EVENTS | VIEW CALENDAR
Machine learning vs. NASH
FOSTER CITY, Calif. & SOUTH SAN FRANCISCO, Calif.—“Go big or go home,” the old adage says, and drug discovery and development company insitro is taking that to heart, having announced that its very first collaboration since being founded is with an industry veteran—Gilead Sciences Inc.
It was just a year ago that insitro launched, in May 2018, with more than $100 million raised in a Series A. Investors included a16z, Arch Venture Partners, Foresite Capital, GV (formerly Google Ventures) and Third Rock Ventures, with additional investments from Mubadala Investment Company, Two Sigma Ventures, Verily, Alexandria Venture Investments, Bezos Expeditions and other undisclosed investors. The company focuses on using cutting-edge machine learning techniques and artificial intelligence to enable predictive modeling and streamline drug development.
This alliance with Gilead is a strategic collaboration under which the companies will be working to discover and develop therapies to treat nonalcoholic steatohepatitis (NASH).
Per the terms of their agreement, insitro will receive $15 million up front, with the potential for additional near-term payments of up to $35 million if certain operational milestones are met. In addition, insitro stands to receive up to $200 million for certain preclinical, development, regulatory and commercial milestones for each of the five targets, as well as up to low double-digit tiered royalties on net sales. Regarding the programs that insitro elects to opt in to, the company has the right to co-develop and co-detail in the United States, receive a profit share in China, and receive milestone payments and royalties on ex-U.S. sales.
“Gilead is committed to researching and developing treatments for patients living with NASH, particularly those with advanced fibrosis who have the greatest unmet need,” said Dr. John McHutchison, chief scientific officer and head of research and development for Gilead Sciences. “We are excited about the opportunity to partner with insitro to tackle the scientific challenges associated with this complex disease. Through this collaboration, we will utilize deep learning to explore the scientific underpinnings of the biology and clinical spectrum of NASH, with the goal of accelerating the development of highly effective treatment options for patients with this disease.”
The collaboration will run three years, during which insitro will apply its proprietary platform to create disease models of NASH and identify targets that affect clinical progression and regression of the disease. That platform—the insitro Human (ISH) platform—combines machine learning, human genetics and functional genomics to create and optimize in-vitro models and accelerate therapeutic discovery and development. In addition to exploring the disease and potential targets, the ISH platform can also predict patient responses to therapies. Gilead will be able to advance up to five targets identified under the collaboration, and will be responsible for handling chemistry and development for its selected targets.
“NASH is a progressive liver disease that can lead to fibrosis, cirrhosis and liver cancer, and will soon be the predominant cause of liver transplantation in the U.S.,” commented Dr. Daphne Koller, CEO and founder of insitro. “We are excited to work with Gilead, a leader in liver disease, in bringing to bear novel tools toward identifying new therapeutics for NASH and helping the many patients in need around the world.”
When insitro launched, it was in conjunction with a blog post by Koller, who noted at the time that drug development is becoming more and more difficult: “[C]linical trial success rates hover around the mid-single-digit range; the pre-tax R&D cost to develop a new drug (once failures are incorporated) is estimated to be greater than $2.5B; and the rate of return on drug development investment has been decreasing linearly year by year, and some analyses estimate that it will hit 0 percent before 2020.”
The insitro approach, Koller wrote in her post, is to train its machine learning models with very large data sets in hopes of tackling the issues hampering drug discovery and development today: “[W]e will use high-quality data that has already been collected, but we will also invest heavily in the creation of our own datasets using high throughput experimental approaches, datasets that are designed explicitly with machine learning in mind from the very start. The ML models that are developed will then help guide subsequent experiments, providing a tight, closed loop integration of in-silico and in-vitro methods (an insitro paradigm).”
“The solution to this problem cannot be that we continue to pay enormous amounts to develop new drugs, most of which fail, and then pass those costs on to our patients,” Koller stated. “This is neither economically sustainable for society, nor is it ethical, since it prices many new drugs out of reach for many people who need them. We must find a different approach to drug development.”
NASH is a chronic liver disease that is characterized by excess fat in the liver, inflammation and liver cell damage, which can cause scarring or fibrosis and eventually lead to cirrhosis or liver cancer. At present, only limited approved treatments exist for the estimated 3 to 12 percent of adults in the United States who have NASH, per the National Institute of Diabetes and Digestive and Kidney Diseases, part of the National Institutes of Health. The NASH Education Program, a Genfit initiative, reports that an estimated 1.5 to 6.45 percent of adults worldwide have NASH (Younossi, Z.M. et al.).