I tested two different HCP ELISA kits on my biopharmaceutical. Why do they give different results and which one is more reliable?
This is an interesting issue that arises quite often.
The short answer is that different HCP ELISAs detect different Host Cell Proteins (HCPs) .
The ELISA assay depends on the specificity of the polyclonal HCP antibodies used. To get the perfect polyclonal HCP antibodies for detection of all HCPs you depend on a “lucky goat”. That is, a goat with the perfect antibody response against all the specific HCPs in your samples .
To explain the problem, let’s first have a look at how HCP ELISAs work:
How ELISAs measure HCP
You can measure the content of HCPs in biopharmaceuticals by ELISA (Enzyme Linked ImmunoSorbent Assay). The results can be used to monitor the consistency of batches and as release tests of clinical GMP batches [1, 2].
The Host Cell Protein ELISA is based on polyclonal HCP antibodies that capture and detect HCPs in the samples .
And this is where you run into trouble!
Because the ELISA results depend highly on the specificity of the polyclonal HCP antibodies .
Generation of polyclonal HCP ELISA antibodies – the demand of a “lucky goat”
So let’s have a look at how you produce antibodies for the ELISA:
The polyclonal HCP antibodies are raised against a null cell sample from the cell line of drug production. Typically an E. coli or CHO lysate containing thousands of proteins.
To raise HCP ELISA antibodies you immunize a range of animals, often goats, rabbits or hens, with the cell line lysate. This triggers their immune system, thus activates multiple B-cells that differentiate into plasma cells that start secreting antibodies against various epitopes in the lysate [3, 4].
Now here’s the problem:
Each goat creates an individual immune response. Therefore, the composition of the polyclonal antibodies varies, both on specificity and affinity, from goat to goat. Thus you need a “lucky goat” in order to generate antibodies with high affinities for the exact HCPs in your specific biopharmaceutical. An individual animal may give a poor immune response to the specific HCPs – and may be considered a “strange bird” [5, 6].
So how do you generate a polyclonal antibody response with a broad epitope specificity?
A good strategy is to fractionate the cell lysate. For instance, you can fractionate the HCPs into a high and low molecular weight fraction. You then immunize different animals with different fractions. This will ensure a higher coverage of the null cell lysate.
Even so, it is not possible to generate antibodies against all the HCPs in a cell lysate. And thus, the ELISA will not monitor all HCPs [1, 2].
What are Jackpot HCPs?
Now here is where it gets interesting:
It turns out that some HCPs are more immunogenic than others, and are effective in hijacking the immune system. Therefore, the B-cells generate a high antibody response towards these HCPs .
This result in an HCP ELISA that is highly sensitive towards these specific “Jackpot HCPs”. This is important because it could lead to a falsely high quantification of HCP content .
Note that the immunogenicity of individual HCPs, and “Jackpot HCPs” vary between goats.
But that’s not all. If you immunize a goat multiple times, its immune response favors certain epitopes. This enhances the generation of “Jackpot HCPs”. It also directs the response towards a more “oligoclonal” antibody response against the most immunogenic subset of HCPs .
Why you should validate the fit-for-purpose of an HCP ELISA for the specific biopharmaceutical
Since the specificity of the polyclonal HCP antibodies is rather coincidental, the different HCP ELISAs will detect and quantify different HCPs. Thus the ELISA measures “immune-equivalents” of the HCP content – and not the exact ng HCP per mg drug substance .
So how do you overcome these challenges? The answer is to combine your ELISA with an orthogonal method like Liquid Mass Spectrometry (LC-MS) [2, 6].
LC-MS analysis of HCPs does not depend on antibodies, and it makes a more precise quantification of Host Cell Protein .
If you’re still with me, you now know that individual HCPs and individual animals can influence the ELISA measurements:
The polyclonal HCP antibodies may not recognize some HCPs, and you may get a false low HCP quantification. And some HCPs might be “Jackpot HCPs” which leads to a false high HCP quantification [1, 2].
So to get back to the initial question: Which HCP ELISA matches your samples best?
The answer is that it relies on the composition of HCPs in the specific biopharmaceutical .
And to validate that the ELISA is “fit-for-purpose”, you should combine it with an orthogonal analysis method, like LC-MS .
Which HCP ELISA matches your samples best?Learn more about Coverage Analysis >>>
 Zhu-Shimoni et al: “Host Cell Protein Testing by ELISAs and the Use of Orthogonal Methods“, Biotechnology and Bioengineering, 2014
 Bracewell et al: “The Future of Host Cell Protein (HCP) Identification During Process Development and Manufacturing Linked to a Risk-Based Management for Their Control“, Biotechnology and Bioengineering, 2015
 Engvall et al: “Enzyme-Linked Immunosorbent Assay, ELISA“, The Journal of Immunology, 1972
 Crowther, J.R.: “Chapter 2: Basic Principles of ELISA“, Methods in Molecular Biologi – ELISA: Theory and Practice, 1995
 Gunawan et al: “Comparison of platform host cell protein ELISA to process-specific host cell protein ELISA.“, Biotechnology and Bioengineering, 2018
 Wang et al: “Host Cell Proteins in Biologics Development: Identification, Quantitation and Risk Assessment“, Biotechnology and Bioengineering, 2009