Application of multivariate data analysis for identification and successful resolution of a root cause for a bioprocessing application

Multivariate Data Analysis (MVDA) can be used for supporting key activities required for successful bioprocessing. These activities include process characterization, process scale-up, process monitoring, fault diagnosis and root cause analysis. This paper examines an application of MVDA towards root cause analysis for identifying scale-up differences and parameter interactions that adversely impact cell culture process performance. Multivariate data analysis and modeling were performed using data from small-scale (2 L), pilot-scale (2,000 L) and commercial-scale (15,000 L) batches. The input parameters examined included bioreactor pCO2, glucose, lactate, ammonium, raw materials and seed inocula. The output parameters included product attributes, product titer, viable cell density, cell viability and osmolality. Time course performance variables (daily, initial, peak and end point) were also evaluated. Application of MVDA as a diagnostic tool was successful in identifying the root cause and designing experimental conditions to demonstrate and correct it. Process parameters and their interactions that adversely impact cell culture performance and product attributes were successfully identified. MVDA was successfully used as an effective tool for collating process knowledge and increasing process understanding.     

Predicting Mab product yields from cultivation media components, using near-infrared and 2D-fluorescence spectroscopies

The yield of monoclonal antibody (Mab) production processes depends on media formulation, inocula quality, and process conditions. As in industrial processes tight cultivation conditions are used, and inocula quality and viable cell densities are controlled to reasonable levels, media formulation and raw materials lot-to-lot variability in quality will have, in those circumstances, the highest impact on process performance. In the particular Mab process studied, two different raw materials were used: a complex carbon and nitrogen source made of specific peptones and defined chemical media containing multiple components. Using different spectroscopy techniques for each of the raw material types, it was concluded that for the complex peptone-based ingredient, near-infrared (NIR) spectroscopy was more capable of capturing lot-to-lot variability. For the chemically defined media containing fluorophores, two-dimensional (2D)-fluorescence spectroscopy was more capable of capturing lot-to-lot variability. Because in Mab cultivation processes both types of raw materials are used, combining the NIR and 2D-fluorescence spectra for each of the media components enabled predictive models for yield to be developed that out-performed any other model involving either one raw material alone, or only one type of spectroscopic tool for both raw materials. For each particular raw material, the capability of each spectroscopy to detect lot-to-lot differences was demonstrated after spectra preprocessing and specific wavelength regions selection. The work described and the findings reported here open up several possibilities that could be used to feed-forward control the process. These include, for example, enabling specific actions to be taken regarding media formulation with particular lots, and all types of predictive control actions aimed at increasing batch-to-batch yield and product quality consistency at harvest.     

De-risking Pharmaceutical Tablet Manufacture Through Process Understanding,Latent Variable Modeling,and Optimization Technologies

In pharmaceutical tablet manufacturing processes, a major source of disturbance affecting drug product quality is the (lot-to-lot) variability of the incoming raw materials. A novel modeling and process optimization strategy that compensates for raw material variability is presented. The approach involves building partial least squares models that combine raw material attributes and tablet process parameters and relate these to final tablet attributes. The resulting models are used in an optimization framework to then find optimal process parameters which can satisfy all the desired requirements for the final tablet attributes, subject to the incoming raw material lots. In order to de-risk the potential (lot-to-lot) variability of raw materials on the drug product quality, the effect of raw material lot variability on the final tablet attributes was investigated using a raw material database containing a large number of lots. In this way, the raw material variability, optimal process parameter space and tablet attributes are correlated with each other and offer the opportunity of simulating a variety of changes in silico without actually performing experiments. The connectivity obtained between the three sources of variability (materials, parameters, attributes) can be considered a design space consistent with Quality by Design principles, which is defined by the ICH-Q8 guidance (USDA 2006). The effectiveness of the methodologies is illustrated through a common industrial tablet manufacturing case study.     

Characterization of mammalian cell culture raw materials by combining spectroscopy and chemometrics

Two of the primary issues with characterizing the variability of raw materials used in mammalian cell culture, such as wheat hydrolysate, is that the analyses of these materials can be time consuming, and the results of the analyses are not straightforward to interpret. To solve these issues, spectroscopy can be combined with chemometrics to provide a quick, robust and easy to understand methodology for the characterization of raw materials; which will improve cell culture performance by providing an assessment of the impact that a given raw material will have on final product quality. In this study, four spectroscopic technologies: near infrared spectroscopy, middle infrared spectroscopy, Raman spectroscopy, and fluorescence spectroscopy were used in conjunction with principal component analysis to characterize the variability of wheat hydrolysates, and to provide evidence that the classification of good and bad lots of raw material is possible. Then, the same spectroscopic platforms are combined with partial least squares regressions to quantitatively predict two cell culture critical quality attributes (CQA): integrated viable cell density and IgG titer. The results showed that near infrared (NIR) spectroscopy and fluorescence spectroscopy are capable of characterizing the wheat hydrolysate's chemical structure, with NIR performing slightly better; and that they can be used to estimate the raw materials’ impact on the CQAs. These results were justified by demonstrating that of all the components present in the wheat hydrolysates, six amino acids: arginine, glycine, phenylalanine, tyrosine, isoleucine and threonine; and five trace elements: copper, phosphorus, molybdenum, arsenic and aluminum, had a large, statistically significant effect on the CQAs, and that NIR and fluorescence spectroscopy performed the best for characterizing the important amino acids. It was also found that the trace elements of interest were not characterized well by any of the spectral technologies used; however, the trace elements were also shown to have a less significant effect on the CQAs than the amino acids. © 2017 The Authors Biotechnology Progress published by Wiley Periodicals, Inc. on behalf of American Institute of Chemical Engineers, 33:1127–1138, 2017     

Towards robust cell culture processes — Unraveling the impact of media preparation by spectroscopic online monitoring

Biopharmaceutical manufacturing processes can be affected by variability in cell culture media, e.g. caused by raw material impurities. Although efforts have been made in industry and academia to characterize cell culture media and raw materials with advanced analytics, the process of industrial cell culture media preparation itself has not been reported so far. Within this publication, we first compare mid‐infrared and two‐dimensional fluorescence spectroscopy with respect to their suitability as online monitoring tools during cell culture media preparation, followed by a thorough assessment of the impact of preparation parameters on media quality. Through the application of spectroscopic methods, we can show that media variability and its corresponding root cause can be detected online during the preparation process. This methodology is a powerful tool to avoid batch failure and is a valuable technology for media troubleshooting activities. Moreover, in a design of experiments approach, including additional liquid chromatography–mass spectrometry analytics, it is shown that variable preparation parameters such as temperature, power input and preparation time can have a strong impact on the physico‐chemical composition of the media. The effect on cell culture process performance and product quality in subsequent fed‐batch processes was also investigated. The presented results reveal the need for online spectroscopic methods during the preparation process and show that media variability can already be introduced by variation in media preparation parameters, with a potential impact on scale‐up to a commercial manufacturing process.     

Guidance for performing multivariate data analysis of bioprocessing data: Pitfalls and recommendations

Biotech unit operations are often characterized by a large number of inputs (operational parameters) and outputs (performance parameters) along with complex correlations among them. A typical biotech process starts with the vial of the cell bank, ends with the final product, and has anywhere from 15 to 30 such unit operations in series. Besides the above‐mentioned operational parameters, raw material attributes can also impact process performance and product quality as well as interact among each other. Multivariate data analysis (MVDA) offers an effective approach to gather process understanding from such complex datasets. Review of literature suggests that the use of MVDA is rapidly increasing, fuelled by the gradual acceptance of quality by design (QbD) and process analytical technology (PAT) among the regulators and the biotech industry. Implementation of QbD and PAT requires enhanced process and product understanding. In this article, we first discuss the most critical issues that a practitioner needs to be aware of while performing MVDA of bioprocessing data. Next, we present a step by step procedure for performing such analysis. Industrial case studies are used to elucidate the various underlying concepts. With the increasing usage of MVDA, we hope that this article would be a useful resource for present and future practitioners of MVDA. © 2014 American Institute of Chemical Engineers Biotechnol. Prog., 30:967–973, 2014     

Fermentanomics: Relating quality attributes of a monoclonal antibody to cell culture process variables and raw materials using multivariate data analysis

Fermentanomics is an emerging field of research and involves understanding the underlying controlled process variables and their effect on process yield and product quality. Although major advancements have occurred in process analytics over the past two decades, accurate real‐time measurement of significant quality attributes for a biotech product during production culture is still not feasible. Researchers have used an amalgam of process models and analytical measurements for monitoring and process control during production. This article focuses on using multivariate data analysis as a tool for monitoring the internal bioreactor dynamics, the metabolic state of the cell, and interactions among them during culture. Quality attributes of the monoclonal antibody product that were monitored include glycosylation profile of the final product along with process attributes, such as viable cell density and level of antibody expression. These were related to process variables, raw materials components of the chemically defined hybridoma media, concentration of metabolites formed during the course of the culture, aeration‐related parameters, and supplemented raw materials such as glucose, methionine, threonine, tryptophan, and tyrosine. This article demonstrates the utility of multivariate data analysis for correlating the product quality attributes (especially glycosylation) to process variables and raw materials (especially amino acid supplements in cell culture media). The proposed approach can be applied for process optimization to increase product expression, improve consistency of product quality, and target the desired quality attribute profile. © 2015 American Institute of Chemical Engineers Biotechnol. Prog., 31:1586–1599, 2015     

Multivariate data analysis on historical IPV production data for better process understanding and future improvements

Historical manufacturing data can potentially harbor a wealth of information for process optimization and enhancement of efficiency and robustness. To extract useful data multivariate data analysis (MVDA) using projection methods is often applied. In this contribution, the results obtained from applying MVDA on data from inactivated polio vaccine (IPV) production runs are described. Data from over 50 batches at two different production scales (700‐L and 1,500‐L) were available. The explorative analysis performed on single unit operations indicated consistent manufacturing. Known outliers (e.g., rejected batches) were identified using principal component analysis (PCA). The source of operational variation was pinpointed to variation of input such as media. Other relevant process parameters were in control and, using this manufacturing data, could not be correlated to product quality attributes. The gained knowledge of the IPV production process, not only from the MVDA, but also from digitalizing the available historical data, has proven to be useful for troubleshooting, understanding limitations of available data and seeing the opportunity for improvements. Biotechnol. Bioeng. 2010;107: 96–104. © 2010 Wiley Periodicals, Inc.     

Development toward rapid and efficient screening for high performance hydrolysate lots in a recombinant monoclonal antibody manufacturing process

Plant‐derived hydrolysates are widely used in mammalian cell culture media to increase yields of recombinant proteins and monoclonal antibodies (mAbs). However, these chemically varied and undefined raw materials can have negative impact on yield and/or product quality in large‐scale cell culture processes. Traditional methods that rely on fractionation of hydrolysates yielded little success in improving hydrolysate quality. We took a holistic approach to develop an efficient and reliable method to screen intact soy hydrolysate lots for commercial recombinant mAb manufacturing. Combined high‐resolution ¹H nuclear magnetic resonance (NMR) spectroscopy and partial least squares (PLS) analysis led to a prediction model between product titer and NMR fingerprinting of soy hydrolysate with cross‐validated correlation coefficient R² of 0.87 and root‐mean‐squared‐error of cross‐validation RMSECV% of 11.2%. This approach screens for high performance hydrolysate lots, therefore ensuring process consistency and product quality in the mAb manufacturing process. Furthermore, PLS analysis was successful in discerning multiple markers (DL‐lactate, soy saccharides, citrate and succinate) among hydrolysate components that positively and negatively correlate with titer. Interestingly, these markers correlate to the metabolic characteristics of some strains of taxonomically diverse lactic acid bacteria (LAB). Thus our findings indicate that LAB strains may exist during hydrolysate manufacturing steps and their biochemical activities may attribute to the titer enhancement effect of soy hydrolysates. © 2012 American Institute of Chemical Engineers Biotechnol. Prog., 28: 1061–1068, 2012     

Advanced Granulation Technology (AGTTM). An alternate format for serum-free, chemically-defined and protein-free cell culture media

To overcome limitations of conventional milling technology, we investigated the application of fluid bed granulation for the production of dry-form nutrient media. Serum-free, protein-free and chemically-defined specialty media were produced in granulated format and compared with identical formulations manufactured by conventional methods. HPLC analysis of multiple lots of granulated materials demonstrated that biochemical constituents were precisely and homogeneously distributed throughout the granules and that nutrient levels were comparable to conventional formats. Comparison of medium performance in cell proliferation and biological production assays demonstrated equivalence with reference media. The fluid bed granulation process meets pharmaceutical quality requirements and may be applied to a broad range of nutrient formulations required for bioproduction applications. This revised version was published online in July 2006 with corrections to the Cover Date.     

Estimation of raw material performance in mammalian cell culture using near infrared spectra combined with chemometrics approaches

Understanding variability in raw materials and their impacts on product quality is of critical importance in the biopharmaceutical manufacturing processes. For this purpose, several spectroscopic techniques have been studied for raw material characterization, providing fast and nondestructive ways to measure quality of raw materials. However, investigations of correlation between spectra of raw materials and cell culture performance have been scarce due to their complexity and uncertainty. In this study, near-infrared spectra and bioassays of multiple soy hydrolysate lots manufactured by different vendors were analyzed using chemometrics approaches in order to address variability of raw materials as well as correlation between raw material properties and corresponding cell culture performance. Principal component analysis revealed that near-infrared spectra of different soy lots contain enough physicochemical information about soy hydrolysates to allow identification of lot-to-lot variability as well as vendor-to-vendor differences. The identified compositional variability was further analyzed in order to estimate cell growth and protein production of two mammalian cell lines under the condition of varying soy dosages using partial least square regression combined with optimal variable selection. The performance of the resulting models demonstrates the potential of near-infrared spectroscopy as a robust lot selection tool for raw materials while providing a biological link between chemical composition of raw materials and cell culture performance.     

Elemental metal variance in cell culture raw materials for process risk profiling

Elemental metals are critical raw material attributes which can impact cell culture performance and associated therapeutic protein product quality profiles. Metals such as copper and manganese act as cofactors and reagents for numerous metabolic pathways which govern cell growth, protein expression, and glycosylation, thus mandating elemental monitoring. The growing complexity of modern cell culture media formulations adds additional opportunities for elemental variance and its associated impact risks. This article describes an analytical technique applying inductively coupled plasma mass spectrometry to characterize a list of common raw materials and media powders used in mammalian cell culture and therapeutic protein production. We aim to describe a method qualification approach suitable for biopharmaceutical raw materials. Furthermore, we present detailed profiles of many common raw materials and discuss trends in raw material subtypes. Finally, a case study demonstrating the impact of an unexpected source of raw material variation is presented along with recommendations for raw material elemental risk profiling and control.     

Effect of copper variation in yeast hydrolysate on C‐terminal lysine levels of a monoclonal antibody

The ability to control charge heterogeneity in monoclonal antibodies is important to demonstrate product quality comparability and consistency. This article addresses the control of C‐terminal lysine processing through copper supplementation to yeast hydrolysate powder, a raw material used in the cell culture process. Large‐scale production of a murine cell line exhibited variation in the C‐terminal lysine levels of the monoclonal antibody. Analysis of process data showed that this variation correlated well with shifts in cell lactate metabolism and pH levels of the production culture. Small‐scale studies demonstrated sensitivity of the cells to copper, where a single low dose of copper to the culture impacted cell lactate metabolism and C‐terminal lysine processing. Subsequent analytical tests indicated that the yeast hydrolysate powder, added to the basal media and nutrient feed in the process, contained varying levels of trace copper across lots. The measured copper concentrations in yeast hydrolysate lots correlated well with the variation in lactate and pH trends and C‐terminal lysine levels of the batches in manufacturing. Small‐scale studies further demonstrated that copper supplementation to yeast hydrolysate lots with low concentrations of copper can shift the metabolic performance and C‐terminal lysine levels of these cultures to match the control, high copper cultures. Hence, a strategy of monitoring, and if necessary supplementing, copper in yeast‐hydrolysate powders resulted in the ability to control and ensure product quality consistency. © 2017 American Institute of Chemical Engineers Biotechnol. Prog., 33:463–468, 2017     

Elucidating Raw Material Variability—Importance of Surface Properties and Functionality in Pharmaceutical Powders

The purpose of this study is to illustrate, with a controlled example, the influence of raw material variability on the excipient’s functionality during processing. Soluble starch was used as model raw material to investigate the effect of variability on its compaction properties. Soluble starch used in pharmaceutical applications has undergone a purification procedure including washing steps. In this study, a lot of commercially available starch was divided into two parts. One was left intact and the other was subjected to an extra washing step. The two resulting lots were subjected to a series of physical characterization tests typical of those used to qualify raw materials. The two resulting lots gave virtually identical results from the tests. From the physical testing point of view, the two lots can be considered as two equivalent lots of the same excipient. However, when tested for their functionality when subjected to a compaction process, the two lots were found to be completely different. The compaction properties of the two lots were distinctly different under all environmental and processing conditions tested. From the functionality point of view, the two lots are two very different materials. The similar physical testing results but different functionality can be reconciled by considering the surface properties of the powders. It was found that the washing step significantly altered the surface energetic properties of the excipient. The washed lot consistently produced stronger compacts. These results are attributable to the measurably higher surface energy of induced by the additional washing step.     

Biomanufacturing process analytical technology (PAT) application for downstream processing: Using dissolved oxygen as an indicator of product quality for a protein refolding reaction

Process analytical technology (PAT) is an initiative from the US FDA combining analytical and statistical tools to improve manufacturing operations and ensure regulatory compliance. This work describes the use of a continuous monitoring system for a protein refolding reaction to provide consistency in product quality and process performance across batches. A small‐scale bioreactor (3 L) is used to understand the impact of aeration for refolding recombinant human vascular endothelial growth factor (rhVEGF) in a reducing environment. A reverse‐phase HPLC assay is used to assess product quality. The goal in understanding the oxygen needs of the reaction and its impact to quality, is to make a product that is efficiently refolded to its native and active form with minimum oxidative degradation from batch to batch. Because this refolding process is heavily dependent on oxygen, the % dissolved oxygen (DO) profile is explored as a PAT tool to regulate process performance at commercial manufacturing scale. A dynamic gassing out approach using constant mass transfer (k_La) is used for scale‐up of the aeration parameters to manufacturing scale tanks (2,000 L, 15,000 L). The resulting DO profiles of the refolding reaction show similar trends across scales and these are analyzed using rpHPLC. The desired product quality attributes are then achieved through alternating air and nitrogen sparging triggered by changes in the monitored DO profile. This approach mitigates the impact of differences in equipment or feedstock components between runs, and is directly inline with the key goal of PAT to “actively manage process variability using a knowledge‐based approach.” Biotechnol. Bioeng. 2009; 104: 340–351 © 2009 Wiley Periodicals, Inc.     

Consequences of trace metal variability and supplementation on Chinese hamster ovary (CHO) cell culture performance: A review of key mechanisms and considerations

Trace metals are supplied to chemically-defined media (CDM) for optimal Chinese hamster ovary (CHO) cell culture performance during the production of monoclonal antibodies and other therapeutic proteins. However, lot-to-lot and vendor-to-vendor variability in raw materials consequently leads to an imbalance of trace metals that are supplied to CDM. This imbalance can yield detrimental effects rooted in several primary mechanisms and pathways including oxidative stress, apoptosis, lactate accumulation, and unfavorable glycan synthesis. Recent research endeavors involve supplying zinc, copper, and manganese to CDM in excess to further maximize culture productivity and product quality. These treatments significantly impact critical quality attributes and furthermore highlight the degree to which trace metal availability can affect CHO cell culture performance. This review highlights the role of trace metal variability, supplementation, and interplay on key cellular mechanisms responsible for overall culture performance and the production and quality of therapeutic proteins.     

Use of near-infrared spectroscopy (NIRs) in the biopharmaceutical industry for real-time determination of critical process parameters and integration of advanced feedback control strategies using MIDUS control

Control of biopharmaceutical processes is critical to achieve consistent product quality. The most challenging unit operation to control is cell growth in bioreactors due to the exquisitely sensitive and complex nature of the cells that are converting raw materials into new cells and products. Current monitoring capabilities are increasing, however, the main challenge is now becoming the ability to use the data generated in an effective manner. There are a number of contributors to this challenge including integration of different monitoring systems as well as the functionality to perform data analytics in real-time to generate process knowledge and understanding. In addition, there is a lack of ability to easily generate strategies and close the loop to feedback into the process for advanced process control (APC). The current research aims to demonstrate the use of advanced monitoring tools along with data analytics to generate process understanding in an Escherichia coli fermentation process. NIR spectroscopy was used to measure glucose and critical amino acids in real-time to help in determining the root cause of failures associated with different lots of yeast extract. First, scale-down of the process was required to execute a simple design of experiment, followed by scale-up to build NIR models as well as soft sensors for advanced process control. In addition, the research demonstrates the potential for a novel platform technology that enables manufacturers to consistently achieve “goldenbatch” performance through monitoring, integration, data analytics, understanding, strategy design and control (MIDUS control). MIDUS control was employed to increase batch-to-batch consistency in final product titers, decrease the coefficient of variability from 8.49 to 1.16%, predict possible exhaust filter failures and close the loop to prevent their occurrence and avoid lost batches.     

Large scale demonstration of a process analytical technology application in bioprocessing: Use of on‐line high performance liquid chromatography for making real time pooling decisions for process chromatography

Process Analytical Technology (PAT) has been gaining a lot of momentum in the biopharmaceutical community because of the potential for continuous real time quality assurance resulting in improved operational control and compliance. In previous publications, we have demonstrated feasibility of applications involving use of high performance liquid chromatography (HPLC) and ultra performance liquid chromatography (UPLC) for real‐time pooling of process chromatography column. In this article we follow a similar approach to perform lab studies and create a model for a chromatography step of a different modality (hydrophobic interaction chromatography). It is seen that the predictions of the model compare well to actual experimental data, demonstrating the usefulness of the approach across the different modes of chromatography. Also, use of online HPLC when the step is scaled up to pilot scale (a 2294 fold scale‐up from a 3.4 mL column in the lab to a 7.8 L column in the pilot plant) and eventually to manufacturing scale (a 45930 fold scale‐up from a 3.4 mL column in the lab to a 158 L column in the manufacturing plant) is examined. Overall, the results confirm that for the application under consideration, online‐HPLC offers a feasible approach for analysis that can facilitate real‐time decisions for column pooling based on product quality attributes. The observations demonstrate that the proposed analytical scheme allows us to meet two of the key goals that have been outlined for PAT, i.e., “variability is managed by the process” and “product quality attributes can be accurately and reliably predicted over the design space established for materials used, process parameters, manufacturing, environmental, and other conditions”. The application presented here can be extended to other modes of process chromatography and/or HPLC analysis. © 2009 American Institute of Chemical Engineers Biotechnol. Prog., 2010     

Modulation of mAb quality attributes using microliter scale fed‐batch cultures

A high‐throughput DoE approach performed in a 96‐deepwell plate system was used to explore the impact of media and feed components on main quality attributes of a monoclonal antibody. Six CHO‐S derived clonal cell lines expressing the same monoclonal antibody were tested in two different cell culture media with six components added at three different levels. The resulting 384 culture conditions including controls were simultaneously tested in fed‐batch conditions, and process performance such as viable cell density, viability, and product titer were monitored. At the end of the culture, supernatants from each condition were purified and the product was analyzed for N‐glycan profiles, charge variant distribution, aggregates, and low molecular weight forms. The screening described here provided highly valuable insights into the factors and combination of factors that can be used to modulate the quality attributes of a molecule. The approach also revealed specific intrinsic differences of the selected clonal cell lines ‐ some cell lines were very responsive in terms of changes in performance or quality attributes, whereas others were less affected by the factors tested in this study. Moreover, it indicated to what extent the attributes can be impacted within the selected experimental design space. The outcome correlated well with confirmations performed in larger cell culture volumes such as small‐scale bioreactors. Being fast and resource effective, this integrated high‐throughput approach can provide information which is particularly useful during early stage cell culture development. © 2014 American Institute of Chemical Engineers Biotechnol. Prog., 30:571–583, 2014     

Enhanced process understanding and multivariate prediction of the relationship between cell culture process and monoclonal antibody quality

This work investigates the insights and understanding which can be deduced from predictive process models for the product quality of a monoclonal antibody based on designed high‐throughput cell culture experiments performed at milliliter (ambr‐15^®) scale. The investigated process conditions include various media supplements as well as pH and temperature shifts applied during the process. First, principal component analysis (PCA) is used to show the strong correlation characteristics among the product quality attributes including aggregates, fragments, charge variants, and glycans. Then, partial least square regression (PLS1 and PLS2) is applied to predict the product quality variables based on process information (one by one or simultaneously). The comparison of those two modeling techniques shows that a single (PLS2) model is capable of revealing the interrelationship of the process characteristics to the large set product quality variables. In order to show the dynamic evolution of the process predictability separate models are defined at different time points showing that several product quality attributes are mainly driven by the media composition and, hence, can be decently predicted from early on in the process, while others are strongly affected by process parameter changes during the process. Finally, by coupling the PLS2 models with a genetic algorithm first the model performance can be further improved and, most importantly, the interpretation of the large‐dimensioned process–product‐interrelationship can be significantly simplified. The generally applicable toolset presented in this case study provides a solid basis for decision making and process optimization throughout process development. © 2017 American Institute of Chemical Engineers Biotechnol. Prog., 33:1368–1380, 2017