Development of new methods to measure and improve HCP ELISA reagent coverage

Protein-based biopharmaceuticals are produced in living host cells, such as Chinese hamster ovary (CHO) cells. Besides the desired product, other host cell proteins (HCPs) are secreted into the supernatant or released through cell lysis. Since these inevitable process-related impurities possibly have immunogenic effects, may destabilize the product or induce adjuvant activity, they need to be removed largely by appropriate purification methods. The most commonly used method to monitor HCP removal over the entire biopharmaceutical manufacturing and to quantify HCPs in the final monoclonal antibody (mAb) product is the Enzyme Linked Immunosorbent Assay (ELISA). This method offers the required sensitivity, sample throughput, selectivity, dynamic range, cost efficiency and ease in application. Reliable HCP quantification relies mainly on the ability of the polyclonal anti-HCP ELISA antibodies to detect the heterogeneous HCP population. In this regard, the ratio of HCPs detected by the polyclonal antibodies (pAbs) relative to the complete HCPs is referred to as coverage. ELISA pAbs are produced by immunizing an appropriate host animal with an HCP standard, containing the vast majority of HCPs present in the current manufacturing process. The resulting pAbs need to be tested regarding their coverage by appropriate analytical methods and for their performance in ELISA based HCP quantification. A major objective in this work was to better understand the influence of the quality of the applied anti-HCP pAbs on their performance in ELISA. The commonly applied method to determine the coverage of HCP-ELISA pAbs, the two-dimensional (2D) western blot (WB), exhibits in general detection gaps. Exact causes for these detection gaps have not yet been clear, but it was suspected that some of these detection gaps originated from inherent technical limitations of the 2D-WB method itself, rather than from a real absence of HCP-specific pAbs. A detailed analysis indicated that the major cause for the detection gaps is the loss of conformational epitopes through protein denaturation during 2D polyacrylamide gel electrophoresis hindering HCP – antibody recognition. Hence, more precise coverage determination can be achieved by using native conditions. An orthogonal method to 2D-WB, the affinity purification based liquid chromatography-tandem mass spectrometry (AP-MS) method, allows pAb – HCP binding under native conditions. Coupled with the sensitivity of MS, it is therefore a promising method to get a deeper understanding of pAb coverage. However, the AP-MS methods used so far have some limitations, which can lead to false positive hits. An optimized AP-MS based method was developed to circumvent the non-specific binding issues to the matrix by separating the immune-complexed HCPs from the matrix prior to digestion. The method proved to identify covered HCPs unambiguously and therefore can enhance the understanding of antibody binding behavior of single HCPs. The analytical toolbox developed within this work is very helpful to shed some light on the details of ELISA antibody coverage and facilitates the selection of the most suitable pAbs for reliable HCP quantification by ELISA. It helps to achieve the other main objective of this work, which is to examine the potential for improving coverage by changing the species or even apply a mixture of species. Thus, the influence of the host animal species on coverage was investigated in detail. Comparison of sheep, goat, donkey, rabbit, and chicken derived anti-CHO-HCP pAbs exhibited similar results for all species besides chicken, which clearly fell short in pAb coverage and ELISA performance. The multispecies pAb mixture achieved slightly improved coverage, but did not perform better in the ELISA for HCP quantification of various samples and therefore provided no clear advantage over the single species pAbs. Overall, 2D-WB to determine coverage should be supplemented by orthogonal methods, which better represent the ELISA conditions for antibody binding and provide additional information. In combination with a performance evaluation of pAbs in ELISAs, this detailed characterization is necessary to make reliable decisions about the suitability of HCP ELISAs, as demonstrated in this work. In turn, the use of appropriate HCP ELISAs helps to increase patient safety by reducing the risk of overlooking individual HCPs or underestimating the total HCP content.