Novel Study Identifies Metabolic Biomarkers That Could Predict Cancer Risk
A groundbreaking study conducted by researchers from the University of South Australia (UniSA) has utilized machine learning to identify a range of metabolic biomarkers that could potentially help predict the risk of cancer. The study, titled “Hypothesis-free discovery of novel cancer predictors using machine learning,” analyzed data from 459,169 participants in the UK Biobank and identified 84 features that could indicate an increased risk of developing cancer.
In addition to predicting cancer risk, several of these biomarkers were also found to be associated with chronic kidney or liver disease. This highlights the importance of investigating the underlying mechanisms of these diseases and their potential connections to cancer.
Lead researcher Dr. Iqbal Madakkatel explains, “We conducted a hypothesis-free analysis using artificial intelligence and statistical approaches to identify cancer risk factors among more than 2800 features. More than 40% of the features identified by the model were found to be biomarkers – biological molecules that can signal health or unhealthy conditions depending on their status – and several of these were jointly linked to cancer risk and kidney or liver disease.”
One of the significant findings of the study was that high levels of urinary microalbumin, a serum protein needed for tissue growth and healing, were the highest predictor of cancer risk after age. While urinary microalbumin is an indicator of kidney disease, it is also a marker for cancer risk. Other indicators of poor kidney performance, such as high blood levels of cystatin C, high urinary creatinine, and lower total serum protein, were also linked to an increased risk of cancer.
The study also found that a greater red cell distribution width (RDW), which refers to the variation in the size of red blood cells, is associated with a higher risk of developing cancer. Discrepancies in the size of red blood cells can indicate higher inflammation and poorer renal function, both of which are connected to an increased risk of cancer.
Furthermore, the researchers discovered that high levels of C-reactive protein, an indicator of systemic inflammation, and high levels of the liver stress-related biomarker gamma glutamyl transferase (GGT) were also connected to an increased risk of cancer.
Chief investigator Professor Elina Hyppönen, Director of the Australian Centre for Precision Health at UniSA, emphasizes the strength of the study’s machine learning approach. “Using artificial intelligence, our model has shown that it can incorporate and cross-reference thousands of features and identify relevant risk predictors that may otherwise remain hidden,” Prof Hyppönen says.
While further studies are needed to confirm causality and clinical relevance, this research suggests that relatively simple blood tests could potentially provide valuable information about an individual’s future risk of cancer. This early detection could enable proactive measures to prevent the disease from progressing.