Delfi Diagnostics' AI-based blood test detects 80% of early-stage lung cancer cases, study finds

blood test tube
Delfi Diagnostics’ liquid biopsy test uses machine learning to spot changes in patients’ DNA fragmentation profiles that indicate the presence of cancerous cells. According to a new study, this method can spot about 90% of lung cancers. (Getty Images)

These days, cancer-spotting liquid biopsy offerings are a dime a dozen, with everyone from Grail to Guardant throwing their hats in the ring and cancer-screening blood test developers making up three of the top five highest-funded medtech companies in 2020.

But while Delfi Diagnostics’ goal of creating a more accurate and affordable method to detect early-stage cancers reflects the aims of its many competitors in the space, it’s taking a different approach to get there.

That approach—which looks for millions of possible abnormalities in the breakages of a patient’s cell-free DNA molecules, rather than looking for cancer in genetic mutations or methylation groups—seems to have paid off.

According to a study of just under 800 patients that was published Friday in the journal Nature Communications, Delfi’s test was able to spot about 90% of lung cancer cases, including 80% of stage 1 cases. The results were validated by low-dose CT (LDCT) scans, which are recommended for high-risk patients ages 50 and up to screen for lung cancer but are rarely performed.

Using the Delfi test in tandem with the CT scan to screen for lung cancer resulted in a 50% reduction in the number of unnecessary procedures that would’ve been performed based on the LDCT’s recommendation alone, the study found.

“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 study author Nic Dracopoli, Ph.D., Delfi’s chief scientific officer.

RELATED: Delfi Diagnostics raises $100M for a new approach to screening blood for cancer

Delfi’s method uses machine learning algorithms to analyze blood samples. During that analysis, the artificial intelligence looks specifically at each patient’s genomewide, cell-free DNA fragmentation profile. Cell-free DNA derived from cancer cells causes greater fragmentation in the genome, meaning it experiences a higher than usual number of the natural breakages that occur in a DNA molecule and its packaging during cell replication.

As a result, Delfi’s algorithms can spot the presence of cancer based on abnormalities in a patient’s fragmentation profile. It can also pinpoint where the cancer originated, as evidenced by the packaging of a cancerous molecule of DNA.

In an interview with Fierce Medtech earlier this year, Delfi CEO Victor Velculescu, M.D., Ph.D., suggested that this method is significantly more accurate than other liquid biopsy-based cancer diagnostics because it looks for the millions of potential changes in DNA fragmentation, while others focus on spotting cancer among dozens of genetic mutations or thousands of possible changes in methyl groups.

“Theirs have many fewer shots on goal, and in contrast, we believe we have many more, and that leads us to be able to have many more opportunities for detecting cancer,” said Velculescu, also the founder of the company and co-director of cancer biology and associate cancer center director for precision medicine at Johns Hopkins University.

RELATED: Liquid biopsy maker Nucleix rakes in $55M to develop blood tests to detect lung cancer early

The promising study results come not long after Delfi raised $100 million to continue developing its technology—upping the stakes in the race against competitors like Nucleix, Resolution Bio and Guardant, all of which have also unveiled liquid biopsy tests to detect lung cancer.

Next up, the company will launch an even larger clinical study to validate its technology’s use as a commercial lung screening test. Enrollment in that 1,700-participant trial has already begun, according to Peter Bach, M.D., Delfi’s chief medical officer.