“Absci really wants to become the Google Index search of protein-based drug discovery and biomanufacturing for large pharmas and biotechs,” founder and CEO Sean McClain declared to GEN Edge.
Since it was established in 2011, Absci’s ambition has encompassed partnering with large biopharmas to use its artificial intelligence (AI) drug discovery platform to help them design better biologics faster.
But this past year, Absci has also concluded that becoming the Google of protein-based drug discovery and biomanufacturing also entails developing its own therapeutics.
“We’ve decided to start to build out our own pipeline as well, because we really believe so much in the platform that it makes a whole lot of sense to start developing out some of our own assets as well,” McClain said. “So, we are both developing our own assets, but then additionally, continuing to rapidly grow the partnerships with large pharma and biotech.”
Absci carries out drug discovery through its own Integrated Drug Creation™ platform, which integrates AI and synthetic biology among other technologies, with the aim of helping users identify novel drug targets, discover optimal biotherapeutic candidates, and generate the cell lines to manufacture them [See below].
The platform is intended to help biopharmas accelerate drug development timelines and decrease attrition in early R&D. In August, Absci researchers posted a preprint on bioRxiv reporting that the Integrated Drug Creation platform rapidly and simultaneously optimized the affinity and naturalness of the HER2-binding antibody trastuzumab (the breast cancer treatment marketed by Roche and its Genentech subsidiary as Herceptin®) in silico using generative AI.
“Additional properties could be added alongside data from their respective assays, such as conditional pH binding, effector function, melting temperature, self-aggregation, viscosity and more,” the researchers asserted. “Harmonizing antibody optimization and de novo generation may be the next big step in data-driven therapeutic design.”
McClain said the preprint showed Absci’s Integrated Drug Creation platform could be used for multi-parametric modeling of antibodies: “You can model for the affinity of the antibody simultaneously with developability attributes, as well as what we call naturalness, which we have shown is inversely correlated to mutagenicity. You can do that in one shot, and you don’t have to go through an iterative process.”
Increasing the success rate
“This is what’s going to really dramatically shorten the time it takes to get drugs into the clinic, and ultimately help with increasing the success rate, because you’re no longer having to give up one attribute for another. You can get all the attributes that you want, all at the same time, and progress those into the clinic,” McClain added. “This was a really big breakthrough for how drug development is done, and also a big breakthrough in just showing how generative AI can be used in drug discovery, and how it can be used to really change the paradigm of how we do drug development.”
Absci reasons that its partnership efforts will be strengthened by having its own pipeline, by demonstrating to large biopharma partners that Absci can also develop its own assets.
“It also shows that we can take drugs to proof of concept,” McClain said. “The more drugs that we that are developed with our platform that show proof of concept, it’s only going to drive more and more partnerships, and so I think it builds a lot of credibility within our partnerships. I see it as actually a huge, huge benefit of developing our own pipeline, just that credibility factor.”
Absci’s pipeline consists of 17 programs, most of them focused on cell line development. Ten of those programs began in 2022, exceeding the company’s investor guidance which called for eight such programs launching this past year.
Absci won’t say what therapeutic areas or modalities it is pursuing, saying the company will disclose that information in 2023. However, McClain did say that Absci’s pipeline will include both wholly-owned candidates and candidates developed through partneships with larger biopharmas.
“We’ll take it to proof of concept in the clinic, and then out-license it, or sell the asset at that point in time,” he said.
The company’s dual focus on manufacturing and drug development is reflected in its nearly year-old partnership with Merck & Co., launched in January 2022 and consisting of up to two collaborations.
One entails applying the Integrated Drug Creation platform to develop production cell lines for antibody-drug conjugate (ADC) candidates, with Merck retaining the option to nominate up to three targets and enter into a drug discovery collaboration agreement. Upon Merck exercising that option, Absci would be eligible to receive up to $610 million in upfront fees and milestone payments for all three targets, as well as research funding and tiered royalties on sales.
“With the target elections for the drug discovery, those discussions are ongoing, and you’re definitely going to be hearing more about that in the future,” McClain vowed.
In the other Absci-Merck collaboration, Absci is deploying its Bionic Protein™ non-standard amino acid (nsAA) technology for targeted incorporation of non-standard amino acids, in return for unspecified upfront and certain other milestone payments.
Technical milestone
During the third quarter, Absci disclosed achieving a technical milestone in the Merck nsAA collaboration that according to McClain accounted for a majority of the $2 million that Absci reported in technology development revenue during Q3. Absci isn’t specifying exactly how much, or detailing the milestone.
“That program has actually been progressing faster than we anticipated,” McClain said. “We couldn’t be more excited about where we sit with that program. It’s progressing extremely well and ahead of schedule.”
Merck is one of a handful of partners disclosed by Absci. Others include:
- Alpha Cancer Technologies, to which Absci licensed its license its E. coli SoluPro® protein expression technology for the production of Recombinant Human Alpha-fetoprotein (hAFP) in 2020.
- EQRx, with which Absci is jointly engineering and developing “several” clinical candidates across therapeutic areas that include oncology and immunology. The discovery partnership combines EQRx’s clinical development expertise and commercial capabilities with Absci’s Drug Creation™ technology, with the goal of advancing next-generation protein-based therapeutics at lower cost.
- PhaseBio Pharmaceuticals, with which Absci is partnering to create a customized cell line to enable high-yield commercial production of an unspecified PhaseBio clinical candidate. The collaboration has remained active though PhaseBio filed in October reorganization under Chapter 11 of the U.S. Bankruptcy Code, and has named Chiesi Farmaceutici as lead bidder with a minimum “floor” bid of up to $100 million at a future auction of assets that include lead candidate bentracimab (PB2452).
In July, Absci announced a partnership with an undisclosed “stealth-mode company” that has developed novel, proprietary warhead linker chemistries that it will target to specified sites on subject antibodies using Absci’s Bionic Protein technology. Under the collaboration, whose value is also undisclosed, Absci has agreed to generate up to seven novel Bionic Antibodies™ incorporating nsAAs as the companies partner on discovery and development of ADCs, initially for oncology indications.
Absci expects its list of biopharma partners to grow in the 2023: “We are into really great discussions with other marquee pharma partners just like Merck.”
Is 2023 too soon to talk about one of Absci’s pipeline candidates getting into the clinic?
“I can’t disclose that. But I would say, yeah, it’s probably a little early,” McClain said.
To oversee development of its own pipeline, Absci has hired Andreas Busch, PhD, to lead its R&D and technical operations, as well as its internal and partnered drug development programs as Chief Innovation Officer.
A veteran biopharma executive, Busch previously held positions at Cyclerion Therapeutics as chief scientific officer and chief innovation officer and head of the innovation center. Earlier, he served as executive vice president, head of research & development and chief scientific officer at Shire before its acquisition by Takeda. Before Shire, he held several senior leadership positions at Bayer, most recently executive vice president and head of Global Drug Discovery for Bayer Pharma, and earlier was Global Head of Cardiovascular Research at Hoechst and Sanofi-Aventis (now Sanofi).
What attracted Absci to Busch was his record of shepherding more than 10 drugs through clinical development to approval. “Andreas brings world-class R&D expertise to our executive leadership team, and continues our momentum toward Absci’s next phase of growth,” McClain said.
Restructuring operations
That next phase began in August 2022 with a restructuring of operations that resulted in Absci eliminating jobs and honing its operations to focus on key technical initiatives including AI drug and target discovery.
Absci won’t say how many jobs or what percentage of its workforce it eliminated, or how much in charges against earnings it has taken to account for the reorganization.
“We were getting much more efficient on the cell line development side of the organization, especially with the integration of AI, and we saw that we didn’t need some of the R&D programs that were going on with the cell line development, and we had enough capacity for the growth that we had,” McClain explained. “We doubled down on our generative AI drug creation platform, and really went all in on that, and then building that out as well as progressing our own assets into the clinic.”
Absci says the reorganization extended its cash and cash equivalents beyond earlier guidance of the end of 2024, enough to fund its operations into late 2025. The company used $19 million in cash for Q3 operating activities, accounting for part of its 48% quarter-to-quarter drop in cash and cash equivalents, which stood at $107.324 million as of September 30—down from $206.021 million as of June 30 and $252.569 million as of December 31, 2021.
However during Q3, Absci gained $73.988 million in short-term investments, giving the company $181.312 million in assets toward funding operations. And for all of 2022, Absci has trimmed its expected total net decrease in cash and cash equivalents, short-term investments and restricted cash to approximately $105 million vs. an earlier estimate of $110 million. The figure includes one-time, time-based disbursements totaling $10.5 million from restricted cash associated with Absci’s acquisitions of AI platform developer Denovium and antibody and target discovery technology developer Totient, paid in the first half of 2022.
Publicly-traded Absci finished the third quarter with a net loss of $27.259 million, up from a net loss of $23.654 in the year-ago quarter, on total revenue that rose 54% to $2.369 million from $1.539 million in Q3 2021. For the first three quarters of 2022, Absci’s net loss jumped 75% year-over-year, to $85.657 million from $48.982 million.
Absci went public in July 2021 through an initial public offering that generated approximately $210 million in net proceeds, at the tail end of an historic bull market for biotechs. Three months later, Absci opened a new 85,000-square-foot headquarters in Vancouver, WA, where the company has been based since 2016. McClain founded Absci in 2011 from a tiny basement lab where he began pursuing a vision of revolutionizing the development of protein-based drugs by using DNA to program E. coli bacteria.
Pursuing AI talent
On the East Coast, Absci has bolstered its AI capabilities—and moved toward its goal of de novo drug and cell line design entirely in silico—by opening the Absci AI Research Lab in New York City. The New York AI team is roughly 15–20 scientists, of which half are based at the lab, located on the 43rd floor of Carnegie Hall Tower, which boasts panoramic views of Central Park; the rest work remotely.
“What really encouraged us to open up an office here in New York is the AI talent that’s here,” McClain said. “Meta is here. Google is here a lot. [AI startup] Hugging Face is here. There’s just an abundance of AI talent that is here in the city. Being able to recruit talent here has definitely made it a lot easier having an office space to house the collaboration of the team.”
Meta’s vice president of connectivity Dan Rabinovitsj joined Absci’s board in November: “We are seeing tech companies get more and more interested in drug development or discovery, biotech/life sciences. And I think that in the future I think that you could see some really interesting partnerships emerge with large tech companies as well. So, having someone like Dan, who is a senior executive at one of the large tech companies in particular, Meta or Facebook, we saw as a huge, huge advantage.”
Other strengths Rabinovitsj brings to Absci, McClain said, include his experience rapidly scaling the work of tech organizations as an executive, and his ability to navigate large tech organizations for recruiting efforts. Absci and other AI-based startups have benefited from talent leaving Meta, which in November announced plans to lay off 13% of its staff or more than 11,000 people.
AI, specifically support for the Integrated Drug Creation platform, is one area of projected job growth for Absci in 2023, along with key hires to Busch’s drug discovery team. The company isn’t specifying how much its workforce will grow beyond its current size of about 200 employees (Absci reported 212 employees, all but one full-time, as of December 31, 2021, according to its Form 10-K annual report for 2021, filed in March.
“At the end of the day, the team is everything, and we just have one of the best teams. And it’s just an amazing team that has really built an amazing platform that is really going to be able to help us see the vision through, which is developing biologics at a click of button,” McClain said.
He and Absci have high hopes for the team in 2023.
“You’re going to see continued innovation on the AI drug creation side. You’re going to see development and publications come out on where we set on our de novo AI platform. And then additionally, you’re going to see continued growth on the partnership side, more marquee deals like the ones with Merck, and I think some progress on our own pipeline as well,” McClain added. “We’re in for a really exciting 2023. I couldn’t be more thrilled with the direction we’re going as a company, the progress we’ve made.”
“To make generative AI drug creation actually work and see our vision through, you have to have that perfect integration of wet lab and wet lab technology, along with the AI and the AI scientists, and being able to collaborate in a very well-oiled fashion, and we really have that.”
Absci’s integrated drug creation™ platform
Absci’s Integrated Drug Creation™ platform designed to unify drug discovery and cell line development into a single efficient process involving several steps:
- Predicting sequences that lie beyond those that exist in nature, using the platform’s Denovium Engine™ AI models.
- Identifying corresponding antigens as potential drug targets starting with disease tissue samples or bulk RNA sequencing data, by reconstructing sequences of human monoclonal antibodies (potential drugs) prevalent in the tissue.
- Designing custom libraries of protein-of-interest variants in the desired scaffold architecture, specifying any desired nsAA placements, and determining any applicable parameters for folding and expression solutions to evaluate.
- Constructing billions of genetically-distinct SoluPro® or Bionic SoluPro® cells, each containing instructions to make one version of the protein of interest, as well as a different assortment of folding and expression solutions, all using synthetic biology.
- Screening via the flow cytometry-based Activity-specific Cell Enrichment (ACE) Assay™ to evaluate and sort billions of cells, then collect subsets of the population that express the best hits—versions of the protein of interest based on target binding, protein quality, and expression titer.
- Selecting leads based on protein function, quality, and titer, as well as additional biophysical attributes impacting development potential, through the HiPrBind™ Assay, which uses automated multiplexed plate-based methods to grow micro-batches of each of the thousands of hits from the ACE Assay™.
- Scaling up the selected lead cell line(s), in order to optimize conditions for productivity and protein quality, and perform comprehensive analytics on the lead drug candidate(s) for evaluation and technology transfer to partners.