New blood test detects more than 50 types of cancer, including at early, more treatable stages

By Sally Robertson, B.Sc.Mar 31 2020

Researchers have developed a simple blood test that can detect more than 50 different types of cancer, often before people have developed any signs or symptoms.

Image Credits: Elpisterra / Shutterstock.com

In a landmark study conducted by scientists from the Dana-Farber Cancer Institute, Harvard Medical School, The Francis Crick Institute and University College London, the test was found to be especially accurate at detecting twelve particularly dangerous forms of cancer including pancreatic cancer, which is generally only identified once the disease is in the late stages.

Writing in the Annals of Oncology, the researchers say “effective screening paradigms exist only for a small subset of cancers, are focused on single cancer types, and have variable adoption and compliance… Thus, diagnoses are often prompted by symptoms and are made at later stages.”

Experts hope the new test can be used to detect early-stage tumors when treatment is likely to be less morbid, more effective and a cure more achievable.

“Hopeful that blood-based cancer detection will be a reality"

Many researchers across the globe are trying to develop blood tests – often called “liquid biopsies” –that can detect cancer.

Study author Professor Geoff Oxnard (Dana Farber Cancer Institute, Boston), told the BBC News. "This blood test seems to have all the features needed to be used on a population scale, as a multi-cancer screening test… Certainly, the field is moving quickly and it makes us hopeful that blood-based cancer detection will be a reality."

For the study, Michael Seiden from the company US Oncology Research and colleagues used blood samples taken from more than 4,000 individuals – some who had cancer and some who did not – to test for cancer based on the DNA sequences that are released into the blood when cells die.

The team writes that using blood-based circulating tumor cell-free DNA (cfDNA) to detect and locate various types of cancer may address the problem that diagnoses are often prompted by symptoms and therefore made at later stages: “In largescale population screening, such a multi-cancer detection approach would require high specificity, clinically useful sensitivity, and highly accurate tissue of origin (TOO) identification to limit the scope, cost, and complexity of evaluating asymptomatic patients.”

What did this study find?

The researchers found that assessing methylation patterns (the adding of methyl groups to inactive genes) was the most promising approach since these patterns are abnormal in cancer cells.

They then used blood samples taken from 1,500 individuals with cancer and 1,500 without cancer to train a machine-learning algorithm to identify cancer.

“A classifier was developed and validated for cancer detection and tissue of origin (TOO) localization,” writes the team.

“Specificity is extremely important”

Next, they tested the classifier among 650 individuals who had cancer and 610 who did not.

The specificity of the test was 99.3%, with only 0.7% of people flagged up as having cancer when they did not.

“Specificity is extremely important because you don’t want to raise false alarm in people who are well,” says Seiden.

Furthermore, the test accurately identified the type of cancer that people had in 96% of cases.

The proportion of cancers detected increased the later the stage of disease. For all types of cancer, the identification rates were 18% for stage I, 43% for stage II, 81% for stage III and 93% for stage IV.

For the twelve most fatal cancers (esophagus, head and neck, anus, bladder, colon/rectum, liver/bile duct, lung, ovary, pancreas stomach lymphoma, and plasma cell neoplasm) the corresponding rates were 39%, 69%, 83%, and 92%.

“In summary, cfDNA sequencing of informative methylation patterns detected a broad range of cancer types at metastatic and non-metastatic stages with specificity and sensitivity performance approaching the goal for population-level screening,” writes Seiden and colleagues.

The test is now being validated among intended-use populations

The researchers say the test is now being validated among “intended-use” populations and one study is already passing its findings on to healthcare providers and patients.

"Everyone asks when will a test like this will be ready for use”… Oxnard told BBC News. "But before this blood test is used routinely, we will probably need to see results from clinical studies like this to more fully understand the test performance.”

Annals of Oncology editor - Fabrice André (Institut Gustave Roussy, France) has called the research a “landmark study” and said it represents a first step towards developing convenient, easy-to-use screening tools:

Earlier detection of more than 50 percent of cancers could save millions of lives every year worldwide."

Fabrice André, Institut Gustave Roussy, France