The OpenAI Foundation has awarded $40 million to UNC Lineberger Comprehensive Cancer Center to support research aimed at improving the design and effectiveness of personalized cancer vaccines using artificial intelligence.
The funding will support the creation of large-scale biological and clinical datasets that researchers plan to use to train AI models to identify stronger tumor targets and help guide the development of more precise cancer vaccines.
The project is part of the OpenAI Foundation’s newly launched Public Data for Health program, its second science initiative under its Life Sciences and Curing Diseases work. The program is initially supporting more than $125 million in grants to universities and nonprofit organizations to create and preserve high-quality scientific datasets that can be made broadly available to researchers.
Using AI to Improve Personalized Cancer Vaccine Design
The UNC effort will be led by Benjamin Vincent, MD, an immunologist at UNC Lineberger, and Alex Rubinsteyn, PhD, a computational biologist and expert in machine learning and computational medicine at the UNC School of Medicine.
The researchers are developing datasets that could allow AI systems to better determine which tumor antigens should be targeted by personalized cancer vaccines.
Personalized cancer vaccines are designed to stimulate a patient’s immune system to recognize specific molecular features of their tumor. However, identifying the most effective targets remains one of the major challenges in developing these therapies. Rubinsteyn said:
“Personalized cancer vaccines are finally starting to show signs of clinical efficacy, but many fail in the development process because they are too arbitrary.”
He added that the team’s goal is to use rich biological data and artificial intelligence to help improve the design of personalized cancer vaccines.
Initiative for Generative Immunotherapy
According to the OpenAI Foundation, the grant will enable UNC to establish the Initiative for Generative Immunotherapy, designed to generate human biological data needed to improve future generations of cancer vaccines.
Rather than relying primarily on tumor sequencing, researchers plan to directly measure proteins present on the surface of tumor cells as well as patients’ T-cell responses. The resulting datasets are intended to be made openly available so other cancer researchers can build on the findings.
Vincent and Rubinsteyn will analyze hundreds of de-identified tumor tissue and immune-cell samples from three biobanks. The data will then be used to develop improved AI-based approaches for selecting tumor antigens for personalized vaccine therapies.
Clinical Research in Triple-Negative Breast Cancer
The initiative will also extend beyond computational modeling.
UNC researchers plan to compare several cancer vaccine formulations in clinical trials to determine how effectively they generate tumor-specific immune responses in patients with triple-negative breast cancer. Vincent said:
“Selecting tumor antigens that are actually good targets, and understanding the relative potency of vaccine formulations for personalized therapy, are big problems in the field which our work will address.”
The aim is to generate data that can help researchers select stronger targets before potential vaccine candidates reach human testing.
The OpenAI Foundation said that, if successful, the project could allow future cancer vaccine candidates to enter clinical development with a stronger set of targets, potentially improving their likelihood of effectiveness while also increasing what researchers can learn from unsuccessful candidates.
OpenAI Foundation Expands Investment in Health Research
The UNC Lineberger grant forms part of the OpenAI Foundation’s broader push to use AI to accelerate biomedical research.
The Public Data for Health program focuses on funding scientific datasets that may be costly or difficult for individual institutions to develop independently but could provide broad value when shared with the research community.
Jacob Trefethen, Head of Life Sciences and Curing Diseases at the OpenAI Foundation, said:
“AI has enormous potential to help researchers design better cancer treatments, but that progress depends on having the right biological data to learn from.”
According to Trefethen, the UNC project is expected to generate data at a scale that is currently unavailable and make the findings accessible so researchers worldwide can use them to support further cancer research.
If successful, the initiative could provide a new foundation for AI-assisted personalized cancer vaccine development and potentially accelerate the translation of vaccine candidates into clinical studies.
Read further on OncoDaily.
