Third Rock-backed startup's blood test can sniff out cancer in patients with no history of disease
The myriad of liquid biopsy companies working on blood tests for early cancer detection have been conducting retrospective studies, looking at how effective and sensitive their technology is in spotting cancer in patients that have already been diagnosed. Now, a Third Rock Ventures-backed startup has also shown its blood test can perform in a large prospective study involving patients with no personal history of cancer.
The company, aptly named Thrive Earlier Detection, is betting on CancerSEEK — its blood test-in-development designed to sniff out a wide variety of cancer types by interrogating genomic mutations in circulating tumor DNA (ctDNA) as well as protein markers in plasma that have been implicated in cancer.
The trial — christened DETECT-A — evaluated the blood test in more than 9,900 women aged 65 to 75 with no evidence or history of cancer.
CancerSEEK is engineered to not only be powered to identify the presence of relatively early cancer, but it uses machine learning to localize the organ of origin of cancer — eventually, the hope is the test will be used as part of the arsenal of routine medical screening tools to complement existing disease-specific screening methods such as mammography and colonoscopy.
“The motivation for this study was, how does this look like in a real-world setting? How does the blood test perform in normal individuals, which means they have comorbidities?” said Christoph Lengauer, the company’s co-founder and chief innovation officer. “It’s the first interventional study, which means that the results of the test are given back to the physician and then shared with the patient and then decisions (on treatment) get made.”
Overall, 96 cases of cancer were identified among the 9,911 participants within 12 months of enrollment. Data showed that CancerSEEK detected 26 cancers — 17 cancers were diagnosed at an early stage (localized or regional to the area it originated from), of which 12 were able to be surgically removed. The diagnoses were confirmed by PET-CT scans.
Standard screening methods, such as mammography or colonoscopy, pinpointed another 24 cancers. The remaining 46 cases were not first detected by either blood testing or by standard screening — in most of those cases diagnostic tests were initiated on the basis of patient symptoms.
“We didn’t know basically how our test fits into reality … and the reality is that today 75% of all cancers that happen are identified by symptoms. With the introduction of colonoscopies and mammographies, that changed, because now you can detect some of those cancers earlier,” Lengauer said. “We wanted to see what our test can add to it in addition to standard-of-care (screening) … that was surprising because our test ended up doubling the number of what can be detected by screening.”
Overall, the specificity of the Thrive test came in at 99.6% — “there were almost no false positives,” said Lengauer.
The results showed CancerSEEK augmented the benefit of standard-of-care screening for breast, colon and lung tumors improving sensitivity from 47% to 71% — and also allowed the detection of seven other cancer types (lymphoma, appendix, uterine, thyroid, kidney, ovary and cancers arising from an unknown primary site) that cannot be screened now, with a sensitivity of 31%.
In 2018, published data from a retrospective study, encompassing 1,005 patients with non-metastatic, clinically detected cancers of the ovary, liver, stomach, pancreas, esophagus, colorectum, lung or breast, showed that CancerSEEK tests were positive in a median of 70% of the eight cancer types.
Armed with these data, Thrive is gearing up to conduct a registrational study, which should enroll an even bigger number of patients than DETECT-A, said Lengauer.
CancerSEEK has been granted the FDA’s breakthrough device status, as has Grail’s multi-cancer detection blood test, which relies on DNA sequencing to assess methylation, an epigenetic change across the genome to expose cancer signals.
Grail, which was spun out of DNA sequencing company Illumina $ILMN in 2016, has raised $1.6 billion in funding to fuel the development of its test, which is now being evaluated in three large-scale studies that will altogether enroll 165,000 individuals. Similarly designed as Thrive’s DETECT-A study, Grail is currently enrolling patients in a prospective PATHFINDER trial, which is designed to assess the utility of its early detection blood test in clinical practice.
Other companies such as Guardant Health and Biocept are also working on their own liquid biopsy tests. Each company is looking to champion consistency and accuracy — false positives induce unnecessary anxiety, and are costly. Another concern is, of course, privacy.
Single disease tests, such as Exact Science’s Cologuard, carry list prices of $600 or $700 and mammographies cost between $100 to $200 — if approved, Thrive’s multi-cancer test CancerSEEK would be priced “in the hundreds of dollars commercially, not thousands of dollars,” company officials suggested in an interview with Endpoints News last year, when Thrive launched and raised $110 million in a Series A financing.
Now, as it works on plans for a large registrational study, raising capital will be imperative, Thrive executives said.
“We have to raise more money because this is very expensive. And this is part of the challenge in the screening space … we just did (a study with) 10,000 patients, I mean, it costs a lot of money,” said Lengauer.
“Therefore this summer, we will raise more money,” he added, without disclosing specifics.