving the desired product quality. This study highlights the influence of multiple parameters of the protein A purification process on critical quality attributes of mAbs, such as the size and charge variants, which has been a very scarcely explored area.

Real time on-line amino acid analysis to explore amino acid trends and link to glycosylation outcomes of a model monoclonal antibody

Bioreactor parameters can have significant effects on the quantity and quality of biotherapeutics. Monoclonal antibody products have one particularly important critical quality attribute being the distribution of product glycoforms. N-linked glycosylation affects the therapeutic properties of the antibody including effector function, immunogenicity, stability, and clearance rate. Our past work revealed that feeding different amino acids to bioreactors altered the productivity and glycan profiles. To facilitate real-time analysis of bioreactor parameters and the glycosylation of antibody products, we developed an on-line system to pull cell-free samples directly from the bioreactors, chemically process them, and deliver them to a chromatography-mass spectroscopy system for rapid identification and quantification. We were able to successfully monitor amino acid concentration on-line within multiple reactors, evaluate glycans off-line, and extract four principal components to assess the amino acid concentration and glycosylation profile relationship. We found that about a third of the variability in the glycosylation data can be predicted from the amino acid concentration. Additionally, we determined that the third and fourth principal component accounts for 72% of our model's predictive power, with the third component indicated to be positively correlated with latent metabolic processes related to galactosylation. Here we present our work on rapid online spent media amino acid analysis and use the determined trends to collate with glycan time progression, further elucidating the correlation between bioreactor parameters such as amino acid nutrient profiles, and product quality. We believe such approaches may be useful for maximizing efficiency and reducing production costs for biotherapeutics.

Continuous integrated manufacturing for biopharmaceuticals: A new paradigm or an empty promise?

Continuous integrated bioprocessing has elicited considerable interest from the biopharma industry for the many purported benefits it promises. Today many major biopharma manufacturers around the world are engaged in the development of continuous process platforms for their products. In spite of great potential, the path toward continuous integrated bioprocessing remains unclear for the biologics industry due to legacy infrastructure, process integration challenges, vague regulatory guidelines, and a diverging focus toward novel therapies. In this article, we present a review and perspective on this topic. We explore the status of the implementation of continuous integrated bioprocessing among biopharmaceutical manufacturers. We also present some of the key hurdles that manufacturers are likely to face during this implementation. Finally, we hypothesize that the real impact of continuous manufacturing is likely to come when the cost of manufacturing is a substantial portion of the cost of product development, such as in the case of biosimilar manufacturing and emerging economies.

In-Line Vis-NIR Spectral Analysis for the Column Chromatographic Processes of the Ginkgo biloba L. Leaves. Part II: Batch-to-Batch Consistency Evaluation of the Elution Process

An in-line monitoring method for the elution process of Ginkgo biloba L. leaves using visible and near-infrared spectroscopy in conjunction with multivariate statistical process control (MSPC) was established. Experiments, including normal operating batches and abnormal ones, were designed and carried out. The MSPC model for the elution process was developed and validated. The abnormalities were detected successfully by the control charts of principal component scores, Hotelling T2, or DModX (distance to the model). The results suggested that the established method can be used for the in-line monitoring and batch-to-batch consistency evaluation of the elution process.

Sensor technologies for quality control in engineered tissue manufacturing

The use of engineered cells, tissues, and organs has the opportunity to change the way injuries and diseases are treated. Commercialization of these groundbreaking technologies has been limited in part by the complex and costly nature of their manufacture. Process-related variability and even small changes in the manufacturing process of a living product will impact its quality. Without real-time integrated detection, the magnitude and mechanism of that impact are largely unknown. Real-time and non-destructive sensor technologies are key for in-process insight and ensuring a consistent product throughout commercial scale-up and/or scale-out. The application of a measurement technology into a manufacturing process requires cell and tissue developers to understand the best way to apply a sensor to their process, and for sensor manufacturers to understand the design requirements and end-user needs. Furthermore, sensors to monitor component cells' health and phenotype need to be compatible with novel integrated and automated manufacturing equipment. This review summarizes commercially relevant sensor technologies that can detect meaningful quality attributes during the manufacturing of regenerative medicine products, the gaps within each technology, and sensor considerations for manufacturing.

Data Augmentation to Support Biopharmaceutical Process Development through Digital Models—A Proof of Concept

In recent years, monoclonal antibodies (mAbs) are gaining a wide market share as the most impactful bioproducts. The development of mAbs requires extensive experimental campaigns which may last several years and cost billions of dollars. Following the paradigm of Industry 4.0 digitalization, data-driven methodologies are now used to accelerate the development of new biopharmaceutical products. For instance, predictive models can be built to forecast the productivity of the cell lines in the culture in such a way as to anticipate the identification of the cell lines to be progressed in the scale-up exercise. However, the number of experiments that can be performed decreases dramatically as the process scale increases, due to the resources required for each experimental run. This limits the availability of experimental data and, accordingly, the applicability of data-driven methodologies to support the process development. To address this issue in this work we propose the use of digital models to generate in silico data and augment the amount of data available from real (i.e., in vivo) experimental runs, accordingly. In particular, we propose two strategies for in silico data generation to estimate the endpoint product titer in mAbs manufacturing: one based on a first principles model and one on a hybrid semi-parametric model. As a proof of concept, the effect of in silico data generation was investigated on a simulated biopharmaceutical process for the production of mAbs. We obtained very promising results: the digital model effectively supports the identification of high-productive cell lines (i.e., high mAb titer) even when a very low number of real experimental batches (two or three) is available.

Application of Supervised and Unsupervised Learning Approaches for Mapping Storage Conditions of Biopharmaceutical Product-A Case Study of Human Serum Albumin

The stability of biopharmaceutical therapeutics over the storage period/shelf life has been a challenging concern for manufacturers. A noble strategy for mapping best and suitable storage conditions for recombinant human serum albumin (rHSA) in laboratory mixture was optimized using chromatographic data as per principal component analysis (PCA), and similarity was defined using hierarchical cluster analysis. In contrast, separability was defined using linear discriminant analysis (LDA) models. The quantitation was performed for rHSA peak (analyte of interest) and its degraded products, i.e., dimer, trimer, agglomerates and other degradation products. The chromatographic variables were calculated using validated stability-indicating assay method. The chromatographic data mapping was done for the above-mentioned peaks over three months at different temperatures, i.e., 20°C, 5-8°C and at room temperature (25°C). The PCA had figured out the ungrouped variable, whereas supervised mapping was done using LDA. As an outcome result of LDA, about 60% of data were correctly classified with the highest sensitivity for 25°C (Aq), 25°C and 5-8°C (Aq with 5% glucose as a stabilizer), whereas the highest specificity was observed for samples stored at 5-8°C (Aq with 5% glucose as a stabilizer).

Automated Bioprocess Feedback Operation in a High-Throughput Facility via the Integration of a Mobile Robotic Lab Assistant

The development of biotechnological processes is challenging due to the diversity of process parameters. For efficient upstream development, parallel cultivation systems have proven to reduce costs and associated timelines successfully while offering excellent process control. However, the degree of automation of such small-scale systems is comparatively low, and necessary sample analysis requires manual steps. Although the subsequent analysis can be performed in a high-throughput manner, the integration of analytical devices remains challenging, especially when cultivation and analysis laboratories are spatially separated. Mobile robots offer a potential solution, but their implementation in research laboratories is not widely adopted. Our approach demonstrates the integration of a small-scale cultivation system into a liquid handling station for an automated cultivation and sample procedure. The samples are transported via a mobile robotic lab assistant and subsequently analyzed by a high-throughput analyzer. The process data are stored in a centralized database. The mobile robotic workflow guarantees a flexible solution for device integration and facilitates automation. Restrictions regarding spatial separation of devices are circumvented, enabling a modular platform throughout different laboratories. The presented cultivation platform is evaluated on the basis of industrially relevant E. coli BW25113 high cell density fed-batch cultivation. The necessary magnesium addition for reaching high cell densities in mineral salt medium is automated via a feedback operation loop between the analysis station located in the adjacent room and the cultivation system. The modular design demonstrates new opportunities for advanced control options and the suitability of the platform for accelerating bioprocess development. This study lays the foundation for a fully integrated facility, where the physical connection of laboratory equipment is achieved through the successful use of a mobile robotic lab assistant, and different cultivation scales can be coupled through the common data infrastructure.

Model Transferability and Reduced Experimental Burden in Cell Culture Process Development Facilitated by Hybrid Modeling and Intensified Design of Experiments

Reliable process development is accompanied by intense experimental effort. The utilization of an intensified design of experiments (iDoE) (intra-experimental critical process parameter (CPP) shifts combined) with hybrid modeling potentially reduces process development burden. The iDoE can provide more process response information in less overall process time, whereas hybrid modeling serves as a commodity to describe this behavior the best way. Therefore, a combination of both approaches appears beneficial for faster design screening and is especially of interest at larger scales where the costs per experiment rise significantly. Ideally, profound process knowledge is gathered at a small scale and only complemented with few validation experiments on a larger scale, saving valuable resources. In this work, the transferability of hybrid modeling for Chinese hamster ovary cell bioprocess development along process scales was investigated. A two-dimensional DoE was fully characterized in shake flask duplicates (300 ml), containing three different levels for the cultivation temperature and the glucose concentration in the feed. Based on these data, a hybrid model was developed, and its performance was assessed by estimating the viable cell concentration and product titer in 15 L bioprocesses with the same DoE settings. To challenge the modeling approach, 15 L bioprocesses also comprised iDoE runs with intra-experimental CPP shifts, impacting specific cell rates such as growth, consumption, and formation. Subsequently, the applicability of the iDoE cultivations to estimate static cultivations was also investigated. The shaker-scale hybrid model proved suitable for application to a 15 L scale (1:50), estimating the viable cell concentration and the product titer with an NRMSE of 10.92% and 17.79%, respectively. Additionally, the iDoE hybrid model performed comparably, displaying NRMSE values of 13.75% and 21.13%. The low errors when transferring the models from shaker to reactor and between the DoE and the iDoE approach highlight the suitability of hybrid modeling for mammalian cell culture bioprocess development and the potential of iDoE to accelerate process characterization and to improve process understanding.

Harnessing the potential of machine learning for advancing "Quality by Design" in biomanufacturing

Ensuring consistent high yields and product quality are key challenges in biomanufacturing. Even minor deviations in critical process parameters (CPPs) such as media and feed compositions can significantly affect product critical quality attributes (CQAs). To identify CPPs and their interdependencies with product yield and CQAs, design of experiments, and multivariate statistical approaches are typically used in industry. Although these models can predict the effect of CPPs on product yield, there is room to improve CQA prediction performance by capturing the complex relationships in high-dimensional data. In this regard, machine learning (ML) approaches offer immense potential in handling non-linear datasets and thus are able to identify new CPPs that could effectively predict the CQAs. ML techniques can also be synergized with mechanistic models as a ‘hybrid ML’ or ‘white box ML’ to identify how CPPs affect the product yield and quality mechanistically, thus enabling rational design and control of the bioprocess. In this review, we describe the role of statistical modeling in Quality by Design (QbD) for biomanufacturing, and provide a generic outline on how relevant ML can be used to meaningfully analyze bioprocessing datasets. We then offer our perspectives on how relevant use of ML can accelerate the implementation of systematic QbD within the biopharma 4.0 paradigm.

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.

Proton-transfer-reaction mass spectrometry (PTR-MS) for online monitoring of glucose depletion and cell concentrations in HEK 293 gene therapy processes

Objectives

The applicability of proton-transfer-reaction mass spectrometry (PTR-MS) as a versatile online monitoring tool to increase consistency and robustness for recombinant adeno-associated virus (rAAV) producing HEK 293 bioprocesses was evaluated. We present a structured workflow to extract process relevant information from PTR-MS data.

Results

Reproducibility of volatile organic compound (VOC) measurements was demonstrated with spiking experiments and the process data sets used for applicability evaluation consisted of HEK 293 cell culture triplicates with and without transfection. The developed data workflow enabled the identification of six VOCs, of which two were used to develop a soft sensor providing better real-time estimates than the conventional capacitance sensor. Acetaldehyde, another VOC, provides online process information about glucose depletion that can directly be used for process control purposes.

Conclusions

The potential of PTR-MS for HEK 293 cell culture monitoring has been shown. VOC data derived information can be used to develop soft sensors and to directly set up new process control strategies.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10529-021-03205-y.

Critical Analysis of cGMP Large-Scale Expansion Process in Bioreactors of Human Induced Pluripotent Stem Cells in the Framework of Quality by Design

Human induced pluripotent stem cells (hiPSCs) are manufactured as advanced therapy medicinal products for tissue replacement applications. With this aim, the feasibility of hiPSC large-scale expansion in existing bioreactor systems under current good manufacturing practices (cGMP) has been tested. Yet, these attempts have lacked a paradigm shift in culture settings and technologies tailored to hiPSCs, which jeopardizes their clinical translation. The best approach for industrial scale-up of high-quality hiPSCs is to design their manufacturing process by following quality-by-design (QbD) principles: a scientific, risk-based framework for process design based on relating product and process attributes to product quality. In this review, we analyzed the hiPSC expansion manufacturing process implementing the QbD approach in the use of bioreactors, stressing the decisive role played by the cell quantity, quality and costs, drawing key QbD concepts directly from the guidelines of the International Council for Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use.

Determination of Chinese hamster ovary (CHO) cell densities and antibody titers from small volumes of cell culture supernatants using multivariate analysis and partial least squares regression of UV-Vis spectra

Antibody titer and viable cell density (VCD) are two important parameters that need to be closely monitored during the process of cell line development and manufacturing of therapeutic antibodies. Typically, determination of each parameter requires 10–100 μL of supernatant sample, which is not suitable for small scale cultivation. In this study, we demonstrated that as low as 2 μL of culture supernatants were sufficient for the analysis using UV-Vis spectrum assisted with partial least squares (PLS) model. The results indicated that the optimal PLS models could be used to predict antibody titer and VCD with the linear relationship between reference values and predicted values at R² values ranging from 0.8 to > 0.9 in supernatant samples obtained from four different single clones and in polyclones that were cultured in various selection stringencies. Then, the percentage of cell viability and productivity were predicted from a set of samples of polyclones. The results indicated that while all predicted % cell viability were very similar to the actual value at RSEP value of 6.7 and R² of 0.8908, the predicted productivity from 14 of 18 samples were closed to the reference measurements at RSEP value of 22.4 and R² of 0.8522. These results indicated that UV-Vis combined with PLS has potential to be used for monitoring antibody titer, VCD, and % cell viability for both online and off-line therapeutic production process.

Digital Twin Application for Model-Based DoE to Rapidly Identify Ideal Process Conditions for Space-Time Yield Optimization

The fast exploration of a design space and identification of the best process conditions facilitating the highest space-time yield are of great interest for manufacturers. To obtain this information, depending on the design space, a large number of practical experiments must be performed, analyzed, and evaluated. To reduce this experimental effort and increase the process understanding, we evaluated a model-based design of experiments to rapidly identify the optimum process conditions in a design space maximizing space-time yield. From a small initial dataset, hybrid models were implemented and used as digital bioprocess twins, thus obtaining the recommended optimal experiment. In cases where these optimum conditions were not covered by existing data, the experiment was carried out and added to the initial data set, re-training the hybrid model. The procedure was repeated until the model gained certainty about the best process conditions, i.e., no new recommendations. To evaluate this workflow, we utilized different initial data sets and assessed their respective performances. The fastest approach for optimizing the space-time yield in a three-dimensional design space was found with five initial experiments. The digital twin gained certainty after four recommendations, leading to a significantly reduced experimental effort compared to other state-of-the-art approaches. This highlights the benefits of in silico design space exploration for accelerating knowledge-based bioprocess development, and reducing the number of hands-on experiments, time, energy, and raw materials.

Towards the Development of Digital Twins for the Bio-manufacturing Industry

The bio-manufacturing industry, along with other process industries, now has the opportunity to be engaged in the latest industrial revolution, also known as Industry 4.0. To successfully accomplish this, a physical-to-digital-to-physical information loop should be carefully developed. One way to achieve this is, for example, through the implementation of digital twins (DTs), which are virtual copies of the processes. Therefore, in this paper, the focus is on understanding the needs and challenges faced by the bio-manufacturing industry when dealing with this digitalized paradigm. To do so, two major building blocks of a DT, data and models, are highlighted and discussed. Hence, firstly, data and their characteristics and collection strategies are examined as well as new methods and tools for data processing. Secondly, modelling approaches and their potential of being used in DTs are reviewed. Finally, we share our vision with regard to the use of DTs in the bio-manufacturing industry aiming at bringing the DT a step closer to its full potential and realization.

Process Analytical Technologies and Data Analytics for the Manufacture of Monoclonal Antibodies

Process analytical technology (PAT) for the manufacture of monoclonal antibodies (mAbs) is defined by an integrated set of advanced and automated methods that analyze the compositions and biophysical properties of cell culture fluids, cell-free product streams, and biotherapeutic molecules that are ultimately formulated into concentrated products. In-line or near-line probes and systems are remarkably well developed, although challenges remain in the determination of the absence of viral loads, detecting microbial or mycoplasma contamination, and applying data-driven deep learning to process monitoring and soft sensors. In this review, we address the current status of PAT for both batch and continuous processing steps and discuss its potential impact on facilitating the continuous manufacture of biotherapeutics.

Multivariate data analysis in cell gene therapy manufacturing

The emergence of cell gene therapy (CGT) as a safe and efficacious treatment for numerous severe inherited and acquired human diseases has led to growing interest and investment in new CGT products. The most successful of these have been autologous viral vector-based treatments. The development of viral vector manufacturing processes and ex vivo patient cell processing capabilities is a pressing issue in the advancement of autologous viral vector-based CGT treatments. In viral vector production, scale-up is a critical task due to the limited scalability of traditional laboratory systems and the demand for high volumes of viral vector manufactured in accordance with current good manufacturing practice. Ex vivo cell processing methods require optimisation and automation before they can be scaled out, and several other manufacturing challenges are prevalent such as high levels of raw material and process variability, difficulty characterising complex materials, and a lack of knowledge of critical process parameters and their effect on critical quality attributes of the viral vector and cell drug products. Multivariate data analysis (MVDA) has been leveraged successfully in a variety of applications in the chemical and biochemical industries, including for tasks such as bioprocess monitoring, identification of critical process parameters and assessment of process variability and comparability during process development, scale-up and technology transfer. Henceforth, MVDA is reviewed here as a suitable tool for tackling some of the challenges faced in the development of CGT manufacturing processes. A summary of some key CGT manufacturing challenges is provided along with a review of MVDA applications to mammalian and microbial processes, and an exploration of the potential benefits, requirements and pre-requisites of MVDA applications in the development of CGT manufacturing processes.

FTIR based kinetic characterisation of an acid-catalysed esterification of 3-methylphthalic anhydride and 2-ethylhexanol

In this study, an inline analytical method was designed and applied in process characterisation and development. The model reaction is the two-step diesterification of 3-methylphthalic anhydride with 2-ethylhexanol consisting of alcoholysis as the first step, followed by an acid-catalysed, second esterification step leading to the corresponding diester. The final product is a potential, alternative plasticiser. For the inline measurements, attenuated total reflection Fourier transformed infrared spectroscopy (ATR-FTIR) was implemented. In order to evaluate the spectra recorded during the reaction, a chemometric model was established. In this work, Indirect Hard Modeling (IHM), a non-linear modeling approach was employed. The respective model was calibrated by using offline samples analysed with gas (GC) and liquid chromatography (HPLC). After successful validation of the chemometric model, the inline measurements were utilized for reaction characterisation. The acid-catalysed, second esterification step was identified as the limiting reaction step. From batch reactions conducted at different temperatures, the energy of activation of this step was determined to be 79.5 kJ mol-1. Additionally, kinetics were shown to follow a pseudo-first order with respect to the monoester formation and a kinetic model was established. The model was validated in simulations with changed reaction conditions.

Monitoring Parallel Robotic Cultivations with Online Multivariate Analysis

In conditional microbial screening, a limited number of candidate strains are tested at different conditions searching for the optimal operation strategy in production (e.g., temperature and pH shifts, media composition as well as feeding and induction strategies). To achieve this, cultivation volumes of >10 mL and advanced control schemes are required to allow appropriate sampling and analyses. Operations become even more complex when the analytical methods are integrated into the robot facility. Among other multivariate data analysis methods, principal component analysis (PCA) techniques have especially gained popularity in high throughput screening. However, an important issue specific to high throughput bioprocess development is the lack of so-called golden batches that could be used as a basis for multivariate analysis. In this study, we establish and present a program to monitor dynamic parallel cultivations in a high throughput facility. PCA was used for process monitoring and automated fault detection of 24 parallel running experiments using recombinant E. coli cells expressing three different fluorescence proteins as the model organism. This approach allowed for capturing events like stirrer failures and blockage of the aeration system and provided a good signal to noise ratio. The developed application can be easily integrated in existing data- and device-infrastructures, allowing automated and remote monitoring of parallel bioreactor systems.

Recent applications of chemometrics in one- and two-dimensional chromatography

The proliferation of increasingly more sophisticated analytical separation systems, often incorporating increasingly more powerful detection techniques, such as high-resolution mass spectrometry, causes an urgent need for highly efficient data-analysis and optimization strategies. This is especially true for comprehensive two-dimensional chromatography applied to the separation of very complex samples. In this contribution, the requirement for chemometric tools is explained and the latest developments in approaches for (pre-)processing and analyzing data arising from one- and two-dimensional chromatography systems are reviewed. The final part of this review focuses on the application of chemometrics for method development and optimization.

Raman spectroscopic based chemometric models to support a dynamic capacitance based cell culture feeding strategy

Multiple process analytical technology (PAT) tools are now being applied in tandem for cell culture. Research presented used two in-line probes, capacitance for a dynamic feeding strategy and Raman spectroscopy for real-time monitoring. Data collected from eight batches at the 15,000 L scale were used to develop process models. Raman spectroscopic data were modelled using Partial Least Squares (PLS) by two methods-(1) use of the full dataset and (2) split the dataset based on the capacitance feeding strategy. Root mean square error of prediction (RMSEP) for the first model method of capacitance was 1.54 pf/cm and the second modelling method was 1.40 pf/cm. The second Raman method demonstrated results within expected process limits for capacitance and a 0.01% difference in total nutrient feed compared to the capacitance probe. Additional variables modelled using Raman spectroscopy were viable cell density (VCD), viability, average cell diameter, and viable cell volume (VCV).

Fermentation for tailoring the technological and health related functionality of food products

Fermented foods are experiencing a resurgence due to the consumers' growing interest in foods that are natural and health promoting. Microbial fermentation is a biotechnological process which transforms food raw materials into palatable, nutritious and healthy food products. Fermentation imparts unique aroma, flavor and texture to food, improves digestibility, degrades anti-nutritional factors, toxins and allergens, converts phytochemicals such as polyphenols into more bioactive and bioavailable forms, and enriches the nutritional quality of food. Fermentation also modifies the physical functional properties of food materials, rendering them differentiated ingredients for use in formulated foods. The science of fermentation and the technological and health functionality of fermented foods is reviewed considering the growing interest worldwide in fermented foods and beverages and the huge potential of the technology for reducing food loss and improving nutritional food security.

Mid-Infrared Spectroscopy and Multivariate Analysis to Characterize Lactobacillus acidophilus Fermentation Processes

The ever-growing competition among global biotech industries has led to high demands on production consistency. A statistical strategy of performance mapping for production optimization is therefore of great economic significance. Process analytical technology (PAT)-based sensors such as mid-infrared (MIR) spectroscopy enable process monitoring through substrate and by-product concentrations that directly represent the physiology of cells. Combined with multivariate statistics, MIR can be employed as a strategy for production performance mapping. This study describes the use of at-line spectroscopy, chemometric modeling, and post-process fitting to characterize Lactobacillus acidophilus fermentations. The emphasis is on alternative arrangements of the data and chemometric methods principle component analysis (PCA), multivariate curve resolution (MCR), and parallel factor analysis (PARAFAC). Two key parameters, rate constant and time of inflection, are extracted by post-process fitting on the outcomes of these different models. Their use as process performance descriptors to characterize the dynamics of substrate consumption, product formation and batch-to-batch variations is suggested. The unconstrained PCA primarily described biomass change, while the constrained models PARAFAC and MCR (both the augmented and individual-run configurations) could model the decrease in sugars and increase in lactic acid over time. It was concluded that MCR on individual batch data, followed by post-process fitting, is the preferred strategy for MIR spectroscopic monitoring.

Decision Tree-PLS (DT-PLS) algorithm for the development of process: Specific local prediction models

This work presents a novel multivariate statistical algorithm, Decision Tree-PLS (DT-PLS), to improve the prediction and understanding of dynamic processes based on local partial least square regression (PLSR) models for characteristic process groups defined based on Decision Tree (DT) analysis. The DT-PLS algorithm is successfully applied to two different cell culture data sets, one obtained from bioreactors of 3.5 L lab scale and the other obtained from the 15 ml ambr microbioreactor system. Substantial improvement in the predictive capabilities of the model can be achieved based on the localization compared to the classical PLSR approach, which is implemented in the commercially available packages. Additionally, the differences in the model parameters of the local models suggest that the governing process variables vary for the different process regimes indicating the different states of the cell under different process conditions.

Risk-Based Bioengineering Strategies for Reliable Bacterial Vaccine Production

Design of a reliable process for bacterial antigen production requires understanding of and control over critical process parameters. Current methods for process design use extensive screening experiments for determining ranges of critical process parameters yet fail to give clear insights into how they influence antigen potency. To address this gap, we propose to apply constraint-based, genome-scale metabolic models to reduce the need of experimental screening for strain selection and to optimize strains based on model driven iterative Design-Build-Test-Learn (DBTL) cycles. Application of these systematic methods has not only increased the understanding of how metabolic network properties influence antigen potency, but also allows identification of novel critical process parameters that need to be controlled to achieve high process reliability.

Comparison of data science workflows for root cause analysis of bioprocesses

Root cause analysis (RCA) is one of the most prominent tools used to comprehensively evaluate a biopharmaceutical production process. Despite of its widespread use in industry, the Food and Drug Administration has observed a lot of unsuitable approaches for RCAs within the last years. The reasons for those unsuitable approaches are the use of incorrect variables during the analysis and the lack in process understanding, which impede correct model interpretation. Two major approaches to perform RCAs are currently dominating the chemical and pharmaceutical industry: raw data analysis and feature-based approach. Both techniques are shown to be able to identify the significant variables causing the variance of the response. Although they are different in data unfolding, the same tools as principal component analysis and partial least square regression are used in both concepts. Within this article we demonstrate the strength and weaknesses of both approaches. We proved that a fusion of both results in a comprehensive and effective workflow, which not only increases better process understanding. We demonstrate this workflow along with an example. Hence, the presented workflow allows to save analysis time and to reduce the effort of data mining by easy detection of the most important variables within the given dataset. Subsequently, the final obtained process knowledge can be translated into new hypotheses, which can be tested experimentally and thereby lead to effectively improving process robustness.

Advanced monitoring and control of pharmaceutical production processes with Pichia pastoris by using Raman spectroscopy and multivariate calibration methods

This contribution includes an investigation of the applicability of Raman spectroscopy as a PAT analyzer in cyclic production processes of a potential Malaria vaccine with Pichia pastoris. In a feasibility study, Partial Least Squares Regression (PLSR) models were created off-line for cell density and concentrations of glycerol, methanol, ammonia and total secreted protein. Relative cross validation errors RMSE_cvrel range from 2.87% (glycerol) to 11.0% (ammonia). In the following, on-line bioprocess monitoring was tested for cell density and glycerol concentration. By using the nonlinear Support Vector Regression (SVR) method instead of PLSR, the error RMSE_Prel for cell density was reduced from 5.01 to 2.94%. The high potential of Raman spectroscopy in combination with multivariate calibration methods was demonstrated by the implementation of a closed loop control for glycerol concentration using PLSR. The strong nonlinear behavior of exponentially increasing control disturbances was met with a feed-forward control and adaptive correction of control parameters. In general the control procedure works very well for low cell densities. Unfortunately, PLSR models for glycerol concentration are strongly influenced by a correlation with the cell density. This leads to a failure in substrate prediction, which in turn prevents substrate control at cell densities above 16 g/L.