Behind Grail, Illumina’s billion-dollar diagnostics startup

With a billion dollars in private financing and a test projected to generate a terabyte of data per person, Grail has seeded a new category of diagnostics and Big Data. Despite this, the company story shared by CEO Jeff Huber is mostly about people.

Damatic sunset sky sunrise

Being a $100 million spin-out of Illumina will surely get you a foot in the door at any life science VC firm. Raising a further $900 million (and counting) — that requires something a little extra.

So who is Grail, the cancer diagnostics company that last week raised what could be the largest-ever equity financing round in biopharma history? How has it shaped a field of diagnostics into a billion-dollar venture in just over a year?

While the company hasn’t been particularly forthcoming with information, a lot can be gleaned from CEO Jeff Huber’s presentation at the Future of Genomic Medicine conference held in San Diego last week.

After over a decade at Google, Huber’s Big Data background led him to Illumina and then Grail, as fate cast him a personal role in the pursuit of early cancer detection.

In the beginning…

Science is full of coincidental success stories, most famously the discovery of penicillin. It’s not clear yet whether Grail will have the same impact, but a similar unintended observation by the Illumina research and discovery team led to its creation.

In 2013, Illumina acquired Verinata Health, a company focused on noninvasive prenatal testing (NIPT). That science came from the discovery that small fragments of fetal DNA cross the placenta during pregnancy. With sensitive tools, scientists learned how to isolate the fetal DNA to safely screen for any chromosomal abnormalities that might impact the infant’s development.

By the time Illumina acquired Verinata, the prenatal test had been run over 100,000 times, Huber said. It worked extremely well, except for a handful of cases — around 20 — where a foreign signal was detected.

The researchers struggled to identify the source of the signal, theorizing that it could be some rare fetal abnormality. Several months later, the 20-odd babies were all born healthy and the team went back to the drawing board.

Around the same time, Illumina hired the former director of the National Cancer Institute, Rick Klausner. He took a look at the data in question and promptly declared that it was cancer, Huber recounted.

Unfortunately for the 20 women involved, Klausner was right. The prenatal test had detected tumor fragments known as cell-free DNA (cfDNA) circulating in the bloodstream.

“They were finding cancers with perfect specificity, literally 100 percent accuracy,” Huber said. 

Other NIPT companies and academic groups have reported similar findings. The catch was that the cancers were late-stage. The test couldn’t change patient outcomes. 

“And that led to the light bulb within Illumina of saying, OK, if this test that was developed for this other purpose, that doesn’t have the kind of sensitivity that it would need to have but is still detecting the signal, what would it take to be able to go from the late-stage diagnoses that are happening here, to be able to detect cancer in its earliest stages,” Huber said.

Illumina’s R&D team began working on the science, embarking on a partnership with Memorial Sloan-Kettering that continues to this day. As the potential of the technology came to light, Huber said the discussions turned from, “wow this is a really good business idea,” to the realization that they had a moral and ethical imperative to make this test a reality.

The vision

Grail has a concise way of explaining its overarching aim. For cancers discovered in the earliest stages, stage one and stage two, the prognosis is typically positive; a 70-90 percent survival rate depending on the cancer type.

Cancer that is discovered late-stage, stage three and stage four, is roughly the inverse, Huber said. Some 80-90 percent of patients die.

Grail wants to drastically shift the odds in the patient’s favor by detecting cancer early. It would take the form of an omnipotent blood test that is routine, preventive and actionable.

Is that a moonshot or a logical extension of the technology?

It’s too early to know. Grail spun out of Illumina just over one year ago. During that time, the company built a dedicated facility in Menlo Park, California, which Huber believes to be the highest capacity sequencing lab in the world. From an initial team of 40 ex-Illumina employees, the staff now number 160.

For a population-wide cancer screening test, Grail knows it needs to work on a large scale. To that end, it has already launched its first multicenter clinical trial, dubbed the Circulating Cell-free Genome Atlas (CCGA) study. The aim is to analyze blood samples from 10,000 participants; 7,000 with cancer and 3,000 controls to build a detailed reference library for what a normal blood profile looks like.

Huber said the studies will ultimately scale to hundreds of thousands of patients to generate enough data on different cancer types and stages.

To detect cancer in its very earliest stages requires sensitivity down to a handful of molecules in a tube of blood. That’s where the ultra-deep, ultra-wide sequencing comes in. Huber described the high-intensity approach as an order of magnitude deeper than traditional liquid biopsy sequencing.

“The net effect of that is that we’re generating on the order of a terabyte of data for every test that we do,” he said.

(This scale is one of the reasons why the company doesn’t refer to its tests as liquid biopsies, a Grail spokeswoman said via email.)

If the company can realize its aim to embed itself in the medical system as a routine test, the data calculations become enormous. Starting with just the adult population in the United States — approximately 100 million people per year — would make Grail the first “zettabyte-scale application,” Huber said.

That’s 1,000,000,000,000,000,000,000 bytes; roughly the size of the likes of Amazon, Google and Facebook.

Collecting, organizing and gleaning insights at that scale will be an immense challenge. However, it does readily explain the unprecedented scale of its Series B financing round.

It’s also partly why Huber is at the helm as well. As a 12-year veteran of Google, he was first recruited to join the Illumina board to inject some Big Data knowledge into the team. That path led to Grail.

It’s not all abstract

Behind the technology and the Silicon Valley idealism is a painful authenticity that stems from Huber himself. Long before the huge financing round was closed, he was burdened with the responsibility to do something meaningful in this space.

In 2014, Huber’s wife Laura was unexpectedly diagnosed with stage four colon cancer. He told the Future of Genomic Medicine audience that there were few warning signs and no risk factors. She was 46, healthy, and did not have a family history of the disease.

After 18 months of intense treatment, Laura passed away. It was November 2015. A month later, Huber was approached by the Illumina team about taking the helm at Grail, given his very personal connection.

It’s a tough topic to discuss with strangers, but undeniably relevant.

“I’m very confident that if Grail had existed three or four years earlier that Laura could have had a very different outcome,” Huber told the audience.

Photo: primeimages, Getty Images