Researchers at the Baltimore-based biotech company, Delfi Diagnostics, have developed a machine learning-based blood testing technology that could greatly help detect early stages of lung cancer.

In their study published in the journal Nature Communications, the team outlines how the diagnostic tool can analyze genome-wide cell-free DNA fragmentation (cfDNAs) profiles, nucleic acid fragments present in the bloodstream that can indicate the presence of tumor cells, with astonishing degrees of accuracy.

The tool was able to detect roughly 90% of cancer cases out of 800 individuals that were screened for lung cancer.

Lung cancer is the leading cause of cancer deaths, with the vast majority of lung cancers detected at late stages when prognosis is poor. LDCT scans have proven to be effective in lowering overall mortality among those at high risk for developing the disease, by detecting cancer early. Recently, the United States Preventive Services Taskforce expanded its existing lung screening guidance to include high risk individuals 50 years old and over, increasing the recommended population to 15 million U.S. adults.

Despite this guidance, fewer than 6% of those individuals receive the recommended screening. Researchers estimate that hundreds of thousands of deaths could be prevented world-wide through improved screening and early detection.

“These results suggest that the Delfi lung cancer screening technology could help reduce lung cancer deaths by offering a convenient, high-performing test to people who are eligible,” Delfi CMO Peter Bach said.

“We have already begun enrollment of a 1,700-patient, prospective, case-control study to generate the clinical evidence that would underpin a commercial lung screening test,” he added.

The blood test is also much easier to administer when compared to low-dose CT scans (LDCT), allowing more individuals to be screened in shorter periods of time.

The study evaluated 365 individuals participating in a seven-year Danish study called LUCAS, with the majority of participants at high-risk for lung cancer and with smoking-related symptoms such as cough or difficulty breathing. The results were validated in a separate study of 431 individuals with and without cancer with similar test performance.

When the Delfi technology was used as a pre-screen to determine if an LDCT should be performed, the combined approach led to detection of 90% of lung cancers, including 80% of stage I cancers, and reduced the number of LDCT induced unnecessary procedures by 50%.

“Because blood tests are so much easier to administer than LDCT, we believe a high-performing, cost effective assay could greatly increase the number of lung cancers that are detected early, when it can make a difference in care and eliminate a great deal of false positives compared to the current standard of care,” said Nic Dracopoli, Delfi’s Chief Scientific Officer and author on the publication. “The test is particularly well suited to lung cancer screening because we’ve shown it can detect both SCLC and NSCLC without inadvertently detecting other lung diseases such as COPD.”