DELFI Diagnostics is developing a novel way to assess circulating tumor DNA and give cancer patients and clinicians the definitive answers they need.

 

July, 2023 - Liquid biopsies are a non-invasive way to assess circulating tumor DNA (ctDNA) and gain actionable information about a patient’s cancer. These emerging diagnostics can identify driving mutations, measure disease progression, detect resistance or monitor a therapy’s efficacy. In the not-too-distant future, liquid biopsies may be sensitive and specific enough to find early-stage cancers and get patients into treatment sooner.

DELFI Diagnostics is developing a new generation of liquid biopsies that could advance these applications. Built on a new method called fragmentomics and amplified by machine learning, DELFI’s approach is already showing promise, providing early detection for lung, liver and other cancers, and could provide essential insights to inform precision medicine.

“DELFI is a next-generation liquid biopsy company working to fulfill the promise and excitement around emerging diagnostics,” said Wouter Meuleman, Partner at Illumina Ventures. “They’re pursuing high-performance tests that are accurate, easily deployable and inexpensive, which means this could be the first liquid biopsy platform with global reach.”

 

Divining the Secrets of Circulating DNA Fragments

Pioneered by founder and CEO Victor Velculescu, M.D., Ph.D., fragmentomics measures ctDNA fragment sizes to identify tumors. While healthy cells produce mostly uniformly-sized fragments, ctDNA is all over the map. With a crucial AI assist, DELFI’s tests analyze these aberrant fragments to provide distinct tumor signatures.

“In 2019, a study was published out of Victor’s lab in Nature that showed fragmentomics could advance single cancer early detection, multi-cancer early detection and minimal residual disease detection, all with a standardized, low-cost approach,” said Doug Schenkel, Chief Financial Officer at DELFI.

The company has refined the technology and is developing a test for lung cancer, which kills around 1.8 million people worldwide each year. “We isolate all cell-free DNA in a blood sample and map it, chromosome by chromosome, at less than ten times coverage,” said Schenkel. “A standard whole genome is 30 times, and a clinical genome is a multiple of that.”

This low sequencing depth simplifies the workflow and reduces costs. Technologies that search blood samples for scarce cancer-associated mutations or epigenetic aberrations in ctDNA fragments must sequence deeply, in highly specific and limited genomics regions, to find them. Fragmentomics essentially mines the data noise mid-sized DNA pieces generate – a signal that is mostly discarded by other methods.

“First-generation approaches spend a lot of time and money trying to find the pebbles that get dropped into the pond,” said Schenkel. “We just look for the ripples, which makes finding that signal across the whole genome a lot easier. We believe we can find a stronger signal with much less sequencing.”

 

DELFI’s Lung Cancer Strategy

When they turn 50, current and former smokers are encouraged to undergo annual, low-dose CT scans to find tumors. Around 15 million people in the U.S. are eligible for this potentially life-saving test. Unfortunately, only about 6% get screened.

People have many reasons to avoid screening: CTs are scary, inconvenient and perhaps unnecessary. DELFI’s lung cancer test could be a first step towards identifying patients who absolutely should be scanned.

“People would start with DELFI, which is just a simple blood draw,” said Schenkel. “We’re designing the test to be highly sensitive, so it won’t miss any cancers. If the test is negative, the patient can be more confident they do not have lung cancer and skip the CT scan just like they would have without the test. If it’s positive, they’re encouraged to seek further testing.”

 

What’s Next

DELFI aims to take the guesswork out of cancer diagnostics. The company is on schedule to complete a 2,500-patient study to potentially launch a lab-developed test (LDT). Within the next two years, they hope to conclude a 15,000-patient study to support FDA approval. Tests for liver and other cancers are also in the works.

“The assays we're developing are designed to give clinicians and patients a real clear direction,” said Schenkel. “We are prioritizing areas where the results, whether positive or negative, provide an easily recognizable path forward.”