Developing individually-tailored immunotherapies is one of the highest aspirations of cancer researchers. Now, new work suggests deep-learning technology can accurately predict which cancer-related protein fragments—neoepitopes—are most likely to trigger an immune system response to a specific tumor.
This tool could help in development of personalized immunotherapies and vaccines, according to the team of Johns Hopkins engineers and cancer researchers who developed it. Their study was published last month in Nature Machine Intelligence. It reports that their deep-learning method (BigMHC) can identify neoepitopes on cancer cells that elicit a tumor cell-killing immune response.
“Cancer immunotherapy is designed to activate a patient’s immune system to destroy cancer cells,” says senior author Rachel Karchin, PhD, professor of biomedical engineering, oncology and computer science. “A critical step in the process is immune system recognition of cancer cells through T cell binding to cancer-specific protein fragments on the cell surface.”
There is a lot to be gained here. The global immunotherapy market is estimated to be more than $110B and expected to grow. But immunotherapy response rates, which can be remarkable, are just 15% to 20%.
Neoepitopes are neoantigens with validated immunogenicity. Neoantigens are novel proteins that form on cancer cells. Each patient’s tumor has a unique set of neoantigens, but determining which of them are immunogenic is the key to developing an effective treatment.
“BigMHC has outstanding precision at predicting immunogenic neoantigens,” says Karchin.
“There is an urgent, unmet clinical need to tailor cancer immunotherapy to the subset of patients most likely to benefit, and BigMHC can shed light into cancer features that drive tumor foreignness, thus triggering an effective anti-tumor immune response,” notes study co-author Valsamo “Elsa” Anagnostou, MD, PhD, associate professor of oncology in the Kimmel Cancer Center.
Current methods for identifying and validating immune response-triggering neoantigens are time-consuming and costly, as these typically rely on labor-intensive, wet laboratory experiments. As a result, there are few data to train deep-learning models. To address this, the researchers trained BigMHC, a set of deep neural networks, in a two-stage process called transfer learning.
First, BigMHC learned to identify antigens that are present at the cell surface, an early stage of the adaptive immune response for which many data are available. BigMHC was then fine-tuned by learning a later stage, T-cell recognition, for which few data exist
The researchers tested BigMHC on a large independent data set, and showed that it was better at predicting antigen presentation than other methods. They further tested BigMHC on data from study co-author Kellie Smith, PhD, associate professor of oncology at the Bloomberg Kimmel Institute for Cancer Immunotherapy, and found that it significantly outperformed seven other methods at identifying neoantigens that trigger T-cell response.
The team is now expanding its efforts in testing BigMHC in several immunotherapy clinical trials to determine if it can help scientists sift through hundreds of thousands of neoantigens to filter to those most likely to provoke an immune response.
Karchin and her team believe BigMHC and machine-learning-based tools like it can help clinicians and cancer researchers efficiently and cost-effectively sift through vast amounts of data needed to develop more personalized approaches to cancer treatment. “Deep learning has an important role to play in clinical cancer research and practice,” she says.
