AI-powered lung cancer atlas identifies cell types linked to higher risk of spread
Recent advances in cancer biology have shown that a tumor is not a uniform mass. It consists of many different cancer cells and is embedded in an environment of immune cells, connective tissue and blood vessels. This variety is a significant factor in determining whether a tumor spreads and how well a therapy works. While single-cell approaches are considered pioneering instruments in cancer…
A groundbreaking single-cell atlas of lung cancer has been developed by researchers from LMU Munich, Charité University Hospital Berlin, and the Berlin Institute for the Foundations of Learning and Data (BIFOLD). This atlas, named LuCaS+, is the result of analyzing tumor tissue from 454 lung cancer patients using single-cell RNA sequencing, which allowed the researchers to examine over 2.5 million individual cells and their gene activity.
The LuCaS+ atlas identifies specific tumor cell types and their interactions with the surrounding microenvironment that contribute to the development of metastases and influence clinical outcomes. Through the integration of this information, the researchers have created a model that significantly improves the prediction of clinical outcomes compared to traditional parameters.
This model accurately identifies patients with a high risk of developing metastases, even in early-stage tumors, where adjuvant therapy is not typically administered.
The study, published in Nature Cancer, also highlights the importance of considering the tumor's microenvironment, i.e., immune, connective tissue, and blood vessel cells, in predicting disease outcomes. By incorporating these profiles, the model can better predict clinical outcomes than conventional parameters alone. The LuCaS+ atlas is now freely available to scientists, aiming to inspire new therapeutic approaches and treatment strategies.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
