Gilead Sciences will use the AI-based platform of insitro to discover and develop treatments for nonalcoholic steatohepatitis (NASH), the companies said today, through a collaboration that could generate more than $1 billion for the South San Francisco, CA, startup focused on drug development using machine learning and data science tools.

The companies have inked a three-year collaboration agreement through which the insitro Human (ISH) platform will be used by Gilead to create disease models for NASH, as well as discover targets deemed to influence the clinical progression and regression of the disease.

ISH is designed to drive therapeutic discovery and development by applying machine learning, human genetics, and functional genomics to generate and optimize unique in vitro models. The platform is expected to offer insights into disease progression, suggest candidate targets, and predict patient responses to potential therapeutic interventions, Gilead and insitro said.

John McHutchison, AO, MD, Gilead’s CSO and head of research and development, said in a statement that the collaboration reflected his company’s commitment to researching and developing treatments for patients living with NASH—particularly those with advanced fibrosis who have the greatest unmet need.

“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,” McHutchinson stated.

Gilead’s liver disease pipeline includes four candidates in clinical phases led by selonsertib (formerly GS-4997), a once-daily, oral inhibitor of apoptosis signal-regulating kinase 1 (ASK-1) now in Phase III trials for NASH and Phase II studies for diabetic kidney disease.

On February 11, however, Gilead acknowledged that selonsertib failed the Phase III STELLAR-4 trial in NASH by failing to meet the pre-specified week 48 primary endpoint of a ≥ 1-stage histologic improvement in fibrosis based on the NASH Clinical Research Network (CRN) classification without worsening of NASH, compared with placebo.

The STELLAR-4 trial (NCT03053063) is one of three Phase III studies of selonsertib. Gilead is awaiting results from the STELLAR-3 (NCT03053050), which is assessing selonsertib in patients with bridging fibrosis (F3) due to NASH; and ATLAS (NCT03449446), a Phase II combination trial of selonsertib and two other Gilead drug candidates, the farnesoid X receptor agonist cilofexor (GS-9674) and acetyl-CoA carboxylase (ACC) inhibitor firsocostat (GS-0976) in patients with advanced fibrosis due to NASH.

Up to five targets

Through the new collaboration with insitro, Gilead can advance up to five targets identified by the companies, and will oversee chemistry and development against these targets.

Gilead agreed to pay insitro $15 million upfront, and up to $35 million in near-term payments tied to achieving operational milestones. Gilead also agreed to pay insitro up to $200 million tied to achieving preclinical, development, regulatory, and commercial milestones for each of the five Gilead targets. Insitro is eligible as well for up to low double-digit tiered royalties on net sales.

For programs where insitro opts in, it will have the right to co-develop and co-detail in the U.S., receive a profit share in China and receive milestone payments and royalties on other ex-U.S. sales, the companies said.

Insitro aims to build predictive models aimed at accelerating target selection and improving the design of effective therapeutics by generating high-throughput, functional genomic data sets that align with patient data, and interpreting those data via novel machine learning methods.

“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,” added Daphne Koller, PhD, insitro’s CEO and founder.

Koller launched insitro last year with over $100 million in Series A financing raised by investors that include some big names—including Jeff Bezos’ venture capital firm Bezos Expeditions, GV (formerly Google Ventures), and Google parent Alphabet’s life science research organization Verily.

Also among investors in insitro are Andreessen Horowitz (also known as a16z), Arch Venture Partners, Foresite Capital, and Third Rock Ventures, with additional investments from Alexandria Venture Investments, Abu Dhabi-based Mubadala Investment Company, Two Sigma Ventures, and other undisclosed investors.

In launching insitro, Koller cited the potential for machine learning to reduce the sky-high cost of traditional drug development—an estimated $2.558 billion, according to a 2016 study by Joseph A. DiMasi, PhD, and colleagues, at Tufts Center for the Study of Drug Development, and published in Journal of Health Economics—and the diminishing annual return on drug development investment.

“Our hope at insitro is that big data and machine learning, applied to the critical need in drug discovery, can help make the process faster, cheaper, and (most importantly) more successful,” Koller stated last year in announcing the launch.

Previous articleAlzheimer’s Disease Toxicity Halted by Synthetic Peptides
Next articleNew Microscopy Technique Measures Activity of Deep Brain Regions