A liquid biopsy company that spun out of Illumina Inc. in January to search for what it considers the holy grail of oncology – a blood test capable of detecting cancer at the earliest, most curable, stages – recently took a big first step toward that end. Grail Inc., of Menlo Park, Calif., launched a study that will eventually include up to four-dozen U.S. sites tasked with analyzing blood and tissue samples from 10,000 patients in an effort to characterize the landscape of cell-free DNA profiles in both cancer and non-cancer patients using a high-intensity sequencing approach that leverages San Diego-based Illumina's technology.

Grail CEO Jeff Huber said the company's ability to detect and characterize tumor DNA, combined with its computer power, will enable Grail to convert vast amounts of genomic data into disease insight. "Our approach will produce more than a terabyte of data per individual, thereby creating datasets of a scale and complexity that are unprecedented in genomic medicine," he said.

Large-scale studies like the Circulating Cell-free Genome Atlas (CCGA) will support the development of a pan-cancer screening test for asymptomatic individuals, which could, according to Grail, make a major dent in global cancer mortality. The company said these studies have to include samples from tens of thousands of people in order for researchers to identify the patterns required to detect many types of cancer. Confirming clinical validity and utility of these tests, however, will be an even bigger feat, which Grail said will require studies of hundreds of thousands of people.

The observational CCGA study will enroll at least 7,000 cancer patients and 3,000 healthy individuals, interrogating the biology of both tumor biopsy tissue samples and the circulating, tumor-derived nucleic acids in blood. Circulating tumor nucleic acids (ctNAs) in the blood are an emerging biomarker for earlier cancer detection. GRAIL and its collaborators will collect clinical outcomes on the enrolled participants for at least five years. The result will be a detailed atlas of cancer genetics that GRAIL will use to support its product development goals. The database, upon analysis, may be expanded to additional enrollment in specific cancers or healthy individuals, Grail noted.

The initial collaborators in the CCGA study include: medical centers of the Guardian Research Network, Mayo Clinic, and Hartford Healthcare Cancer Institute, a member of the Memorial Sloan Kettering Cancer Alliance.

"For too many patients and their families, a late diagnosis of incurable cancer is devastating, and the complexity of cancer has made it challenging to find biomarkers for early-stage detection when the cancer could be cured," said Jose Baselga, physician-in-chief at Memorial Sloan Kettering Cancer Center and chairman of Grail's scientific advisory board. "The CCGA study will provide a critically important library of knowledge about cell-free nucleic acid profiles in cancer patients and new insights into the biology of cancer at its earliest stages."

Baselga, who also is serving as a member of the CCGA advisory board, said it is equally important that the study characterize the heterogeneity of the population of individuals without cancer, and "thus enable the development of models which distinguish people with and without cancer with unprecedented accuracy."

Detecting cancer at an earlier stage when it can be cured is an ambitious goal, admitted Timothy Yeatman, president and chief scientific officer of the Guardian Research Network, which has already begun enrolling participants across its network of cancer centers and U.S. hospitals. But it's a goal that could potentially offer "immeasurable benefit to society," Yeatman said.

Grail said its approach is to sequence circulating nucleic acids at unprecedented breadth and depth to optimize the detection of early-stage cancer. The company also said it will use the latest tools of data science, including approaches from machine learning such as hierarchical neural networks. Grail also said it will apply such methods to all steps of its data-generating pipeline including the ultimate challenge of classifying patients according to the presence, type, and severity of cancer.

Grail has secured more than $100 million in series A financing from Illumina and Arch Venture Partners, with participating investors including Bezos Expeditions, Bill Gates, Sutter Hill Ventures,and GV (formerly Google Ventures).

THE PROMISE OF LIQUID

The true clinical impact of liquid biopsy technology remains to be seen, but the potential to represent a major shift in cancer treatment has attracted interest from a growing number of companies of all sizes. The promise of liquid biopsy technology is so strong, in fact, that 20 stakeholders, including biopharma companies, diagnostic players, and academic institutions, are working to create a large database for cancer genomic profiling data as part of the U.S. Vice President Joe Biden's Cancer Moonshot effort. This liquid biopsy database is expected to serve as the basis for the development of blood-based cancer tests, pooling data from assays that use circulating tumor cells, circulating tumor DNA, and exosomes.

The fact that there are so many different approaches being explored in liquid biopsy is one of the things that makes this emerging field so interesting to follow. While both blood-based and urine-based assays qualify as liquid biopsy tests, the majority of players in the space are focused on blood, leaving San Diego-based Trovagene Inc. one of few contenders on the urine-based side of the market.

On the blood-based side, the market is segregated based on specific components of blood that these tests are designed to capture – circulating tumor cells, circulating tumor DNA, nucleosomes, and exosomal RNA. Beyond that, the space can be split into subcategories based on whether or not the test in question is intended to find cancer or just follow it to monitor how well a prescribed therapy is working for an individual patient.