Using Protein Characterization to Support Regulatory Submissions

Published on: 
BioPharm International, BioPharm International-01-01-2021, Volume 34, Issue 01
Pages: 34–37

Clear understanding of what exactly the biomolecule entails is essential.

Functional characterization of protein therapeutics is an expectation of regulatory agencies. The critical quality attributes (CQAs) that impact function and efficacy of the protein therapeutic should therefore be identified and characterized for a regulatory submission. Throughout the development process, notes Khanh Ngo Courtney, director of biologics for Avomeen, bioanalytical testing must be dynamic and tailored to both the protein and the manufacturing approach. “Given the speed of enzyme and protein therapy R&D, rapidly assessing biomolecule efficacy (or potency) and determining an accurate list of CQAs are key,” she states.

Varying data requirements

Before any biologic drug enters into the clinic, it is essential to have a clear understanding of what exactly the biomolecule entails, according to Vu Truong, CEO of Aridis Pharmaceuticals. That includes both structural and physicochemical characteristics.

General structural features that must be described for a protein therapeutic include the primary amino acid sequence, amino acid composition, and any available secondary and tertiary structures, Courtney notes. Peptide mapping can be used to determine the amino acid composition, protein sequence—including N- and C-termini status—post-translational modifications (PTMs), disulfide bridges, and sulfhydryl status, for instance, although additional methods are often needed to fully characterize all these structural elements, according to Andrew Hanneman, scientific advisor for biologics testing at Charles River Laboratories.

The molecular weight profile, isoform distribution patterns, and molar absorptivity are other physicochemical properties that must be determined to support clinical studies and product release.

The specific part of the biomolecule that mediates activity should be identified as well, adds Truong. Stability profile assays must also be conducted to demonstrate under what conditions that activity is preserved and lost due to fragmentation/cleavage, unfolding, and in-vivo immune responses, he says. Information on impurities and the acceptable levels at which they do not adversely impact activity and safety must also be developed.

It must be kept in mind, though, Courtney stresses, that structural and physicochemical properties vary depending on the CQAs for efficacy of each drug substance. For instance, if the drug requires higher-ordered structures, such as oligomerization or other protein binding partners, for activity, showing evidence of quaternary structure might be expected. If certain PTMs are required for a protein to be internalized, such as glycosylation, then elucidation of the glycan structure important for such activity might be helpful in a filing.

There may also be varied requirements for different regulatory authorities, observes Hanneman. “Although drug development processes and submissions are increasingly harmonized among the major markets, there are some differences that need to be understood early in drug development. The characterization methods needed are developed and conducted or validated with knowledge of market-specific guidance and the expected level of oversight and data review,” he says.

MS gains importance

A wide variety of analytical methods are needed to fully characterize therapeutic proteins. “Physiochemical properties are investigated via structural characterization and confirmation studies described in the International Council for Harmonization (ICH) Q6B Guidance (1) initiated in collaboration between FDA, the European Medicines Agency (EMA), and the Japan Pharmaceutical Manufacturers Association (JPMA),” states Hanneman.

In addition to peptide mapping, electrophoresis, high-performance liquid chromatography (HPLC), size-exclusion chromatography, and various forms of spectroscopy (e.g., Fourier transform infrared) are widely used. “The more methods that are employed, especially if they are truly orthogonal, the greater confidence the drug developer and regulators have in the nature and quality of the protein therapeutic,” Truong says. He also notes that by performing a very comprehensive set of analyses, companies set a standard that developers of biosimilars will have to meet, which raises the hurdles.

More recently, nucleic acid sequencing technologies, X-ray crystallography, cryomicroscopy, and mass spectrometry (MS) have attracted increasing attention. MS, says Hanneman, has proven highly flexible in biopharma applications because the soft ionization methods used to ionize large molecules provide a significant opportunity to manipulate large biomolecules like proteins, leading to direct and unambiguous readout of primary and, increasingly, higher order structural elements.

Protein MS, either intact or bottom-up, is commonly used in the characterization of protein therapeutics because it is a single method capable of characterizing many attributes at a time, according to Courtney. “Using high sensitivity instrumentation such as the Orbitrap, one can elucidate batch-to-batch primary sequence consistency, post-translational modifications, such as phosphorylation or glycosylation, and also damages to the protein upon stressed treatment, such as deamidation or oxidation,” she explains. A quadrupole time-of-flight (QTOF) MS system would be useful for verifying accurate intact mass of the biomolecule for use as an identity test, Courtney adds.

Quantitative peptide mapping conducted as a multi-attribute method (MAM), Hanneman agrees, holds significant promise to displace a variety of redundant HPLC and/or electrophoretic methodologies to speed up protein drug development.

Bottom up mass spectrometry is not new, but different ways to use this powerful tool are more frequently employed, Courtney says.

“MS continues to see increased applications, with newer and streamlined approaches used to manipulate large gas-phase ions. MS offers the significant advantage of precise mass measurement (ppm mass accuracy) as its output,” Hanneman agrees. He points to ion mobility and charge detection MS as two innovative methods.

Room for improvement

Some aspects of therapeutic protein characterization still create challenges today. Truong points to demonstration of glycoform consistency and how changes in glycoform profiles can impact safety, efficacy, and immunogenicity as one area where improvements in speed and accuracy are needed.

Potency assays, such as an enzymatic activity assays and cell-based potency assays, also present difficulties, according to Courtney. She notes that both are necessary for characterization of potency and in-vitro cellular efficacy, but cells are intrinsically tricky “reagents” and can be difficult to control and troubleshoot, while, enzymatically, assays can be difficult to develop if the mechanism of the protein is not fully understood.


“These issues lead to higher variability for enzymatic and cellular potency assays than chromatographic methodologies,” Courtney says. The challenge, she adds, is to design such methods so they are fit-for-purpose. “Often these assays are over-designed, resulting in added complexity that leads to increased variability or being under-designed, and, thus, lack sufficient specificity. Achieving the right balance is challenging and the most difficult part of the process,” she explains.

The time it takes to complete cell-based assays is also an issue, according to Truong, but advances in nucleic-acid based methods (next-generation sequencing technologies) should ultimately overcome this difficulty. Similarly, the need to demonstrate genetic (cell-line) stability through numerous passages takes significant time.

Molecular analytics

Advances in analytical technologies are helping to increase the understanding of protein structure and function at the molecular level, providing greater insight into how proteins actually bind to targets and their detailed modes of action. “Elucidating structural and physicochemical properties of proteins and relating them to biologic activity is increasingly possible and at an ever-increasing pace,” Truong asserts.

Courtney refers again to MS as a powerful tool not only for understanding the molecular characteristics of proteins, but also as a read-out method for bioactivity and cellular potency. “If a protein therapeutic can elicit a cellular response, then perhaps a representative biomarker resulting from that response could be monitored by MS using selective ion monitoring or multiple-reaction monitoring,” she observes.

Significant effort has also been underway to use “native mass spectrometry” for direct MS measurement of proteins in their native conformation, including complexes of increasingly higher molecular weight. Improved separation modes allow for direct analysis of native proteins, including charge variant species, according to Hanneman.

There are hurdles to using MS for potency and routine testing, however, including the need for specialized equipment and personnel not usually found in a quality control (QC) laboratory. “I look forward to seeing how mass spectrometers can be simplified as bench-top instruments that aren’t so intimidating for routine use and could be run by non-specialized personnel,” Courtney comments.

Hanneman notes that instrument companies and software developers continue to work closely to provide advanced “push-button” MS data applications under Code of Federal Regulations Title 21 Part 11 (21 CFR Part 11) (2)-compliant electronic data platforms for routine use in QC labs. For instance, MS tool kits include improved workflows and streamlined data analysis approaches for profiling and quantifying impurities, including host-cell proteins (HCP) and other process impurities, which will continue to grow in their implementation.

New formats and improved techniques for conducting in-vitro bioassays are also evolving, including new labeling and imaging approaches, such as innovative time-resolved fluorescence methods and reagents, Hanneman observes. “These tools for measuring target binding typically offer increased signal-to-noise ratio resulting in higher sensitivity and greater confidence in demonstrating specificity,” he says.

New and improved methodologies for aggregate and particle separation, including field flow fractionation (FFF), as well as instruments providing high-definition particle imaging and detection will see increased application, according to Hanneman. These new tools and approaches for detecting and measuring sub-visible particles and determining the size of protein aggregates (as well as viral particles) are undergoing innovation and developments to improve rate of throughput.

Necessary techniques

Regardless of what characterization methods are selected, they are targeted and tuned toward uncovering and reliably measuring drug product CQAs because they ultimately impact protein bioactivity in vivo and safety, including potential immune reactions. The methods chosen, therefore, should provide overlapping and correlative information without “knowledge gaps”, Hanneman asserts.

“Comprehensive protein structure determination ranging from primary to tertiary/quaternary elements currently requires a wide variety of complementary techniques (some truly ‘orthogonal’), and each may fill a different piece of an overall large puzzle. The overlapping data sets then need to be synthesized into an overall description, which can be a complex process for large, heterogeneous, or multimeric protein drugs,” says Hanneman.

Truly comprehensive structural determination remains an elusive goal, however, according to Hanneman, due to protein higher order conformation as well as proteoform heterogeneity within a given sample. “[On a positive note], the development of computational bioinformatics tools and shared database resources go a long way toward simplifying the processing, visualization, and sharing of large protein characterization and comparability data sets among collaborating stakeholders,” he observes.

Consider the intended use

The advantages and disadvantages of various methods and approaches should, in addition, be weighed in terms of the “intended use” of each analytical method and whether each method needs to be validated, according to Hanneman. “Information-rich characterization methods need to be developed and applied as early as possible in a drug development program and used to direct best use of effort, then continually refined (or re-defined) all the way through to drug commercialization,” he says.

Furthermore, with cell and gene therapy programs gaining ground in the biologic marketplace, Hanneman believes the intended use and implementation of protein characterization methods will see changes and refinements in response to the need to detect the presence of genetic material and potentially characterize oligonucleotides as well as to analyze increasingly smaller sample aliquots produced using new and varied manufacturing processes.

Right talent, best practices

Underlying the need to select the right set of methods for each given protein therapeutic is the critical importance of having the right talent working on a project. “People with the right skillsets and experience are needed to design studies that can overcome the various analytical challenges faced with respect to protein characterization,” Courtney states. “The right team,” she says, “will include someone with a fundamental understanding of protein biochemistry, enzyme kinetics, and/or cell biology to enable development of the right model system and a person who is QC-centric to help edit the method to ensure that it is fit-for-purpose.”

The team should also leverage best practices and, wherever possible, a set of analytical methods that have been successfully demonstrated in prior filings, according to Truong. “Evaluation of optimal sets of assays used in past applications can be very informative when selecting assays for a new protein candidate that will afford the best specificity and speed,” he says.

Going forward, adds Hanneman, cloud-based computing and “paperless” corporate infrastructure will greatly improve drug development timelines by supporting faster clearer data sharing. “These developments take significant time and resources to implement but once in place are highly advantageous,” he says.


1. ICH, Q6B Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products, Step 4 version (1999).
2. CFR Title 21, 11 (Government Printing Office, Washington, DC) pp. 110–114.

About the author

Cynthia A. Challener, PhD, is a contributing editor to BioPharm International.

Article Details

BioPharm International
Vol. 34, No. 1
January 2021
Pages: 34–37


When referring to this article, please cite is as C. Challener, “Using Protein Characterization to Support Regulatory Submissions,” BioPharm International 34 (1) 2021.