2019-nCoV drug discovery: HHS funds Regeneron mAbs and BenevolentAI deploys machine learning

Regeneron and BenevolentAI unveiled a deal and a platform, respectively, on Tuesday that could yield therapies against 2019-nCoV that go beyond anti-inflammatory drugs for other viral infections.

The first nine clinical trials of medicines for 2019-nCoV acute respiratory disease are evaluating approved HIV, HCV, flu or HBV drugs and anti-inflammatory steroids (see “Clinical Trials for Coronavirus”).

By contrast, Regeneron Pharmaceuticals Inc. (NASDAQ:REGN) is developing mAbs to neutralize the virus with new funding from HHS, and BenevolentAI is using its machine-learning platform to identify drugs such as rheumatoid arthritis therapy Olumiant baricitinib that can be repurposed to stop cells from internalizing 2019-nCoV.

Regeneron’s spike shield

Regeneron and the Biomedical Advanced Research and Development Authority (BARDA) expanded a 2017 mAb discovery and development collaboration to include mAbs against 2019-nCoV.

Shares of the company rose $14.88 to $356 Tuesday, gaining $1.6 billion in market cap.

Christos Kyratsous, Regeneron’s VP of research for infectious diseases and viral vector technologies, told BioCentury last week that it had already begun discovery work (see “The Race Is On”).

Regeneron declined to disclose the financial terms of the deal expansion. But under the original deal terms, HHS had agreed to fund 80% of Regeneron’s R&D and manufacturing costs for antibodies chosen for advancement, for up to 10 pathogens.

Prior to the 2017 deal, the company had validated the utility of its mAb platform, VelocImmune, to rapidly yield clinical antibodies candidates against antigens of emerging pathogens. In a matter of months, Regeneron was able to generate antibody candidates against Middle East respiratory syndrome coronavirus (MERS-CoV) and Ebola virus. Regeneron is co-developing with NIH a three-mAb Ebola therapy -- which reduced mortality from 67% to 29% in a clinical trial in the Democratic Republic of the Congo -- and a MERS therapy comprising REGN3048 and REGN3051 (see “Two Ebola Therapies Best ZMapp in Congo Trial”).

REGN3048 and REGN3051 both target MERS-CoV S protein, the spike protein on the virus’ surface.

Regeneron spokesperson Alexandra Bowie told BioCentury that mAbs against 2019-nCov “will likely target the spike protein, as we have seen this to be effective against other coronaviruses.”

She did not provide a firm development timeline, but noted that Regeneron generated the Ebola mAbs and validated their efficacy in animal models in a “six-month time frame” and added that it took “about six more months from there to be ready to test in humans.”

While the company also has its MERS mAbs in preclinical testing for 2019-nCoV, “We can’t say there is a high likelihood of them working against 2019-nCoV” because it’s “fairly distinct from MERS,” said Bowie.

According to a genome analysis published Jan. 30 in The Lancet, 2019-nCoV has only about 50% overlap in sequence identity with MERS-nCoV.

BenevolentAI learns how to block entry

In a Lancet correspondence published Tuesday, an Imperial College London and BenevolentAI team used a machine-learning platform to unveil therapeutics that could be repurposed for 2019-nCoV acute respiratory disease, marking the entry of artificial intelligence technology into 2019-nCoV drug discovery race.

BenevolentAI’s machine-learning platform uses private and public genomic databases, as well as the scientific literature and databases, to predict therapeutic targets or candidates against a target.

The virus is thought to enter cells through ACE2 binding, which triggers endocytosis, a process promoted by AAK1.

Application of the platform to 2019-nCoV identified six compounds previously tested in the clinic that were predicted to bind AAK1 with high affinity, including Olumiant baricitinib, a dual JAK-1 and JAK-2 inhibitor from Eli Lilly and Co. (NYSE:LLY) and Incyte Corp. (NASDAQ:INCY).

BenevolentAI’s platform also predicted Olumiant binds GAK, another endocytosis promoter.

The authors suggested testing Olumiant in 2019-nCoV acute respiratory disease as a way to hinder viral entry while reducing the inflammation resulting from infection.

Further analysis of the coronavirus crisis can be found at https://www.biocentury.com/coronavirus. The 2019n-CoV content is free to all who visit the site.

Targets

AAK1 - AP2 associated kinase 1

ACE2 - Angiotensin-converting enzyme 2

GAK - Cyclin G associated kinase (GAK)

JAK-1 - Janus kinase-1

JAK-2 - Janus kinase-2

Trial Subscription

Get a two-week free trial subscription to BioCentury

SIGN UP

Article Purchase

This article may not be distributed to non-subscribers
More Info >PURCHASE