Cancer: Award goes to researcher behind the ‘cancer glucose monitor’ that can predict relapses

Liquid biopsy seeks to provide answers by analysing fragments of circulating tumour DNA (ctDNA) in the blood. In some haematological cancers, monitoring residual disease is already standard clinical practice. In solid tumours, however, the field is still at an earlier stage of development.

The innovation Landau is working on is to make this monitoring increasingly sensitive, accessible and suitable for large-scale use. At present, many tests require a genetic analysis of an individual patient’s tumour in order to develop a personalised test capable of detecting mutations in the blood. However, this is a sophisticated and costly process, and therefore not feasible on a large scale.

The new frontier lies in using standardised platforms that can be easily applied to a large number of patients, rather than developing a bespoke test for each individual patient. However, the reliability of these tools across different cancers and clinical situations remains to be demonstrated, and, above all, it remains to be seen whether basing treatment decisions on their results provides a real benefit in terms of survival and quality of life.

The second key player in this revolution is AI. A test capable of collecting extensive genomic information, in accordance with standardised procedures, could feed into large databases. And algorithms, trained on this information and on clinical outcomes, could learn to recognise increasingly early signs of disease progression, resistance or recurrence.

This is the ‘virtuous circle’ of data: the more high-quality information that is collected, the greater the models’ ability to identify biological patterns that are difficult to detect using traditional methods. But even here, having large amounts of data at one’s disposal does not guarantee accurate predictions: independent validation, rigorous checks and prospective clinical trials are required.

But Landau goes even further. In the future, molecular blood tests could help not only to monitor the progression over time of a tumour that has already been diagnosed, but also to identify people at particularly high risk of developing one. This would make it possible to tailor screening programmes and, in selected cases, to consider preventive measures before the disease develops.

Moving from treating cancer to detecting it before it appears. This is the most fascinating prospect, but also the one that calls for the greatest caution. Identifying molecular markers associated with ‘risk’ is not the same as predicting with certainty that a tumour will develop in the future. And offering large-scale screening to the general population, without having demonstrated its accuracy and usefulness, could lead to false alarms, unnecessary tests and unnecessary treatments.

This is why the ‘cancer glucose meter’ is not a device you can buy from a chemist’s, ready for home use, nor is it a universal test capable of diagnosing any type of cancer. The aim is to transform cancer monitoring from a series of check-ups spaced out over time into a molecular assessment carried out much more frequently, in a personalised manner. Perhaps not too far in the future: Landau envisages that the convergence of molecular biology and artificial intelligence could transform medicine over the next 10–15 years.

The next revolution in oncology will lie in getting to know the tumour so well that we can anticipate it, rather than chasing after it.