EXOME

all the information, none of the junk | biotech • healthcare • life sciences

To Catch Early Cancer, and Rivals, Thrive Banks on Hopkins Blood Test

Xconomy Boston — 

The earlier cancer is diagnosed, the better the odds of survival, and the race is on to detect the earliest traces of the disease from an otherwise healthy person’s blood sample. A new entrant in that race has emerged today.

Thrive Earlier Detection has raised $110 million in financing from biotech investors Third Rock Ventures and others to develop a low-cost test—with a price tag in the hundreds, not thousands of dollars—that primary care doctors can use to screen healthy patients for the first signs of a variety of cancers.

It’s a daunting task. Blood tests can currently help classify previously diagnosed tumors, and they can monitor patients after treatment for signs of recurrence. But screening healthy people is fraught with risk of a misleading or mistaken diagnosis. Some cancer screening techniques, for example, have pushed people toward treatment they might not have needed. Memorial Sloan Kettering Cancer Center urologist Behfar Ehdaie, referring to a widespread prostate cancer test, told Xconomy in 2016, “Probably one million men in the PSA-test era have gotten surgery that did no good, at best, or perhaps did them harm.”

Thrive CEO Steve Kafka began a discussion earlier this week by stating the company was “leading a shift toward early cancer detection,” only to step back from that language. Thrive, based in Cambridge, MA, is a bit late to the game, in fact. A main competitor, Grail, launched three years ago with the same aim and has since raised more than $1 billion from Bill Gates, Jeff Bezos, a host of big drug companies, and Chinese investors, much of that going into large-scale clinical trials.

Guardant Health and Foundation Medicine, a division of Roche, are also well-funded developers of so-called liquid biopsies. They have first focused on tests that help doctors better match cancer patients with drug regimens and that monitor patients for cancer recurrence. But Guardant, riding last year’s blockbuster IPO wave, has made no secret of its ambition to eventually provide cancer screening for healthy people with its Lunar product, which is still in the research-only phase.

Thrive believes it has a leg up because the test it is using, CancerSEEK, analyzes both genes and proteins. Other tests are DNA-only.

It has an exclusive license for CancerSEEK from Johns Hopkins University in Baltimore, where it was developed at the Hopkins School of Medicine. In a paper published in Science last year, Hopkins researchers and their colleagues compared blood samples from 1,005 cancer patients, across eight types of cancer, and 812 healthy patients. Of the eight cancer types, five have no screening tests available.

The test did well by the famous medical dictum “First do no harm.” It had a false-positive rate of less than 1 percent, misdiagnosing only seven of the healthy 812 volunteers with cancer.

In the participants with cancer, CancerSEEK returned a positive result in a median of 70 percent. For the earliest cancers, however, the positive hit was much lower (43 percent). When asked if Thrive needs to boost those rates to an acceptable threshold, Kafka said, “There is no magic number for thresholds, and looking at sensitivity oversimplifies.” (Returning a correct diagnosis of disease—“yes, this person has cancer”—is known as sensitivity.)

Thrive must improve CancerSEEK in other ways before it can be widely useful and reliable. For example, it must take into account regional, ethnic, and other differences across populations. The first large trial to help refine CancerSEEK has already recruited 10,000 healthy women between 65 and 75. It’s a joint effort of Johns Hopkins and the healthcare provider Geisinger. Thrive will need other trials as well to refine the test and the predictive software that underpins it.

(The Hopkins researchers who developed CancerSEEK wrote as much in their paper: “To actually establish the clinical utility of CancerSEEK and to demonstrate that it can save lives, prospective studies of all incident cancer types in a large population will be required.”)

Thrive plans to add other layers of analysis to CancerSEEK, what Kafka called an “integrated service,” such as a review group to interpret positive results. “The goal is to wrap as much of an integrated service as possible into our service to the physician,” he said.

Convincing doctors to rely on such a crucial test is a matter of price and of trust. “I’ve lived through some of those battles,” said Kafka, who is a partner at Third Rock and was previously a top executive at Foundation Medicine. It took years for Foundationto gain insurance coverage for its $5,000 tissue biopsy test. It has also been tough to persuade oncologists in community clinics and other places that don’t have the same resources as major cancer centers to use cancer DNA tests like Foundation’s.

When asked how Thrive’s proposed service–which would seem to require human expertise and analysis, akin to the tumor boards that hospitals convene to mull over difficult cases—could meet the goal of costing no more than a few hundred dollars per test, Kafka said that the clinical review could eventually be “virtual.”

Photo of Gilman Hall, Johns Hopkins University by Callison-Burch via Creative Commons.