Researchers at the Massachusetts Institute of Technology identified that blocking the enzyme caspase-1 with an existing human-tested drug reduced lung tumor risk in mice, offering a potential early prevention strategy for patients at high risk of developing lung cancer.
Targeting Caspase-1 to Block Early-Stage Lung Cancer
Scientists at the Massachusetts Institute of Technology traced how lung tumors form in their earliest stages, focusing on the biological mechanisms that drive inflammation. By targeting a specific enzyme, researchers demonstrated a way to intercept the disease before it establishes itself in tissue.
In laboratory experiments on mice, animals treated with a caspase-1 blocker before tumors appeared developed fewer and smaller lesions than untreated animals. When researchers combined the enzyme inhibitor with an antibody against IL-1 beta, nearly 20 percent of the mice avoided tumor development entirely.
If one observes the cancer deaths in the world, lung cancer causes the majority, and much of that is driven by smoking. In addition, also appear cases in people who never smoked. You can imagine a future in which you get a test and, if you are considered high risk, receive preventative medicine. This concept is called cancer interception and could help millions of people. Sangeeta Bhatia, professor of Health Sciences and Technology and Electrical Engineering and Computer Science at the MIT, and senior author of the study
Tracing the Biological Origins of Early Inflammation
The MIT study builds on a trail of clinical clues left by earlier research. In 2017, a clinical trial run by Novartis known as CANTOS evaluated whether an anti-inflammatory drug could reduce strokes and heart attacks. The trial unexpectedly revealed lower rates of lung cancer among patients treated with an antibody blocking IL-1 beta, a protein that regulates immune response.
Subsequent studies confirmed that while the IL-1 beta antibody had minimal effect on patients who already had established lung cancer, it retained potential for halting progression in high-risk individuals. A recent study from the laboratory of Charles Swanton in the Francis Crick Institute also identified a set of proteins that could help predict which patients might respond to treatment with an IL-1 beta antibody. The MIT team moved further back in the biological chain to examine proteases—enzymes that cut other proteins to activate inflammation—knowing that IL-1 beta requires cutting by a protease to become its mature, active form.
From there, the group led by Sangeeta Bhatia set out to determine which of those enzymes participate most intensely during the early development of lung cancer. For several years, the laboratory of Sangeeta Bhatia has developed tools to track and visualize proteases and their role in cancer, helping tumor cells leave their origin site by cutting extracellular matrix proteins while also intervening in inflammatory cell migration.
Combining Prevention Research with Advanced AI Screening
Preventative strategies arrive as researchers also tackle the challenge of identifying high-risk patients earlier. Lung cancer is the No. 1 deadliest cancer in the world, resulting in 1.7 million deaths worldwide in 2020, killing more people than the next three deadliest cancers combined. Florian Fintelmann, MGCC thoracic interventional radiologist and co-author on the new work, notes that lung cancer is the biggest killer because it is relatively common and hard to treat once advanced. While early detection yields a five-year survival rate closer to 70 percent, advanced detection drops that rate to just short of 10 percent.
While low-dose computed tomography (LDCT) scans represent the most common screening method, researchers at MIT’s Abdul Latif Jameel Clinic for Machine Learning in Health (Jameel Clinic), Mass General Cancer Center (MGCC), and Chang Gung Memorial Hospital (CGMH) developed an artificial intelligence tool named Sybil. The name Sybil traces its origins to the oracles of Ancient Greece, known as sibyls, who relayed divine knowledge of the unseen and omnipotent past, present, and future.
Sybil analyzes LDCT image data without the assistance of a radiologist to predict the risk of a patient developing a future lung cancer within six years. Published in the Journal of Clinical Oncology, researchers demonstrated that Sybil obtained C-indices of 0.75, 0.81, and 0.80 over six years from diverse LDCT scans taken from the National Lung Cancer Screening Trial (NLST), Mass General Hospital (MGH), and CGMH, respectively. Co-author Peter Mikhael, an MIT PhD student in electrical engineering and computer science and affiliate of Jameel Clinic and the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), likened building the 3D model to trying to find a needle in a haystack. Co-author Jeremy Wohlwend, an MIT electrical engineering and computer science PhD student and Jameel Clinic and CSAIL affiliate, and co-author Lecia V. Sequist, a medical oncologist, lung cancer expert, and director of the Center for Innovation in Early Cancer Detection at MGH, contributed to the work alongside Florian Fintelmann.
