A new collaborative study led by researchers from The Ohio State University, Carnegie Mellon University, and the University of Pittsburgh has developed a new way to identify and track aging cells in the human lung, offering insights that could reshape how scientists study—and eventually treat—age-related diseases.
As we age, some cells enter a state called cellular senescence—they stop dividing but don’t die. These “zombie-like” cells accumulate throughout the body and release inflammatory signals that can damage tissues and impair organ function. Studies in animal models point to a causal role for senescent cells in many age-related conditions, including cancer, dementia, and chronic lung diseases such as COPD and fibrosis. Despite their importance, senescent cells have been difficult to study because they are relatively rare in healthy tissues and vary widely depending on cell type, location, and disease state. This has made it challenging to pinpoint where these cells arise, how they expand, and how they drive disease.
In this study, published in The EMBO Journal in May, the research team describes an AI-based approach called SenSet to identify senescent cells. The method uses molecular data curated from more than half a million cells across 106 donors to define high-confidence signatures of cellular senescence. Using advanced 3D human lung cultures exposed to agents that trigger cellular aging, such as chemotherapy drugs and radiation, the team showed that SenSet can detect senescent cells across a range of stress conditions. They also found that the molecular features of senescence differ across key lung cell types, including fibroblasts, basal cells, and alveolar epithelial cells.
This work provides one of the most detailed maps to date of aging cells in the human lung. These findings will help guide efforts to understand the causes of diseases such as pulmonary fibrosis, clarify how environmental exposures accelerate lung aging, and improve disease models to better reflect aging human tissue. Because SenSet performs well across multiple cell types, it may also be applied to other tissues to study a broad range of age-related conditions, including frailty, dementia, cancer, cardiovascular disease, and diabetes.