Getting Host Cell Protein Detection Right

Host cell protein (HCP) testing is a cornerstone of biologics quality control, providing critical assurance that process- and product-related impurities are cleared to safe, well-characterized levels. As biologic pipelines diversify and analytical expectations rise, researchers face an increasingly nuanced set of decisions around assay coverage, matrix interference, cross-lab variability, and how to reconcile ELISA results with orthogonal methods like mass spectrometry.

To help researchers work through these questions, we asked three experts to share their insights: Stéphane Martinez, Senior Product Manager, Immunoassay Reagents, Revvity; Peng Zhang, Research Scientist, Raybiotech; and Jenna Frame, Ph.D., Product Marketing Manager, Cygnus Technologies.

Biocompare: What are the most common sources of false negatives in HCP ELISAs, and what workflow steps do labs most often overlook when trying to address them?

Stéphane Martinez, Revvity: “False negatives in HCP ELISAs most commonly arise from the intrinsic limitations of antibody-based detection. These assays rely on polyclonal antibodies raised against a representative HCP mixture, which means that not all proteins present in a given process will necessarily be recognized. If certain HCP species are missing from the immunogen, or if they are weakly immunogenic, they may not be detected in the final assay.

“In addition, HCP profiles are dynamic and can change with upstream conditions, cell line evolution, or downstream purification steps. This can result in the presence of new or enriched proteins that fall outside of the assay’s detection capability. Some HCPs also co-purify with the therapeutic product or interact strongly with it, reducing their accessibility to antibodies. Finally, the dilution steps required to bring samples into the ELISA working range can further reduce the detectability of low-abundance species.

“Together, these factors mean that ELISA results should be interpreted in the context of assay coverage and process knowledge, rather than as an absolute measure of total HCP content.”

Jenna Frame, Cygnus Technologies: “In most cases, a lack of HCP antibody reactivity could contribute to false negative results. Cygnus recommends performing HCP antibody coverage analysis by Antibody Affinity Extraction combined with mass spectrometry (AAE-MS) on the upstream harvest, in-process samples and the final DS to confirm that the chosen ELISA demonstrates coverage to the majority of HCPs present in a given process cell line, including problematic HCPs of concern. If AAE-MS reveals a significant gap in coverage, users can consider generating a custom, process-specific HCP ELISA for their analytical needs.

“Issues with positive controls can alert the user to the potential for false negative readings. Matrix interference is a common source of false negatives. To avoid matrix interference, we suggest that users determine the minimum required dilution for their sample matrix prior to ELISA.”

Peng Zhang, Raybiotech: “As in other ELISA methods, several factors can lead to false-negative results in HCP ELISAs. The specificity of the chosen HCP ELISA assay is one of the most critical issues—this means whether or not it is suitable for the cell source of interest. Depending on the cell line-specific protein expression pattern, culture conditions, and methods, a generic ELISA may not provide 100% coverage of all HCPs. In some rare cases, severe matrix effects, for example, for those biologics that may contain excess lipids, hemoglobin or bilirubin, can also cause false-negative results.

“On the other hand, the possibility of false-positive results needs to be evaluated. For example, recombinant human tissue plasminogen activator (rhTPA) produced in CHO cells may contain endogenous hamster TPA originating from the host cells. Although the hamster and human proteins are not identical, separating the host-derived TPA from the recombinant product can be technically challenging. Hence, a positive result from a generic CHO HCP ELISA does not necessarily indicate failure of protein purification or high HCP contamination, as the outcome researchers examine with HCP ELISAs, but rather the existence of a side product.”

Biocompare: How do you determine whether a generic assay provides adequate coverage for a specific process, and at what point does a lab need to consider a process-specific assay?

Peng Zhang, Raybiotech: “HCP coverage depends heavily on the antibodies used in the ELISA assay. While it’s technically difficult to evaluate those antibodies on the user’s end, one alternative is to compare raw total protein and purified protein from a specific process. A reference HCP can be included for comparison, if desired. In parallel, orthogonal methods such as mass spectrometry and 2D-DIGE can be used alongside to confirm HCP ELISA results.”

Jenna Frame, Cygnus Technologies: “Antibody affinity extraction combined with mass spectrometry is the most sensitive and informative assay to estimate percent coverage of an HCP antibody. AAE-MS also identifies all HCPs in an upstream harvest and HCPs reactive with an HCP antibody, and also provides the molecular weight and isoelectric point of HCPs present in a given process. Comparing data before and after AAE enrichment allows manufacturers to assess fitness-for-purpose, as there is no singular regulatory standard for coverage. It is important to note that an HCP ELISA is considered to be qualified as fit for purpose when all other criteria such as dilution linearity, accuracy, precision, assay range are also met. If your process shows unique HCP profiles and a generic kit cannot be qualified as fit-for-purpose, it is recommended to develop a custom assay to reduce residual risk for pivotal clinical and commercial stages.”

Stéphane Martinez, Revvity: “Generic HCP assays are designed to provide broad coverage across a range of CHO-derived HCP populations and are therefore well suited for early-stage development, when processes are still being optimized and may change frequently.

“Assessing their suitability involves confirming that the antibody population adequately represents the HCP profile of the specific process. This is typically evaluated through coverage studies and by verifying assay performance in relevant matrices using approaches such as spike-and-recovery or dilution linearity experiments. In practice, a generic assay is considered appropriate if it consistently detects HCP clearance trends and provides reproducible data across process steps.

“However, as the process matures and becomes more defined, expectations increase regarding the specificity and relevance of HCP detection. At later development stages, particularly for validation and regulatory submissions, it becomes important to demonstrate that the assay reflects the actual HCP population associated with the process. In such cases, transitioning to a platform or process-specific assay can provide greater confidence that critical or process-related HCPs are effectively monitored.”

Biocompare: What are the most significant sources of variability between labs running the same HCP ELISA, and how do you recommend addressing them?

Jenna Frame, Cygnus Technologies: “The most common source of variability is poor plate loading and/or washing technique. There are several tips that we provide our users to eliminate such issues. Stamping your samples into the ELISA plate with a multichannel pipette allows your samples to have a similar reaction time across the ELISA plate, minimizing positional impact that could cause higher CVs with a slower loading procedure. We also recommend that users avoid washing the ELISA plate too aggressively, as it can cause variable dissociation of the antibody/analyte, leading to higher CVs in your assay.”

Stéphane Martinez, Revvity: “Inter-laboratory variability in HCP ELISA results is a well-recognized challenge and can arise from a combination of reagent, procedural, and sample-related factors. One major contributor is the use of polyclonal antibodies, which can vary between lots and influence both sensitivity and specificity of detection.

“In addition, ELISA workflows involve multiple incubation, washing, and handling steps, all of which can introduce operator-dependent variability. Small differences in timing, washing efficiency, or plate handling can affect signal intensity and overall assay performance. Differences in calibration standards, which are rarely perfectly matched to the sample matrix, can also impact quantification.

“Variations in sample preparation, storage conditions, and dilution practices further contribute to differences between laboratories. For these reasons, minimizing variability typically involves implementing standardized protocols, using well-characterized reagents and controls, and performing bridging or comparability studies when transferring methods between sites.”

Biocompare: When HCP ELISA results conflict with orthogonal methods like mass spectrometry, what are the most likely explanations?

Stéphane Martinez, Revvity: “Differences between HCP ELISA and mass spectrometry (MS) results are expected, as the two techniques rely on fundamentally different detection principles. ELISA provides a measure of total immunoreactive HCPs based on antibody binding, whereas MS quantifies individual HCP species based on their molecular signatures, often in a relative or semi-quantitative manner.

“Because ELISA depends on antibody recognition, any HCP not captured by the antibody pool will not contribute to the measured signal, even if present. In contrast, MS can detect a broader range of proteins, particularly when enrichment strategies are used, although it may be influenced by dynamic range limitations or masking by highly abundant proteins such as the therapeutic product.

“As a result, ELISA may underestimate total HCP content relative to MS when antibody coverage is incomplete, while MS may identify low-level proteins that have minimal impact on the overall ELISA signal. These differences underline the complementary nature of the two methods and support the common practice of using orthogonal techniques to obtain a more comprehensive view of HCP content and composition.”

Jenna Frame, Cygnus Technologies: “Though both assays are quantitative, results from HCP ELISA and mass spectrometry data cannot be directly compared. The units in MS are molar in nature, while ELISA measures the total mass of HCPs in the sample. To convert, we need the molar ratio between peptides and proteins or use the protein molecular weight.

“ELISA continues to be the gold standard method for process monitoring and release testing for HCPs. However, HCP ELISA is a multianalyte ELISA based on polyclonal antibodies and measures total HCP content, without providing HCP identities and number of HCPs in a given DS sample. This concentration is also impacted by assay reactivity to various HCP antigens.

“LC-MS advantages include broad applicability and reduced time required to perform HCP analysis. While the signal is not a direct readout of protein abundance as in the ELISA, it is more direct than ELISA and is not affected by antibody—HCP reactivity. The biggest advantage of LC-MS is the ability to identify and quantify individual HCPs—whereas ELISA produces a cumulative HCP response.

“The LC-MS approach also has limitations. From a technical perspective, LC-MS contains a number of stochastic elements during sample prep and acquisition, including digestion of proteins into peptides, ionization of peptides, and fragmentation of those peptides into ions for peptide sequencing.

“Perhaps most importantly, the limited dynamic range of mass spectrometers prevents the reliable measurement of HCPs in downstream bioprocess samples. This necessitates the use of enrichment methods, which can bias the overall quantification.

“Obtaining accurate LC-MS quantification also necessitates use of standards that simulate the diversity of HCPs in the sample. When quantification is assessed via the standards themselves, this further complicates how well you can actually measure an individual HCP.”

Biocompare: What sample matrix effects are most likely to interfere with accurate HCP quantification, and what mitigation strategies work best?

Peng Zhang, Raybiotech: “Sample matrix interference is not uncommon to encounter when measuring bioprocess contaminants, such as in the use of HCP ELISAs, across different sample types and may result in falsely decreased or elevated analyte concentrations. Although the product protein itself may contribute to this interference, other matrix components—including buffer salts, detergents, pH, solvents, special treatment, endogenous components, and additional formulation additives—can also affect assay performance. Most sample matrix effects can be evaluated through standard spike-and-recovery and dilutional-linearity studies.

“One of the easiest ways to overcome sample matrix interference is to further dilute the sample and dilute it into a more assay-compatible buffer. In cases where diluting samples is not feasible due to assay sensitivity limitations, consider sample processing to remove interfering components, for example, using a purification spin column. Samples that are known to contain excess lipids, hemoglobin, bilirubin or any other potential interferents need to be specifically validated for matrix interference.”

Stéphane Martinez, Revvity: “Sample matrix effects are a significant consideration in HCP analysis and can influence both the accuracy and reliability of ELISA results. High concentrations of the therapeutic protein can interfere with antibody access to HCPs, for example through masking or steric hindrance, potentially reducing signal. Residual components from upstream processing, such as media constituents, buffers, or impurities, may also affect antibody-antigen interactions or assay readout.

“Formulation excipients, including surfactants or stabilizers, can further complicate detection by altering protein interactions or affecting enzyme-based signal generation. In addition, certain HCPs that strongly associate with the product during purification may become less accessible to antibodies, leading to underestimation of their levels.

“To address these effects, it is common practice to evaluate dilution linearity and perform spike-and-recovery experiments to confirm that the assay performs reliably in the specific matrix. Where necessary, additional sample preparation steps or orthogonal analytical methods can be used to ensure that HCP measurements accurately reflect the true impurity profile.”

Jenna Frame, Cygnus Technologies: “Extremes in pH, detergents, organic solvents, high product protein concentration, and high buffer salt concentrations are known interference components. When the levels of analyte are well above the assay Limit of Quantitation (LOQ), dilution of the sample is often the easiest way to overcome interference. If dilution cannot be performed or does not solve the problem, consider performing a buffer exchange of your samples into a more compatible matrix.”