Besides serving as tiny factories that churn out valuable biopharmaceuticals, living cells release substances that we call byproducts, process-related impurities, or (less euphemistically) contaminants. Particularly troublesome contaminants include host cell proteins (HCPs). Because they can reduce product quality and even endanger patients, HCPs attract the interest of regulatory agencies. HCPs, the agencies insist, must be analyzed and brought down to acceptable levels.
HCP analysis has traditionally relied on the enzyme-linked immunosorbent assay (ELISA), Western blotting, and other immunoassays. These techniques are sensitive, but they may overlook weakly immunogenic HCPs as well as HCPs for which anti-HCP antibodies are lacking.
To overcome these limitations, HCP analysis incorporates orthogonal methods that can identify and quantify individual HCPs, including low-abundance HCPs that may be overshadowed by high-abundance HCPs. These methods, which include liquid chromatography with tandem mass spectrometry (LC-MS/MS), are being used more frequently, especially during the development and purification stages of biomanufacturing.
Of course, detecting, quantifying, and identifying HCPs are useful activities only if they help analysts establish that monoclonal antibodies and other bioprocess-derived pharmaceuticals are safe and effective. Analysts, then, need to establish whether the HCPs that are present might react with biopharmaceutical products or excipients, or whether they are immunogenic.
What are HCPs?
HCPs are the predominant class of process-related impurities in biomanufacturing. As the name implies, they are comprised principally of endogenous proteins produced as by the host cells during the process of culturing or fermentation. “Even after multiple sophisticated purification steps, some HCPs remain or copurify with the drug protein,” says Olaf Stamm, PhD, senior business development director, Charles River Laboratories.
While most HCPs are considered safe in minute quantities, some have been shown to reduce product quality or even threaten patient safety.1 They may demonstrate enzymatic activity directly in the drug formulation or in the patient, for example, or act as a receptor agonist or antagonist. They can destabilize the product or excipient. Or they can elicit the production of antibodies or act as an adjuvant.
The risk for adverse effects does not necessarily correlate with the levels of HCPs. “Even small traces of certain HCPs,” Stamm warns, “can be highly immunogenic.”
It is incumbent upon the manufacturer to detect and minimize, or ideally eliminate, HCPs in the final product. As such, regulatory guidelines “mandate that impurities should be measured at every step of a biologic’s process development and manufacturing,” points out Lisa Sapp, biopharma market manager, Agilent Technologies. Doing so allows the purification process to be adjusted at each step to reduce the HCPs to a safe level.
ELISA and Western blot
ELISA is still the dominant assay format. “It provides a fast, cost-effective, and quantitative method with good assay sensitivity, that is, with limit of quantification (LoQ) in the ng/mL range,” explains Weihong Wang, PhD, technology development manager, Eurofins Lancaster Laboratories.
Early in the development phase, a commercial HCP kit—raised against a generic Chinese hamster ovary (CHO) line, for example—is typically used. It will generally be acceptable for early clinical studies, but it may become inadequate by Phase III, when regulatory agencies, Wang observes, often demand “a process-specific HCP ELISA in which the HCP antigen and anti-HCP antibody are generated from the same manufacturing process, but with a null/unengineered cell line.”
By itself, an anti-HCP ELISA can reveal only so much. It uses a polyclonal antiserum through the manufacturing process. Individual proteins will not be differentiated. A one- or two-dimensional SDS-PAGE or Western blot is most commonly used to separate the constituents from one another, says David P. Chimento, PhD, vice president of client solutions, Rockland Immunochemicals.
To help with the arduous task of comparing gels, Rockland has collaborated with Cytiva in the development of an HCP coverage assay using differential in blot electrophoresis (DIBE) technology. The companies produced the Amersham HCP DIBE CHO kit, which uses the same antibodies as the Amersham HCPQuant CHO ELISA kit.
DIBE builds on differential in gel electrophoresis (DIGE), fluorescently labeling the HCPs and following up with antibody labeled with a second fluorescent dye. The kit “gives a high level of accuracy in 2D,” says Chimento, “and analysis is infinitely easier.”
Antibody Affinity Extraction (AAE™) is a method developed by Cygnus Technologies to overcome the technical limitations of Western blot–based orthogonal methods such as 2D Western blot or 2D-DIBE in assessing the coverage of a polyclonal antibody to total HCP. In this method, the purified polyclonal antibody is covalently immobilized on a chromatography support. The column is then conditioned to prevent significant leaching of the antibody and to greatly minimize any nonspecific binding.
The HCP sample in its native, nondenatured state is passed over the column for binding and then eluted several times. All HCP elution fractions are pooled and concentrated back to the original sample volume. The final sample is then separated by 2D-PAGE and analyzed by a comparison to a silver stain of starting, unextracted sample. AAE is more predictive of how the anti-HCP antibody will perform in HCP ELISA and provides sufficient sensitivity to evaluate individual HCPs that persist through purification process and may copurify with a drug substance.
One of the limitations of MS is that IgG drug substances often mask HCPs by a factor of 104–106. To improve MS sensitivity, AAE can be used as a sample preparation method to enrich HCPs and eliminate most of the drug substance in a sample. Enrichment of HCPs by AAE and detection by MS are two complimentary orthogonal methods to identify individual HCPs present in a drug substance and facilitate the decision on whether immunoassay is fit for process monitoring and product lot release.
In addition, integration of orthogonal methods for comprehensive HCP analysis provides data throughout drug substance purification to inform process development teams of how to modify their purification processes to improve drug substance purity. Such integration also ensures comprehensive quality control data for regulatory agencies.
Separate and identify
In most instances, regulatory authorities do not require the identification of HCPs. Still, knowing the identity of HCPs can facilitate several activities. These include validation of ELISA platform results, process optimization, and risk assessment.
ELISA and polyclonal immunoreagent-based techniques lack the specificity and coverage to accomplish the identification and quantification of individual HCPs. “HPLC and LC/MS are rapidly becoming the preferred choice for HCP analysis,” notes Sapp. “Although the upfront cost of the LC/MS system is high, reagent costs are low because no antibodies are involved.”
Larger groups with larger budgets (about $20–30,000 per experiment) and the required expertise are incorporating 2D MS into their process development workflows, Chimento adds. “The forward-thinking ones,” he points out, “are doing it tactically,” anticipating that such assays may help with regulatory filings, especially for biosimilars.
Samples can be somewhat unwieldy for many MS protocols. Early in the purification process, samples consist of very heterogenous mixtures, whereas at later stages, there can be as much as a six-log or greater difference in concentration between the product and what is left of the HCP, making the latter more difficult to identify.
“You want to separate the sample’s constituents before they go into the MS instrument,” advises Michelle Busch, scientist, Bioanalytics Characterization, Sanofi. The separation is typically accomplished using liquid or gas chromatography.
“One approach to increase coverage and improve confidence in results is 2D-LC prior to MS/MS,” notes Sapp. Affinity-based methods may precede protocols to help purify and concentrate the sample as well. Researchers have also shown that filtration using a 50-kDa molecular weight cutoff prior to proteomic analysis allows detection of low-molecular-weight HCPs down to a concentration of one part per million.2
Next step
“Untargeted MS lets us discover which HCPs are present, without any prior knowledge and without bias,” explains Busch. Although quantitating a rare protein in an otherwise pure product sample may be difficult, it can be, she remarks, “extremely hard” to identify what that protein is.
To address the issue, Sanofi takes samples from across the purification process and analyzes them. This procedure, Busch says, makes it possible “to identify an HCP where it’s very abundant, and then track that one HCP through your whole process. You can quantitate very low, but you can’t usually get an identification that low.”
“Identifying HCPs in the product sample is just a first step in understanding the impact to the product and to the patient,” Busch continues. “After the identity has been established, it’s necessary to do a risk assessment—you don’t get to just throw it over the wall and be done with it.”
Busch insists that there are pitfalls to be aware of and questions to be answered before a “final understanding” can be reached, “especially if you aren’t a cell biologist.” How is the protein processed? Which forms need to be monitored? Knowing the answers to such questions may change immunogenicity scores.
“The key part,” she elaborates, “is figuring out what each protein does. Will it react with your product or something in the product? Is it immunogenic? You really need the annotations. (We go to UniProt as a website database.) Either the annotations will be in the CHO or you have to go look up the comparable human one and see what this protein does.”
Busch cautions that several proteins in an analysis may have the same name, or that a single protein may have multiple names. “And then there are ways to analyze how similar the CHO sequence is to human,” she notes. “And that’s where you start to figure out the immunogenicity.”
HCP-ELISA is still the required release assay for biopharmaceuticals, with orthogonal assays used along the way to validate them and gain additional information about process and product.
“With MS, an unbiased method has become available to cross-qualify the reagents and the results generated by the traditional immunoassay used in QC release testing,” maintains Stamm.
“Combining high-performance and product- and platform-specific immunoassays with high-resolution MS proteomics illuminates what had been the black box of process/assay development. The risk for adverse effects is minimized or eliminated by this holistic approach.”
References
1. Vanderlaan M, Zhu-Shimoni J, Lin S, et al. Experience with Host Cell Protein Impurities in Biopharmaceuticals. Biotechnol. Prog. 2018; 34(4): 828–837. DOI: 10.1002/btpr.2640. Epub 2018 May 10.
2. Chen I-H, Xiao H, Daly T, Li N. Improved Host Cell Protein Analysis in Monoclonal Antibody Products through Molecular Weight Cutoff Enrichment. Anal. Chem. 2020; 92(5): 3751–3757. DOI: 10.1021/acs.analchem.9b05081. Epub 2020 Feb 12.
