Alex Zhavoronkov (Insilico)

'Longevi­ty as a ser­vice': In­sil­i­co spins off AI biotech that trains deep learn­ing on an­ti-ag­ing treat­ments

Since its in­cep­tion in 2014, In­sil­i­co has racked up deal af­ter deal on the promise that ar­ti­fi­cial in­tel­li­gence can has­ten drug de­vel­op­ment, tout­ing their work on dis­cov­er­ing small mol­e­cules. The biotech’s founder, Alex Zha­voronkov, has pub­lished sev­er­al sci­en­tif­ic pa­pers quan­ti­fy­ing just how im­pact­ful this tech­nol­o­gy can be.

On Tues­day, Zha­voronkov ex­tend­ed his AI ex­per­tise in­to the field of an­ti-ag­ing.

In­sil­i­co is spin­ning off a new biotech, Deep Longevi­ty, fo­cus­ing on de­vel­op­ing and fine-tun­ing such AI tech­niques where Zha­voronkov will al­so serve as founder and CEO. Cou­pled with the launch, Deep Longevi­ty an­nounced the com­ple­tion of Se­ries A fund­ing and a col­lab­o­ra­tion with Hu­man Longevi­ty, which owns the world’s largest data­base of se­quenced genomes. Fi­nan­cial terms of the deal were not dis­closed, but Deep Longevi­ty said in a press re­lease it falls some­where in the “few mil­lion” dol­lar range.

To­geth­er, the biotechs will de­ploy an ex­ten­sive range of AI-pow­ered an­ti-ag­ing ser­vices.

“Age is a uni­ver­sal fea­ture that every liv­ing be­ing has, un­like dis­eases, so you can train from re­al­ly mas­sive da­ta to pre­dict age,” Zha­voronkov told End­points News. “In the past we have ex­per­i­ment­ed with oth­er col­lab­o­ra­tion mod­els, but I think that Hu­man Longevi­ty is by far the most cred­i­ble ex­pert in pre­ven­ta­tive care.”

The an­ti-ag­ing in­dus­try is rel­a­tive­ly new. One promi­nent area re­volves around uti­liz­ing “bi­o­log­i­cal clocks” to de­ter­mine one’s bi­o­log­i­cal age rel­a­tive to their ac­tu­al age to flag po­ten­tial com­pli­ca­tions or rec­om­mend pre­ven­ta­tive treat­ments. Most bi­o­log­i­cal clocks that cur­rent­ly ex­ist mea­sure bi­o­log­i­cal age with DNA methy­la­tion, but Zha­voronkov said such meth­ods have not proven very ac­cu­rate.

In­stead, Deep Longevi­ty will feed sev­er­al dif­fer­ent met­rics in­to its AI sys­tem to pro­duce a com­pre­hen­sive ag­ing re­port. The da­ta can come from blood tests, mi­cro­bio­m­ic in­fo and even voice and imag­ing tests. Fi­nal re­ports, which Deep Longevi­ty has dubbed Age­Met­ric, will pro­duce a sin­gle num­ber that rep­re­sents an in­di­vid­ual’s true bi­o­log­i­cal age and break down their age by da­ta cat­e­go­ry.

“Methy­la­tion da­ta is not very in­ter­pretable, so it’s very dif­fi­cult to un­der­stand what’s re­al­ly go­ing on, why a per­son is ag­ing and what to do about it, what to change to, quote-un­quote ‘get younger,’” Zha­voronkov said. “[Ours] is a deep neur­al net­work where you have 41 pa­ra­me­ters of in­put and just one fea­ture on out­put — your age — and we re­al­ized that this sys­tem is pret­ty pre­dic­tive.”

De­spite heavy ear­ly in­vest­ment in biotechs like Uni­ty and Cal­i­co, an­ti-ag­ing treat­ments have strug­gled to get off the ground. No­var­tis spin­out resTOR­bio reg­is­tered a high-pro­file flop last No­vem­ber af­ter the Phase III study for its lead clin­i­cal pro­gram showed its can­di­date — a TORC1 in­hibitor — per­formed slight­ly worse than a place­bo in pre­vent­ing clin­i­cal­ly symp­to­matic res­pi­ra­to­ry ill­ness in pa­tients 65 and old­er.

But Deep Longevi­ty isn’t fo­cus­ing on drug dis­cov­ery or de­vel­op­ment. By us­ing its AI to cre­ate these re­ports, Zha­voronkov said he hopes the biotech can pro­vide an es­sen­tial ser­vice to its cus­tomer base and the “ul­tra-high-end” hos­pi­tals that treat them.

“The idea is to pro­vide what we call longevi­ty as a ser­vice, how like soft­ware is a ser­vice,” Zha­voronkov said. “You send the da­ta in, you get the num­ber back, and you al­so get this num­ber in­ter­pret­ed. We al­so tell you what has con­tributed to you look­ing old­er or younger and pos­si­ble be­hav­ioral mod­i­fi­ca­tions that the doc­tor could pre­scribe to be able to re­verse this dif­fer­ence.”

As the an­ti-ag­ing in­dus­try grows, so will its cred­i­bil­i­ty, Zha­voronkov said. The biotech chief added that he hopes the new spin-off will be­come the stan­dard-bear­er of the field’s mea­sure­ment tools.

“Sev­en years ago, most com­pa­nies that are nowa­days not on­ly me­dia dar­lings, but now are ac­tu­al­ly run by very cred­i­ble sci­en­tists and drug de­vel­op­ers, did not ex­ist,” Zha­voronkov said. “We’re wit­ness­ing the birth of an en­tire in­dus­try. The rea­son Deep Longevi­ty is cool is be­cause we can ser­vice this in­dus­try. You can­not in­ter­vene if you can­not mea­sure.”

Eli Lilly CEO David Ricks (Evan Vucci/AP Images)

A P val­ue of 0.38? NE­JM re­sults raise new ques­tions for Eli Lil­ly's vaunt­ed Covid an­ti­body

Generally, a P value of 0.38 means your drug failed and by a fair margin. Depending on the company, the compound and the trial, it might mean the end of the program. It could trigger layoffs.

For Eli Lilly, though, it was part of the key endpoint on a trial that landed them a $1.2 billion deal with the US government to supply up to nearly 1 million Covid-19 antibodies.

So what does one make of that? Was the endpoint not so important, as Lilly maintains? Or did the US government promise a princely sum for a pedestrian drug?

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 92,800+ biopharma pros reading Endpoints daily — and it's free.

Noubar Afeyan, Flagship founder and CEO (Victor Boyko/Getty Images)

UP­DAT­ED: Flag­ship launch­es Sen­da Bio­sciences with $88M in back­ing, look­ing to pi­o­neer the field of 'In­ter­sys­tems Bi­ol­o­gy'

Flagship Pioneering has a fresh company out this week, one that aims to lay the groundwork for a whole new discipline.

Senda Biosciences launched Wednesday with $88 million in Flagship cash. The goal? Gain insights into the molecular connections between people and coevolved nonhuman species like plants and bacteria, paving the way for “Intersystems Biology.”

Guillaume Pfefer has been tapped to run the show, a 25-year biotech veteran who comes from GSK after leading the development of the company’s shingles vaccine.

Daphne Koller, Getty

Bris­tol My­er­s' Richard Har­g­reaves pays $70M to launch a neu­rode­gen­er­a­tion al­liance with a star play­er in the ma­chine learn­ing world

Bristol Myers Squibb is turning to one of the star upstarts in the machine learning world to go back to the drawing board and come up with the disease models needed to find drugs that can work against two of the toughest targets in the neuro world.

Daphne Koller’s well-funded insitro is getting $70 million in cash and near-term milestones to use their machine learning platform to create induced pluripotent stem cell-derived disease models for ALS and frontotemporal dementia.

Eli Lilly CEO David Ricks at the Rose Garden, May 26, 2020 (Evan Vucci/AP Images)

Eli Lil­ly lines up a block­buster deal for Covid-19 an­ti­body, right af­ter it failed a NI­AID tri­al

Two days after Eli Lilly conceded that its antibody bamlanivimab was a flop in hospitalized Covid-19 patients, the US government is preparing to make it a blockbuster.

The pharma giant reported early Wednesday that it struck a deal to supply the feds with 300,000 vials of the drug at a cost of $375 million — once it gets an EUA stamp from the FDA. And once that 2-month supply deal is done, the government has an option on another 650,000 doses on the same terms — which could potentially add another $812 million.

CMO Merdad Parsey (Gilead)

Gilead hits the brakes on a tri­fec­ta of mid- and late-stage stud­ies for their trou­bled fil­go­tinib pro­gram. It's up to the FDA now

Gilead $GILD execs haven’t decided exactly what to do with filgotinib in the wake of the slapdown at the FDA on their rheumatoid arthritis application, but they’re taking a time out for a slate of studies until they can gain some clarity from the agency. And without encouraging guidance, this drug could clearly be axed from the pipeline.

In their Q3 report out Wednesday afternoon, the company says researchers have “paused” a Phase III study for psoriatic arthritis along with a pair of Phase II trials for ankylosing spondylitis and uveitis. Late-stage studies for ulcerative colitis and Crohn’s are continuing, but you can see for yourself how big a hole this leaves in the inflammatory disease pipeline, with obvious implications if the company abandons filgo altogether.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 92,800+ biopharma pros reading Endpoints daily — and it's free.

Q32 Bio grabs $60M to kick off hu­man stud­ies for next-gen com­ple­ment drugs — with some Covid-19 tweaks along the way

For a company that launched in the early months of the pandemic, Q32 Bio had its fair share of run-ins with the new normals under Covid-19.

The original plan, for instance, was to conduct first-in-human studies of the IL-7 receptor antibody it licensed from Bristol Myers Squibb in the Netherlands. But they realized shortly after that while the country was beginning to open up clinical trials, there were additional restrictions on drugs that tampered with immunological mechanisms.

Hal Barron, GSK R&D chief

GSK's Hal Bar­ron ax­es a once-prized drug from J&J, con­tin­u­ing shift away from res­pi­ra­to­ry

Hal Barron’s revamp of the GlaxoSmithKline pipeline continued yesterday, as the British pharma announced they axed an asthma drug they once promised over $200 million to acquire.

Then led by CEO Andrew Witty and R&D chief Patrick Vallance, GSK picked up the drug, known elegantly as GSK3772847, from J&J in 2016, hoping to expand on the beachhead in asthma they had established the year prior with Breo Ellipta. They promised up to $227 million in upfront payments and milestones.

Konstantin Poukalov

Per­cep­tive re­cruits A-list in­vestors to back its in-house Chi­na start­up with a mam­moth $310M raise

It took two years for Perceptive Advisors to conceive and boot up LianBio, its big bet on a new kind of in-licensing model for China, seeding it with enough cash to set up two anchoring deals with MyoKardia and BridgeBio. The result was a startup that was all ready to go, reaping $310 million just a little over two months after official launch.

Homegrown Chinese biotechs — many of them boasting of US ties and execs with overseas credentials — have been raking in mega-venture rounds in 2020, both from influential local backers and overseas VC firms that have been loading up new cash. As with IPOs, the deal flow might be slower but the amounts are often more staggering. LianBio’s latest round, unusually, is branded both a Series A and crossover.

Ar­cus and As­traZeneca part­ner on a high stakes an­ti-TIG­IT/PD-L1 PhI­II can­cer study, look­ing to im­prove on a stan­dard of care

For AstraZeneca, the PACIFIC trial in Stage III non-small cell lung cancer remains one of the big triumphs for AstraZeneca’s oncology R&D group. It not only made their PD-L1 Imfinzi a franchise player with a solid advance in a large niche of the lung cancer market, the study — which continues to offer data on the long-range efficacy of their drug — also helped salve the vicious sting of the failure of the CTLA-4 combo in the MYSTIC study.