ImmunoScape Uses Machine Learning to Accurately Predict Antigen Specificity Based on T-Cell Phenotypes in New Peer-Reviewed Research

×

Error message

  • Notice: Undefined property: stdClass::$ds_changed in eval() (line 16 of /var/www/7/modules/php/php.module(80) : eval()'d code).
  • Notice: Undefined property: stdClass::$ss_search_api_url in eval() (line 20 of /var/www/7/modules/php/php.module(80) : eval()'d code).
  • Notice: Undefined property: stdClass::$tm_title in eval() (line 20 of /var/www/7/modules/php/php.module(80) : eval()'d code).

The research could have profound implications for the use of machine learning in the discovery and development of T-cell based immunotherapies.

Metadata
Provider: 
Business Wire, Inc.Business Wire, Inc.
Retrieved on: 
Thursday, October 19, 2023
Information
Tags: 
Research, General Health, Pharmaceutical, Oncology, Technology, Genetics, Health Technology, Science, Biotechnology, Other Technology, Other Science, Health, Therapy, SARS-CoV-2, RNA, Antigen, TCR, Allele, COVID-19, Machine learning, CMV, Cancer, EBV, Patient, Cell Reports, HLA, Vaccine, Medical imaging
Geotags: 
CALIFORNIA, SINGAPORE, UNITED STATES, NORTH AMERICA, ASIA PACIFIC
Organisation: 
IMMUNOSCAPE
Summary
Key Points: 
  • The research could have profound implications for the use of machine learning in the discovery and development of T-cell based immunotherapies.
  • Machine learning models trained on immune profile features of virus-specific cells were built to predict viral T-cell specificity.
  • Machine learning was used to predict antigen specificity from T-cell phenotypes.
  • Machine learning can be used as a statistically rigorous and unbiased method to accurately predict antigen specificity based on T-cell phenotype information.


ImmunoScape, a biotechnology company focused on next-generation immunotherapies, published new peer-reviewed research that demonstrates the use of machine learning to predict the virus specificity of human T-cells. The research, led by ImmunoScape Scientist Michael Fehlings, Ph.D., was published in the scientific journal, Cell Reports. The research could have profound implications for the use of machine learning in the discovery and development of T-cell based immunotherapies.

The Study

  • The study, which analyzed more than 500 different antigen-specific T-cell responses using high-dimensional mass cytometry and single cell RNA sequencing technologies, identified unique phenotypic profiles of T cells specific for antigens from different virus categories.
  • Researchers performed in-depth profiling of CD8+ T cells binding to CMV, EBV, Influenza and SARS-CoV-2 (COVID-19) derived antigens in peripheral blood cell samples from 114 healthy donors and 55 cancer patients.
  • Machine learning models trained on immune profile features of virus-specific cells were built to predict viral T-cell specificity.

The Findings

  • Unique phenotypes of T cells specific for different virus antigens were identified.
  • Machine learning inferred phenotypic signatures from virus-specific T cells.
  • Machine learning was used to predict antigen specificity from T-cell phenotypes.
  • Machine learning predictions were functionally validated.

The Impact

  • The peer-reviewed research provides a comprehensive and unprecedented phenotypic atlas of human peripheral CD8+ T cells specific for antigens from different virus categories.
  • Machine learning models trained on phenotypic profiles of antigen-specific T cells allow the delineation of phenotypic signatures linked to T-cell antigen specificity.
  • Machine learning can be used as a statistically rigorous and unbiased method to accurately predict antigen specificity based on T-cell phenotype information.

This study opens novel avenues to address the challenges of finding TCRs that can be transformed into therapeutics against different diseases and indications. ImmunoScape has built a discovery engine that offers 360-degree views of lab-validated data from millions of T cells, which serves as the foundation for its machine learning platform. By identifying the characteristics of tumor-reactive TCRs across hundreds of solid tumor targets and multiple HLA alleles, ImmunoScape aims to leverage deep T-cell profiles, linking TCR sequence, phenotypes and antigen information to unlock the faster discovery of novel and efficacious TCR-based therapies across a variety of tumor indications.

“Our models are trained on high quality and functionally-validated data produced by our wet-lab – and the better the data that goes into machine learning, the better the predictions are,” Fehlings said. “This study demonstrates that we are on the right path to find novel and specific T cells and build our TCR discovery pipeline as we extend into the oncology field.”

ImmunoScape developed and trained machine learning models to identify and learn common phenotypic signatures inherent to T cells specific for different antigen categories in the study. With the validation of its platform, ImmunoScape will expand the use of machine learning to identify tumor-specific TCRs to support its pipeline for the development of novel TCR-based therapeutics against solid tumors.

“Our proprietary multi-omics database includes an unparalleled set of T-cell phenotypes and corresponding TCRs against viral, cancer and other antigens and builds the foundation for our computational models,” said Choon-Peng Ng, co-founder and CEO, ImmunoScape. “Our mission is to harness the power of machine learning to accelerate the discovery of cancer-specific T cells while expanding access to more effective therapeutics.”

To learn more about ImmunoScape, please visit https://immunoscape.com/.

About ImmunoScape

ImmunoScape is a biotechnology company focused on the discovery and development of next-generation TCR cell therapies in the field of oncology. The company's proprietary Deep Immunomics technology and machine learning platforms enable highly sensitive, large-scale mining and immune profiling of T cells in cancer patient samples to identify novel, therapeutically relevant TCRs across multiple types of solid tumors. ImmunoScape has multiple discovery programs ongoing and will be progressing towards IND-enabling studies and entry into the clinic. For more information, please visit https://immunoscape.com/.

Copyright: 
Copyright Business Wire 2023