AI Flags 'Immune Cold' Lobular Breast Cancer as Potentially Responsive to Immunotherapy
核心洞察
The Allen Institute for AI (搜索)'s AutoDiscovery tool identified invasive lobular carcinoma (搜索) (ILC), long considered unresponsive to immunotherapy (搜索), as having a stronger immune signature than previously known.
The finding was validated across an independent patient dataset and through laboratory analysis of tumor samples, lending significant weight to its potential clinical impact.
ILC affects roughly 15 percent of U.S. breast cancer (搜索) patients each year, and the discovery could open existing immunotherapies to a patient population previously excluded from such treatment.
The Allen Institute for AI (搜索) (Ai2) and the Paul G. Allen Research Center (PARC) at Providence Swedish Cancer Institute (搜索) have announced a surprise discovery: invasive lobular carcinoma (搜索) (ILC), a breast cancer (搜索) subtype long considered unresponsive to immunotherapy (搜索), appears to have a stronger immune signature than previously known. The finding emerged from AutoDiscovery, an open-source experimental AI tool that analyzes massive scientific datasets to uncover new lines of inquiry, and was validated across an independent patient dataset and through a lab analysis of tumor samples.
ILC affects roughly 15 percent of U.S. patients each year. Tumors like these are classified as "immune cold"—the immune system isn't engaging with them, so the drugs that work by unleashing that response have nothing to unleash. The AI's identification of a stronger immune signature in ILC challenges that longstanding assumption and could open existing immunotherapies to a patient population previously excluded from such treatment.
According to Dr. Kelly Paulson, PARC's lead for the Center for Immuno-Oncology: "There's immune therapies that are already available today that we could think about testing on this type of cancer, and we can also use this insight to help develop and apply new forms of immunotherapy (搜索) treatments." She added that it was "exciting" to witness the discovery of something that could be immediately translatable into a new cancer treatment.
How AutoDiscovery Works
Built from Ai2's Asta framework, AutoDiscovery autonomously explores complex datasets to generate and evaluate surprising hypotheses with large language models. It is designed to work collaboratively with scientists, allowing them to guide promising directions, apply domain expertise, and assess which findings should be probed further.
"We are taking away the goal," said Bodhisattwa Majumder, Ai2's senior research scientist. "We're saying that it is open-ended. It doesn't wait for the user to give a goal. It looks at the data and starts figuring it out, [coming] up with its own goal and exploring autonomously."
The tool was applied to The Cancer Genome Atlas, one of the most comprehensive cancer datasets. "Cancer researchers have access to extraordinary datasets, but the challenge is no longer collecting data; it's understanding everything those datasets have to tell us," Paulson said in a statement. "AutoDiscovery helped us identify a promising signal that we may not have otherwise investigated, and from there we were able to validate the finding through additional datasets and laboratory research."
Two Findings, 65,000 Hypotheses
The ILC revelation is the second publicly disclosed hypothesis coming from AutoDiscovery. When the tool was introduced in February, Ai2 revealed an unexpected signal documented in a paper submitted to the Conference on Neural Information Processing Systems (NeurIPS): among patients with a PIK3CA gene mutation, mutations in another gene, TP53 (搜索), occurred less often than chance would predict. Because they rarely occur together, they might serve the same biological function—either that, or cancer cells carrying both couldn't survive.
AutoDiscovery has already generated 65,000 hypotheses by scientists working in domains like oncology, neuroscience, and social science, according to Majumder. "Local deployments like the one Providence Swedish Cancer Institute (搜索) is doing now, will unlock even more discoveries as they begin to apply our tools to their private clinical data."
Paulson noted that before now, AutoDiscovery was validated by Ai2 on multiple different data sets, but wasn't applied specifically to cancer. "When we get an AutoDiscovery run back we get information back on surprising hypotheses, but the system also tests hypotheses that are very likely to be true," she said. "Most of those true hypotheses validate as real, and the data and code are provided. These 'true positives' help to provide support."
From Discovery to Clinical Validation
Details of the discovery are documented in a paper titled "Surprisal-based large language models reveal immunologic insights in breast cancer (搜索)." The team said the manuscript has been submitted to MedRxiv, the most relevant preprint server for the work—it hasn't been peer-reviewed yet—and there are plans to provide it to leading medical and science journals. Researchers suggest that ILC "may warrant broader investigation" in future immunotherapy (搜索) research.
The finding still needs to progress through rigorous clinical trials, testing the hypothesis in larger patient cohorts and comparing outcomes against standard treatments. The role of AI in this journey won't diminish, as it can help design more efficient clinical trials, identify patient subgroups most likely to benefit, and monitor real-world evidence once treatments are approved.
AutoDiscovery Expands to Active Cancer Research
Since its launch, AutoDiscovery has been limited to public research datasets. Its partnership with the Providence Swedish Cancer Institute (搜索) marks the start of a new chapter—the tool is now being used in active cancer research programs at a leading oncology research center.
The Providence Swedish Cancer Institute (搜索) said it's bringing AutoDiscovery into its cloud environment, which will give the AI tool access to the institute's private research and clinical data. Doing so keeps the data secure and ensures that PARC's computational research team handles installation, operation, and support for internal researchers.
"Our early research gave us confidence that AutoDiscovery could complement the way our scientists already work by helping surface hypotheses that thus far have stood up to rigorous validation," PARC's Lead Data Scientist, Zachary Reitz, said in a statement.
Beyond local deployment, Providence Swedish Cancer Institute (搜索) plans to eventually import Ai2's agents locally, enabling them to analyze private clinical data in the hopes of empowering new drug and treatment discoveries.
"Scientific discovery depends on trust, and trust is earned through close collaboration with researchers and rigorous validation of every finding," Peter Clark, Ai2's interim chief executive, said. "We believe partnerships like this will help define how AI is used to accelerate discovery across medicine."