Promising new blood test can detect 50 cancers, scientists say

Research indicates the test could even reveal location of disease with impressive reliability

Researchers have published their findings in the latest issue of Annals of Oncology. Getty Images

A simple blood test able to detect some cancers even before symptoms appear has been developed by researchers.

The test looks set to revolutionise the treatment of the disease, where the chances of recovery often depend on early detection.

Based on studies of samples from thousands of patients with more than 50 different types of cancer, the test revealed the existence and even location of cancers with impressive reliability.

Reporting their findings in the latest issue of Annals of Oncology, researchers found the test has a false positive rate of less than 1 per cent, making it highly unlikely to give false results for patients free of the disease.

The test works by detecting tell-tale signs of cancer in fragments of DNA from dead cells circulating in the blood.

Our results show that this approach to testing cell-free DNA in blood can detect a broad range of cancer types at virtually any stage of the disease

The behaviour of all cells is controlled by their genes. When cells become cancerous, these instructions become corrupted, making them proliferate uncontrollably.

Until now, attention has focused on developing tests that detect cancerous changes in the DNA making up the genes of normal cells.

But as well as changes in their DNA, cancer cells have different arrangements of chemicals called methylation patterns that affect how the cells behave. Detecting these differences is the key to the reliability of the new test.

“Our previous work indicated that methylation-based tests outperform traditional DNA-sequencing approaches to detecting multiple forms of cancer in blood samples,” said Dr Geoffrey Oxnard of the Dana-Farber Cancer Institute, Boston, and one of the study’s authors.

The researchers exploited this insight using an artificial intelligence technique known as machine learning.

Early detection of cancer can have a significant impact on chances of recovery. The National

They took blood containing fragments of DNA and its methylation patterns from thousands of patients with specific cancers as well as from healthy patients.

The results were then analysed using a computer that used algorithms to discover the chemical “fingerprint” that reveals whether a cell has become cancerous.

The results published by the team are striking. Of the dozen cancers responsible for nearly two-thirds of cancers in the US, the test failed to detect fewer than 1 in 10 cases when cancer was present.

Even at the earliest stages of the disease, the test correctly detected more than a third of all cases.

For lung and breast cancer – the most common forms among men and women respectively in the UAE  – the test successfully identified around half of cases even in their early stages.

The test can also identify the organ or tissue where the cancer originated, as revealed by the nature of the cell that has become cancerous.

When cancer was detected, the test correctly identified the original site of the disease in more than 90 per cent of cases. This information is key for clinicians planning treatment.

“Our results show that this approach to testing cell-free DNA in blood can detect a broad range of cancer types at virtually any stage of the disease” said Dr Oxnard.

The test is now the focus of a large clinical trial involving more than 6,000 patients across the United States.

This will include cancer survivors and others who are at an increased risk of developing the disease.

The test could prove to be a landmark in the treatment of cancer, which is the second leading cause of death globally after cardiovascular disease, and responsible for around 10 million deaths each year.

Early diagnosis can transform the chances of successful treatment. The test is likely to first see use among patients already suspected of having some form of cancer. Its potential as a routine screening tool for family doctors remains to be seen.

Robert Matthews is Visiting Professor of Science at Aston University, Birmingham, UK