LAUSR

in terms of the identification of accurate biomarkers for a specific cancer type. As we move into an era of personalized and precision medicine that involves deep and multi-omic profiling of each cancer patient, this current limitation can be addressed through large data sets and more advanced machineand deep-learning algorithms, presuming there is a will among all concerned to share the required information more openly than is currently practised7. As the current study has successfully demonstrated, the combination of a range of platforms is inevitably essential for more precise and informed diagnosis of cancers. A recent study utilized a combination of conventional tumour DNA assays, bioinformatics tools, as well as cognitive computational systems for the analysis of deep cancer genomic datasets to identify actionable variants and therapeutic options in a clinically timely manner for individual patients8. Despite the limitations of these studies, such as the need for longer-term patient surveillance, they take a major step forward in highlighting the progress of cancer diagnosis, predicting prognosis and identifying patients that would likely require aggressive therapeutic options. With our aging populations comes the dramatic increase in the incidence of cancer. According to Cancer Research UK, one in two individuals will develop cancer at some point in their lives9. However, through the use of biomarkers such as ctDNA methylation markers and prediction of prognosis, the goal of defeating cancer, or at least turning it into a treatable disease through early detection, is evidently coming within reach. ❐