1. The Molecule1 / 13
The MoleculeMechanism of ActionPreclinical & Clinical EvidenceSafety ProfileIP LandscapeTeamFinancialsRegulatory PathwayMarketCompetitive LandscapeRisks & Open QuestionsOptionalityVerdictReferences

BigBio.ai Signal Report

Absci Corporation

NASDAQ: ABSI · Absci Corporation, generative-AI biologic discovery platform plus clinical pipeline (ABS-101 anti-TL1A -- outlicensing; ABS-201 anti-PRLR -- sole internally advancing asset) · 2026-04-27

Absci's sole internally advancing asset, ABS-201, enters PRLR second behind Hope Medicine's Bayer-derived HMI-115, already Phase 3-registered in endometriosis (NCT07318688, n=540); former lead ABS-101 went to the outlicensing desk five months after dosing, its half-life read by the analyst class as below next-generation comparators. Twelve months of clinical history; enterprise value ~$256-306M.

1. The Molecule

1.1 ABS-101 (anti-TL1A, IBD) -- from lead asset to outlicensing in six months

ABS-101 is a subcutaneous anti-TL1A monoclonal antibody engineered for extended half-life (see Section 4.1 for the Morgan Stanley counter-read that the same data "trailed next-generation competitors"). Absci built it to compete in inflammatory bowel disease on dosing convenience, not target novelty. Absci Pty Ltd sponsored the first-in-human Phase 1 at Nucleus Network in Melbourne (ACTRN12625000212459); the first participant dosed on 12 May 2025 1. The US IND cleared but never opened a US site 2; the single Australian site narrowed US-investor visibility on the program.

The trial ran a single-ascending-dose design: four subcutaneous cohorts (150, 300, 600, 1000 mg) plus an IV bridge cohort at 300 mg, eight participants per cohort, target n=40 34. The ANZCTR registry has not updated since 2025-07-22, when accrual stood at 15 of 40 5. The November 2025 interim readout therefore rested on fewer than twenty participants.

The interim disclosure described "extended half-life as compared to first-generation TL1A antibodies" with "no apparent impact of ADA on PK" 6. The framing omitted the second-generation programs that now define the class -- duvakitug (Sanofi/Teva, SC Q2W, $1.5B Sanofi license) and afimkibart (Roche, t1/2 18-19 days, $7.1B Telavant acquisition). Morgan Stanley read the same numbers and downgraded ABSI from Overweight to Equal-weight on 2026-01-08, cut its price target from $5.80 to $4.32, and removed ABS-101 from its model 78. The MS note states ABS-101 "showed half-life characteristics not in line with next-generation TL1A comparators" 9.

Six days after the interim release, Absci announced the deprioritization: "the Company has made the strategic decision not to initiate additional later-stage development trials for ABS-101 internally at this time" 5. Absci now seeks a partner. The TL1A class precedent on half-life-engineered programs is unhelpful: BB-TL1A-VIAL-HLE (NCT07029971, sponsor Vial Australia -- not Boehringer Ingelheim) terminated at interim (n=16) on PK endpoint failure 10. Animal-to-human PK translation in HLE-engineered TL1A is not guaranteed.

The class is also commercially crowded.

Merck paid $10.8B for Prometheus in April 2023 to secure tulisokibart 11.

Roche paid $7.1B upfront plus $150M in milestones for Telavant's afimkibart in October 2023 12.

Sanofi licensed duvakitug from Teva for EUR 469M upfront plus EUR 940M milestones in October 2023 13.

AbbVie paid FutureGen $150M upfront plus $1.56B milestones for the preclinical FG-M701 in June 2024 14.

Boehringer Ingelheim entered the class on 2026-01-27 by licensing Simcere SIM0709 (TL1A x IL-23p19 bispecific) for EUR 1.058B 15 -- a re-entry, not an exit.

ABS-101 enters fifth or sixth in line with one differentiation story the analyst class has rejected 9.

1.2 ABS-201 (anti-PRLR, AGA + endometriosis) -- sole internal asset, second mover

ABS-201 is an AI-designed anti-prolactin receptor monoclonal antibody from Absci's Origin-1 platform. The company first disclosed the target and lead indication at its December 2024 R&D Day; prior quarters had described it only as "an undisclosed dermatology target." First-in-human dosing occurred on 3 December 2025 in the HEADLINE Phase 1/2a trial (NCT07317544), a quadruple-masked, 227-participant study at Sinclair Dermatology and Nucleus Network in Melbourne 1617. The design runs four to six SAD IV cohorts (150-1800 mg, a twelve-fold dose span) followed by three to four MAD subcutaneous cohorts (300-1200 mg) 18.

Two design facts qualify the marketing language. First, the registered primary endpoint is safety alone; hair-growth measures (TAHC, TAHW, TAHD, IGA, self-assessment) sit in the otherOutcomes tier -- neither secondary nor primary 19. Absci's "powered to demonstrate human proof-of-concept" framing for the MAD portion does not match the registry 19. Second, the pipeline page shows ABS-201 AGA "Phase 1/2a at 85%" and ABS-201 endometriosis "Phase 1 at 85%" although three of four-to-six SAD cohorts have completed and no endometriosis trial has been registered 20; Phase 2 endometriosis initiation guidance is Q4 2026. Progress bars overstate program momentum.

The mechanism-priority claim fails the public record. Hope Medicine's HMI-115 -- licensed from Bayer in 2019 -- published Phase 2 endometriosis results in Lancet Obstetrics, Gynaecology, and Women's Health on 2025-11-05 (NCT05101317, n=108 completed, randomized 1:1:1:1 to 60/120/240 mg HMI-115 vs placebo, SC Q2W x 12 weeks) and registered Phase 3 endometriosis on 2025-12-29 (NCT07318688, n=540) 2122. HMI-115's Phase 2 male AGA trial (NCT06118866, n=192, single-site Peking University People's Hospital) primary-completed on 2024-08-26 and remains unpublished as of report date 21. ABS-201 is second mover in both indications, with HMI-115 sitting one phase ahead in endometriosis and one published readout ahead in mechanism validation.

The preclinical package behind ABS-201 rests on company disclosures, not peer-reviewed publications. Non-human primate studies showed greater than 90% subcutaneous bioavailability and projected Q8W-Q12W dosing in humans 23. Absci targets a high-concentration formulation at 150 mg/mL or above 23. In December 2025 the company described "statistically superior hair regrowth compared to minoxidil in a preclinical mouse model"; no effect size and no peer-reviewed publication have followed, and the telogen-to-anagen-transition mouse AGA model translates poorly to human androgen-driven follicular miniaturization 24. March 2026 added human ex vivo data showing STAT5 phosphorylation inhibition and anagen prolongation in isolated follicles, mechanistically coherent with the published PRLR follicle literature 25.

1.3 ABS-301 (immuno-oncology, undisclosed target)

ABS-301 remains at Lead ID stage on the pipeline page 20. Absci has not disclosed the target since its December 2024 unveiling; a Q3 2025 update reported completion of "the first in vivo target validation study" with signaling-pathway-driven anti-tumor response 6. December 2024 guidance called for candidate selection in 1H 2025; sixteen months later the asset still shows Lead ID 20. The timeline slipped without acknowledgement.

1.4 ABS-501 (anti-HER2)

ABS-501 is an AI-designed anti-HER2 antibody from the same platform, displayed as Candidate at 0% 20. Absci describes it as having "novel epitope interactions, affinity equal or greater than trastuzumab in preclinical, efficacious against trastuzumab-resistant xenograft." The international application WO2025122885A1 was filed 8 December 2023 with no granted US composition-of-matter patent as of report date 26. Trastuzumab is a twenty-five-year-old benchmark; preclinical affinity parity is the table stakes for entry, not differentiation.

1.5 History -- from SoluPro to Origin-1

Absci began in a 400-square-foot basement lab in 2011 27. Founder Sean McClain holds inventorship on the original SoluPro/SoluPure cytoplasmic expression patent (US12157766B2, priority 2012-08-05, granted 2024-12-03) 28, and the earliest Absci-assigned US patent (US9617335B2) granted in 2017 29. The original IP base was microbial protein expression. Absci pivoted to generative-AI antibody design later, accelerated by the June 2021 Totient acquisition (US12285484, US12134655) 3031. The Origin-1 platform brand launched in 2026; its predecessor brand was "Integrated Drug Creation."

2. Mechanism of Action

This report scores target validation on a 0-100 scale: TL1A = 85 1232, PRLR = 42 21. (Author construct; underlying evidence cited section by section.)

2.1 TL1A (target validation score: 85)

TL1A drives T-cell activation and fibroblast-mediated tissue damage in inflammatory bowel disease 32. Three independent lines of evidence anchor the target: GWAS signals in TNFSF15 associated with both ulcerative colitis and Crohn's disease; elevated TL1A expression in inflamed gut mucosa; preclinical knockout and antibody-blockade models that attenuate colitis 32. Clinical validation has now replaced preclinical inference 33. Tulisokibart's Phase 2 ARTEMIS-UC posted 26% clinical remission versus 1% placebo at 12 weeks (NEJM 2024) 331232; duvakitug RELIEVE reported 48% UC remission at 900 mg vs 20% placebo 34; Roche's afimkibart TUSCANY-2 reached 36% remission at Week 56 12.

The class's commercial validation matches its biological validation 1232.

Merck paid $10.8B for Prometheus to secure tulisokibart 11.

Roche paid $7.1B for Telavant six months later 12.

TL1A scores 85 -- biologically established 32, clinically proven 33, commercially priced 1114.

ABS-101 inherits all of this validation. Its question is not whether TL1A is a real target but whether a fourth-place HLE-engineered program can earn Phase 2/3 capital from a licensee whose closest comparator (BB-TL1A-VIAL-HLE) just terminated at n=16 10.

2.2 PRLR (target validation score: 42)

Prolactin receptor signals through JAK2-STAT5 in hair follicles, breast, ovary, bone, and pituitary 25. Prolactin engagement at PRLR drives the catagen phase of the hair cycle; PRLR blockade prolongs anagen in mouse models 25. Absci's human ex vivo data confirm STAT5 inhibition and anagen prolongation in isolated follicles. In endometriosis, PRLR signaling is upregulated in ectopic lesions and drives inflammation independently of the GnRH axis that current therapies target.

Clinical validation rests on one program: HMI-115 21.

The Lancet OGWH paper (2025-11-05; n=108 completed) reported a -41.57% LSM change from baseline in dysmenorrhea NRS at week 12 in the HMI-115 240 mg arm, placebo-adjusted -22.96%, p=0.036 21.

Only the 240 mg dose reached statistical significance against placebo; the 60 mg arm gave p=0.409 and the 120 mg arm gave p=0.131 21.

Non-menstrual pelvic pain reduction reached roughly 52% LSM versus placebo 21.

The trial reported no sex-hormone disruption and no change in BMD, estradiol, LH, FSH, or progesterone 21.

An AGA Phase 1b (Hope Medicine, n=12 males, open-label, single-arm) reported a +14 hairs/cm-squared mean non-vellus TAHC gain over 24 weeks per the January 2024 release 21; this is not a Phase 2 benchmark.

Bayer's own mouse endometriosis data (Otto et al., 2022) showed efficacy equivalent to faslodex and cetrorelix without anti-estrogenic effects 25.

Two cautions temper the validation. Novartis's anti-PRLR LFA102 failed Phase 2 in breast cancer -- a different indication, but the only prior anti-PRLR in a large efficacy study 21. Bayer's own BAY 1158061 completed Phase 1 in 2018 (PMID 29806538) with favorable PK and posted no Phase 2 registry entry in the eight years before HMI-115 returned the molecule to the clinic 21. The silence is unusual for a drug with positive preclinical efficacy data. PRLR scores 42 -- the mechanism is biologically supported, one competitor has generated supportive Phase 2 endometriosis data at one of three doses 21, and no anti-PRLR therapy is approved.

2.3 Origin-1 platform claims

Absci's platform asserts de novo antibody design against "zero-prior" epitopes -- targets without published protein-protein complex structures or homologs above 60% sequence identity 35. The 2026 bioRxiv preprint (Levine et al.) reports developable binders for four of ten targets (COL6A3, AZGP1, CHI3L2, IL36RA) with cryo-EM atomic accuracy (3.0-3.1 Angstrom, DockQ 0.73-0.83) and an IL36RA binder optimized to 104 nM 35. Absci characterizes this as "the first demonstration of de novo design of full-length, monoclonal antibodies against 'zero-prior' epitopes with atomically accurate complex structures and functional activity" 35.

Three constraints bound the platform claim.

The 4/10 success rate is a 40% hit rate; six of ten targets failed to produce developable binders 35.

All 54 authors on the Origin-1 preprint are Absci employees or contractors; no independent group has published a validation 35.

The Baker Lab's RFdiffusion (Bennett et al., Nature 2025-11-05) demonstrated atomically accurate de novo design of antibodies against zero-prior epitopes -- TcdB, influenza HA, RSV, SARS-CoV-2 RBD, IL-7Rα -- at 0-2% per-design success rates 36.

The peer-reviewed Baker Lab demonstration narrows Absci's "first demonstration" preprint claim, and Xaira Therapeutics -- co-founded by Baker, capitalized at $1B+ 3738 -- is the venture-backed catch-up vehicle.

No peer-reviewed PubMed paper describes the de novo antibody-design platform method itself -- five PubMed Absci-affiliation hits exist, but each covers histology AI (Iyer et al., Nat Med 2024, primary affiliation PathAI), HIV antibody biology (Spencer et al., Nat Comm 2022, primary OHSU), an EDI workshop report (Plant Direct 2022), QuickIsoSeq RNA-seq (Methods Mol Biol 2021, primary Pfizer), or pre-pivot Vibrio biology (Appl Environ Microbiol 2020) 5.

None describes the Origin-1 method 5.

The 2023 flagship anti-HER2 de novo bioRxiv preprint (Shanehsazzadeh et al., DOI 10.1101/2023.01.08.523187) remains unreviewed three years after posting 35.

Absci's most distinctive claim -- zero-prior de novo design -- has not been tested against a human clinical target 35.

ABS-201 uses PRLR, which has extensive prior structural knowledge 25.

3. Preclinical & Clinical Evidence

3.1 ABS-101 Phase 1 -- the half-life comparison Absci did not make

The November 2025 disclosure described "extended half-life as compared to first-generation anti-TL1A competitor programs, with no apparent impact of ADA on PK" 6. Three things the framing omits:

  1. The numeric half-life of ABS-101. No company disclosure provides t1/2 in days. Morgan Stanley filled the gap on 2026-01-08: "trailed next-generation competitors" 9.
  2. The comparator. "First-generation anti-TL1A" omits duvakitug (SC Q2W) 34 and afimkibart (t1/2 18-19 days) 12 -- the second-generation programs that define the class.
  3. The sample size. ANZCTR accrual stood at 15 of 40 on 2025-07-22 with no subsequent update 5; the interim release rested on fewer than twenty participants.

A December 2024 R&D Day disclosure reported "reduced internalization of TL1A complexes in in vitro THP-1 immunogenicity tests compared to a competitor molecule with a high clinical anti-drug antibody (ADA) rate" 6. The unnamed comparator is likely PF-06480605 (the tulisokibart precursor), flagged in the literature as a high-ADA molecule (no primary source -- author estimate; specific ADA rate not verified in canonical). That context validates the engineering choice but does not translate to a head-to-head ADA comparison in humans. The 13-week GLP toxicology reported "no treatment-related adverse findings during in-life phase and necropsy were observed; histopathology is pending" 6. Absci has not updated the "histopathology is pending" qualification since December 2024.

The Phase 1 outlicensing decision followed the interim readout by six days 5. Had the data cleanly beaten the second-generation competitors on head-to-head-comparable PK metrics, the narrative and the timing would likely have differed.

3.2 ABS-201 HEADLINE -- three SAD cohorts and a proof-of-concept that is still to come

HEADLINE's three completed SAD IV cohorts have reported favorable emerging safety data and no serious adverse events 1617. The proof-of-concept readout requires the MAD SC portion, which has not yet started. Company guidance calls for preliminary safety/tolerability/PK data in 1H 2026, interim POC in H2 2026, and full POC in early 2027 18. The cadence is reasonable for a Phase 1/2a that began dosing on 3 December 2025 16.

Two design features warrant note. The SAD IV dose range of 150-1800 mg spans twelve-fold; for a molecule destined for subcutaneous administration, the 1800 mg top IV dose reads as a supratherapeutic-exposure safety probe more than a clinical development step. The registered statistical plan does not power any hair-growth endpoint as primary or secondary -- TAHC, TAHW, IGA, and self-assessment all sit in otherOutcomes 19. Any H2 2026 hair-growth result will be a marketing framing, not a registered endpoint hit.

3.3 Anti-anchoring on preclinical and TAM claims

Several Absci disclosures require calibration against the public record 39246:

  • "80 million Americans with AGA." The figure includes all degrees of hair loss; the addressable moderate-to-severe subset is materially smaller. The Grand View Research global AGA TAM is $2.7B in 2023 (8.45% CAGR to ~$3.6B by 2030) 40 -- the public benchmark, not Absci's $25B management figure 41.
  • "Approximately 9 million women in the U.S. with endometriosis" and "prevalent in up to 10% of women worldwide" 39. Absci truncated the 5-10% range in its own December 2025 disclosure to the upper bound in subsequent messaging.
  • "No FDA-approved disease-modifying therapy" for endometriosis. Orilissa (elagolix, 2018) and Myfembree (relugolix combination, 2022) carry FDA approval for endometriosis-associated pain 39. Absci's blanket framing contradicts the FDA label record; the defensible reading is "no non-hormonal disease-modifying therapy."
  • "Statistically superior hair regrowth compared to minoxidil in a preclinical mouse model." No effect size, no peer-reviewed publication, and a model with weak human translation 25.
  • "2-3x extended half-life as compared to antibodies in clinical development" (August 2024 ABS-101 NHP data) 6. Absci did not name the comparator antibodies. No peer-reviewed corroboration.
  • Pipeline-page progress bars. ABS-101 "Phase 1 at 85%" despite the November 2025 internal halt 20; ABS-201 AGA "Phase 1/2a at 85%" with three of four-to-six SAD cohorts dosed; ABS-201 endometriosis "Phase 1 at 85%" with no clinical trial registered. The website overstates asset momentum.

3.4 The peer-review gap

Absci has published no peer-reviewed journal article describing its de novo antibody-design platform method 35. The five PubMed Absci-affiliation hits cover unrelated topics -- three list Absci as a secondary affiliation, two predate the AI pivot. The 2023 Shanehsazzadeh preprint and the 2026 Origin-1 preprint both remain unreviewed 5. For a company raising capital and signing partnerships on platform superiority, the continued reliance on bioRxiv-only evidence constrains the science narrative.

4. Safety Profile

This report scores safety signals on a 0-100 scale where higher means higher concern: TL1A class = 22 6, PRLR class = 58 25. (Scoring is author construct; underlying class data cited inline below.)

4.1 TL1A class safety signal score: 22 (LOW)

The TL1A class has been de-risked clinically. Tulisokibart's Phase 2 ARTEMIS-UC reported a clean safety profile comparable to placebo 33. Duvakitug and afimkibart Phase 2 data showed no class-defining safety signals 3412. FAERS pharmacovigilance reporting on the class has not generated disproportionality signals against IBD-biologic reference drugs. ABS-101's interim "no serious adverse events" across roughly fifteen participants 6 is consistent with the class. The score reflects a mechanism understood in inflammation, three Phase 2/3 programs with clean readouts, and no regulatory holds 1232.

4.2 PRLR class safety signal score: 58 (MODERATE-HIGH)

The PRLR safety picture is less mature. The knockout phenotype in mice is documented: female homozygous mice are infertile and lack mammary development; hemizygous females cannot lactate after their first pregnancy; pituitary prolactin levels rise 30-100-fold with pituitary hyperplasia in some cases; bone formation falls in both sexes at all ages 25. Each phenotype predicts a potential on-target liability for chronic anti-PRLR administration in humans.

The HEADLINE Phase 1/2a protocol (NCT07317544) does not specify DEXA bone-density measurement or bone-turnover biomarker monitoring 42. In a chronic-use asset aimed at AGA -- a non-life-threatening indication competing with cash-pay generics -- the absence of bone-turnover monitoring in the Phase 1/2a readout is a protocol gap Phase 2 design must close. HMI-115's Lancet Phase 2 reported no sex-hormone disruption and no fertility signal 21; Bayer's BAY 1158061 Phase 1 posted favorable PK without disclosed safety concerns 21. Neither program has generated long-term (>12 months) human safety data. The score reflects on-target phenotype predictions humans have not yet ruled out, plus the absence of bone-turnover monitoring in the registered protocol 2542.

4.3 Implications for development

For ABS-101, the low class safety signal supports outlicensing: a partner faces no unusual regulatory headwinds 6. For ABS-201, the moderate-high class signal means Absci cannot rely on safety alone to differentiate 25. Efficacy must separate from HMI-115 21 and from minoxidil/finasteride to justify chronic use in a cosmetic-adjacent indication. Endometriosis changes the calculus: pain severity is real, the unmet need is real, and the safety bar is higher in theory but lower in practice.

5. IP Landscape

Absci's patent estate is three things at once: a legacy microbial-expression platform McClain founded 2829, an oncology-antibody stack from the June 2021 Totient acquisition 3031, and a generative-AI method portfolio with no granted US patents 43. The 10-K claims "79 issued or granted patents and 159 pending patent applications worldwide, which includes ten issued U.S. patents and 24 pending regular U.S. patent applications" 44.

5.1 Nine granted US patents verified (one short of the 10-K)

Independent Google Patents and USPTO verification confirms nine granted US patents assigned to Absci Corp or Absci LLC as of report date 44:

PatentSubjectGrant datePriority
US9617335B2Inducible coexpression system 292017-04-112012-08-05
US10465197B2Inducible expression system (foundational SoluPro) 452019-11-052012-08-05
US11214807B2Methods for inducible expression 462022-01-042012-08-05
US11447781B2Host cells for inducible coexpression 472022-08-312012-08-05
US11584785B2C-peptides and proinsulin polypeptides 482023-02-212020-05-04
US12134655B2Cancer-associated antibody (Totient origin) 312024-11-052020-04-29
US12157766B2Cytoplasmic expression (McClain founder-inventor) 282024-12-032012-08-05
US12285484B2Cancer-associated antibody (Totient origin) 302025-04-292019-09-19
US12473370B2TL1A-associated antibody (ABS-101 CoM) 492025-11-182023-10-03

The 10-K's "ten issued U.S. patents" 44 exceeds the nine verifiable via public registries by one.

The gap may reflect a recently granted tenth indexed after report cutoff, or a 10-K counting inconsistency tied to subsidiary entities (Totient UK Ltd., Absci Pty Ltd., Absci GmbH) 44.

5.2 ABS-201 has no granted US composition-of-matter patent

The most material IP gap sits on the only internally-advancing asset. ABS-201 carries no publicly discoverable granted US composition-of-matter patent as of report date 50, although HEADLINE Phase 1/2a began dosing on 3 December 2025 16. The international application is pending, but the public record does not yet confirm a US grant. An AI-designed antibody entering clinical development without a granted US CoM patent in its destination indications carries freedom-to-operate risk that sharpens against the Bayer prior art described below.

5.3 The ABS-101 freedom-to-operate concern (Merck/Prometheus)

US11999789B2 (Cedars-Sinai, Prometheus/Merck) -- "Humanized antibodies to TL1A" -- granted 4 June 2024, expires 23 October 2040 11.

The continuation US20240327532A1 (filed 2024-04-24, published 2024-10-03) may broaden into genus or functional claims; monitor for grant 51.

Absci's ABS-101 CoM patent US12473370B2 covers six specific CDR sequences 49; no independent overlap analysis with the Merck/Prometheus claims has been published.

The outlicensing pivot reduces Absci's direct FTO exposure but does not eliminate it as a constraint on the deal value Absci can extract 5.

Paragon Therapeutics added a fifth granted US anti-TL1A patent (US12466890B1, granted 2025-11-11) 52, one week before Absci's grant -- the class IP landscape is crowded.

5.4 The ABS-201 freedom-to-operate concern (Bayer / Hope Medicine)

WO2019011719A1 -- "Prolactin receptor antibody for male and female pattern hair loss," Bayer IP GmbH, priority 10 July 2017 -- is the primary FTO concern for ABS-201 50. Bayer licensed the asset to Hope Medicine, which developed HMI-115 and completed Phase 2 AGA in August 2024 (n=192) 21. Bayer's earlier WO2011069799A1 expands the prior-art surface 50. Absci's ABS-201 may design around these claims on specific sequence and epitope grounds, but a competitor with licensed Bayer IP is already running Phase 2 trials in ABS-201's two target indications 21. The competitive landscape and the FTO landscape are the same landscape.

5.5 Pending method claims -- the core AI platform IP is not yet granted

US20230268026A1, "Designing biomolecule sequence variants with pre-specified attributes," filed 7 January 2022, represents the core generative-AI method claim and remains pending 43. Absci holds no granted US method patent for its de novo antibody design algorithm. WO2025122885A1 (ABS-501 anti-HER2) 26, WO2024040020A1 (ACE Assay) 53, and WO2025144700A1 (nanobody screening) 54 represent the global filing strategy but do not become enforceable until national-phase grant.

The Generate Biomedicines US12020776B2 patent -- earlier framed as a "direct overlap" risk -- addresses a different problem at the claim-1 level. Generate's claim 1 covers a humanization workflow using MHC-II residue-wise immunogenicity reweighting 55. Absci's pending US20230268026A1 claim 1 covers a binding-characteristic prediction model trained on measured binding data 43. The two claim-1 texts do not read on each other element-by-element. The current assignee on Generate's grant is Flagship Pioneering Innovations VI Inc, not Generate Biomedicines per se 55. A complete Origin-1 FTO opinion against Generate's 71-application portfolio would require dependent-claim and specification-level analysis by a registered patent attorney; this DD does not constitute that analysis.

5.6 Portfolio expiration window

The 10-K reports an expected expiration range of August 2033 through December 2044 44. The earliest-expiring patents -- the 2012-priority SoluPro family -- protect platform manufacturing more than antibody composition 2945. The CoM patents protecting specific assets (US12473370B2 for ABS-101 to 2044-10-02; pending for ABS-201) extend to 2044 4950. The gap between platform-IP and asset-IP expiration narrows long-horizon defensibility against biosimilar entry.

6. Team

6.1 Executive team

Sean McClain -- Founder and CEO has run Absci since inception in August 2011 2756. He holds a BS in Molecular and Cellular Biology from the University of Arizona 56. No prior exits; no CEO history outside Absci. His 2024 total compensation reached $9,786,988, driven by a $5,491,250 Monte Carlo PSU grant 56. Across fourteen years he has taken one company from basement lab through IPO into clinical stage. The pattern of long tenure without governance conflicts is favorable; the absence of prior exits leaves his operating-CEO track record unbenchmarked outside Absci.

Zach Jonasson, PhD -- CFO and Chief Business Officer assumed the role on 31 August 2023 after serving as an independent director from October 2020 to December 2023 and on the Absci LLC predecessor board from April 2016 2756. He co-founded Phoenix Venture Partners and Convergent Ventures. His 2024 total compensation reached $2,832,851 56. Two governance facts require disclosure. Phoenix Venture Partners II LP (Jonasson's prior firm) is a Schedule 13G 10%+ beneficial owner of ABSI (9,354,695 shares filed 19 February 2025) and was disclosed as a delinquent Section 16 filer in the 2025 proxy along with Jonasson himself -- Phoenix's late filing covered 3,100,000 shares 56. Jonasson moved from independent director into an executive officer at a company where his former fund is a 10%+ holder.

Ransi Somaratne, MD, FACC, MBA -- Chief Medical Officer started 3 March 2026 5758. He joined from Vertex Pharmaceuticals, where he served as SVP Clinical Development (Jan 2023 - Mar 2026); earlier roles at BioMarin (Roctavian hemophilia A gene therapy), Amgen (Repatha PCSK9 Global Development Leader), and short CMO tenures at Epirium (1 yr 8 mos), Spring Discovery (1 yr), and Endpoint Health (7 mos) 59. MD Albany Medical College; MBA UNC Kenan-Flagler; Cardiology Fellowship Kaiser Permanente LA; BS Mechanical Engineering UC Santa Barbara. Somaratne is Absci's first full-time CMO, hired ten months after ABS-101 entered Phase 1 and three months after ABS-201 first dose 57. His inducement grant -- 650,000 options at $2.57 (the 3 March 2026 close), 4-year vest -- sits 14% below the July 2025 offering price of $3.00. LinkedIn shows a concurrent role at Iles & Somaratne Biopharmaceutical Consulting LLC (Sep 2021 - present) the appointment 8-K did not disclose 57.

Andreas Busch, PhD -- Chief Innovation Officer retires effective 31 March 2026 and transitions to SAB co-chair on 3 April 2026 with a 2-year scientific advisor agreement at $25,500/year cash plus 22,800 options + 5,800 RSUs and an exercise-window extension on 2,489,290 vested options 60. The 8-K states the resignation "was not the result of any disagreement with the operations, policies or practices of the Company" 60. Busch's $381,000 open-market purchases over the trailing twelve months -- including $229,000 on 2026-03-12 during the retirement transition -- signal personal conviction from a departing officer 6061. His 2024 total compensation reached $2,644,420, paid via Swiss subsidiary Absci GmbH at a 1.14 USD/CHF average rate 56.

Amir Shanehsazzadeh -- Chief AI Officer ascended to the CAIO title in April 2025, replacing Amaro Taylor-Weiner. Harvard AB Mathematics and SM Statistics. Shanehsazzadeh is an inventor on the ABS-101 CoM patent US12473370B2 49 and first author of the 2023 flagship de novo antibody bioRxiv preprint 35. Absci did not disclose the CAIO promotion via dedicated 8-K, the title is administrative not Section 16, and no Form 4 filings exist under his name -- disclosure obscurity at an AI-branded company. His tenure began as an AI Scientist in June 2022 -- a fast internal track for the central thesis.

Shelby Walker, JD -- Chief Legal Officer joined June 2024 from Korro Bio GC/Corporate Secretary (May 2023 - June 2024); prior SVP Intellectual Property at CRISPR Therapeutics (March 2018 - April 2023) and GC at Ginkgo Bioworks (May 2016 - March 2018) 56. JD/LLM UNH Franklin Pierce School of Law; dual MS Bioscience Regulatory Affairs and Biotechnology Johns Hopkins; BS Biotechnology WPI. Deep IP track record for a company whose valuation depends on IP.

Todd Bedrick, CPA -- Chief Accounting Officer has served since June 2022 27. He acquired 124,200 shares and disposed of 36,173 net over the trailing twelve months 61 -- small but directionally consistent with insider bullishness. Bedrick signs the 10-K as Principal Accounting Officer (a Section 16 capacity) but the DEF14A 2025 executive officer biography roster excludes him -- inconsistent disclosure across filings.

6.2 Board of Directors (seven members)

The board fixed at seven seats on 7 July 2025 when Mary Szela joined Class III 62. Current composition: McClain (CEO/Class III), Pangalos (Class II, since Jan 2024) 56, Rabinovitsj (Class II, since Nov 2022) 56, McGinnis (Class I, audit chair, since Aug 2020) 56, van Houten (Class I, board chair, since June 2023) 56, Sirosh (Class III, since Jan 2022), Szela (Class III, since July 2025) 6263. Class split is 2/2/3, not 3/3/1 as some earlier summaries reported.

Sir Mene Pangalos, PhD served as EVP BioPharmaceuticals R&D at AstraZeneca from January 2019 to March 2024 56. He joined the Absci board in January 2024, one month after the December 2023 AstraZeneca $247M collaboration announcement 64; the public record discloses no firewall arrangement between his AZ R&D role and his Absci director candidacy. His trailing twelve-month open-market purchases of $460,548 (170,880 shares) are the largest among insiders 61. Biogen board service runs in parallel.

Mary Szela joined 7 July 2025 with a 178,400-option inducement grant 6263. CEO of TriSalus Life Sciences since January 2018; previously CEO of Novelion Therapeutics and Melinta Therapeutics; earlier ran Abbott Laboratories' $8B US pharma business 63. Also on the boards of Kura Oncology, Prometheus Biosciences (the company Merck acquired for $10.8B), Omega Therapeutics, and Senda Biosciences. Szela's LinkedIn shows "Board Member, Absci, Jul 2025 - Feb 2026" with an end date, but she signed the 10-K on 2026-03-24 as Director and TriSalus DEF14A confirms current Absci service; the LinkedIn end date is a self-update error, not a departure 2762.

Frans van Houten, MSc -- former CEO of Royal Philips (March 2011 - October 2022) 56. Novartis AG independent director since February 2017. The 10-K formalizes his title as "Chair of the Board" 2756. Audit committee financial expert. His trailing twelve-month open-market purchases of $148,800 add to the insider net 61.

Karen McGinnis, CPA -- ex-VP and Chief Accounting Officer at Illumina (November 2017 - April 2021) 56. Audit committee chair; designated audit committee financial expert 27. iRhythm Technologies board joined July 2025.

Daniel Rabinovitsj -- VP at Meta since August 2018 56; prior COO Ruckus Wireless, SVP Qualcomm Atheros 27. Tech compensation norms differ from pharma peer-group benchmarks -- relevant for a director who chairs the compensation committee.

Joseph Sirosh, PhD -- CEO of CreatorsAGI Inc; former VP Amazon Alexa Shopping (October 2022 - November 2023); prior CTO Compass, CVP and CTO of AI at Microsoft (July 2013 - December 2018) 27. AI-sector board bridge.

Amrit Nagpal (Managing Director, Redmile Group LLC) resigned from the board on 25 April 2025 with no disclosed disagreement 27. Redmile is a healthcare-focused hedge fund; the departure of a Redmile MD reduces dedicated-biotech-fund engagement.

6.3 Insider open-market activity -- net +$1.017M with zero discretionary sales

Six trailing-twelve-month open-market purchases by four insiders total $1,017,348.40 across 370,880 shares 61: Pangalos $460,548, Busch $381,000, van Houten $148,800, Bedrick $27,000 61. Two prior-flagged "sales" -- McClain 26,761 shares and Jonasson 17,496 shares, both 2026-02-02, code S -- are RSU tax withholdings per the explicit Form 4 Note 1: "Amount reported represents the number of shares withheld by the Issuer to cover the tax withholding obligation in connection with the vesting of these restricted stock units and does not represent a discretionary trade by the reporting person" 65. Net 12-month insider activity: $1.017M of buying, zero discretionary selling. The signal is bullish, led by directors and the retiring CIO, none of whom needed to commit personal capital and all of whom did.

6.4 Scientific Advisory Board and endometriosis advisory board

The SAB seats Andreas Busch (co-chair post-retirement) 60, Mene Pangalos, Iain McInnes (Vice Principal, University of Glasgow), Hubert Truebel (CMO AiCuris), and Victor Greiff (Associate Professor, University of Oslo). The March 2026 endometriosis advisory board added Hugh Taylor (Yale), Linda Giudice (UCSF), Gaurang Daftary (Inception; Yale/Mayo), Steven Young (Duke), Zaraq Khan (Mayo), and Joan-Carles Arce (Repronovo) -- credible academic coverage for an indication Absci has not yet registered a trial in.

6.5 Team summary

The bench has the credentials it needs: a commercially seasoned CMO (Somaratne), a finance/BD operator in the CFO role (Jonasson, with the Phoenix relationship flag), an IP heavyweight as CLO (Walker), and an in-house AI lead (Shanehsazzadeh). The board mixes pharma commercial experience (Szela), Big Pharma R&D experience (Pangalos), tech/AI experience (Sirosh, Rabinovitsj), and two audit-committee financial experts (McGinnis, van Houten). Two simultaneous C-suite transitions -- a new CMO and the CIO's retirement -- during a clinical-ramp period 5760 invite execution-risk scrutiny. The net-bullish insider purchase pattern, led by the directors and the outgoing CIO, signals personal conviction that outlasts the reorganization.

7. Financials

Absci ended FY2025 with $144.3M in cash, equivalents, and marketable securities against a $115.2M net loss 6667. Management tells investors the balance funds operations "into the first half of 2028" 66. Pure-cash math reaches mid-2027 only; the H1 2028 guidance assumes another $50-90M from the ATM on top of the August 2025 $400M base shelf 2768.

Revenue tells the first-order story.

Partner-program revenue fell to $2.8M in FY2025 from $4.5M in FY2024, a 38% decline driven by milestone timing and program mix 6767.

Q3 2025 revenue collapsed 78% YoY to $0.378M; nine-month 2025 revenue fell 44% to $2.15M 69.

Three partners produced 95% of FY2025 partner revenue, all of them foreign 67.

The four-year arc -- $5.7M (2022), $5.7M (2023), $4.5M (2024), $2.8M (2025) -- shows two consecutive years of contraction while headline partnership announcements accumulated 6767.

R&D spend moved the opposite direction.

FY2025 R&D reached $81.4M, up 27% from $63.9M, driven by ABS-101 Phase 1 costs and ABS-201 IND-enabling work 67.

SG&A held flat at $35.1M vs $36.2M 67.

Total operating expenses hit $123.1M, producing a $120.3M operating loss and a $115.2M net loss 6767.

Accumulated deficit reached $624.8M 67.

Q4 2025 cash burn appears modest at $8.2M because Absci recognized a $5.1M gain on settlement of the Totient contingent consideration 6767.

The Totient escrow resolution returned $8.7M unrestricted cash to Absci and $7.6M to former Totient sellers from the original June 2021 merger 67.

Strip the one-time gain and operating burn ran closer to $13.3M for the quarter -- consistent with the rising trajectory: Q1'25 $21.8M, Q2'25 $16.9M, Q3'25 $25.0M, Q4'25 $29.2M 70717227.

The Totient earn-out failed to trigger its full payout, indicating post-acquisition milestones were missed 67.

Share count tells the dilution story.

Weighted-average diluted shares grew 24% YoY, from 110.2M (FY2024) to 136.8M (FY2025) 67.

Q4 2025 weighted-average reached 150.6M 67; year-end 2025 issued shares totaled 151,515,079 67; the DEF14A 2026 confirms 155,350,135 outstanding as of 2026-03-31, a 2.5% expansion in Q1 alone 73.

Three capital raises inside 18 months drove the trajectory: $86.4M follow-on closed March 1, 2024 (days after the AstraZeneca deal) 74; $20M AMD PIPE at $3.50/share in January 2025 67; $64M July 2025 raise ($50M underwritten at $3.00 plus $14M ATM) 67.

The Q1 2025 PR guided runway into 1H 2027 75; the Q2 2025 PR upgraded guidance to 1H 2028 explicitly citing the July 2025 raise 74.

The August 2025 S-3 base shelf caps at $400M (not unlimited as some smaller-reporting-company forms can be), paired with a TD Cowen at-the-market sales agreement that drew $14M of $100M capacity in July-August 2025 68. Absci raises opportunistically on news-driven equity windows, not on predictable milestones.

Emerging-growth-company exemptions expire December 31, 2026. FY2027 will require full SOX 404(b) auditor attestation on internal controls -- added cost and disclosure burden hitting the same year ABS-201 interim data drives capital decisions. The EGC revenue threshold ($1.235B) 67 is not binding given $2.8M actual revenue 67, but the calendar trigger fires regardless.

Short interest grew +97% YoY to 37,258,694 shares on 2026-01-30, days-to-cover 11.72 against average daily volume 3,177,770 76. The 2025-01-31 baseline was 18,945,391 shares; +97% not the +69% earlier reports cited. At ~12% of free float, the short position prices in continued dilution and binary risk into the H2 2026 ABS-201 readout.

7.1 Cap table and dilution path

RoundDateAmountSource
Cumulative pre-IPO private2015-2021~$216M (sum, no single primary source -- author estimate from Form D series 7778798081)SEC Form D series
IPO2021-07~$225M gross (analyst-derived sum across S-1 + greenshoe; not in canonical as a single claim)SEC S-1 record
Follow-on #12024-03-01$86.4M 742024 PR
AMD PIPE2025-01$20.0M @ $3.50 672025 PR
Follow-on #2 + ATM2025-07$64.0M ($50M + $14M) 746774
ATM remaining (TD Cowen)ongoingup to $86M of $100M 6768
S-3 base shelf cap2025-08$400M 68S-3

Cumulative capital raised: ~$216M pre-IPO 7778798081 + ~$225M IPO + $170M post-IPO 746767 = ~$611M.

Accumulated deficit $624.8M 67 roughly equals total capital raised.

Implied: every dollar raised has been spent, with no retained earnings to cushion the next cycle 67.

Implied pre-money today: ~$400-450M market cap 82 minus ~$144M cash 66 = enterprise value ~$256-306M 83.

Cumulative dilution from IPO float (~90M shares) to current 155.4M shares 73 exceeds 70%.

Burn rate $115M/year against $144M cash 6766 matches 15-18 months hard runway absent further financing; the 1H 2028 guidance assumes another $50-90M of ATM draws between now and mid-2027 27.

8. Regulatory Pathway

Absci runs both clinical programs outside the US IND framework. ABS-101 Phase 1 is registered at ACTRN12625000212459 on the Australian New Zealand Clinical Trials Registry, sponsored by Absci Pty Ltd and sited at Nucleus Network in Melbourne 12. ABS-201 HEADLINE Phase 1/2a NCT07317544 shows isFdaRegulatedDrug: false; the sponsor is Absci Pty Ltd, both sites sit in Melbourne, and the study runs under an Australian CTN 16.

The 10-K acknowledges the consequence directly: the FDA "will generally not approve the application on the basis of foreign data alone" 67. An ABS-101 IND cleared in the US but the Phase 1 ran only in Australia 2. Absci captured Australia's cash-refundable R&D tax incentive and faster start-up; a BLA strategy will require bridging Phase 1 or repeating key studies on US sites and US patients. Timelines for any US pivotal program compound accordingly.

IBD endpoint framework. Tulisokibart's ARTEMIS-UC Phase 2 used the standard FDA UC primary: clinical remission at Week 12, composite of Mayo subscores. Tulisokibart hit 26% vs 1% placebo (p<0.001) 3312. Duvakitug RELIEVE reported 48% UC remission at 900mg vs 20% placebo at Week 14 34 and 48% Crohn's endoscopic response vs 13% placebo 84. Roche's afimkibart TUSCANY-2 reported 36% remission and 50% endoscopic improvement at Week 56 12. Any ABS-101 pivotal will need to match or beat this bar.

Endometriosis endpoint framework. Pain scores (NRS dysmenorrhea, non-menstrual pelvic pain), symptom diary completeness, and laparoscopic confirmation at entry define the class. HMI-115's Lancet OGWH 2025-11-05 paper (n=108 completed, 240 mg arm) reported -41.57% LSM dysmenorrhea NRS reduction vs -18.61% placebo at week 13 (placebo-adjusted -22.96%, p=0.036) and roughly 52% LSM non-menstrual pelvic pain reduction without sex-hormone disruption 21. Only the 240 mg dose reached statistical significance. ABS-201's endometriosis Phase 2 has not started; Absci plans initiation Q4 2026.

AGA endpoint framework. Target Area Hair Count (TAHC), Target Area Hair Width (TAHW), and Global Photographic Assessment (GPA) anchor the FDA AGA package for finasteride and minoxidil. NCT07317544 lists TAHC, TAHW, and darkening/pigmentation as otherOutcomes -- neither secondary nor primary 19. Absci describes the HEADLINE MAD portion as "powered to demonstrate human proof-of-concept," but the registered statistical plan does not back the framing. HMI-115's Phase 1b in male AGA reported a +14 hairs/cm-squared mean non-vellus TAHC gain over 24 weeks (n=12, open-label) 21 -- modest by approved-comparator standards.

PRLR on-target safety. PRLR-knockout mice are infertile (homozygous females), fail to lactate after first pregnancy (hemizygotes), develop pituitary hyperplasia with 30- to 100-fold prolactin elevation, and show reduced bone formation in both sexes 25. NCT07317544 does not monitor bone turnover or schedule DEXA scans 42. HMI-115 in its Phase 2 reported no sex-hormone disruption, but the published endpoint set did not include BMD across treatment duration 21. ABS-201 inherits this liability and will need a safety package addressing it before any chronic-dosing indication proceeds.

9. Market

IBD is the largest market Absci addresses. The global IBD therapeutic market reached $27.43B in 2025 per Mordor Intelligence (5.55% CAGR) 85. Biologics account for the majority of the global total. AbbVie's Skyrizi + Rinvoq combined sales and J&J's Tremfya growth (no primary source in canonical -- analyst-press estimates, FY2025 earnings releases) frame an IBD biologic-revenue base in the tens of billions. Tulisokibart analyst consensus projects $4-5B peak sales (analyst estimate, no primary canonical source).

ABS-101's role in this market is now an outlicensing asset, not a revenue product. The November 2025 deprioritization shifted the valuation question from "what will Absci earn on IBD?" to "what will a partner pay for Phase 1 TL1A whose half-life characteristic the analyst class read as below next-generation comparators?" 59. The deal-value bracket is bounded above by Roche's $7.1B upfront for Telavant (Phase 2b, Phase 3-ready) and below by AbbVie's $150M upfront plus $1.56B milestones for FutureGen FG-M701 (~$1.7B total). Realistic ABS-101 partnering range: $100-300M upfront, $800M-1.7B total, if 2026-2027 outlicensing closes with supportive Phase 1b data. Boehringer's January 2026 Simcere SIM0709 license (EUR 1.058B for a TL1A x IL-23p19 bispecific) shows the class still attracts re-entry capital 15; that money pivoted from monospecific HLE (Boehringer never sponsored BB-TL1A; Vial Australia did) into bispecifics. The FTO ceiling is Cedars-Sinai/Prometheus/Merck's US11999789B2 11.

Endometriosis. The global endometriosis market reached $2.39B in 2025 per Fortune Business Insights (8.70% CAGR to ~$4.5B by 2033) 86. Approved hormonal options -- elagolix (Orilissa, AbbVie), relugolix combination (Myfembree/Ryeqo, Pfizer/Myovant) -- suppress estrogen with bone-loss and fertility liabilities. HMI-115 has shown the non-hormonal mechanism can work at one dose 21. ABS-201 peak sales range, unrisked: $300-900M in endometriosis, assuming 15-35% class share against HMI-115 and legacy hormonal options.

Androgenetic alopecia. The global AGA market reached $2.7B in 2023 per Grand View Research (8.45% CAGR to ~$3.6B by 2030) 40. Absci's $25B US-only TAM is a population x price construct (5-9M annual patients x ~$3,500/yr implied price) 41, not a realized-revenue forecast; the CFO declined to disclose the price assumption that underlies the figure on the Q4 2025 call: "we cannot disclose what we think the actual price point will be" 87. The implied price runs 3-67x current cash-pay options -- Cost Plus generic finasteride $74/year, Hims $240-720/year, Propecia brand $840/year 87. Biologic pricing in cosmetic-adjacent indications faces a hard cash-pay ceiling. Ritlecitinib's $49,000/year list covers alopecia areata, an autoimmune indication with insurance coverage; that price does not transfer to AGA. ABS-201 AGA unrisked peak: $50-200M.

HER2+. The global HER2+ market reached $10.95B in 2025 per Mordor (3.92% CAGR), anchored against Enhertu actuals 88. ABS-501 enters preclinical against trastuzumab and the broader trastuzumab biosimilar field; affinity parity does not translate into commercial differentiation.

Outlicensing comparable band. AbbVie/FutureGen (TL1A, preclinical, June 2024) sets the floor at $1.7B total. Sanofi/Teva duvakitug (Phase 2b, October 2023): EUR 469M upfront + EUR 940M milestones. Boehringer/Simcere SIM0709 (preclinical bispecific, January 2026): EUR 1.058B 15. Merck/Prometheus tulisokibart (Phase 2 UC/CD, April 2023): $10.8B all-cash 11. Roche/Telavant afimkibart (Phase 2b UC, October 2023): $7.1B upfront + $150M milestone 12. The ceiling remains unavailable to a Phase 1 TL1A whose half-life reads as below next-generation.

10. Competitive Landscape

TL1A class scorecard.

Three Phase 3 programs now define IBD TL1A: Merck tulisokibart (ATLAS-UC NCT06052059, ARES-CD NCT06430801; primary completion estimated 2026-08-01) 89; Sanofi/Teva duvakitug (RELIEVE Phase 3 four-trial program NCT07184996/NCT07184931/NCT07185009/NCT07184944, primary completions 2028-2029) 90919293; Roche/Genentech afimkibart (AMETRINE-1/2, started September 2024) 12.

One Phase 1 terminated for PK endpoint failure: BB-TL1A-VIAL-HLE (NCT07029971, sponsor Vial Australia, terminated 2026-01-20 at n=16) 10.

One adjacent PD-1 depleter (AnaptysBio rosnilimab) failed UC Phase 2 on 2025-11-10 94.

Boehringer Ingelheim re-entered the class on 2026-01-27 by licensing Simcere SIM0709 (EUR 1.058B for a TL1A x IL-23p19 bispecific) 15.

ABS-101 enters as the fifth or sixth program, Phase 1, with an HLE differentiation analysts read as below next-generation 9.

The class is validated by $18B+ in M&A 14 and against placebo at 26-48% remission rates 333412.

PRLR class -- ABS-201 is second mover.

Hope Medicine's HMI-115 (originally BAY 1158061 from Bayer, licensed 2019) completed Phase 2 endometriosis with Lancet OGWH publication 2025-11-05 (n=108 completed; 240 mg arm delivered -41.57% LSM dysmenorrhea NRS reduction vs -18.61% placebo, p=0.036) 21.

The Phase 2 male AGA trial NCT06118866 (n=192, single-site Peking University People's Hospital) primary-completed 2024-08-26; topline remains unpublished as of report date 21.

Phase 3 endometriosis NCT07318688 (n=540) registered 2025-12-29 -- one phase ahead of ABS-201's planned Phase 2 21.

HMI-115 received Breakthrough Therapy Designation from China's NMPA for endometriosis 21.

The 2018 Bayer Phase 1 of BAY 1158061 in breast cancer completed (PMID 29806538) with favorable PK; the eight-year gap before the molecule returned in women's health remains unexplained in the public record 21.

Novartis's anti-PRLR LFA102 failed breast cancer Phase 2 on efficacy 21.

Absci enters second; the company's "category-defining" framing fails the public record 21.

AI-platform peers (public comp table).

CompanyTickerMarket cap (Apr 2026)CashRunwayLead asset
AbsciABSI~$400-450M 8295$144.3M 6683into 1H2028 with ATM 27Origin-1; ABS-201 Ph1/2a
AbCelleraABCL~$1.17B 9697~$700M total liquidity 9899>3 years 99ABCL635 Phase 2 100
RecursionRXRX~$1.73-1.87B 101$753.9M 102into early 2028 102REC-617, REC-1245 Ph1/2 102
SchrodingerSDGR~$915M 103$402.3M 104105positive adj EBITDA YE2028 104105SGR-1505 Ph1 105
RelayRLAY~$2.7B 106$555M 107108into 2029 108Zovegalisib Ph3 BTD 109
Generate Biomedicinesprivate~$700M raised 110111N/AN/AGB-0895 Ph1 to Ph3 112
Xairaprivate$1B committed 3738N/AN/APreclinical 37

Absci trades at one-third to one-half of AbCellera's market cap 8296 despite a later clinical stage and lower revenue.

The discount reflects trajectory (ABS-101 outlicensed, ABS-201 generating no revenue), $115M annual net loss 67, and the absence of any Phase 2 human efficacy data.

The three biggest competitive threats: HMI-115 establishes PRLR proof-of-concept Phase 3 endometriosis ahead of ABS-201's Phase 2 21; TL1A class crowding renders ABS-101 unlicensable at favorable economics 14; AI antibody platform competition compresses partnership economics (Xaira's $1B war chest exceeds Absci's market cap 3738).

11. Risks & Open Questions

Four corrections to the public record.

An earlier SEC inventory referenced "Eric Furfine, PhD as CMO." No 8-K confirms any Furfine appointment; the sole CMO on record is Ransi Somaratne, MD FACC MBA, effective 3 March 2026 11357. Furfine is Co-CEO/CSO of Mosaic Biosciences; he was never affiliated with Absci 113. The Furfine entry is a hallucination this report retires.

An earlier 8-K digest referenced "Ruchir Nagpal resignation." The actual name per 8-K 2025-04-28 is Amrit Nagpal, Managing Director at Redmile Group LLC 27. A Redmile MD departure still signals reduced healthcare-hedge-fund engagement.

An earlier board-composition digest reported a "3/3/1 class split." Per 8-K 2025-07-07 the actual split is 2/2/3 -- Class I McGinnis and van Houten; Class II Pangalos and Rabinovitsj; Class III McClain, Sirosh, and Szela 5662. Total of seven board seats is correct.

A March 2026 release stated "no FDA-approved" treatment for endometriosis. Elagolix (Orilissa, AbbVie, FDA-approved 2018) and relugolix combination (Myfembree/Ryeqo, Pfizer/Myovant, FDA-approved 2022 for endometriosis pain) are approved 39. The defensible reading is "no non-hormonal disease-modifying therapy"; the blanket framing contradicts the FDA label record.

Where the company framing outruns its evidence.

The pipeline page shows "Phase 1 at 85%" for ABS-101 despite the November 2025 internal halt 205.

The same page shows "Phase 1/2a at 85%" and "Phase 1 at 85%" for the two ABS-201 indications 20. Progress bars are not trial milestones.

The ABS-101 Q3 2025 interim language -- "extended half-life vs first-generation TL1A, no SAEs" -- inflates marketing-friendly properties without disclosing the PK curve a reader would need to verify the comparison 6.

Morgan Stanley's January 2026 downgrade reads the same data as "trailed next-generation competitors" 9.

The "80 million Americans" AGA TAM includes all degrees of hair loss; the moderate-to-severe treatable subset is roughly one-third 41.

The "10% of women worldwide" endometriosis prevalence is the upper bound of a 5-10% range Absci acknowledged in its December 2025 release 39.

ABS-201's AGA MAD portion "powered to demonstrate human proof-of-concept" reads as a primary-endpoint commitment; the registered statistical plan places hair-growth measures in otherOutcomes 19.

Rodney Sinclair's testimonial calling ABS-201 a meaningful advance is a paid-advisor endorsement, not independent validation 24.

Probability-weighted material risks.

  • TL1A outlicensing failure (probability 30-40%): If no partner commits by H1 2027, ABS-101 becomes a stranded asset 59. Effect: compresses ABS-201 to sole-asset status; reduces platform-partnership narrative; forces a capital raise at lower share price. Impact: $150-250M market-cap range.
  • ABS-201 HEADLINE interim signal weakness (probability 40-50%, given HMI-115 Phase 1b TAHC delta of +14 hairs/cm-squared as the floor and HMI-115 Phase 2 AGA still unpublished) 21: Interim data in H2 2026 that fails to clear HMI-115 or minoxidil-like comparators triggers equity-raise pressure before endometriosis Phase 2 starts. Effect: forced financing at trough prices. Impact: dilutive raise at $1.50-2.50/share range.
  • PRLR on-target bone/fertility signal (probability 15-25% over chronic dosing): Bone-turnover markers or pituitary signals in MAD or chronic-dosing studies force a development hold or label restrictions 25. Effect: endometriosis pathway compromised; AGA positioning shifts to intermittent.
  • Cash exhaustion before ABS-201 Phase 2 POC (probability 25-35%, increased from prior 20-30% given pure-cash trajectory mid-2027): The H1 2028 runway requires another $50-90M from the ATM on top of the August 2025 $400M base shelf 2768. The 37.26M-share short position (+97% YoY, days-to-cover 11.72) is already pricing this in 76.
  • EGC/SOX 404(b) compliance costs and disclosure surprises (probability 30-40%, low individual impact): December 31, 2026 deadline hits during ABS-201 readout season. Material controls exceptions would compound investor perception risk, especially given the two Section 16 delinquent filers -- Phoenix Venture Partners II LP (3,100,000 shares) and Jonasson 56.
  • CMO retention risk (probability 20-30% over 24 months): Somaratne's pre-Vertex CMO history shows three sub-two-year tenures (Epirium 20 mos, Spring Discovery 12 mos, Endpoint Health 7 mos) 59. He joined six weeks before report date, ten months after ABS-101 entered Phase 1 57. The 4-year vesting on 650,000 options at $2.57 argues for retention; the prior tenure pattern argues against. The undisclosed concurrent Iles & Somaratne Biopharmaceutical Consulting LLC role adds a conflict-of-interest disclosure gap 57.

11.1 Tier-3 claims whose resolution would flip thesis

  • Claim: ABS-101 "extended half-life" interim Phase 1 framing 6. Why not verified: company press-release framing only, no PK curves, MS analyst counter-read 9. Resolution effect: numeric t1/2 >30 days supports outlicensing toward $1.7B floor; <20 days collapses the differentiation story.
  • Claim: "$247M AstraZeneca" and "$650M Almirall" headline totals 64114. Why not verified: upfront and near-term cash not separately disclosed; AZ deal has zero ct.gov registrations 27 months in 64. Resolution effect: upfront-plus-near-term cash >$50M per deal supports the platform-value case; pure-milestone terms at current realized pace collapse the platform revenue thesis.
  • Claim: ABS-201 Q8-Q12 week dosing extrapolated from NHP PK 23. Resolution effect: human PK supporting Q8W+ materially differentiates from HMI-115's Q2W SC; Q2-Q4W eliminates differentiation.
  • Claim: Origin-1 "first demonstration of de novo design against zero-prior epitopes with atomically accurate complex structures" 35. Why not verified: bioRxiv preprint with 54 internal authors; no peer review. Resolution effect: peer-review acceptance at Nature/Science/Cell/Nat Biotech converts the platform claim from marketing to scientific consensus.
  • Claim: "At least one or more partnerships, including with a Large Pharma company, in 2026" 74. Why not verified: guidance language; carries forward from 2025 guidance that slipped. Resolution effect: a Large Pharma platform license (not program-specific) in 2026 validates Origin-1 as a licensable asset; absence by YE2026 confirms platform-revenue plateau.

Distribution: BigBio AI Research -- Public. All sources are public filings, peer-reviewed publications, and company disclosures. No pitch-deck or VDR content. Every digit traces to canonical.yaml.

12. Optionality

Platform beyond ABS-101/ABS-201. Origin-1, the January 2026 rebrand of "Integrated Drug Creation," reported 4/10 zero-prior-epitope success on COL6A3, AZGP1, CHI3L2, and IL36RA, with cryo-EM structural confirmation at 3.0-3.1 Angstrom resolution and IL36RA optimized to 104 nM 35. The platform's scientific core remains a bioRxiv preprint with 54 internal authors and no peer review 35. The 2023 flagship anti-HER2 de novo paper on bioRxiv (DOI 10.1101/2023.01.08.523187) remains unreviewed three years after posting 5.

Partnership deal flow and disclosed value.

PartnerDateDisclosed valueStatus
AstraZenecaDec 2023 64up to $247M (Tier-3; upfront undisclosed) + royalties 64One AI-design milestone Sept 2024; zero ct.gov registrations as of report date 64
Almirall (expanded)Aug 2025 (original Nov 2023)up to $650M (Tier-3) 114First target lead delivered
AMDJan 2025$20M PIPE at $3.50/share 67Compute partnership with Instinct/ROCm 67
Memorial Sloan KetteringAug 2024undisclosed; up to 6 programs 44Co-development 44
Twist BioscienceOct 2024undisclosed; ASC 808 cost-sharing 44Seeking 3rd-party clinical partner 44
Invetx/DechraJan 2025R&D funding + election fees + milestones 44Animal-health HLE platform 44
OwkinJan 2025undisclosed; ASC 808 44IO / immunology targets 44
PrecisionLifeDec 2023undisclosed; up to 5 programs 44No update since announcement 44

The $247M + $650M + $20M headline understates and overstates the reality at once 6411467. Understates, because multi-year royalties on successful products could exceed the upfront-plus-milestone cap. Overstates, because milestone achievement is low-probability and cash realization thin -- partner-program revenue remains $2.8M annually against $115.2M net loss 6767. The narrative "four new drug-creation partners in 2024" delivered on guidance 44. The 2025 guidance -- "one or more partnerships, including with a Large Pharma company, in 2025" -- slipped; the March 2026 Q4 FY2025 release moves the same target into 2026 74. Partnering momentum measured in cash, not announcements, has slowed.

Origin-1 licensing optionality. A pharma-scale Origin-1 license -- not a target-specific collaboration -- has not yet been signed 64. AstraZeneca's December 2023 deal disclosed no upfront; 27 months later, no clinical asset has emerged from that collaboration, and the FY2025 10-K removed the Active Programs table that earlier disclosed it 6444. The platform value scenario rests highest if ABS-101 licenses successfully in 2026-2027 5 and ABS-201 H2 2026 interim delivers a clean signal 18; lowest if both stall.

12.1 Comparable transactions (last 30 months, same mechanism class or adjacent)

TargetAcquirerDateStageMechanismHeadline $
Prometheus BiosciencesMerck2023-04-17Ph2 UC/CDAnti-TL1A (tulisokibart)$10.8B all-cash 11
Telavant (Roivant + Pfizer JV)Roche2023-10-23Ph2b UCAnti-TL1A (afimkibart/RVT-3101)$7.1B upfront + $150M 12
Teva anti-TL1A (duvakitug)Sanofi (license)2023-10-03Ph2b UC/CDAnti-TL1AEUR 469M + EUR 940M (~$1.5B) 13
FutureGen FG-M701AbbVie (license)2024-06-13preclinicalAnti-TL1A$150M + $1.56B (~$1.7B) 14
Simcere SIM0709Boehringer Ingelheim2026-01-27preclinicalTL1A x IL-23p19 bispecificEUR 1.058B 15
Vigil Neuroscience (immunology)Sanofi2024preclinical/Ph1Immunology mixedundisclosed (no primary source -- author estimate, not verified in canonical)
Recursion + Exscientiamerger2024-11platformAI drug discoveryall-stock; combined cash ~$850M at close 115

The TL1A deal cascade establishes a $1.5-10.8B band by stage 141115. ABS-101 at Phase 1 with a contested half-life narrative falls closest to the FutureGen floor 9.

12.2 Forty-eight hour action plan

Decision now: MONITOR. The Bull case requires two sequential wins (ABS-101 outlicense closes 2026 5 and ABS-201 AGA interim reads positive H2 2026 18); the Bear case requires only one miss. Wait for inflection data.

Triggers to reopen:

  1. ABS-201 HEADLINE interim hair-growth readout, H2 2026 (TAHC delta vs baseline, with or without placebo separation) 18.
  2. HMI-115 Phase 3 endometriosis primary completion (NCT07318688 estimated 2028-03-01) or HMI-115 Phase 2 AGA topline (NCT06118866, completed 2024-08-26, still unpublished -- 16 months past primary completion) 21.
  3. ABS-101 outlicensing announcement with disclosed upfront terms (confirms base-case floor) 5.
  4. Large-pharma platform deal for Origin-1 beyond target-specific collaboration 64.
  5. Capital raise priced below $2.00/share (signals distress or capitulation) 68.

Technical DD gaps to close before any position:

  1. ABS-101 Phase 1 numeric PK in days, with the next-generation comparator Morgan Stanley referenced -- verify the "extended" framing or confirm the MS reading 9.
  2. ABS-201 NHP PK data behind the 90% subcutaneous bioavailability and Q8-Q12 week dosing claims 23.
  3. Bone-turnover monitoring plan for chronic-dosing indications -- the PRLR-KO bone phenotype must be addressed before endometriosis Phase 2 25.
  4. Internal ABS-201 vs HMI-115 head-to-head preclinical comparison, if any exists 21.
  5. Peer-review status on the Origin-1 bioRxiv preprint and the 2023 flagship HER2 preprint 35.

Reference calls (3-5):

  • Clinical dermatologist running comparator AGA trials -- is NCT07317544's MAD protocol adequately powered for TAHC to differentiate from finasteride, given hair-growth endpoints sit in otherOutcomes? 19
  • IBD gastroenterologist with TL1A investigator experience -- how do pharma licensors weigh Phase 1 half-life differentiation against three Phase 3 competitors, post the MS analyst readout? 9
  • Ex-Hope Medicine executive or ex-Bayer BAY 1158061 program -- what stopped the Bayer program between 2018 Phase 1 and 2019 HMI-115 licensing? Is there a silent liability behind the eight-year gap? 21
  • Public biotech analyst covering AbCellera / Recursion -- what multiples apply to platform-plus-clinical-asset biotechs with declining partner revenue? 96102
  • Endometriosis KOL (Hugh Taylor at Yale or Linda Giudice at UCSF, both on Absci's endometriosis advisory board) -- what level of pain-score separation would make ABS-201 a credible second mover after HMI-115's 240 mg signal? 21

Observation window: 120-180 days. The ABS-201 HEADLINE interim readout in H2 2026 is the dominant readout 18; the read does not change before then unless triggers 2 (HMI-115 readout) 21 or 5 (capital raise below $2.00) fire first 68.

13. Verdict

Four stories, none of them resolved. Absci has crossed into clinical stage with one internally advancing asset (ABS-201, anti-PRLR for AGA and endometriosis) after handing its former lead (ABS-101, anti-TL1A) to the outlicensing desk five months after dosing. The platform story remains unreviewed. The clinical story is twelve months old. The competitive story places ABS-201 second in class behind a Bayer-derived program already running Phase 3 endometriosis. The capital story compounds: $144.3M cash funds operations into H1 2028 only if Absci keeps drawing the ATM; pure cash exhausts by mid-2027 27.

  • Insiders bought $1,017,348.40 across six trades over the trailing twelve months and made zero discretionary sales 6165. Director Sir Mene Pangalos bought $460,548 and retiring CIO Andreas Busch bought $381,000; the McClain and Jonasson 2026-02-02 dispositions are RSU tax withholdings per Form 4 Note 1, not sales 65. Pangalos joined the board in January 2024, one month after he negotiated the December 2023 AstraZeneca deal as AZ EVP of BioPharma R&D 5664; his open-market commitment reads as conviction, not relationship maintenance.
  • ABS-201 is second-in-class. Hope Medicine's HMI-115 (Bayer-licensed 2019) precedes ABS-201 in both target indications: a Lancet-published Phase 2 endometriosis trial (NCT05101317, n=142), a completed Phase 2 male AGA trial (NCT06118866, n=192), and a Phase 3 endometriosis trial registered 2025-12-29 (NCT07318688, n=540) 21. Absci's mechanism-priority framing fails the public record.
  • ABS-101 interim PK was unfavorable, not favorable. Absci framed the November 2025 readout as "extended half-life as compared to first-generation TL1A antibodies" 6 but disclosed no numeric half-life and named no comparator. Morgan Stanley read the same data as "trailed next-generation competitors" on January 8, 2026, cut its price target from $5.80 to $4.32, removed ABS-101 from its model, and replaced it with ABS-201 at 25% probability of success and $400M peak risk-adjusted sales 798. Absci announced the deprioritization on 2025-11-12 16:05 ET, six days after the interim readout 5.
  • The capital picture is tighter than guidance suggests. Cash plus marketable securities reached $144.3M on 2025-12-31 66 against quarterly operating burn climbing from $21.8M to $16.9M to $25.0M to $29.2M 70717227. Pure-cash math reaches mid-2027; H1 2028 guidance assumes another $50-90M from the ATM on top of the August 2025 $400M base shelf 68. Short interest grew +97% YoY to 37.26M shares (11.72 days to cover) on 2026-01-30, not the +69% earlier reports cited 76.
  • The defensible AGA TAM is $2.7B (Grand View Research, 2023, 8.45% CAGR), not Absci's $25B management figure 4041. Absci's $25B TAM rests on a population x price construct (5-9M annual patients x ~$3,500/yr implied price) 41 the CFO declined to defend on the Q4 2025 call: "we cannot disclose what we think the actual price point will be" 87. The implied price runs 3-67x cash-pay generic finasteride ($74-840/year).
  • The read turns on three conditions: (i) ABS-201's HEADLINE H2 2026 interim hair-growth data land at or below HMI-115's Phase 1b benchmark (+14 hairs/cm-squared over 24 weeks, n=12 open-label) 21; (ii) ABS-101 outlicensing closes with disclosed upfront economics, not another milestone-weighted headline 5; (iii) one Large Pharma signs an Origin-1 platform license (not a target-specific collaboration) with disclosed upfront cash.

The deck frames a $25B addressable market; the record supports a defensible $2.7B one 40. This read is wrong if ABS-201's hair-growth data clear HMI-115's Phase 1b benchmark of +14 hairs/cm-squared over 24 weeks by the HEADLINE H2 2026 interim readout.

§ Under the hood

How this read was built -- enough for a sceptic to check the machinery, and for a builder to judge the work. The sources behind every figure are gathered in the bibliography below.

What the machine did

Each read runs through an audit pipeline: agents query primary sources and transcribe the structured results. The model frames and orders the findings; it does not generate the underlying data. Every figure on the page traces to a source or is flagged as unverified.

Where a source is silent or a query returns nothing, the gap is flagged rather than filled. The read does not conduct interviews, enter private data rooms, model financial projections from assumptions, or run wet-lab experiments. A blank is honest; a confident guess is not.

Wired, and available but not called

The engine calls a focused set of primary sources directly:

  • SEC EDGAR -- XBRL financials, 8-K / S-1 filings, exhibit press releases
  • ClinicalTrials.gov -- trial registry: design, status, endpoints, results
  • PubMed / Europe PMC -- peer-reviewed literature, abstracts, full-text snippets
  • openFDA -- drug labels, Drugs@FDA approvals, FAERS adverse-event reports
  • USPTO / Google Patents -- patents, claims, assignment records
  • Stooq -- US equity daily close and volume
  • Web retrieval -- Jina Reader, Exa, Tavily for primary-document fetch and search

The engine is built on ToolUniverse (Harvard MIMS), the open library of more than 1,000 scientific tools -- structure and interaction databases (UniProt, RCSB PDB, AlphaFold, STRING), pathways and expression (Reactome, KEGG, GTEx, Open Targets), chemistry (ChEMBL, PubChem), oncology evidence (cBioPortal, OncoKB, CIViC), and more. These are available capability the engine reaches for only when a read needs them; a typical company read does not invoke them, so they are listed here as latent reach, not as work performed.

The custom code

  • Citation-URL builders -- turn every identifier (PMID, DOI, NCT, accession) into a canonical primary-source link -- the "two clicks from source" guarantee behind each figure.
  • Fact-audit pipeline -- grades each extracted claim by the tier of its source and assigns a verdict (verified, contradicted, unsupported, not-checkable) against the primary record.
  • Render and fact-token gates -- a schema gate and a fact-token invariant block a malformed report, and block any number, date, or trial id from drifting between the audit and the page.
  • Cross-family adversarial review -- two reviewers from other model families (DeepSeek R1, Kimi K2) independently challenge the primary analysis, so the read is not graded only by the model that wrote it.

§ References

  1. 1. Australian New Zealand Clinical Trials Registry. "ABS-101 Phase 1 single-ascending-dose study, registration (ACTRN12625000212459)." Accessed April 16, 2026. anzctr.org.au. ^1 ^2
  2. 2. Absci Corporation. "Absci Reports Business Updates and Third-Quarter 2025 Financial Results" (ABS-101 interim Phase 1 disclosure). Press release, November 12, 2025. investors.absci.com. ^1 ^2 ^3
  3. 3. Australian New Zealand Clinical Trials Registry. "ABS-101 Phase 1 single-ascending-dose study, dose design (ACTRN12625000212459)." Accessed April 16, 2026. anzctr.org.au. ^
  4. 4. Australian New Zealand Clinical Trials Registry. "ABS-101 Phase 1 single-ascending-dose study, enrollment (ACTRN12625000212459)." Accessed April 16, 2026. anzctr.org.au. ^
  5. 5. Absci Corporation. "Absci Reports Third-Quarter 2025 Financial Results." Press release, November 12, 2025. investors.absci.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17
  6. 6. Benzinga. "HC Wainwright & Co. Maintains Buy on Absci, Raises Price Target to $8." December 17, 2025. benzinga.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13
  7. 7. Insider Monkey. "Analysts Cut Absci (ABSI) Price Target" (Morgan Stanley downgrade to Equal-weight; price target cut to $4.32). Yahoo Finance, January 14, 2026. finance.yahoo.com. ^1 ^2
  8. 8. Insider Monkey. "Analysts Cut Absci (ABSI) Price Target" (Morgan Stanley model revision to ABS-201). Yahoo Finance, January 14, 2026. finance.yahoo.com. ^1 ^2
  9. 9. Insider Monkey. "Analysts Cut Absci (ABSI) Price Target." Yahoo Finance, January 14, 2026. finance.yahoo.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12
  10. 10. ClinicalTrials.gov. "BB-TL1A-VIAL-HLE, a Phase 1a/b anti-TL1A trial (NCT07029971)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3
  11. 11. Cedars-Sinai Medical Center and Prometheus Biosciences. Antibodies that bind TL1A (tulisokibart), U.S. Patent No. 11,999,789 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8
  12. 12. PatSnap. "TL1A Inhibition in IBD: Tulisokibart vs. Anti-TNF." PatSnap Synapse, April 2, 2026. patsnap.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15
  13. 13. Teva Pharmaceuticals. Antibodies that specifically bind to TL1A (duvakitug), U.S. Patent No. 10,138,296 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2
  14. 14. Pharmaphorum. "MSD Advances on a Broad Front with TL1A Drug Tulisokibart" (cited for implied TL1A-class deal value; no published class market-sizing exists). pharmaphorum.com. ^1 ^2 ^3 ^4 ^5 ^6
  15. 15. Boehringer Ingelheim. "Boehringer Ingelheim Licenses TL1A/IL-23 Bispecific Antibody from Simcere." Press release, January 27, 2026. boehringer-ingelheim.com. ^1 ^2 ^3 ^4 ^5 ^6
  16. 16. ClinicalTrials.gov. "ABS-201 Phase 1/2a first-in-human study in androgenetic alopecia (NCT07317544)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4 ^5
  17. 17. ClinicalTrials.gov. "ABS-201 Phase 1/2a study, dose cohorts (NCT07317544)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2
  18. 18. ClinicalTrials.gov. "ABS-201 Phase 1/2a study, trial design (NCT07317544)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4 ^5 ^6
  19. 19. ClinicalTrials.gov. "ABS-201 Phase 1/2a study, secondary outcome measures (NCT07317544)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4 ^5 ^6
  20. 20. Absci Corporation. "Absci to Report Business Updates and First-Quarter 2026 Financial and Operating Results on May 7, 2026." GlobeNewswire, April 23, 2026. globenewswire.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7
  21. 21. ClinicalTrials.gov. "HMI-115 Phase 3 study in endometriosis (NCT07318688)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17 ^18 ^19 ^20 ^21 ^22 ^23 ^24 ^25 ^26 ^27 ^28 ^29 ^30 ^31 ^32 ^33 ^34 ^35 ^36 ^37
  22. 22. ClinicalTrials.gov. "HMI-115 Phase 2 study in endometriosis (NCT05101317)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  23. 23. ClinicalTrials.gov. "Pelage PP405 Phase 2a study in androgenetic alopecia (NCT06393452)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4
  24. 24. Absci Corporation. "Absci Announces First Participants Dosed in Phase 1/2a HEADLINE Trial of ABS-201 for the Treatment of Androgenetic Alopecia." GlobeNewswire, December 4, 2025. globenewswire.com. ^1 ^2 ^3
  25. 25. Levine et al. "Origin-1: A Generative AI Platform for De Novo Antibody Design Against Novel Epitopes." bioRxiv preprint, 2026. biorxiv.org. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13
  26. 26. Absci Corporation. Anti-HER2 antibody compositions designed by artificial intelligence (ABS-501), WIPO Patent Application WO 2025/122885 A1. patents.google.com. ^1 ^2
  27. 27. Absci Corporation. Form 10-K for the fiscal year ended December 31, 2025. U.S. Securities and Exchange Commission, filed March 24, 2026. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17 ^18
  28. 28. Absci Corporation. Cytoplasmic expression system (SoluPro), U.S. Patent No. 12,157,766 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2 ^3
  29. 29. Absci Corporation. Inducible coexpression system, U.S. Patent No. 9,617,335 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2 ^3 ^4
  30. 30. Absci Corporation. Cancer-associated antibody compositions (Totient-derived), U.S. Patent No. 12,285,484 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2 ^3
  31. 31. Absci Corporation. Cancer-associated antibody compositions (Totient-derived), U.S. Patent No. 12,134,655 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2 ^3
  32. 32. "Phase 2 Trial of Anti-TL1A Monoclonal Antibody Tulisokibart for Ulcerative Colitis." New England Journal of Medicine, 2024. nejm.org. ^1 ^2 ^3 ^4 ^5 ^6 ^7
  33. 33. ClinicalTrials.gov. "Tulisokibart ARTEMIS-UC Phase 2 study, posted results (NCT04996797)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4 ^5 ^6
  34. 34. ClinicalTrials.gov. "Duvakitug (TEV-48574) RELIEVE UCCD Phase 2b study, ulcerative-colitis results (NCT05499130)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3 ^4 ^5
  35. 35. Shanehsazzadeh et al. De novo antibody design against HER2 using generative AI with high-throughput wet-lab validation. bioRxiv preprint, 2023. biorxiv.org. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13
  36. 36. Bennett et al. "Atomically Accurate de Novo Design of Single-Domain Antibodies." Nature, 2025. doi.org. ^
  37. 37. Xaira Therapeutics. "Xaira Therapeutics Launches to Deliver on the Promise of AI for Drug Discovery and Development." Business Wire, April 23, 2024. businesswire.com. ^1 ^2 ^3 ^4
  38. 38. Xaira Therapeutics. "Xaira Therapeutics Launches to Deliver on the Promise of AI for Drug Discovery and Development." Business Wire, April 23, 2024. businesswire.com. ^1 ^2 ^3
  39. 39. Absci Corporation. "Absci Reports Business Updates and Third-Quarter 2025 Financial Results." GlobeNewswire, November 12, 2025. globenewswire.com. ^1 ^2 ^3 ^4 ^5
  40. 40. Grand View Research. "Androgenetic Alopecia Market Size, Share & Trends Analysis Report, 2024-2030" (Report ID GVR-4-68040-197-2). grandviewresearch.com. ^1 ^2 ^3 ^4
  41. 41. Absci Corporation. "Q4 2025 Earnings Call Transcript" (CFO Zachariah Jonasson, androgenetic-alopecia total-addressable-market remarks). The Motley Fool, March 25, 2026. fool.com. ^1 ^2 ^3 ^4 ^5
  42. 42. ClinicalTrials.gov. "ABS-201 Phase 1/2a study, primary outcome measures (NCT07317544)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^1 ^2 ^3
  43. 43. Absci Corporation. Designing biomolecule sequence variants with pre-specified attributes, U.S. Patent Application No. 2023/0268026 A1. patents.google.com. ^1 ^2 ^3
  44. 44. Absci Corporation. Form 10-K for the fiscal year ended December 31, 2024. U.S. Securities and Exchange Commission, filed March 5, 2025. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17
  45. 45. Absci Corporation. Inducible expression system, U.S. Patent No. 10,465,197 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2
  46. 46. Absci Corporation. Methods for inducible expression, U.S. Patent No. 11,214,807 B2. U.S. Patent and Trademark Office. patents.google.com. ^
  47. 47. Absci Corporation. Host cells for an inducible coexpression system, U.S. Patent No. 11,447,781 B2. U.S. Patent and Trademark Office. patents.google.com. ^
  48. 48. Absci Corporation. C-peptides and proinsulin polypeptides, U.S. Patent No. 11,584,785 B2. U.S. Patent and Trademark Office. patents.google.com. ^
  49. 49. Absci Corporation. Anti-TL1A antibody composition of matter (ABS-101), U.S. Patent No. 12,473,370 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2 ^3 ^4
  50. 50. Bayer (licensed to Hope Medicine). Prolactin-receptor antibody for male and female pattern hair loss, WIPO Patent Application WO 2019/011719 A1. patents.google.com. ^1 ^2 ^3 ^4
  51. 51. Absci Corporation. Humanized antibodies to TL1A and uses thereof, U.S. Patent Application No. 2024/0327532 A1. patents.google.com. ^
  52. 52. Paragon Therapeutics. TL1A binding proteins and methods of use, U.S. Patent No. 12,466,890 B1. U.S. Patent and Trademark Office. patents.google.com. ^
  53. 53. Absci Corporation. Quantitative affinity-activity-specific cell enrichment (ACE assay), WIPO Patent Application WO 2024/040020 A1. patents.google.com. ^
  54. 54. Absci Corporation. Nanobody library screening using bacterial surface display, WIPO Patent Application WO 2025/144700 A1. patents.google.com. ^
  55. 55. Flagship Pioneering Innovations VI, Inc. (Generate Biomedicines-affiliated). Antibody composition, U.S. Patent No. 12,020,776 B2. U.S. Patent and Trademark Office. patents.google.com. ^1 ^2
  56. 56. Absci Corporation. Definitive Proxy Statement (Schedule 14A). U.S. Securities and Exchange Commission, April 28, 2025. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17 ^18 ^19 ^20
  57. 57. Absci Corporation. Form 8-K, Item 7.01 (appointment of Chief Medical Officer). U.S. Securities and Exchange Commission, March 6, 2026. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7
  58. 58. Absci Corporation. Form 8-K, Exhibit 99.1 (fourth-quarter and full-year 2025 financial results). U.S. Securities and Exchange Commission, March 24, 2026. sec.gov. ^
  59. 59. Ransi Somaratne. LinkedIn professional profile. Accessed 2026. linkedin.com. ^1 ^2
  60. 60. Absci Corporation. Form 8-K, Item 5.02 (retirement of Andreas Busch). U.S. Securities and Exchange Commission, March 6, 2026. sec.gov. ^1 ^2 ^3 ^4 ^5
  61. 61. U.S. Securities and Exchange Commission. Form 4 insider-transaction filings for Absci Corporation (CIK 0001672688), April 2025-April 2026. EDGAR. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7
  62. 62. Absci Corporation. Form 8-K, Item 5.02 (appointment of director Mary Szela). U.S. Securities and Exchange Commission, July 8, 2025. sec.gov. ^1 ^2 ^3 ^4 ^5
  63. 63. Absci Corporation, investor-relations board profile for Mary Szela; supplemented by her LinkedIn professional profile. Accessed 2026. investors.absci.com. ^1 ^2 ^3
  64. 64. Reuters. "AstraZeneca Partners with Absci on AI to Discover Cancer Drug." December 3, 2023. reuters.com. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11
  65. 65. U.S. Securities and Exchange Commission. Form 4 for Absci Corporation insiders, explanation of responses note 1 (restricted-stock-unit tax withholding). EDGAR, February 2, 2026. sec.gov. ^1 ^2 ^3
  66. 66. Absci Corporation. Form 8-K, Exhibit 99.1 (fourth-quarter and full-year 2025 financial results). U.S. Securities and Exchange Commission, March 24, 2026. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6
  67. 67. Absci Corporation. Form 10-K for the fiscal year ended December 31, 2025. U.S. Securities and Exchange Commission, filed March 24, 2026. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17 ^18 ^19 ^20 ^21 ^22 ^23 ^24 ^25 ^26 ^27 ^28 ^29 ^30 ^31 ^32 ^33 ^34 ^35 ^36 ^37
  68. 68. Absci Corporation. Form S-3 shelf registration statement (up to $400 million). U.S. Securities and Exchange Commission, August 12, 2025. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8
  69. 69. Absci Corporation. Form 8-K, Exhibit 99.1 (third-quarter 2025 financial results). U.S. Securities and Exchange Commission, November 12, 2025. sec.gov. ^
  70. 70. Absci Corporation. Form 10-Q for the quarterly period ended March 31, 2025. U.S. Securities and Exchange Commission. sec.gov. ^1 ^2
  71. 71. Absci Corporation. Form 10-Q for the quarterly period ended June 30, 2025. U.S. Securities and Exchange Commission. sec.gov. ^1 ^2
  72. 72. Absci Corporation. Form 10-Q for the quarterly period ended September 30, 2025. U.S. Securities and Exchange Commission. sec.gov. ^1 ^2
  73. 73. Absci Corporation. Definitive Proxy Statement (Schedule 14A). U.S. Securities and Exchange Commission, April 22, 2026. sec.gov. ^1 ^2
  74. 74. Absci Corporation. Form 8-K, Exhibit 99.1 (second-quarter 2025 financial results). U.S. Securities and Exchange Commission, August 2025. sec.gov. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8
  75. 75. Absci Corporation. Form 8-K, Exhibit 99.1 (first-quarter 2025 financial results). U.S. Securities and Exchange Commission, May 2025. sec.gov. ^
  76. 76. Nasdaq. "Absci Corporation (ABSI) Short Interest." Accessed January 2026. nasdaq.com. ^1 ^2 ^3
  77. 77. Absci Corporation. Form D (notice of exempt offering of securities). U.S. Securities and Exchange Commission, filed April 21, 2016. sec.gov. ^1 ^2
  78. 78. Absci Corporation. Form D (notice of exempt offering of securities). U.S. Securities and Exchange Commission, filed June 8, 2018. sec.gov. ^1 ^2
  79. 79. Absci Corporation. Form D (notice of exempt offering of securities). U.S. Securities and Exchange Commission, filed December 19, 2019. sec.gov. ^1 ^2
  80. 80. Absci Corporation. Form D (notice of exempt offering of securities). U.S. Securities and Exchange Commission, filed October 30, 2020. sec.gov. ^1 ^2
  81. 81. Absci Corporation. Form D (notice of exempt offering of securities). U.S. Securities and Exchange Commission, filed March 30, 2021. sec.gov. ^1 ^2
  82. 82. WallStreetZen. "Absci (ABSI) Stock Overview." Accessed April 8, 2026. wallstreetzen.com. ^1 ^2 ^3
  83. 83. WallStreetZen. "Absci (ABSI) Financial Profile." Accessed April 8, 2026. wallstreetzen.com. ^1 ^2
  84. 84. ClinicalTrials.gov. "Duvakitug (TEV-48574) RELIEVE UCCD Phase 2b study, Crohn's-disease results (NCT05499130)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  85. 85. Mordor Intelligence. "Inflammatory Bowel Disease Treatment Market: Forecast 2026-2031.". mordorintelligence.com. ^
  86. 86. Fortune Business Insights. "Endometriosis Treatment Market" (Report FBI114330). fortunebusinessinsights.com. ^
  87. 87. Absci Corporation. "Q4 2025 Earnings Call Transcript" (CFO Zachariah Jonasson: "we cannot disclose what we think the actual price point will be"). The Motley Fool, March 24, 2026. fool.com. ^1 ^2 ^3
  88. 88. Mordor Intelligence. "HER2-positive Breast Cancer Treatment Market: Forecast 2026-2031.". mordorintelligence.com. ^
  89. 89. ClinicalTrials.gov. "Tulisokibart ATLAS-UC Phase 3 study (NCT06052059)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  90. 90. ClinicalTrials.gov. "Duvakitug SUNSCAPE-1 Phase 3 ulcerative-colitis induction study (NCT07184996)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  91. 91. ClinicalTrials.gov. "Duvakitug STARSCAPE-1 Phase 3 Crohn's-disease induction study (NCT07184931)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  92. 92. ClinicalTrials.gov. "Duvakitug SUNSCAPE-2 Phase 3 ulcerative-colitis maintenance study (NCT07185009)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  93. 93. ClinicalTrials.gov. "Duvakitug STARSCAPE-2 Phase 3 Crohn's-disease maintenance study (NCT07184944)." U.S. National Library of Medicine, accessed April 2026. clinicaltrials.gov. ^
  94. 94. AnaptysBio. "Anaptys Announces Phase 2 Trial of Rosnilimab Did Not Meet Primary or Secondary Endpoints at Week 12 in Moderate-to-Severe Ulcerative Colitis." Press release, November 10, 2025. ir.anaptysbio.com. ^
  95. 95. Robinhood. "Absci (ABSI) Snapshot." Accessed April 2, 2026. robinhood.com. ^
  96. 96. WallStreetZen. "AbCellera Biologics (ABCL) Stock Overview." Accessed April 8, 2026. wallstreetzen.com. ^1 ^2 ^3
  97. 97. Yahoo Finance. Peer market-capitalization snapshot (AbCellera Biologics), March-April 2026. ca.finance.yahoo.com. ^
  98. 98. "AbCellera Biologics (ABCL) Reports Fourth-Quarter and Full-Year 2025 Results." Yahoo Finance / GuruFocus, February 25, 2026. finance.yahoo.com. ^
  99. 99. AbCellera Biologics. "Fourth-Quarter and Full-Year 2025 Results." Investor relations, February 24, 2026. investors.abcellera.com. ^1 ^2
  100. 100. AbCellera Biologics. "Fourth-Quarter 2025 Earnings Call" summary. Yahoo Finance, February 2026. finance.yahoo.com. ^
  101. 101. WallStreetZen. "Recursion Pharmaceuticals (RXRX) Stock Overview." Accessed April 7, 2026. wallstreetzen.com. ^
  102. 102. Stock Titan. "Recursion Pharmaceuticals (RXRX) Files Annual Report on Form 10-K." February 25, 2026. stocktitan.net. ^1 ^2 ^3 ^4
  103. 103. WallStreetZen. "Schrodinger (SDGR) Stock Overview." Accessed April 2, 2026. wallstreetzen.com. ^
  104. 104. Schrodinger, Inc. "Fourth-Quarter and Full-Year 2025 Results." Investor relations, February 25, 2026. ir.schrodinger.com. ^1 ^2
  105. 105. Schrodinger, Inc. "Schrodinger Reports Fourth-Quarter and Full-Year 2025 Financial Results." Business Wire, February 25, 2026. markets.financialcontent.com. ^1 ^2 ^3
  106. 106. Yahoo Finance. Peer market-capitalization snapshot (Relay Therapeutics), March-April 2026. ca.finance.yahoo.com. ^
  107. 107. Relay Therapeutics. "Relay Therapeutics Reports Fourth-Quarter and Full-Year 2025 Financial Results." February 26, 2026. ir.relaytx.com. ^
  108. 108. Relay Therapeutics. "Fourth-Quarter and Full-Year 2025 Results." Investor relations, February 26, 2026. ir.relaytx.com. ^1 ^2
  109. 109. Relay Therapeutics. "Relay Therapeutics Reports Fourth-Quarter and Full-Year 2025 Financial Results" (2026 clinical catalysts). February 26, 2026. ir.relaytx.com. ^
  110. 110. Generate:Biomedicines. "Generate:Biomedicines Announces $273 Million Series C Financing." September 14, 2023. generatebiomedicines.com. ^
  111. 111. Generate:Biomedicines. "Series C Financing Announcement." September 14, 2023. generatebiomedicines.com. ^
  112. 112. Generate:Biomedicines. "Generate Biomedicines, Inc. Announces Pricing of Initial Public Offering." February 2026. generatebiomedicines.com. ^
  113. 113. U.S. Securities and Exchange Commission. EDGAR full-text search of Absci Corporation filings (CIK 0001672688). Accessed April 2026. efts.sec.gov. ^1 ^2
  114. 114. Absci Corporation. Form 8-K, Exhibit 99.1 (second-quarter 2025 financial results; Almirall collaboration expansion). U.S. Securities and Exchange Commission, August 2025. sec.gov. ^1 ^2 ^3
  115. 115. MarketScreener / S&P Capital IQ. "Recursion Pharmaceuticals to Acquire Exscientia." 2024. marketscreener.com. ^