Stopping Breast Cancer with Help from AI
The Department of Defense is backing an effort to use machine learning to find clues about tumor biology.
The U.S. government wants to find out if artificial intelligence can help doctors diagnose and treat breast cancer more effectively.
In an effort to find targeted treatments for particularly invasive types of breast cancer that don’t respond well to existing drugs, the Department of Defense announced this week that it is enlisting the biopharma company Berg Health to use AI for drug discovery. The partnership supports the White House’s Cancer Moonshot initiative to screen up to 250,000 patient samples in search of new biological indicators, or biomarkers, of the earliest signs of cancer. While the death rate from breast cancer has dropped steadily over the past two decades, it remains the second-biggest killer among cancers in U.S. women, according to the National Cancer Institute.
Under the partnership, Berg will have access to the DoD’s Clinical Breast Care Project, a bank of 13,600 samples of both healthy and diseased tissue from nearly 8,000 patients.
Berg will start by sequencing samples from healthy donors and those with various breast cancer subtypes, which will generate genomic and other information on the mutations, proteins and cellular processes present in cancerous and healthy cells. That data will then be combined with patients’ known medical histories and fed into Berg’s AI-based platform, which will produce different models of healthy and diseased tissue using trillions of data points. The platform’s algorithms will then help spot patterns—hot spots or hubs—in molecular signatures across these models. Such patterns could represent biomarkers or drug targets.
Berg starts with data and allows the data to generate hypotheses—the reverse of the process common in drug discovery, says Niven Narain, Berg’s cofounder, president, and CEO.
Startups including AtomWise, Insilico Medicine, and TwoXAR are taking similar approaches, using custom-built AI platforms to help eliminate some of the guesswork involved in traditional drug discovery.
Narain believes there are other subtypes of breast cancer that researchers have not identified yet, and he hopes Berg can help identify those as well as drug targets for known subtypes. Key biomarkers discovered under this collaboration could lead to a blood test for breast cancer, a much less invasive procedure than the biopsies required today.
Targeted cancer therapies like the breast cancer treatment trastuzumab, known as Herceptin, have shown incredible promise, but they don’t work on all patients because they are designed for specific genetic mutations within tumors. About 25 percent of breast cancer patients have a subtype known as HER2-positive disease, which can in some cases be treated effectively with Herceptin in its early stages. But not all patients on Herceptin respond to the drug, an indication that other biological factors are in play, Narain says.
Berg has already used its AI platform to identify and advance an experimental drug that has the potential to slow or reverse cancer cell growth by changing a driver believed to be involved in different types of cancer. The investigational drug is currently in a phase II clinical trial for advanced pancreatic cancer, in combination with a common cancer drug.
Couldn't get to Cambridge? We brought EmTech MIT to you!Watch session videos here