Detecting cancer before it spreads too far means that treatment is more likely to be successful, thus significantly increasing the patient’s chances of survival. However, for many types of cancer, early diagnosis is very complicated, which makes the disease a leading cause of death worldwide.
Seeking to provide a solution scientists developed an AI-supported blood test that can detect more than 50 types of cancer before clinical symptoms emerge. Now, the researchers behind the medical innovation say that the test is accurate enough to be rolled out as a screening test.
Expected to be piloted by NHS England this fall, the blood test is primarily intended for people at higher risk of the disease, including patients over 50 years old. As reported by The Guardian, it can accurately identify multiple types of cancers that are difficult to detect in the early stages. Among these are head and neck, ovarian, pancreatic, and some blood cancers.
The test works by examining chemical changes (methylation patterns) in fragments of genetic code — known as cell-free DNA (cfDNA) — that tumors shed into the bloodstream. The entire process is conducted using a machine learning algorithm — a type of artificial intelligence.
As part of the latest study, the scientists analyzed the performance of the test in 2,823 people with cancer and 1,254 people without. The findings showed that the test correctly identified when cancer was present in 51.5 percent of cases, across all stages of the disease, and wrongly detected cancer in only 0.5 percent.
The team also found that the test’s sensitivity for solid tumors without screening options, including pancreatic, esophageal, and liver cancers, was almost double (65.6 percent) than for solid tumors with screening options such as breast, cervical, and prostate cancer. The test also correctly identified the origin of the cancer in 88.7 percent of cases.
“Finding cancer early, when treatment is more likely to be successful, is one of the most significant opportunities we have to reduce the burden of cancer,” said Dr. Eric Klein, first author on the research. “These data suggest that, if used alongside existing screening tests, the multi-cancer detection test could have a profound impact on how cancer is detected and, ultimately, on public health.”
The results of the NHS trial, which will include 140,000 participants, are expected by 2023.
Study source: Annals of Oncology — Multi-cancer early detection test using an independent validation set
