Six Genes Can Predict Survival in Acute Myeloid Leukemia Patients

Acute myeloid leukemia patient survival can now be predicted using a six-gene scoring system called PHFScore, according to research published in Science Advances. Researchers at the Centre for Genomic Regulation in Barcelona led the study, combining laboratory experiments on mice with clinical patient records to establish a reliable method for forecasting disease progression.

Six genes that determine acute myeloid leukemia survival

The study identifies six distinct genes that behave consistently across both laboratory mouse models and human clinical data. Three of these genes show heightened activity in patients who experience an unfavorable disease trajectory, while three others show lower activity in those same patients.

The genes involved in the scoring system are CARMIL1, HOPX, and ALDH1A1, which exhibit higher expression levels in patients with poor outcomes. Conversely, CALR, STAR, and PHF19 demonstrate lower activity in those unfavorable cases. The combined output of these six biological markers forms the PHFScore, which sorts patients into distinct survival groups.

Laboratory experiments and clinical patient records

To build the scoring system, researchers alternated repeatedly between mouse laboratory tests and human clinical histories. The five-year project was directed by Luciano Di Croce and Gianni Gamarra at the Centre for Genomic Regulation, with participation from Lars Velten’s single-cell analysis group alongside the center’s genomics and flow cytometry units.

As part of the investigation into cellular differentiation, scientists introduced a leukemia-causing gene into mouse blood cells with and without the PHF19 factor. Cells lacking PHF19 formed more colonies in laboratory conditions and implanted with greater ease when transplanted into mice. Y esto se ve reflejado también en pacientes con bajos niveles de expresión de este factor, explained a researcher, noting that data integration successfully separated patients into severe and favorable prognosis groups.

Simplified diagnostic testing and future clinical validation

Analyzing six specific genes offers a more streamlined and cost-effective approach than conducting broader genetic profiling. Because clinicians already extract bone marrow samples from individuals with suspected acute myeloid leukemia, testing this specific panel can utilize existing diagnostic workflows. Patients with the highest expression levels achieved up to an 85 percent survival rate, compared to less than half in lower expression groups.

Several of the identified genes had never previously been linked to blood cancers, opening new pathways for research. However, the PHFScore remains a research proposal that requires further clinical validation before standard implementation.