Feliza Mirasol is the science editor for 

Feliza Mirasol

The complexity of biologic drug substances requires specialized analytics and the training to do the analyses. It is not enough to know how to run the assays, but also how to interpret the data to give meaningful clinical value to the analyses. To gain insight into methods with which biologics drug substance testing results must be analyzed and interpreted as well better understand the challenges inherit in dealing with complex biological molecules, 

 spoke with Khanh Ngo Courtney, senior director of Biologics at Element, and Mahesh Bhalgat, chief operating officer of Syngene International.

 Is there currently a lack in lab personnel with the specific skill sets to interpret data from biologic drug substance testing? If so, what skills training is in most dire need?

 More and more, we are seeing that the characterization and routine testing of biological therapeutics, particularly of advanced therapeutics such as gene and cell therapies, require advanced analytical techniques combined with a thorough understanding of biology, which requires specialized lab personnel. For example, enzyme-linked immunosorbent assay (ELISA) is becoming a less effective way to quantify host cell protein (HCP) clearance due to emerging data showing that antibodies generated against the HCPs do not provide sufficient host cell proteins coverage. Hence, a more comprehensive analysis of HCPs is required. Liquid chromatography–tandem mass spectrometry (LC–MS/MS) is emerging as a needed analytical method to quantify and characterize HCP clearance. Finding talent who could develop and perform a robust, chemistry, manufacturing, and controls (CMC)-appropriate, quantitative LC–MS/MS method for HCPs is challenging.

Similarly, biological therapeutics require an 

 cellular assay to show potency during characterization and for release and stability testing. The design and development of the appropriate cellular potency methods, including the method for read-out, require personnel who understand the mechanism of action of the therapeutic, and how to exploit the cell biology toolbox to show 

 efficacy of the drug in a robust and CMC-appropriate manner. The appropriate skillset required to perform this type of work is not easy to find in one individual.

 Biological drug substance testing is rapidly evolving, and new platforms and technologies are being used for testing purpose to suit the needs of the assays and drug mechanisms. While many labs and lab personnel are technically sound, there is still a lack of understanding and decision-making on the use of analysis models for biological product characterization and for the mechanism of action (MoA) studies. We find the greatest shortage in laboratory personnel is scientists trained on the statistical designs, analysis, and validation of the qualitative assays.

Additionally, there is a lacuna in getting professionals with biologics analytical training, especially in the structural and functional characterization areas. Apart from the physicochemical area, there is a need to strengthen the knowledge base on analytical ion exchange chromatography and N-glycan analysis for glycosylated biologics.

 What have been some of the major challenges in interpreting and analyzing data generated from drug substance testing of biologic APIs?

Bhalgat (Syngene): Drugs, including APIs and biologics, must be manufactured in compliance with appropriate quality standards. Some of the major opportunitiesassociated with analysis and interpretation of data from testing of biologics include product- and process-related impurity identification and analysis, which is always challenging due to the specialized nature of testing; assay changes during stability studies due to noncovalent interaction-driven oligomer formation; and HCP analysis of in-process samples.

The major challenge lies in interpreting data for its significance when it lacks large data sets generated over a sustained period. Since monitoring and analyzing the trend of the data is important for biologics and other drugs, we apply quality-by-design (QBD) approaches for ensuring data interpretation and comparison needs are met.

 The heterogeneity of the biological molecule impacts how methods are developed, data are generated, and interpreted. Large molecules are tens, if not hundreds, of thousands of small molecules chemically interacting with one another to form tertiary structures, which translate to function. Yet, not all the molecules in a solution of the drug substance are going to be exactly the same—they vary in the amount of post-translational modifications such as glycosylation, phosphorylation, and other chemical modifications as well as endured damages, such as amino acid deamidation, reduction, and oxidation. All of these chemical modifications together impact the structure of the biological molecule, its function, and how it interacts with other molecules in the cell and how it behaves in analytical methodologies.

 What methods or approaches have thus far been successful in generating clinically meaningful data and analyses of that data?

 Meaningful data in the clinic require that the appropriate biomarkers are selected, and the methods for detecting and quantifying said biomarkers are robust and sensitive. The Meso Scale Discovery (MSD) and Quanterix immunological technologies are common methodologies and have been successful in the analysis of clinical samples. LC–MS/MS, if available, remains a powerful tool for the analysis of clinical data for its sensitivity and robustness in detecting and quantifying not only large molecules, but small-molecule biomarkers.

 In clinical research, obtaining meaningful data is built on several factors. Best practices that have been successful in generating clinically meaningful data and have aided the interpretation/analysis of the generated data include drug-specific pharmacokinetic (PK), pharmacodynamic (PD), and immunogenicity assays, rather than generic/off-the-shelf kit assays or methods. Validated methods are also required in assuring assay performance, as per expectations for the set dosage, administration route, and intended clinical interventions. In addition, a careful review of the clinical trial protocol early in assay development, including a two-way dialogue between the clinician and the bioanalytical scientist, helps in setting up fit-for-purpose methods for the relevant trial. Finally, paying special attention to the performance of the method for a particular trial subject population, their age, disease state, ethnicity, etc., helps in generating reliable data.

 Where is there the most unmet need in term of analytical solutions for complex and challenging biologic drug substance testing?

 The bioanalysis of biologics requires the ability to use many different approaches, so there are many unmet needs. For instance, there is a need for high throughput automation and practical biostatistics for complex and challenging biologic drug substance testing. Meanwhile, immunoassays for biologics are labor-intensive and people-dependent. The move to much more automated technology for conducting sensitive and specific immunoassays is a clear opportunity. There is also a need for the availability of assays for biophysical characterization in the presence of various process matrices, host-specific, HCP assay, and non-destructive assay to understand the tertiary structure of drugs at the atomic level (for example, cryo-electron microscopy or single-molecule fluorescence techniques are yet not available for industrial use). Additionally, due tolack of proper guidelines, and there are differences in regulation across different geographies resulting in the development of varying approaches to testing products.

 Highly complex biological drug substances, such as genetic information in the form of viral or nanoparticle encapsulated therapeutics (e.g., lentivirus, adeno-associated virus [AAV], nanoparticles), are extremely difficult analytically due to the heterogeneity in the quality of the drug substance. One example is the analytical challenge of separating and quantifying fill-amounts of viral particles (i.e., empty vs. partial vs. full). When AAV or lentiviral APIs are produced, the drug substance solution will comprise partially of viral particles containing the correct number of copies of genetic information, particles with too much material, particles with too few, and particles containing no material at all. Having the analytical capability and technology to obtain resolution between these different molecular compositions of the drug substance is still an unmet need. Current methods all have their own pitfalls. Analytical ultracentrifugation is expensive, not robust, and requires a significant amount of sample. Transmission electron microscopy is time-consuming and lacks quantitative power. High-performance liquid chromatography has poor resolution. Finding the right analytical solution for the intended purpose is unmet need for many quality-defining methods for advanced therapeutic molecules at the moment.

Articles

Personnel need to be trained in various analytical skill sets for biologic drug substance testing.

Analytical Skills Are Needed in Biologics Testing

Biopharmaceutical analysis is one area in biopharma drug development that requires improvement, namely to manual processes that take up significant amounts of time and labor. Sample preparation remains a major bottleneck point to productivity because of the number of manual steps involved and, in some cases, the outdated technology used to ready samples for analysis.

Glycosylation monitoring is one aspect of biopharma analysis that falls under the strain of laborious manual processes. However, advances made in technology and approaches to automate the workflows for glycosylation monitoring offer quicker sample preparation and can cut down on overall time needed for sample analysis.

Because many recombinant biotherapeutics, such as monoclonal antibodies (mAbs), are glycoproteins, monitoring glycosylation activity is necessary for biopharma drug development. Bioprocessing requires precise control and monitoring of multiple parameters, including cell line stability, product yield, protein folding, and post-translational modifications (PTMs), specifies Ning Zhang, PhD, senior product marketing manager, Chemistry Technology Center, Waters Corporation.

Among the parameters of interest, the host cell’s biosynthesis of glycans is routinely monitored as a critical quality attribute (CQA) because glycan composition and structure are crucial for the biological or clinical activity of the drug. “Slight changes in manufacturing conditions can alter the glycosylation patterns of recombinant proteins and, consequently the biological activity, safety, stability, efficacy, and immunogenicity of the end drug product,” explains Zhang. The manufacturing of therapeutic glycoproteins, therefore, requires careful monitoring and characterization of their glycosylation profile to achieve consistent glycan composition and meet desired quality, clinical safety, and effectiveness targets.

“Protein glycosylation plays an important role in bioactivity, stability, biological half-life, immunogenicity, and pharmacokinetics of a biopharmaceutical,” confirms Radoslaw Kozak, PhD, head of Glycoprofiling at Ludger. Kozak explains that glycosylation is a PTM and, unlike transcription, is a non-template-driven enzymatic modification process. Glycosylation of proteins can therefore change with alterations to production conditions.

Glycosylation could be considered the “canary in the coalmine” for monitoring the consistency and reliability of a manufacturing process, adds Amy Claydon, PhD, Scientific Account Manager, Genedata. She explains that even slight changes to the bioreactor conditions during protein expression can lead to measurable differences in the glycosylation profile.

“The presence, or absence, of glycosylation modulates antibody activity by enhancing either antagonist (blocking, inhibiting) or agonist (activation) functions. Subsequently, glycans may impact the safety, efficacy, structure, and circulating half-life of biotherapeutics; for example, a mAb with no glycosylation will not fold correctly and so will be ineffective as a treatment. This effect on the immunogenicity of a therapeutic antibody increases the risks for patients,” says Claydon.

Claydon further explains that pharmacokinetics, especially clearance and circulating half-life, are also affected by N-glycosylation. For instance, higher levels of sialic acid-containing glycans can increase the circulating half-life because these glycans become more resistant to removal by hepatic asialoglycoprotein receptors.

Dealing with processes in sample preparation for glycosylation monitoring that currently depend on manual labor tends to be a bottleneck point in productivity. To start with, there are several analytical approaches used for the analysis and characterization of protein N-glycans during process development, notes Zhang.

One common approach involves the enzymatic release of N-glycan chains from the protein of interest and their derivatization with a fluorescent label. “A manual procedure, this conventional deglycosylation method requires the incubation of a glycoprotein sample for about one hour, while many researchers generally employ an overnight (16 hours) incubation. This is followed by a lengthy two-to-four-hour labeling step based on reductive amination reactions,” explains Zhang.

The analysis of released N-glycans is challenging because of the number of steps that are involved in preparing the glycoprotein samples and processing the released glycans to prepare them for analysis, adds Richard Gardner, PhD, lead scientist, Ludger. The release of peptide-N-glycosidase F (PNGase F) and the fluorescent labeling of N-glycans, followed by a cleanup, are necessary steps for N-glycans to be analyzed. Gardner points out that these steps are challenging and time consuming when performed manually, especially when the numbers of samples increase beyond a certain amount. “This makes it more difficult for the user to accurately pipette low volume solutions and to efficiently track which samples have been processed,” he states.

Each stage needed for optimizing sample preparation for N-glycan analysis requires on-going evaluation to determine effectiveness, says Claydon. This can subsequently lead to lengthy assay development times if rapid data analysis techniques are not available. “Of course, there are also numerous methods to implement automation (robotic liquid handling, for example) in sample preparation itself, but these often can highlight process inefficiencies further along the process too,” Claydon says.

Screening necessarily large numbers of samples can be time-consuming and monotonous, says Zhang. Furthermore, methods for glycan sample preparation and analysis must be robust and reproducible to ensure accuracy and consistency, she adds.

Automating different aspects of a workflow is one method to alleviate the bottlenecks caused by processes dependent on manual labor. Automated data analysis, for example, can improve many areas of glycosylation monitoring workflows, says Claydon. “When automated data analysis is employed, it can quickly facilitate understanding the effects of different sample preparation optimization steps—the impact of altering digestion conditions and so on. This, in turn, can lead to more rapid improvements to the digestion and labeling protocols during the method development process and an overall increase in assay throughput,” she states.

Bottlenecks often go “hand-in-hand”, Claydon also notes, and solving one bottleneck often leads to another in a different part of the process. “This is why we like to look at data solutions as part of facilitating the whole analytical workflow. In this example, we can easily predict that making sample preparation faster in some way (e.g., automated robotics) can create a lot more data that then needs to be analyzed. So, it is not a simple case of adding automation in just one part of the workflow that ultimately makes the challenges easier.”

Another instance of laborious manual work centers on the fact that many of these workflows have multiple pipetting steps in a range of volumes, which takes time. Automation can save time and allow for more reproducible liquid handling, says Zhang. In addition, automation allows multiple samples to be prepared in parallel, thus improving the throughput.

“Standardized sample preparation is essential to minimize user variations and pipetting errors and to ensure that glycan data is accurate and reproducible. Minimal user intervention—or none at all—allows users to stay away from the bench and work on documentation or setting up an instrument while preparing the samples,” Zhang emphasizes.

Gardner concurs that automation would alleviate this bottleneck by allowing a higher throughput of samples where user intervention is minimized. “Automation would improve the repeatability, reproducibility, and robustness of N-glycan processing by removing the user pipetting steps, thereby decreasing possible human error when transferring the small amounts of liquids that are used in these methods. One outcome would be a harmonized method that many users can operate that removes potential differences in the way users work and interpret SOPs [standard operating procedures],” he notes.

Some technological advances have been helpful in easing the incorporation of automation in the glycosylation monitoring workflow. For instance, specialty sample preparation kits (GlycoWorks RapiFluor–MS, Waters) available commercially and an associated workflow can make fast work of de-glycosylation and labeling. This reduces sample preparation time from a day to less than one hour while enhancing the sensitivity for both fluorescent and mass spectrometry (MS) detections, notes Zhang. “Reagents and reaction conditions that come with these kits have been optimized to enable a fast and complete de-N-glycosylation of glycoproteins and subsequently the rapid and efficient labeling of released N-glycans. Following a HILIC [hydrophilic interaction liquid chromatography]–SPE [solid phase extraction] sample clean-up step, the samples are then ready for analysis by LC–MS [liquid chromatography–mass spectrometry],” Zhang says.

The main technological advance in the field of automation, notes Gardner, is the introduction of robotic liquid handling platforms. Gardner explains that the incorporation of robotic liquid handlers such as Hamilton’s Microlab STAR into glycosylation analysis workflows has allowed many of the processes to be automated, from the release of N-glycans using PNGase F to fluorescent labeling of N-glycans all the way through to preparation of samples for LC–MS analysis.

“Many larger robotic platforms including Hamilton’s Microlab STAR not only incorporate liquid handling and vacuum manifolds but also incorporate plate sealers, heater shakers, and the ability to move plates off deck for centrifuge steps. This ability to incorporate different lab equipment into a robotic workflow enables the move towards full automation of glycosylation analysis. Laboratories require an automated system that is simple to operate, scalable, and repeatable,” Gardner states. To facilitate the move to higher throughout and automated analysis, a range of products have been introduced onto the market that are compatible and adaptable to many robotic platforms, including a 96-sample recombinant PNGase F kit, a 96-well protein binding membrane cleanup plate, a 96-sample procainamide glycan labeling kit, and a procainamide cleanup plate (Ludger).

Meanwhile, Claydon says that automation is best thought of as an end-to-end process, and we must therefore consider how the introduction of automation in one area may move bottlenecks to another part of the overall workflow. “If one considers that an automated data analysis platform actually facilitates both upstream and downstream areas of the entire assay, then it can be a tool to help fully understand where bottlenecks may have moved to, and, in doing so, then accelerate optimization to reduce the time spent in other areas, if they do not significantly affect the end result. For example, long chromatography runs can be avoided if the challenges introduced by co-eluting peaks can be later simplified using software that can display the data two-dimensionally, and deconvolute them,” she states.

“The number one request we get is for help with automating the liquid handling steps, essentially the pipetting of sample preparation protocols,” adds Zhang. “Liquid handling devices now come in several forms including robots (Andrew+ Pipetting Robot, Waters) that automate many sample preparation steps including pipetting which, as everyone knows, is tedious.”

In addition, according to Zhang, robotic pipetting systems can prepare up to 48 samples in under 2.5 hours. Greater automation is achieved by adding connected devices for heating/cooling and a vacuum for sample clean up with SPE.

There are many ways that samples can be prepared for glycosylation analysis, according to Kozak, who explains that some samples may need filtering and/or buffer exchange while other samples may need an enrichment step to extract the glycoprotein to be analyzed. Technologies using 96-well plates that incorporate filters or protein capture technology have allowed this type of processing to happen on automated platforms, Kozak states.

Employing a fully automated liquid handling robot is one innovative approach for automating and streamlining routine sample preparation steps, Zhang adds. This improves both reproducibility and sample throughput. Another innovative approach is by means of a guided pipetting system (Andrew Alliance Pipette+, Waters) that features Bluetooth-connected “smart” pipettes connected to a personal computer or tablet via software for designing, sharing, and executing protocols with audit trails, she also notes.

Meanwhile, the ability to routinely monitor many glycopeptides independently using automated LC–MS analysis offers the ability to trend and compare various aspects of glycosylation, says Claydon. “If one were to observe a change in the sample-to-sample response for specific glycopeptides, for example, the signal for one glycopeptide remains stable but that of a different glycopeptide (or a combination of glycopeptides) gives a varying response, then site-specific information can be derived,” she states.

Sample preparation tends to be manually labor intensive, but automating this step helps streamline the glycosylation monitoring workflow.

Automation Helps Streamline Glycosylation Monitoring Workflow


Vol. 46, No. 1
January 2022
Pages: 38–40

When referring to this article, please cite it as F. Mirasol, “Biologics Testing Highlights Need for Analytical Skills,” 

There is need for training personnel in various analytical skill sets for biologic drug substance testing.

Biologics Testing Highlights Need for Analytical Skills

Advances in Molecule Characterization and Biomanufacturing for Emerging Therapies was part of a keynote series at INTERPHEX 2021. This is Part 2 of 2.

The progression of increasingly complex biological molecules in early drug development poses unique challenges to their analytical characterization, which, in turn, impacts the design and approach of biomanufacturers to the bioprocesses needed to produce these molecules, both in the lab and at scale up/scale out. This session will explore current advances in analytical methodologies and equipment starting at the benchtop and carrying through to process analytics and process development.

Pharmaceutical Technology, Pharmaceutical Technology Europe, and BioPharm International

Gregory Berger, Product Manager, Virus-based Therapeutics—Gene Therapy, Sartorius

Greg is part of the Viral Based Therapeutic product management team at Sartorius bringing a wealth expertise in viral vector/cell line development, vector production in bioreactors and development of appropriate analytical tools for characterization and functionality assessment. Greg obtained his PhD in Virology and Immunology from the Ecole Normale Superieure de Lyon (ENS Lyon) in France, focusing upon the development of new Lentiviral vectors for the development of Dendritic cell-based immunotherapy. He then joined the department of Immunobiology and Infectious Disease at King’s College London as a post-doctoral researcher where his work focused predominantly on viral vector integration and immune cell dysfunction upon transduction. Prior Joining Sartorius, Greg worked at the Cell and Gene Therapy Catapult where he led the development of an End-to-End AAV manufacturing platform to address the current challenges regarding viral vector manufacturing.

Ahmed Besheer, Head Formulation Development, Lonza Drug Product Services

Ahmed Besheer has extensive experience in drug product development, with ten years of industry experience. Before joining Lonza in 2018, Ahmed worked at Novartis and Merck Serono in parenteral formulation development as DP project lead and later as head of NBE formulation development. At Lonza, Ahmed leads a team that develops drug products of monoclonal antibodies, recombinant protein, fusion proteins, bioconjugates, antibody fragments, nanobodies, and viral vectors. He has extensive experience in the formulation of liquid and lyophilized early- and late-phase Drug Product for various administration routes.

Ahmed earned his pharmacy degree from Ain Shams University, Cairo (EG), holds a PhD in pharmaceutical technology from the University of Halle (DE), and worked as an assistant professor at the University of Munich (DE)

Lowell Brady, Head Drug Product Phys-Chem Analytical Development, Lonza Drug Product Services

Lowell Brady leads the Phys-Chem analytical development team at Lonza Drug Product Services, which is responsible for analytical method development and support of specification setting and the analytical control strategy for customer products. Prior to joining Lonza, Lowell worked in analytical and CMC roles at large biotech and pharmaceutical companies. Lowell has a broad range of expertise ranging from in-depth characterization of protein therapeutics using mass spectrometry to the development of routine quality control methods for product release and stability testing. Lowell has experience defining strategies for supporting all stages of product and process development, from early pool/clone screening through BLA filings.

Sébastien Ribault, PhD, Vice-President and Head of End-to-End Solutions, part of the BioReliance® portfolio, MilliporeSigma

Sébastien joined MilliporeSigma in 2005 and is vice-president and head of End-to-End Solutions and the managing director of the Biodevelopment Center located in Martillac, France, where he leads the CDMO branch dedicated to large molecules within MilliporeSigma. Sébastien also supports the development of new and innovative biomanufacturing technologies and solutions in collaboration with the company’s process development team.

Prior to joining MilliporeSigma, he was a gene therapy development scientist at Transgene and was head of the R&D laboratory and member of the board of directors at Hemosystem, an in-vitro diagnostics company. Sébastien received his microbiology engineer degree from the Engineer School of Luminy, Marseille, France and his PhD in molecular and cellular biology from the University of Strasbourg.

Marc Hummersone, Senior Director, Research & Development, Astrea Bioseparations

Marc Hummersone has over 20 years of project and managerial experience in both big and small technology companies, including industry leaders AstraZeneca, General Electric Healthcare, and Cytiva. He has a broad experience base across a wide range of technologies including the development of novel chromatographic media for the purification of complex biological feed streams as well as drug discovery and development. He was a key contributor to the development of the Fibro chromatography family of products which resulted in the acquisition of Puridify, by GE Healthcare in 2017. Additionally, he was a key part of the management team who validated the concept of synthetic lethality and who identified and developed Lymparza, which is an approved drug for the treatment of BRCA-mutated advanced ovarian cancer which resulted in the acquisition of KuDOS Pharmaceuticals in 2005, by AstraZeneca. Marc has held positions including Company Director, Chief Scientific Officer and Department Director.He gained a PhD and Post-Doctoral experience in chemistry which was focused on the total synthesis of complex natural products at University College, London.

Watch other videos in this keynote series:

Equipment Trends in Advanced Solid-Dosage Drug Manufacturing

Serialization on a Global Scale: Deep Dive into the US DSCSA & EU FMD

Part 2—Industry experts from MilliporeSigma, Lonza, Astrea Bioseparations, and Sartorius provide insight and examples about solving critical challenges in both analytics and process development for emerging therapies.

Advances in Molecule Characterization and Biomanufacturing for Emerging Therapies: Part 2

Advances in Molecule Characterization and Biomanufacturing for Emerging Therapies was part of a keynote series at INTERPHEX 2021. This is Part 1 of 2.

Part 1—Industry experts from MilliporeSigma, Lonza, Astrea Bioseparations, and Sartorius discuss key issues affecting the development of emerging biotherapies.

Advances in Molecule Characterization and Biomanufacturing for Emerging Therapies: Part 1 

The use of data has grown increasingly crucial to both the molecular development and manufacturing process development of new biologic drugs. With the growing use of automated platforms that generate vast amounts of data, some in real time, the issue of managing data to arrive at meaningful evaluations is concurrently more challenging.

In the industrial and clinical life sciences laboratory, reliance on manual lab protocols becomes impractical and inconvenient. The growing use of automated platforms in the lab can streamline operational efficiency as well as lend stronger sterility assurance by minimizing the amount of human interference in sampling and running analyses. Another meaningful benefit of automation in the lab is the reduction in time during drug discovery processes or data generation for product characterization. Automation can improve reproducibility, researcher efficiency, clinical translation, and safety (1).

With the use of automation, data generation happens much quicker, and the vast amount of data generated requires more sophisticated data management. Integrating data systems via software systems has become more of a requirement. Software systems such as laboratory information management systems (LIMS), electronic laboratory notebooks (ELNs), sample management solutions, and hybrid systems have been made available to coordinate these data (2). Further, a growing number of laboratories now include automated computer-controlled instruments that generate data in various formats (2).

Although solutions exist for managing the breadth of data, they are limited by the fact that data are often in proprietary formats, making it difficult to connect with other systems. While vendors may sell end-to-end solutions, such integrative systems are typically costly to implement in terms of time, investment, and infrastructure (2). While most commercial systems are designed for bigger labs or production laboratories, for studies conducted in academic research labs, parameters are subject to change due to high staff turnover, which can have an effect on how assays are performed and observational outcomes are reported. Recently, ELNs have been shown to adapt better to a changing environment than LIMS, but they remain an incomplete solution to current data management needs. Particularly in research labs, data still need to be managed and shared even as standardized protocols are being developed (2).

In addition, the use of bioinformatics in defining experimental parameters poses further challenges to data management because data need to be shared prior to concluding study results. What is needed is management of data at both the production-laboratory level and research-laboratory level.

Scientific research has shifted in recent years from being process driven to being data driven, and scientists are now able to infer the results of their studies through data analysis because more data are generated nowadays (3). Managing the data generated from research remains a challenge. “R&D generates enormous amounts of not only operational data, but also high-value scientific data, all stored mostly in isolated databases throughout a vast array of locations and systems,” says Bob Voelkner, vice-president, Sales and Marketing, LabVantage Solutions. “The data, which can be in unstructured and structured formats, are very difficult to manage and maintain,” he confirms.

Voelkner also notes that accessing all the data and combining them into one system can be very complex. “R&D labs are also faced with the struggle of securing distributed system data stores and later finding and using that data,” he states.

The use of automated platforms in laboratories has helped by harmonizing and structuring the data and making them available, Voelkner points out. Making useful data accessible is an important factor in making informed data-driven decisions. Automation alone, however, is not the full solution to lab data management, Voelkner cautions. “There can be potential data integrity issues in accessing the data. An automated platform requires processes and controls around it for a complete solution,” he adds.

There still remain significant challenges to data-driven research, however, starting with the fact that such generating data-rich research is time consuming, consuming as much as 80% of a researcher’s time on mundane data processing tasks (3). Secondly, data exchange is inefficient when reliant on manual curation on one end and manual data extraction at the receiver’s end. Such data exchange processes create inefficiencies and opens opportunities for transcription errors. Finally, a lack of reliability still exists. For instance, with the increasing use of artificial intelligence and machine learning models as a part of decision making, data errors can impact modeling results, potentially compromising reproducibility. A published report has suggested that 50% of deep learning experiments in pharmaceuticals were not reproducible. Of those, 25% were attributed to data integrity (3).

The vast amounts of data now able to be generated via automated lab platforms have made it more vital to create a data management strategy. According to Voelkner, having deep knowledge on how an organization can make the best use of its data, who needs access to that data, and who the data needs to be shared with outside R&D are important considerations. “This will drive which technology and approach is the best solution for a specific organization. Eliminating accessibility issues with siloed data and moving toward a harmonized solution is solved with an analytics solution that can bring value to an organization’s data,” he states.

Not having a well-thought-out data management strategy can become a barrier to lab data workflow, impeding the success of analytical studies. “Knowing where your data is, how your data will be used, and by whom is key to success,” says Voelkner.

Voelkner points out that leading-edge companies have recognized the fact that they have silos of valuable data that are difficult to access and make meaningful use of. He recommends that establishing a centralized laboratory informatics solution that can support an organization’s enterprise R&D workflow and data-capture needs as a first step in tackling data management issues. Next, an organization should utilize a powerful analytics solution that can ingest data from multiple sources. This will allow the organization to consolidate its data into a single data lake, to which business intelligence tools can be applied to provide context and visualization. “Additionally, the analytics tool should apply machine learning and artificial intelligence algorithms to predict possible outcomes and make informed decisions,” Voelkner states.

 8, 571777 (2020).
2. K.A. Hobbie, et al., 

 17 (4) 275–283 (2012).
3. R. Brown, “The Value Of Lab Data Automation To Facilitate Data-Centric Research,” 

The increased flow of data from laboratory studies requires a data management strategy to optimize study success. 