A method found to predict liver disease 16 years before symptoms appear

An international team of scientists has identified five proteins in the blood that allow for highly accurate prediction of the risk of developing a severe liver disease - metabolic-associated fatty liver disease (MAFLD) - long before the first symptoms appear. The researchers reported this at Digestive Disease Week (DDW).

MAFLD is the most common form of liver disease in the world. It is spreading rapidly and can double the risk of premature death.

"Imagine that we can predict the risk of MAFLD many years before the disease begins," noted Dr. Shii Yu, a resident physician at the Department of Gastroenterology at Guangdong Provincial People's Hospital. - "Too often people learn about the disease when symptoms are already interfering with their lives. Our results open new possibilities for early diagnosis and prevention."

Analysis of blood from more than 50 thousand participants from the UK Biobank with subsequent follow-up over 16 years allowed researchers to discover five proteins - CDHR2, FUOM, KRT18, ACY1, and GGT1 - whose levels signaled the risk of developing MAFLD long before its manifestations. The combined level of these proteins allowed predicting the disease with 83.8% accuracy 5 years before onset and with 75.6% accuracy 16 years before. When clinical indicators such as body mass index and physical activity level were added to the analysis, accuracy increased to 90.4% and 82.2% respectively.

The model was also tested on an independent Chinese sample, where it gave similar results. This confirms its stability and applicability to different populations.