Connect with us

Medtech

Harbinger Health raises $140M Series B to launch liquid biopsy for early cancer detection by 2025

Harbinger Health, a startup with ambitious goals to develop liquid biopsies that spot signs of cancer early, has raised $140 million in Series B funding…

Published

on

This article was originally published by Endpoints

Harbinger Health, a startup with ambitious goals to develop liquid biopsies that spot signs of cancer early, has raised $140 million in Series B funding and announced plans to begin selling its first commercial blood tests in 2025.

The Cambridge, MA-based company, which was launched by Flagship Pioneering with $50 million in 2021, has remained tight-lipped about exactly how its experimental test works. But Harbinger believes it can get the cost roughly an order of magnitude cheaper than its competition, potentially placing its price tag under $100.

Stephen Hahn

“It’s aspirational. I can certainly imagine a world where over the next three to five years that actually occurs,” Harbinger CEO Stephen Hahn told Endpoints News in an interview. “But we can’t have a screening test that costs $1,000, that your insurance won’t pay for, and that folks aren’t willing to pay out of pocket for.”

Harbinger will use the new funds — which come from Flagship, Pictet, Partners Investment, M&G Investments and other undisclosed investors — to finish a 10,000-person clinical trial of its blood tests, expand its 60-person team to roughly 100 over this year and next, and conduct additional studies of its cancer tests in high-risk populations.

Doug Cole, co-founder and chairman of Harbinger and managing partner at Flagship, said that the company was created on the vision that most cancers could be treated more effectively if they were caught earlier, much like blood pressure and cholesterol checkups are used as early warning signs for cardiovascular disease.

“It’s not a test to diagnose cancer, but to test risk for it,” said Hahn, who was a former FDA commissioner and chief medical executive of the MD Anderson Cancer Center before joining Harbinger.

Detecting a ‘pan-cancer signature’

The startup faces competition from better-funded companies that have a head start developing their own liquid biopsies, including startup Freenome, which has raised more than $1.1 billion; Exact Sciences, which acquired liquid biopsy startup Thrive Earlier Detection; and Illumina, whose subsidiary Grail sells the cancer blood test Galleri for $949.

Harbinger’s tests measure methylation patterns in circulating tumor DNA to detect genes that are turned on or off early in the formation of cancer, Hahn said. Although methylation normally changes with age, “in cancer, it’s a specific pattern,” he added.

“Strikingly, the pathways that are turned on in a very regulated process for developing the fetus and placenta are very similar to many of the pathways that are associated with what’s dysregulated in cancer,” Hahn said. It’s not quite a “universal signature,” but appears to be a “pan-cancer signature,” he added.

Kieran Chacko, head of strategy at Harbinger and a principal at Flagship, noted that “machine learning is obviously a core component” of the company but didn’t specify how or when it was used.

Hahn said that the test needs to have a high positive predictive value (PPV), the likelihood that if the test says a patient has cancer, the doctor can find it. PPV scores of 50% or less have been the subject of criticism for other liquid biopsies that aim to spot a broad array of cancers, sometimes called multi-cancer early detection (MCED) tests.

“The whole concept of overdiagnosis, which is been plaguing screening for a significant amount of time, will probably be exacerbated by a blood-based multi-cancer early detection approach,” Hahn said.

Hahn believes that a PPV “in the 70% range would be very valuable” for an MCED test. But he thinks a lower sensitivity could still be useful for certain kinds of cancers, including pancreatic and esophageal cancers, where most patients are diagnosed with later-stage diseases that are often lethal.

“There are a number of cancers for which there are no good tests, and there the bar is much lower,” Cole said.

‘We want it to be like cholesterol’

Harbinger’s large 10,000-person CORE-HH study, where half the participants are already diagnosed with cancer, and the other half are presumably healthy, is a key step toward refining the design of its blood tests and proving their worth before commercialization.

The trial is running slightly behind schedule. Last summer, Hahn told Business Insider that data from the study were expected in the third quarter of 2023. Hahn told Endpoints this week that the company “didn’t have the accrual that we wanted” last winter and spring.

An initial readout of the “discovery” phase of the trial is now expected in the fourth quarter, Hahn said. Those data will help the company “lock down the assay” before fully moving into the second part of the study, where the company will conduct blinded tests of the liquid biopsy’s performance in roughly eight to 20 kinds of cancer. Those results are expected in the first half of 2024.

Harbinger is also planning to assess its tests in high-risk populations, such as people with a family history of cancer, with a high BMI, or who smoke, Hahn said. The company is also considering “disease-based bundles,” such as tests for cancers that commonly affect women or cancers found in a certain part of the body, such as the gastrointestinal tract. That could help lower costs and increase the chances that insurers will cover it, Hahn said.

The company will initially launch a so-called laboratory-developed test, a less stringent pathway that many developers of Covid-19 tests used to quickly commercialize their products during the pandemic.

Hahn said that real-world evidence collected after commercializing the tests will help support a submission to the FDA through the premarket approval (PMA) pathway for medical devices, potentially opening the door to broader markets and decentralized tests that don’t have to be run in Harbinger’s labs — key to the vision of making it as easy to get as a cholesterol test.

“We want it to be like cholesterol, where you can go to any of the laboratories and get it done. And it can be done closer to home rather than being mailed in,” Hahn said. “That is a really important part of our future.”



devices
machine learning

Medtech

ETF Talk: AI is ‘Big Generator’

Second nature comes alive Even if you close your eyes We exist through this strange device — Yes, “Big Generator” Artificial intelligence (AI) has…

Continue Reading
Medtech

Apple gets an appeals court win for its Apple Watch

Apple has at least a couple more weeks before it has to worry about another sales ban.

Continue Reading
Medtech

Federal court blocks ban on Apple Watches after Apple appeal

A federal appeals court has temporarily blocked a sweeping import ban on Apple’s latest smartwatches while the patent dispute winds its way through…

Continue Reading

Trending