Atom­wise's X-37 spin­out gets $14.5 mil­lion to launch AI dis­cov­ery ef­forts

The folks be­hind Atom­wise’s spin­out X-37 like to think in cos­mo­log­i­cal metaphors, and you can think of their AI drug de­vel­op­ment mod­el as probes sent in­to space from a cen­tral sta­tion. That sta­tion just got $14.5 mil­lion in Se­ries A fund­ing from DCVC Bio, Al­pha In­tel­li­gence Cap­i­tal and He­mi Ven­tures to back those mis­sions.

X-37 us­es Atom­wise’s AI plat­form to iden­ti­fy drug tar­gets and – un­like the par­ent com­pa­ny, which large­ly sticks to com­put­ers  – bring those in­to a wet lab and pre­clin­i­cal test­ing.  In ad­di­tion to AI pro­fes­sion­als, it’s led in by part by drug de­vel­op­ers from Ve­loc­i­ty Phar­ma­ceu­ti­cal De­vel­op­ment.

Their busi­ness struc­ture is note­wor­thy and be­com­ing in­creas­ing­ly com­mon for biotechs. Each drug de­vel­op­ment pro­gram is housed as its own vir­tu­al com­pa­ny. CEO David Col­lier used the mod­el at Ve­loc­i­ty and said it al­lows X-37 to cut any failed tar­gets with­out af­fect­ing the rest of the com­pa­ny, or li­cens­ing out the drug and po­ten­tial­ly run­ning in­to in­tel­lec­tu­al prop­er­ty is­sues down the road.

“Twen­ty years ago, if you fund­ed a biotech, and you had 3 to 4 pro­grams, you ran in­to the prob­lem that a phar­ma com­pa­ny want­ed one of them but not the rest,” Col­lier told End­points News.  “Your op­tions were to li­cense the drug or sell the biotech and give up on every­thing else. So peo­ple have been mov­ing to the LLC struc­ture.”

The name X-37 is a ref­er­ence to Boe­ing X-37, a reusable space­craft that has made five trips in­to or­bit and re­turned last month af­ter a record 780 days in out­er space. On their web­site, they ex­plain their vi­sion in space meme for­mat:

In that theme, you can imag­ine their tar­gets as dif­fer­ent probes out to dis­ease plan­ets, ones they could cut ties with if they fail. Those still very ear­ly probes are: Zap-70 for au­toim­mune dis­ease, cGAS/STING for au­toim­mune dis­ease and can­cer, SHP2 for can­cer, PIM 3 for can­cer and Fac­tor XI­Ia for an­ti­co­ag­u­la­tion.

The process for de­vel­op­ment goes like this: Use AI to screen vast li­braries of com­pounds for an ef­fec­tive mol­e­cule, prefer­ably ones for dis­eases or as­says that phar­ma com­pa­nies have un­der­stood for some time but have had dif­fi­cul­ty tar­get­ing. Test 100-300 of the com­put­er’s re­sults in a lab­o­ra­to­ry. Feed the lab re­sults back in­to the com­put­er. Test the com­put­er’s new re­sults. Re­peat.

Col­lier said they’ve gone through one or two cy­cles for most of those probes. They’re aim­ing for a clin­i­cal tri­al by 2022, based on a time­line of a cou­ple more rounds of screen­ing and then a year of IND-en­abling test­ing.

“That’s as­sum­ing suc­cess,” Col­lier said.

Like some oth­er phar­ma­ceu­ti­cal ar­ti­fi­cial in­tel­li­gence com­pa­nies, Atom­wise us­es neur­al net­works to iden­ti­fy com­pounds with the promise of speed­ing up pre­clin­i­cal work and cut­ting at the ex­pand­ing costs, in dol­lars and time, of drug de­vel­op­ment.

It’s an idea that’s been around since at least the 1980s, but one that has at­tract­ed ven­ture dol­lars as com­put­ing pow­er has caught up to the vi­sion. An in­com­plete list com­piled from Bench­Sci, last up­dat­ed this week, counts 167 AI drug dis­cov­ery com­pa­nies, in­clud­ing over 50 that, like Atom­wise, try to gen­er­ate nov­el drug can­di­dates.

Still, these com­pa­nies have yet to gen­er­ate much clin­i­cal im­pact and one of the most high-pro­file com­pa­nies, Benev­o­len­tAI, got tak­en down a peg and a bil­lion dol­lars amid the Neil Wood­ford de­ba­cle. Writ­ing up a re­cent high-pro­file study that claimed to use AI to find a promis­ing new drug can­di­date in 21 days, Derek Lowe wrote the vaunt­ed tech was com­ing – but be pa­tient.

“The good news, though, is that there is no rea­son that vir­tu­al screen­ing can’t do great things, even­tu­al­ly,” Lowe wrote. “We just have to get a lot bet­ter at it than we are now, and that’s as true as it was when I first heard about it in the mid-1980s.”

Un­pack­ing the Aduhelm de­ci­sion, Ver­tex's half full glass, a $525M J&J breakup, and more

Welcome back to Endpoints Weekly, your review of the week’s top biopharma headlines. Want this in your inbox every Saturday morning? Current Endpoints readers can visit their reader profile to add Endpoints Weekly. New to Endpoints? Sign up here.

By now you have surely read about the FDA’s controversial approval of Biogen’s Alzheimer’s drug and all its reverberations. But I’d still recommend checking out the meaty recap below to make sure you didn’t miss all the angles that the Endpoints team has covered. If you’d rather look ahead, look no further than our three-day virtual panels next week at BIO, where we will discuss what the new normal means for every part of the industry.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 109,100+ biopharma pros reading Endpoints daily — and it's free.

What does a clear ma­jor­i­ty of the bio­phar­ma in­dus­try think of the FDA ap­proval of ad­u­canum­ab? 'Hor­ri­fy­ing' 'Dan­ger­ous' 'Con­fus­ing' 'Dis­as­ter'

Over the years, we’ve become used to seeing a consensus emerge early in our industry polls at Endpoints News. And when we took the pulse of drug hunters on the heels of a controversial FDA approval for aducanumab this week, it became immediately apparent that the vast majority of our readers — heavily concentrated among biopharma staffers and execs — were incensed by what they had just witnessed.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 109,100+ biopharma pros reading Endpoints daily — and it's free.

Aaron Kesselheim (Scott Eisen/AP Images for AIDS Healthcare Foundation)

Har­vard’s Aaron Kessel­heim re­signs from ex­pert pan­el in wake of ad­u­canum­ab OK, blast­ing FDA for ‘worst drug ap­proval de­ci­sion in re­cent U.S. his­to­ry'

A third member of the FDA’s Peripheral and Central Nervous System Drugs Advisory Committee has resigned in the wake of Biogen’s controversial Aduhelm approval, slamming the agency as he left and further deepening the controversy surrounding the decision.

Harvard University professor Aaron Kesselheim quit in protest Thursday afternoon, calling the Aduhelm OK “probably the worst drug approval decision in recent U.S. history.” Kesselheim follows both Joel Perlmutter, a neurologist from Washington University in St. Louis, and David Knopman, a neurologist from the Mayo Clinic, out the door.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 109,100+ biopharma pros reading Endpoints daily — and it's free.

David Knopman (Mayo Clinic via YouTube)

A sec­ond ad­comm mem­ber aban­dons his post in af­ter­math of con­tro­ver­sial ad­u­canum­ab de­ci­sion

As the fallout from the FDA’s approval of Alzheimer’s med aducanumab grows, a second member of the adcomm overseeing that drug’s review has walked away. But even with two experts now having resigned from that committee in protest, is there enough broad-level outrage to prevent another aducanumab from getting approved?

The FDA on Wednesday lost another member of its Peripheral and Central Nervous System Drugs Advisory Committee as Mayo Clinic neurologist David Knopman hit the exit over the agency’s decision to approve Biogen’s Alzheimer’s drug Aduhelm despite the committee’s near-unanimous vote against it.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 109,100+ biopharma pros reading Endpoints daily — and it's free.

Reshma Kewalramani, Vertex CEO (BIO via YouTube)

UP­DAT­ED: Ver­tex strikes out on its lat­est big shot at a rare ge­net­ic dis­ease. But they're go­ing to keep on swing­ing

It’s been several months since Vertex culled one of its small molecules for alpha-1 antitrypsin deficiency (AATD), taking a big hit after evidence of liver damage surfaced in a key Phase II trial. Now we learned that the company has whiffed on its second shot, and there’s nothing left in the clinic to treat the rare genetic disease — but that won’t stop it from trying.

Despite avoiding the safety issues that plagued the last candidate, Vertex $VRTX is taking the axe to VX-864 after Phase II results revealed the magnitude of the drug’s response is “unlikely to translate into substantial clinical benefit.” As a result of the news, the company’s stock fell 12.5% after hours.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 109,100+ biopharma pros reading Endpoints daily — and it's free.

FDA au­tho­rizes about 10M J&J vac­cine dos­es, trash­es 60M more from trou­bled Emer­gent plant

The FDA on Friday released about 10 million doses of J&J’s vaccine for use, and disposed of another 60 million doses that were manufactured at the now-shuttered Emergent BioSolutions facility in Baltimore where cross-contamination occurred.

The agency said it’s not yet ready to allow the Emergent plant to be included in the J&J EUA, but that may occur soon. FDA came to the decision to authorize some of the doses after reviewing facility records and quality testing results.

Paul Hudson, Sanofi CEO (Eric Piermont/AFP via Getty Images)

Months af­ter FDA re­jec­tion, Sanofi touts piv­otal win for rare dis­ease drug su­tim­limab as it preps to re­file

One of the pillar drugs of Sanofi’s $11.6 billion pickup of Bioverativ hit a big setback late last year when the FDA sent its application for approval back. Now, as Sanofi gears up to resubmit the drug for review, the drugmaker is touting pivotal data it hopes will help take it over the finish line.

Sanofi’s sutimlimab nailed all three of its primary endpoints in its Phase III CADENZA study for patients with cold agglutinin disease, a rare disorder that can cause severe anemia, without a recent history of blood transfusion, the French drugmaker said Friday. The topline results will be presented at this weekend’s virtual EHA meeting.

Endpoints News

Keep reading Endpoints with a free subscription

Unlock this story instantly and join 109,100+ biopharma pros reading Endpoints daily — and it's free.

Ver­tex and CRISPR Ther­a­peu­tics un­veil more pos­i­tive gene ther­a­py da­ta, but busul­fan again casts a shad­ow over the field

Less than 12 hours after revealing a flop on its second shot for alpha-1 antitrypsin deficiency, Vertex plowed ahead with another data drop from its partnership with CRISPR Therapeutics. And though the topline proved positive, concerns over conditioning agents continue to linger over the collaboration, as well as the entire gene therapy space.

Presenting data from two trials at the European Hematology Association annual meeting, the pair announced that follow-up data of at least three months for 22 patients with genetic blood disorders indicated a “consistent and sustained” response to the experimental drug CTX001. All 15 patients with transfusion-dependent beta thalassemia did not need further blood transfusions and all seven with severe sickle cell disease were pain free, the biotechs announced.

Janet Woodcock, acting FDA commissioner, at Thursday's Senate Appropriations hearing (Bill Clark/CQ Roll Call via AP Images)

Sen­a­tors lam­bast new Alzheimer’s drug’s price but give Janet Wood­cock a free pass on the ap­proval de­ci­sion

Senate Finance Democrats took aim at Biogen’s pricey new Alzheimer’s drug on Thursday, but members on both sides of the aisle at a separate appropriations hearing didn’t question acting FDA commissioner Janet Woodcock on the approval.

“I was appalled that Biogen priced their Alzheimer’s drug approved by the FDA at $56,000 per year — I’m not going to debate whether this is effective or not, but it’s double the household median income for Michiganders over the age of 65,” Sen. Debbie Stabenow (D-MI) said at the finance hearing.