Implementation of chemometrics, design of experiments, and neural network analysis for prior process knowledge assessment, failure modes and effect analysis, scale-down model development, and process characterization for a chromatographic purification of Teriparatide.

Process understanding and characterization forms the foundation, ensuring consistent and robust biologics manufacturing process. Using appropriate modeling tools and machine learning approaches, the process data can be monitored in real time to avoid manufacturing risks. In this article, we have outlined an approach toward implementation of chemometrics and machine learning tools (neural network analysis) to model and predict the behavior of a mixed-mode chromatography step for a biosimilar (Teriparatide) as a case study. The process development data and process knowledge was assimilated into a prior process knowledge assessment using chemometrics tools to derive important parameters critical to performance indicators (i.e., potential quality and process attributes) and to establish the severity ranking for the FMEA analysis. The characterization data of the chromatographic operation are presented alongwith the determination of the critical, key and non- key process parameters, set points, operating, process acceptance and characterized ranges. The scale-down model establishment was assessed using traditional approaches and novel approaches like batch evolution model and neural network analysis. The batch evolution model was further used to demonstrate batch monitoring through direct chromatographic data, thus demonstrating its application for continuos process verification. Assimilation of process knowledge through a structured data acquisition approach, built-in from process development to continuous process verification was demonstrated to result in a data analytics driven model that can be coupled with machine learning tools for real time process monitoring. We recommend application of these approaches with the FDA guidance on stage wise process development and validation to reduce manufacturing risks.

Analytical tools for monitoring changes in physical and chemical properties of chromatography resin upon reuse.

Protein A resins are often reused for multiple cycles to improve process economy during mAb purification. Significant reduction in binding capacity and product recovery are typically observed due to the presence of unwanted materials (foulants) deposited on the resin upon reuse. In this paper, we have used a wide spectrum of qualitative and quantitative analytical tools (particle size analysis, HPLC, fluorescence, SEM, MS, and FTIR) to compare the strengths and shortcomings of different analytical tools in terms of their capability to detect the fouling of the resin and relate it to chromatographic cycle performance. While each tool offers an insight into this complex phenomena, fluorescence is the only one that can be used for real-time monitoring of resin fouling. A correlation could be established between fluorescence intensity and the process performance attributes (like yield or binding capacity) impacted upon resin reuse. This demonstration of the application of fluorescence for real-time monitoring correlated empirically with process performance attributes and the results support its use as a PAT tool as part of a process control strategy. While the focus of this paper is on fouling of protein A chromatography resin, the approach and strategy are pertinent to other modes of chromatography as well.

Process development in the Quality by Design paradigm: Modeling of Protein A chromatography resin fouling.

Protein A chromatography is quite commonly used for capture of monoclonal antibodies from the clarified cell culture broths. Protein A resins are expensive and economic feasibility demands that the resin be reused for 50-300 cycles. Resin reuse is, however, accompanied by resin fouling, impacting both the binding and mass transfer characteristics of the resin. In the present study, we attempt to model the variations in binding and mass transfer characteristics of a commercially available Protein A resin, mAbSelect SuRe™, as a function of resin's reuse. Simplified linear driving force modeling and kinetic modeling of Protein A chromatography step elution cycling data has been successfully used to predict resin performance up to 100 cycles based on fouling data up to 50 cycles. Fouling factor for Protein A resin has been empirically modeled as a function of binding and mass transfer characteristics of resin and the resin's reuse using a combination of Buckingham's π theorem and statistical analysis. The proposed empirical model enables reliable prediction of performance of Protein A resin as well as offers an improved understanding of the underlying mechanism behind the decline in resin performance during fouling.

Protein A chromatography resin lifetime-impact of feed composition.

Adsorbent lifetime during protein A chromatography is not readily predicted or understood, representing a key challenge to be addressed for biopharmaceutical manufacturers. This article focuses on the impact of feed composition on the performance of a typical agarose-based protein A resin across a lifetime of 50 cycles. Cycling studies were performed using three different feed materials with varying levels of feed components including proteases, histones, DNA, and nonhistone proteins. Changes in the process and quality attributes were measured. The DBCs were not seen to vary between conditions although there was a reduction in particle porosity in all cases. Fluorescence spectroscopy and LC-MS/MS were used to identify the contribution and extent of fouling to the observed capacity loss. Residual protein A ligand density and deposition of foulants (HCP, residual mAb, and DNA) varied between the three feed materials. Resins cycled in feed materials containing high concentrations of HCP and histones were seen to have greater extents of capacity loss. The mode of performance loss, capacity loss, or impact on product quality was seen to vary depending on the feed material. The results indicate that feed material composition may be correlated to the rate and mode of resin aging as a basis for improved process understanding. © 2018 American Institute of Chemical Engineers Biotechnol. Prog., 34:412-419, 2018.

Fluorescence based real time monitoring of fouling in process chromatography.

A real time monitoring of fouling in liquid chromatography has been presented. The versatility of the approach has been proven by successful implementation in three case studies with an error <1%. The first application demonstrates the monitoring of protein A ligand density and foulant concentration for assessing performance of protein A chromatography resin during purification of monoclonal antibodies. The observations have been supported from LC-MS/MS studies that were independently performed. The second application involves monitoring of foulant deposition during multimode cation exchange chromatography based purification of human serum albumin. Finally, in the third application, monitoring of foulants during multimodal hydrophobic interaction chromatography of recombinant human granulocyte colony stimulating factor is demonstrated. In all three cases, it is observed that the fluorescence intensity consistently increases with resin reuse as more foulants are deposited over time. The proposed approach can be readily used for real time monitoring of fouling and process control.

Enabler for process analytical technology implementation in Pichia pastoris fermentation: Fluorescence-based soft sensors for rapid quantitation of product titer.

Rapid quantitation of product titer is a critical input for control of any bioprocess. This measurement, however, is marred by the myriad components that are present in the fermentation broth, often requiring extensive sample pretreatment before analysis. Spectroscopy techniques such as fluorescence spectroscopy are widely recognized as potential monitoring tools. Here, we investigate the possibility of using fluorescence of the culture supernatant as a potential at-line monitoring tool to measure the concentration of a recombinant therapeutic protein expressed in a Pichia pastoris fed-batch fermentation. We propose an integrated method wherein both the target protein and total protein concentrations are predicted using intrinsic riboflavin fluorescence and extrinsic fluorescence, respectively. The root mean square error for estimating the concentrations of the target protein (using riboflavin fluorescence) and total protein (using extrinsic fluorescence) have been estimated to be <0.1 and <0.2, respectively. The proposed approach has been validated for two different biotherapeutic products, human serum albumin and granulocyte colony stimulating factor, that were expressed using Mut+ and Muts strains of P. pastoris, respectively. The proposed approach is rapid (1 min analysis time, 10 min total with at line sampling) and thus could be a significant enabler for process analytical technology implementation in Pichia fermentation.

Residual on column host cell protein analysis during lifetime studies of protein A chromatography.

Capacity reduction in protein A affinity chromatography with extended cycling during therapeutic antibody manufacture is well documented. Identification of which residual proteins remain from previous cycles during the lifetime of these adsorbent materials is required to understand their role in this ageing process, but represents a significant metrological challenge. Scanning electron microscopy (SEM) and liquid chromatography mass spectrometry (LC-MS/MS) are combined to detect and map this phenomenon of protein carry-over. We show that there is a morphological change at the surface of the agarose resin, revealing deposits on the polymer fibres increasing with cycle number. The amount of residual host cell proteins (HCPs) by LC-MS/MS present on the resin is shown to increase 10-fold between 50 and 100 cycles. During this same period the functional class of the predominant HCPs associated with the resin increased in diversity, with number of proteins identified increasing 5-fold. This ageing is observed in the context of the product quality of the eluate HCP and protein A leachate concentration remaining constant with cycle number.

Mechanistic understanding of fouling of protein A chromatography resin.

This paper aims to provide a thorough understanding of how fouling of Protein A resin takes place. Binding and mass transport properties of widely used agarose-based Protein A resin, MabSelect SuRe™, have been examined to understand the mechanism of resin fouling. There could be various factors that impact resin fouling. These include product/impurity build-up due to components in the feed material and ligand degradation due to the use of harsh buffers. To unravel their contributions, cycling studies were performed with and without product loading. The results presented in this paper provide a lucid understanding of the causative factors that limit Protein A chromatographic resin lifetime. The capacity fall for protein A resin at the end of 100th cycle due to use of feed material was found to be five times greater than that without using feed material. Compared to the fresh resin, the cycled resin samples shows 24% reduction in particle porosity and 51% reduction in pore mass transfer coefficient. Transmission electron microscopy (TEM) was used to qualitatively monitor accumulation of foulants on the cycled resin. Fouled resin sample contained a dense residue in the interior and exterior of resin particle both as a film at the bead surface and as granules. The surface activation energy increased five times in the case of fouled resin sample. The major event in fouling was identified as the non-specific adsorption of the feed material components on resin, signaling that pore diffusion is the rate limiting step. It is anticipated that these findings will assist in development of a more robust and economical downstream manufacturing process for monoclonal antibody purification.

Analytical characterization of in vitro refolding in the quality by design paradigm: Refolding of recombinant human granulocyte colony stimulating factor.

Protein based therapeutics dominate most pharmaceutical pipelines today. For a therapeutic product to be effective, it is important that it is in its native form as slight modifications have been known to result in significantly different performance in the clinic. When expressed in hosts such as Escherichia coli, formation of inactive insoluble aggregates of proteins popularly known as inclusion bodies occurs in most cases. This necessitates the need for in vitro refolding to generate the native (and active) form of the therapeutic protein. This paper aims to provide an approach to generate a deeper understanding of refolding of a therapeutic protein and then to use it for its optimal production commercially. Recombinant human granulocyte colony stimulating factor has been chosen as the model protein. Seven orthogonal analytical tools have been used to elucidate the refolding process. By strategically using these tools protein refolding has been segregated into a series of well-defined sequence of events, starting from the unfolded random coil and ending with the uniquely folded metastable state. The study also suggests the choice of tools that can be used to monitor each event. We believe that this paper successfully demonstrates an approach to generate deeper understanding of the protein refolding process as per the expectations laid out in the Quality by Design paradigm.

Monitoring Quality of Biotherapeutic Products Using Multivariate Data Analysis.

Monitoring the quality of pharmaceutical products is a global challenge, heightened by the implications of letting subquality drugs come to the market on public safety. Regulatory agencies do their due diligence at the time of approval as per their prescribed regulations. However, product quality needs to be monitored post-approval as well to ensure patient safety throughout the product life cycle. This is particularly complicated for biotechnology-based therapeutics where seemingly minor changes in process and/or raw material attributes have been shown to have a significant effect on clinical safety and efficacy of the product. This article provides a perspective on the topic of monitoring the quality of biotech therapeutics. In the backdrop of challenges faced by the regulatory agencies, the potential use of multivariate data analysis as a tool for effective monitoring has been proposed. Case studies using data from several insulin biosimilars have been used to illustrate the key concepts.

Process development in the QbD paradigm: Role of process integration in process optimization for production of biotherapeutics.

Biotherapeutics have become the focus of the pharmaceutical industry due to their proven effectiveness in managing complex diseases. Downstream processes of these molecules consist of several orthogonal, high resolution unit operations designed so as to be able to separate variants having very similar physicochemical properties. Typical process development involves optimization of the individual unit operations based on Quality by Design principles in order to define the design space within which the process can deliver product that meets the predefined specifications. However, limited efforts are dedicated to understanding the interactions between the unit operations. This paper aims to showcase the importance of understanding these interactions and thereby arrive at operating conditions that are optimal for the overall process. It is demonstrated that these are not necessarily same as those obtained from optimization of the individual unit operations. Purification of Granulocyte Colony Stimulating Factor (G-CSF), a biotherapeutic expressed in E. coli., has been used as a case study. It is evident that the suggested approach results in not only higher yield (91.5 vs. 86.4) but also improved product quality (% RP-HPLC purity of 98.3 vs. 97.5) and process robustness. We think that this paper is very relevant to the present times when the biotech industry is in the midst of implementing Quality by Design towards process development. © 2015 American Institute of Chemical Engineers Biotechnol. Prog., 32:355-362, 2016.

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.

Continuous processing for production of biopharmaceuticals.

The merits of continuous processing over batch processing are well known in the manufacturing industry. Continuous operation results in shorter process times due to omission of hold steps, higher productivity due to reduced shutdown costs, and lowers labor requirement. Over the past decade, there has been an increasing interest in continuous processing within the bioprocessing community, specifically those involved in production of biotherapeutics. Continuous operations in upstream processing (perfusion) have been performed for decades. However, recent development of continuous downstream operations has led the industry to envisage an integrated bioprocessing platform for efficient production. The regulators, key players in the biotherapeutic industry, have also expressed their interest and willingness in this migration from the traditional batch processing. This paper aims to review major developments in continuous bioprocessing in the past decade. A discussion of pros and cons of the different proposed approaches has also been presented.

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.

Analytical QbD: development of a native gel electrophoresis method for measurement of monoclonal antibody aggregates.

This paper presents a quality by design (QbD) based development of a novel native PAGE (N-PAGE) method as a low-cost analytical tool for analysis of aggregates of monoclonal antibodies. Comparability to the present gold standard of SEC has been established. The motivation is the fact that SEC requires relatively expensive equipment and consumables, thus making N-PAGE relevant to those academicians and other small companies involved in early-stage development of biotherapeutics that do not have access to SEC, especially in developing countries. Furthermore, SEC suffers from certain disadvantages including the possibility of secondary interactions between the stationary phase and analyte resulting in higher elution time and therefore underestimation of the analyte size. The proposed N-PAGE method can also serve as an orthogonal analytical method for aggregate analysis. A QbD-based approach has been used for development and optimization of the protocol. First, initial screening studies were carried out with parameters including the running buffer pH, running buffer molarity, gel buffer pH, loading dye, sample concentration, and running voltage. Next, optimization of operating parameters was performed using principles of design of experiments. The final optimized protocol was compared to the traditional SEC method and the results were found to be comparable. While N-PAGE has been in use for protein analysis for several decades, use of N-PAGE for analysis of mAb aggregates with data comparable to SEC such as the case presented here is novel.

Development of a low-cost, high-throughput native polyacrylamide gel electrophoresis (N-PAGE) protocol for lipoprotein sub-fractionation using Quality by Design approach.

Ratio of low density to high density lipoprotein concentration is critical for normal functioning of human body. Deviation in this ratio has been linked to various diseases, many of which are fatal if not diagnosed at early stages. For example, cardiovascular diseases (CVD) have been linked to the level of low density lipoprotein (LDL). Henceforth, detection of the lipoprotein subtractions is crucial for health of an individual. To date, methods like ultracentrifugation, nuclear magnetic resonance (NMR), high performance liquid chromatography (HPLC) and gradient gel electrophoresis (GGE) have been used for separation and identification of lipoprotein types and subtypes. However, these methods are expensive, time consuming and require specialized equipments and expertise. This paper aims to propose a low-cost, high-throughput native polyacrylamide gel electrophoresis (N-PAGE) based protocol for analysis of lipoproteins. Quality by Design (QbD) based approach has been utilized. The initial screening of parameters was followed by a fractional factorial design to optimize the protocol. The lipoprotein subtractions obtained by the optimized protocol were compared with the commercially available and commonly used Lipoprint(®) Lipoprotein Subfractions Testing System from Quantimetrix. The proposed method gave comparable results to those obtained with the commercial system. The proposed method is capable of analysis of up to forty different samples in two hours at a cost of approximately 2$/sample. This is an order of magnitude better than the present cost of 265$/sample when using the commercial system. We think that the proposed method would be of particular interest to the developing and under-developed economies of the world, where this cost differential would be deemed quite significant and would make testing affordable to the majority of the population.