Intel-Edico partnership aims to break a medical logjam
DNA RxIn 2003, a team of researchers working for the Department of Energy and the National Institutes of Health completed a 13-year long effort called the Human Genome Project. The goal of the project was, among other things, to identify and map every gene in human DNA. For a while, the scientific community has been looking at the data and wondering, "So, what's next?" One answer is targeted therapy, medical treatments that do not only address symptoms, but also respond to a patient's genetic makeup. The past couple of years have yielded breakthroughs in this field. In 2011, a team of scientists at the University of Pennsylvania effectively treated leukemia patients by engineering their own white blood cells to fight the disease. Also in 2011, researchers in Britain achieved promising results in treating the genetic disease hemophilia B by injecting patients with a re-engineered version of the gene. 2012 brought even more advancements, using targeted therapy to treat inherited blindness. "Talk about transforming an industry," says George Day, co-director of the Mack Center at Wharton. "Big Pharma has always been pill-based," he says, meaning patients need daily dosage, whereas "gene therapy is one and done. "I think finally after 20 years, the promise of gene therapy is starting to be realized."
Genome sequencing has the potential to enable huge advances in “precision medicine”–genetically tailored treatments for leukemia, cystic fibrosis, and other devastating diseases. A new partnership between Intel Corp. and startup Edico Genome, announced this week, could make the distance between the sequencing and the treatment considerably shorter.
The partnership would pair Edico Genome’s Dragen high-speed DNA sequence analyzer with Intel’s high-powered Xeon processors to accelerate the speed at which genomic sequences can be analyzed. Announcing the partnership this week, the companies estimated that Dragen and Xeon, working together, would be able to analyze a patient’s entire genome in real time. Until recently, it typically took 10 to 20 hours to analyze data from a single genome, and in practice, backlogs and other delays can mean a long wait between sequencing and diagnosis.
Edico Genome rolled out the Dragen processor chip in 2014, and made its first sale to Sequenom (SQNM), which conducts non-invasive prenatal testing; Dragen working on its own was able to complete genome analysis in as little as 20 minutes, according to its manufacturer. The Intel (INTC) partnership could substantially expand Edico Genome’s reach, since Xeon chips are already widely used within the servers that perform analysis alongside DNA sequencing equipment.
“In genomics, an unprecedented amount of big data is being generated,” says Pieter van Rooyen, CEO of Edico Genome. “It made sense to combine our expertise [with Intel’s] to create a solution that could analyze next-generation sequencing data in real time.” The companies declined to disclose financial details of the partnership.
Precision medicine, which involves tailoring specific therapies or drugs to a patient’s genetic makeup, is typically a two-step process. First comes the initial sequencing of the nucleotides in a patient’s DNA, which essentially spit out an enormous alphabet soup of As, Ts, Gs and Cs; next comes the analysis that pinpoints the relevant mutations on a sequenced genome. There’s a host of biotech and tech companies competing to make this second step happen faster, including behemoths like Amazon (AMZN) and upstarts like Bina Technologies.
“Our call to the industry–both technical and medical community alike–is: Could we do precision medicine in a day by 2020?” says Ketan Paranjape, general manager of life sciences at Intel. “The middle phase, the analysis, is what we’re trying to cut down in partnership with Edico.” The vision is that, within a day, a doctor could sequence a patient’s genome (or a tumor’s genome), analyze that sequence and provide a diagnosis, and then design a tailored treatment.
Precision medicine has enjoyed a higher public profile in the U.S. since President Obama launched a $215 million initiative to support research and investment into the industry. Global consulting firm McKinsey recently called next-generation genomics, which forms the foundation of precision medicine, one of the 12 technologies that will transform life, business and the global economy and anticipates that it will have a global impact of $1.6 trillion by 2025.
The advent of next-generation sequencing (NGS) has given rise to faster and more affordable DNA decoding. The first sequencing of a human genome took 13 years and $2.7 billion to accomplish. Today, using NGS, that task can be completed in a few hours at the cost of about $1,000, by data-analysis companies like Illumina (ILMN).
However, doctors still face the challenge of what to do with that data deluge, and how to respond to the mutations once they’re identified. To date, scientists have identified 3,600 genes related to relatively uncommon single-gene disorders; 4,000 genes connected to more common illnesses like heart disease and diabetes; and several hundred genes responsible for cancer. Only a small portion of those genetic connections have a distinctive treatment that corresponds to them. The known genetic links only account for a minority of the nearly 25,000 genes in the average human, and of course, many diseases affect more that one gene.
Accurately cataloging a person’s or tumor’s genetic profile is relatively easy given decades of technologic advances, but applying that knowledge to predict, prevent or treat disease is significantly more complicated, explained Dr. Eric Lander in an article for the New England Journal of Medicine. Lander, the founding director of Harvard’s Broad Institute (which also partners with Intel), summed it up: “Genetic discoveries are often devilishly hard to apply in practice.”
Clarification, April 23, 2015: An earlier version of this article stated that the goal of the Intel-Edico partnership was to reduce genome-analysis time to 20 minutes; the goal is actually to enable real-time analysis.