An international team of oncologists has found that cancer-related mutations may remain dormant in the body for years before being activated by factors such as smoking and exposure to air pollution. The researchers identified a protein signature in blood plasma that may signal an increased risk of rapidly developing lung cancer. The finding could eventually help doctors identify changes in the lungs years before conventional screening detects a tumor.
A study published in the journal Cell reports that researchers used machine learning to analyze plasma protein data from more than 48,000 participants in the UK Biobank. They then matched the data with cancer registry records to identify participants who subsequently developed lung cancer.
Protein signature may predict cancer risk
The AI analysis identified 14 proteins in the blood that could predict a future lung cancer diagnosis up to five years before the disease was diagnosed. The researchers validated the protein signature using eight datasets from around the world and found that its levels were higher in people who later developed lung cancer.
The analysis suggested that the signature was not produced directly by tumors. Instead, it appeared to reflect changes in the inflammatory environment of the lungs that occur before cancer develops.
The researchers also found evidence that exposure to air pollution, along with smoking, may accelerate the development of lung tumors.
