Researchers from the Johns Hopkins Kimmel Cancer Center in the US, along with other institutions, have developed and validated an AI-based liquid biopsy to help significantly improve early detection of lung cancer.

A prospective study indicated that the technology can discern patterns of DNA fragments in the blood associated with the disease.

The study involved around 1,000 participants, both with and without cancer, who were eligible for traditional lung cancer screening.

The new blood test aims to enhance lung cancer screening rates and potentially lower mortality rates through early diagnosis.

Johns Hopkins Cancer Genetics and Epigenetics programme co-director Victor Velculescu said: “We have a simple blood test that could be done in a doctor’s office that would tell patients whether they have potential signs of lung cancer and should get a follow-up CT scan.”

The researchers trained AI software to detect DNA fragment patterns in the blood of 576 individuals.

This method was then validated in a separate group of 382 people. With a negative predictive value of 99.8%, the test is highly reliable, potentially missing only two in one thousand cases of lung cancer.

The blood test is currently available through DELFI Diagnostics for laboratory use under the Clinical Laboratory Improvement Amendments.

The team is seeking US Food and Drug Administration approval for lung cancer screening.

Further research is planned to explore the test's applicability to other cancer types.

The study's co-author, Robert Scharpf of Johns Hopkins contributed to this research with colleagues from various institutions. The study received support from several foundations and the National Institutes of Health.