The future of host cell protein (HCP) identification during process development and manufacturing linked to a risk-based management for their control

Abstract

The use of biological systems to synthesize complex therapeutic products has been a remarkable success. However, during product development, great attention must be devoted to defining acceptable levels of impurities that derive from that biological system, heading this list are host cell proteins (HCPs). Recent advances in proteomic analytics have shown how diverse this class of impurities is; as such knowledge and capability grows inevitable questions have arisen about how thorough current approaches to measuring HCPs are. The fundamental issue is how to adequately measure (and in turn monitor and control) such a large number of protein species (potentially thousands of components) to ensure safe and efficacious products. A rather elegant solution is to use an immunoassay (enzyme-linked immunosorbent assay [ELISA]) based on polyclonal antibodies raised to the host cell (biological system) used to synthesize a particular therapeutic product. However, the measurement is entirely dependent on the antibody serum used, which dictates the sensitivity of the assay and the degree of coverage of the HCP spectrum. It provides one summed analog value for HCP amount; a positive if all HCP components can be considered equal, a negative in the more likely event one associates greater risk with certain components of the HCP proteome. In a thorough risk-based approach, one would wish to be able to account for this. These issues have led to the investigation of orthogonal analytical methods; most prominently mass spectrometry. These techniques can potentially both identify and quantify HCPs. The ability to measure and monitor thousands of proteins proportionally increases the amount of data acquired. Significant benefits exist if the information can be used to determine critical HCPs and thereby create an improved basis for risk management. We describe a nascent approach to risk assessment of HCPs based upon such data, drawing attention to timeliness in relation to biosimilar initiatives. The development of such an approach requires databases based on cumulative knowledge of multiple risk factors that would require national and international regulators, standards authorities (e.g., NIST and NIBSC), industry and academia to all be involved in shaping what is the best approach to the adoption of the latest bioanalytical technology to this area, which is vital to delivering safe efficacious biological medicines of all types.

Keywords: biopharmaceuticals; bioprocessing; host cell proteins; process-related impurities.

Figure 1

Overview of the HCP landscape in the context of biologics manufacture. The left side of the figure shows a typical ELISA‐based approach where a summed value of the HCPs detected by the polyclonal antibody used is generated with a probability of error (usually based on a Gaussian distribution as indicated). On the right side, a mass spectrometry driven approach is depicted indicating the far larger datasets of multiple species in a first mass spectra which themselves be further fragmented to create spectra for identification. The differing challenges for these differing approaches are highlighted.

Figure 2

The risk assessment continuum is presented in the context of biopharmaceutical process development. In the risk assessment phase, risk is identified and assessed using prior and existing knowledge to identify which parameters may impact the critical quality attributes (CQA). Then a risk mitigation or reduction action is taken. This could be acceptance of the risk or process modification to ensure risk reduction. Then the risk assessment is repeated to determine if acceptable or not. As indicated, this is all mediated by the organization's analytical capability.

Figure 3

Proposed approach to a summed risk index for a measured HCP profile. In this scenario, RA would be derived from mass spectrometry‐based techniques, and the indices PI, Im, and BA would be based on agreed standards.