Roche leads Freenome Inc US$257m financing

Roche led the financing that will be used by San Francisco-based Freenome Inc to finance testing its early cancer detection platform in the Vallania Study that will recruit 6.200 healthy and cancer patients to find cancer-specific multi-omics patterns in the blood using a self-learning AI algorithm. The US$254m funding round was joined by a16z Life Sciences Growth Fund, the American Cancer Society’s BrightEdge Ventures, ARK Investments, ArrowMark Partners, Artis Ventures, Bain Capital Life Sciences, Cormorant Capital, DCVC, Eventide Asset Management LLC, Intermountain Ventures, Perceptive Advisors, Polaris Partners, Pura Vida Investments, Quest Diagnostics (NYSE: DGX), RA Capital Management, Sands Capital, Section 32, Squarepoint Capital, with funds and accounts advised by T. Rowe Price Associates, Inc., and others.

Despite technical hurdles in liquid biopsy-based early cancer detection tests, Swiss Roche AG sees potential in blood tests that are designed to detect cancer and tissue-specific biomarkers in cancer stages where the disease does not mean certain death sooner or later. If it were possible to keep cancer in check through lifelong medication, this would be good for both business and patients. Distinguishing healthy from malignant stages by blood draws is ambitous as shown by Illumina’s current US$7.1bn divestement from Craig Venter’s start-up GRAIL.

However, as classical liquid biopsy tests were only able to detect cancer biomarkers in late cancer stages (III/IV) new technologies that focus on DNA methylation patterns in genes such as PAX that regulate a cell’s fate in early development stage. Alex Meissner’s Harbinger Health, which raised US140m in a Series B financing last September, for example, thus focuses on tissue specific detection of such cancer-specific patterns with a self-learning algorithm. According to its academic founder, it is 5 magnitudes of orders more sensitive than classical cancer liquid biopsy that suffers from the low count of detectable cancer mutations detectable in the blood.

Freenome Inc is leveraging a multiomics platform, which uses computational biology, machine learning and other technologies to develop screening tools to detect cancer in its earliest, most treatable stages (I + II). The platform is currently being evaluated Freenome’s biopharma and diagnostic company partners to almost non-invasively detect minimal residual disease (MRD) augmented with biological insights derived from the multiomics platform.

Freenome’s initial programs are focused on deadly and actionable cancers with high mutational count — colorectal and lung with a pipeline of single-cancer and multi-cancer tests under development. The company has two registrational studies underway:

PREEMPT CRC: A >40,000-participant prospective clinical study with comprehensive longitudinal real-world data (RWD) evaluating Freenome’s blood-based screening test among adults at average risk for colorectal cancer (CRC). PROACT LUNG: A prospective observational clinical study enrolling up to 20,000 participants. The study is intended to validate Freenome’s lung screening test in current and former smokers 50 years and older who are eligible for screening with an LDCT scan.

Freenome is also developing tests to screen for other cancers as part of its multi-cancer research programs, including the Vallania Study and others. Combined, the multi-cancer studies will involve >10,000 participants with paired RWD.

In the Vallania Study, Freenome will enrol 6,200 participants including risk-matched control participants to reflect intended use populations. The study will compare blood samples from both cancer and non-cancer individuals to understand patterns associated with lung and other priority cancers.