Aimee Mattei | Director of Immunoinformatics
EpiVax Inc
Aimee Mattei, MS. is the Director of Bioinformatics at EpiVax. She manages a team of analysts and project managers providing in silico immunogenicity risk assessments supporting research, collaborative, and commercial projects. She leads the development of novel in silico methods and tools for assessing the immunogenic risk of peptide impurities, including sequences containing unnatural amino acids. She holds an M.S. in Pharmaceutical Chemistry and has previous experience in the design, synthesis, purification, and characterization of peptides supporting research programs from concept assessment through lead optimization in the endocrinology and oncology therapeutic areas.
Appearances:
Festival of Biologics Day 2 @ 15:00
In Silico Immunogenicity Risk Assessments: AI- Enhanced Prediction of Clinical Immunogenicity Outcomes with EpiVax’s ISPRI Toolkit
The development of biologic therapeutics necessitates comprehensive strategies to assess immunogenicity, a key factor influencing drug efficacy and safety. This presentation will focus on the innovative use of in silico immunogenicity risk assessment tools, particularly the AI-enhanced ISPRI (Interactive Screening and Protein Reengineering Interface) toolkit from EpiVax. ISPRI’s computational methods identify effector and regulatory T cell epitopes and predict anti-drug antibody (ADA) responses, offering insights into minimizing risks during the drug development process.
As biologic formats increase in complexity, assessing the immunogenic potential of various constructs becomes critical. The adaptation of established immunogenicity analysis approaches to evaluate CD4+ T cell epitope content in multifaceted biologic formats, including non-antibody scaffolds and bispecific antibodies, will be discussed. This adaptation offers valuable insights into the distinct characteristics of individual components that may influence immunogenic responses. In a retrospective analysis of monoclonal and bispecific antibodies, ISPRI predictions provided consistent results in a fraction of the time and cost compared to published immunogenicity risk assessments made by combining observations from three in vitro assays and public in silico tools. In addition, nearly all (92%) promiscuous T cell epitopes predicted by ISPRI aligned with peptides identified in MHC-Associated Peptide Proteomics (MAPPS) assays, considerably outperforming traditional public T cell epitope prediction tools.
AI and machine learning (ML) have recently been integrated into ISPRI, significantly enhancing its predictive accuracy by a six-fold increase in the correlation between predicted and observed ADA rates, as well as an 85% reduction in false negatives. The integration of AI into in silico immunogenicity assessments represents a significant leap forward in the biopharmaceutical development landscape. By enhancing the predictive capabilities of existing tools like ISPRI, researchers can make informed decisions early in drug development, ultimately leading to safer and more effective therapeutic options.
Created by
Terrapinn is proud to be a member of isla.
Working together to build sustainable events