DELFI Diagnostics, founded by a stellar team of top-flight scientists and entrepreneurs, and backed by top-tier investors, is working to make early detection of cancer accessible, affordable, and accurate using a new approach to liquid biopsy. It has just released promising early data, but it joins a crowded field that is years away from proving clinical utility.
Blood-based screening tests that detect cancer early are to medicine what SpaceX’s Dragon and other privately funded spacecraft are to space travel: high stakes, expensive, multi-decade initiatives, which at first seem like pet projects of ambitious, super-wealthy investors and entrepreneurs.
In the past decade, a staggering amount of money and resources has been devoted to these once pie-in-the-sky screening efforts to extend human survival. And whoever succeeds could be at the forefront of one of medicine’s great challenges of the early 21st century: reducing cancer mortality rates.
That said, early detection testing and screening technologies, even more than other diagnostic tools, require exquisite precision and accuracy if they are to become standard of care. And to date, obtaining reliable information from the targets available in blood, urine, or other easy-to-collect liquid biopsy samples has been challenging. Pioneers, such as Grail and Guardant, have suffered business and scientific setbacks in recent years for a variety of reasons, and for the most part have yet to produce convincing data that their screening tests save lives.
Enter DELFI Diagnostics, which is developing a new class of liquid biopsy tools that look for patterns among fragments of cell free DNA (cfDNA), rather than tools that identify precise but more difficult to find and extremely rare genetic alterations. The company believes that by reading patterns of DNA fragments, regardless of their actual composition, it can enable detection of cancer in small blood samples with equal or better accuracy than deep coverage sequencing required by competitors. This is accomplished by applying advanced machine learning to whole-genome sequencing data. Therefore, its testing services can be faster and cheaper, in short, providing the kind of life-prolonging and accessible early interventions medicine is seeking.