With Third Rock leading the way in a $110 million series A, newly launched Thrive is banking that a multi-analyte approach to liquid biopsy could give it an edge in creating a test that is affordable enough for routine use in primary care. Thrive Earlier Detection Corp. launched on May 30 with an exclusive license to CancerSEEK, a multi-analyte test that has been in development by the company’s scientific co-founders Bert Vogelstein, Kenneth Kinzler and Nickolas Papadopoulos at Johns Hopkins University for over 25 years.
The company will seek to find its niche in a space that features two other cancer detection plays with extremely deep pockets, Grail Inc. and Guardant Health Inc.
According to Thrive CEO and Third Rock partner Steven Kafka, what most attracted VCs to the start-up was the potential to shift the point of intervention for cancer much earlier, when disease is non-metastatic, increasing the likelihood of a cure via resection or pharmacotherapy.
“So much investment is going into addressing late-stage cancer patients,” Kafka told BioCentury. “If we can move the stage I, stage II or even stage III survival curve for a patient, it is a fundamental transformation where we are talking about years of quality life, and in most cases, curing the cancer because you can either take it out or treat it up front with very effective chemo or radiation.”
Prior to joining Third Rock, Kafka was president and COO of cancer diagnostics company Foundation Medicine Inc. Prior to its acquisition by Roche in 2018, Foundation launched the FoundationOne pan-cancer diagnostic to identify treatment options and clinical trials for patients. The company has since launched FoundationOne Liquid, a blood-based circulating tumor DNA (ctDNA) test to match tumor mutations to the best therapy and available trials.
Thrive’s management team includes Christoph Lengauer, a Third Rock partner and former CSO of Blueprint Medicines Corp., as chief innovation officer; Isaac Kinde, co-inventor of the technology enabling CancerSEEK, as head of research and innovation; Geoff Otto, former Foundation Medicine VP of product development, as SVP and head of product development; Isaac Ro, former VP and lead medical technology analyst at Goldman Sachs, as CFO; and Samantha Singer, former COO at the Broad Institute, as COO.
CancerSEEK analyzes ctDNA for 61 amplicons across 16 genes along with expression levels for eight protein biomarkers, all of which are fed into a classifier machine learning algorithm to determine whether a patient has cancer and the most likely tissue of origin for the disease.
Because Thrive is targeting the test toward the primary care setting, Kafka said it needs to extract enough information to provide a reliable diagnosis, but do so in a cost-efficient manner.
“We really want to drive this into routine medical care, and so it needs to be both affordable and high quality,” Kafka said.
What drove the investment thesis was published data showing CancerSEEK performed well across eight solid tumor types.
Last year, Vogelstein, Kinzler, Papadopoulos and colleagues published in Science data from a retrospective study of the CancerSEEK test in 1,005 patients with non-metastatic, clinically detected gastrointestinal stromal tumors (GIST) and ovarian, liver, pancreatic, esophageal, colorectal, lung and breast cancers.
The data showed CancerSEEK correctly identified the cancer in a median of 70% of cases across the eight tumor types, while specificity was >99% with only seven false-positives among 812 healthy controls. The paper did not disclose the specific tumor types of the false positive results.
The sensitivity was 69-98% for five tumor types that have no standard screening tests: ovarian, liver, esophageal and pancreatic cancers and GIST. Ovarian and liver had the best sensitivity at 98%, while the other three were closer to 70%.
“Not only are we detecting cancers, but as a multicancer test, we are able to identify cancers for which there are no screening tests,” Kafka said. “So sensitivity is effectively zero today.”
The study authors noted that the feature most important to the algorithm was the combination of a ctDNA mutation and elevations of one of the eight protein biomarkers. Less than 20% of the cancer patients had no detectable genomic mutations. The authors didn’t break out what the median sensitivity of the test was across all eight tumor types without one of the components.
The authors estimated the cost of the test to be less than $500; Kafka confirmed that this is what Thrive is aiming for. For other cancers where screening protocols already exist such as breast or colorectal cancer, Kafka said CancerSEEK will be positioned to complement, not replace existing tests. “We are not looking to replace mammography, colonoscopy or even low-dose CT in heavy smokers. We are really looking to augment those approaches,” he said.
Kafka said the majority of the series A financing will go toward additional clinical trials to support FDA approval, establish the test’s positioning in cancer care guidelines and generate health economic data for payers. Other investors in the A round included Section 32, Casdin Capital, Biomatics Capital, BlueCross BlueShield Venture Partners, The Invus Group, Exact Sciences Corp., Cowin Venture, Camden Partners and Gamma 3.
Future studies would be conducted to identify the most appropriate target population for the test, such as whether it can be used broadly in a healthy population of a particular age, or whether it is better aimed at individuals with certain risk factors, such as a family history of disease.
Data from the first prospective trial could be coming in the near future, as the DETECT study, conducted in partnership among Johns Hopkins, Geisinger Health System and Thrive, has enrolled 10,000 women. The study is only enrolling women to remove the gender variable, while future studies will include both genders.
The study will screen each individual using the CancerSEEK test and follow them for five years. One year follow-up data could come as early as next year for the study.
Kafka added that Thrive will also incorporate real-world evidence into its development program. For example, he said Thrive could analyze in silico cohorts from Geisinger or other healthcare systems using electronic healthcare records to understand how cancer incidence evolves over time with and without testing.
Thrive is certainly not alone in developing a liquid biopsy test for the early detection of cancer.
Grail, which has raised at least $1.3 billion since its founding in 2016, appears to be the closest comparator, as it is developing an NGS-based liquid biopsy test that is now focusing on methylation markers for early detection of cancer. Grail has enrolled about 115,000 out of a targeted 165,000 participants across three large-scale trials that are designed to create an atlas of genomic cancer signals in the blood (see “Sequential Insights”).
Last year Grail reported the first data from its Circulating Cell-free Genome Atlas (CCGA) study, which showed sensitivities ranging from 15% for HER2-negative breast cancer to 90% for ovarian cancer across 11 tumor types.
Guardant launched in 2012 and has raised over $1.1 billion, including $273 million in a 2018 IPO. Its Guardant360 test is aimed at aiding cancer patients with treatment selection, while the GuardantOMNI test is for companies to use to accelerate clinical trials.
The company’s LUNAR-2 test is in development for the early detection of cancer. Guardant has said early results from the test show somatic mutations alone may be insufficient for detecting early stage cancer, so the company is exploring epigenetic and other signatures to enhance the sensitivity.
While Kafka declined to comment on competing tests, he said one possible differentiator is that Thrive plans to build out an integrated care service that would guide patients and physicians through the care continuum following a positive test result, such as follow-up with a CT scan or the most appropriate confirmatory test for the given tumor type.
“We really believe strongly that we can’t stop with the test result. We have to build a comprehensive solution that helps guide the physician through the next steps after a positive detection result,” Kafka said. “Because it is a primary care audience, we have to think about it differently and provide those extra services.”
Thrive Earlier Detection Corp.
Technology: CancerSEEK genomic and protein-based liquid biopsy test
Disease focus: Cancer
Clinical status: Pilot
Founded: 2019 by Bert Vogelstein, Kenneth Kinzler, Nickolas Papadopoulos, Isaac Kinde, Christoph Lengauer
University collaborators: Johns Hopkins University
Corporate partners: Geisinger Health System
Number of employees: 40
Funds raised: $110 million
Investors: Third Rock Ventures, Section 32, Casdin Capital, Biomatics Capital, BlueCross BlueShield Venture Partners, The Invus Group, Exact Sciences Corp., Cowin Venture, Camden Partners, Gamma 3, other undisclosed investors
CEO: Steven Kafka
Blueprint Medicines Corp. (NASDAQ:BPMC), Cambridge, Mass.
Broad Institute of MIT and Harvard, Cambridge, Mass.
Exact Sciences Corp. (NASDAQ:EXAS), Madison, Wis.
Geisinger Health System, Danville, Pa.
Grail Inc., Menlo Park, Calif.
Guardant Health Inc. (NASDAQ:GH), Redwood City, Calif.
Johns Hopkins University, Baltimore, Md.
Roche (SIX:ROG; OTCQB:RHHBY), Basel, Switzerland
Thrive Earlier Detection Corp., Cambridge, Mass.