A recent study conducted by the University of South Australia has utilized artificial intelligence to identify a range of metabolic biomarkers that could potentially predict the risk of cancer. By analyzing data from 459,169 participants in the UK Biobank using machine learning techniques, the study discovered 84 features that could indicate an increased likelihood of cancer.
Interestingly, some of these markers also indicated chronic kidney or liver disease, suggesting a possible connection between these conditions and the development of cancer. The researchers emphasized the importance of exploring the underlying mechanisms of these diseases to better understand their potential relationship with cancer.
The study, titled “Hypothesis-free discovery of novel cancer predictors using machine learning,” was carried out by Dr. Iqbal Madakkatel, Dr. Amanda Lumsden, Dr. Anwar Mulugeta, and Professor Elina Hyppönen from the University of South Australia, along with Professor Ian Olver from the University of Adelaide. The findings have been published in the European Journal of Clinical Investigation.
Dr. Madakkatel explained that the study utilized artificial intelligence and statistical approaches to analyze more than 2,800 features and identify potential cancer risk factors. More than 40% of the features identified were found to be biomarkers, which are biological molecules that can indicate either healthy or unhealthy conditions depending on their status. Furthermore, several of these biomarkers were associated with both cancer risk and kidney or liver disease.
One of the key findings of the study was that high levels of urinary microalbumin, a protein found in urine, were the strongest predictor of cancer risk after age. Additionally, indicators of poor kidney function, such as high blood levels of cystatin C and high urinary creatinine, were also linked to cancer risk. The study also revealed that a higher red cell distribution width (RDW), which refers to the variation in the size of red blood cells, was associated with an increased risk of cancer.
Furthermore, the study highlighted the connection between high levels of C-reactive protein, an indicator of systemic inflammation, and increased cancer risk. Elevated levels of gamma glutamyl transferase (GGT), an enzyme related to liver stress, were also found to be associated with higher cancer risk.
Professor Hyppönen, the chief investigator of the study, emphasized the strength of using artificial intelligence for this research. The model was able to analyze thousands of features and identify relevant risk predictors that might not have been apparent through other methods. The researchers believe that with simple blood tests, it may be possible to gain information about an individual’s future risk of cancer, allowing for early intervention and prevention of the disease.
While further research is needed to confirm causality and the clinical relevance of these findings, this study demonstrates the potential of AI in identifying cancer risk factors and has paved the way for future investigations in this field.