Duvakitug

Structure-free AI platform delivers picomolar leads while cutting wet-lab validation to 20-50 samples per round Ainnocence Inc., an AI-driven pharmaceutical & material IP-generating platform company, today reported performance results from 75 completed protein design programs and announced that it is seeking licensing and partnering agreements for 28 antibody, bispecific and ADC programs that have generated positive preclinical data across oncology, immunology and metabolic indications. Across 58 affinity maturation campaigns, Ainnocence's SentinusAI® platform succeeded in 53, a 91.4% program success rate, with only five campaigns failing to produce an improved binder. On novel targets attempted with no structural information and no prior binder, 9 of 17 de novo campaigns (53%) yielded validated hits. Performance without structure Unlike 3D structure-dependent approaches, SentinusAI® predicts binding directly from sequence, removing the requirement for a resolved co-crystal or predicted complex. That design choice underpins both the platform's throughput and its economics: Ainnocence reports a roughly 10,000-fold reduction in computational cost relative to conventional 3D modeling workflows, a computation cycle of one to two weeks per round, and capacity to run more than 100 programs concurrently. The practical effect is on the wet lab. Because each round is narrowed to 20-50 designs before synthesis, partners test tens of molecules rather than thousands: · De novo design: 7.5% positive rate on round-1 zero-shot prediction, rising to 30% on round-2 reinforcement-learning iteration · Affinity maturation: 21.5% positive rate on round-1 zero-shot prediction, rising to 43.8% on round-2 iteration · Affinity gains from tens- to thousands-fold over the parent molecule, with leads reaching picomolar potency Picomolar leads across formats Delivered programs span Fabs, scFvs, mAbs, bispecifics and ADC-enabling binders . Of the campaigns with quantified KD readouts, 28 reached sub-nanomolar affinity and seven reached 100 pM or better . Leading results include an IL-15×IL-21 bispecific at approximately 0.09 pM on the IL-21 arm, a TNFα-OX40L bispecific at 8.6 pM, an anti-C1q program at 1-20 pM and a TL1A×p40 bispecific at 23 pM. An anti-VEGF scFv-IgG measured 27.8-34.6 pM by SPR, outperforming the bevacizumab reference run in the same experiment. Twenty-eight programs available for licensing Ainnocence's internal pipeline comprises 37 programs, of which 28 have produced positive data and are now available for licensing or partnering, individually or as a portfolio . Highlights include a TL1A×p40 bispecific (KD 0.16 nM; benchmark duvakitug, Phase 3), an anti-IL-36R Fab with 13 of 21 matured variants binding below the 1 pM detection limit (benchmark spesolimab-class HB-0034), an anti-TROP2 ADC lead at 0.37-0.43 nM with confirmed internalization (benchmark Trodelvy) and an anti-B7-H3 program at approximately 0.56 nM (benchmark ifinatamab deruxtecan, Phase 2). Seven ADC-ready internalizing oncology targets have been validated by FACS, and internalization assay and leads have expressed at titers up to approximately 440 mg/L at greater than 99% SEC purity. “The number that matters to a partner is not how many molecules our model can score, it is how many they have to make,” said Dr.Lurong Pan, PhD, Founder and CEO of Ainnocence . “Twenty to fifty designs per round, one to two weeks of compute and a 91.4% program success rate across 58 maturation campaigns is a different operating model for antibody discovery. And because we never depended on a resolved structure, the targets other platforms set aside are the ones we take on.” Collaborate with Ainnocence Ainnocence is seeking licensing and partnering relationships to advance the pipeline through affinity finalization, ADC conjugation, in vivo efficacy and clinical development. Additional non-confidential and confidential materials are available on request. About Ainnocence Inc. Founded in 2021 and headquartered in California, Ainnocence is a next-generation biotechnology company transforming drug discovery and synthetic biology through AI-based, sequence-first engineering. Ainnocence’s self-evolving platform evaluates up to 10 billion molecules spanning proteins, antibodies, small molecules, nucleic acids, and chemical formulations within hours to weeks, enabling rapid, multi-objective design across therapeutic, biological, and chemical systems . By reducing R&D timelines and costs while increasing success rates, Ainnocence empowers industry and academic partners to pursue complex biological innovation with greater precision and control. Contact: Dr. Lurong Pan, PhD Ainnocence Inc. +1 205-249-7424 lurong.pan@ainnocence.com service@ainnocence.com Visit us on social media:  LinkedIn  |  YouTube  |  X

A phase 2 trial of Merck & Co.’s anti-TL1A monoclonal antibody has missed its primary endpoint, closing off one avenue for expansion of an asset acquired in a $10.8 billion takeover. The setback was offset by a midphase win in another indication. Merck acquired the antibody, tulisokibart, when it bought Prometheus Biosciences in 2023. Racing Roche’s afimkibart and Sanofi and Teva’s duvakitug, Merck expanded on Prometheus’ work by launching a broad R&D program. The U.S. Big Pharma inherited a phase 2 study in systemic sclerosis associated with interstitial lung disease (SSc-ILD) and started trials in indications including hidradenitis suppurativa (HS). In its second-quarter earnings update, the drug developer revealed that the SSc-ILD study missed its primary endpoint. As a result, Merck is ending the SSc-ILD trial, although the company pointed out that no new safety concerns were seen in the failed study. Merck reported the failure alongside news of the success of its phase 2b trial in HS, which hit its primary and key secondary endpoints. The drugmaker plans to share HS data at an upcoming medical meeting. In June, Merck reported a phase 3 victory for tulisokibart in ulcerative colitis (UC). With hits in UC and HS, Merck still has a clear path to turning tulisokibart into a major product that could help offset the loss of exclusivity on Keytruda in 2028. But the failure in SSc-ILD nonetheless dents efforts to establish TL1A as a “node” for treating a wide range of diseases, as happened with TNF and IL-23. The broad clinical development program is designed to validate TL1A as a node in immunofibrosis. “We are hoping that as we see data from this, can those different indications make that concept of immunofibrosis come alive?” Dean Li, M.D., Ph.D., president of Merck Research Laboratories, explained (PDF) at an event back in June. “And if it does, then it gives us a beachhead to expand, deepen and extend with other biomarkers and other combinations, all being different depending on what the indication is.” The coming year could shed more light on Merck’s chances of delivering Li’s vision. Merck is scheduled to publish phase 3 maintenance data in UC this year. Phase 2 trials in rheumatoid arthritis and radiographic axial spondyloarthritis have primary completion dates of next month and June 2027 , respectively, per the federal trials database.