ISPRI-HCP: CHO protein impurity immunogenicity risk prediction for improving biosimilar product development and assessing product interchangeability - ABSTRACT
The identification and removal of host cell proteins (HCP) from biologic products is a critical step in biosimilar
drug development. While the sequence of a biosimilar may be identical to the innovator, the process used to
produce the biosimilar will be different, and as a result, new HCPs may be introduced into the product. Despite
recent improvements to purification processes, biologics that are manufactured in different cell lines and purified
using different processes contain variable HCP impurities, making it necessary to identify and quantify impurities
for each product, be it a reference innovator product or a proposed biosimilar product. In this U01 program, we
propose to develop a predictive model for HCP immunogenicity that can facilitate assessment of clinically
meaningful immunogenicity risk for biologics and assess interchangeability risk between a biosimilar and an
innovator product. We have developed a web-based tool called ISPRI-HCP (formerly called CHOPPI) that
predicts the immunogenic potential of HCP sequences by evaluating T cell epitope count and density, and
relative conservation with other epitopes in the human genome. Building on previous studies of monoclonal
antibody and biologic protein immunogenicity using silico methods and our FDA generic peptide immunogenicity
research experience, we hypothesize that ISPRI-HCP can accurately classify candidate HCP impurities
according to their immunogenicity risk. To address this hypothesis, we will train, test and cross-validate an
immunogenicity prediction model for Chinese hamster ovary (CHO) HCPs using a T cell immunogenicity dataset
generated for 87 common CHO HCP impurities in licensed monoclonal antibody products with our method for
stimulating de novo T cell responses in vitro, named In Vitro Immunization Protocol (Aim 1). We will also test
the quality of the antigen stimulation strategy that is used to stimulate de novo T cell immunogenicity data for
development of the ISPRI-HCP predictive model (Aim 2). This research program will improve the accuracy of
the ISPRI-HCP platform, providing drug developers with a rapid and efficient means to reduce HCP-associated
immunogenicity risk to address this critical quality attribute in biosimilar development.
