InSilico Medicine Hong Kong Ltd.

Just as call-and-response apps like ChatGPT are trained on the various connections between words of the English language, BioNeMo toolkits can now draw lines among the DNA, RNA and proteins that ultimately govern cell function, according to Nvidia. A little more than a year after announcing plans for its molecule design programs to set up shop in the field of generative AI, Nvidia is opening its doors.  Just as call-and-response apps like ChatGPT are trained on the various connections between words of the English language, the company’s BioNeMo toolkits can now draw lines among the DNA, RNA and proteins that ultimately govern cell function, according to Nvidia—with the goal of helping drug discovery researchers to conduct more experiments virtually before pursuing them in the physical world. Nvidia plans to offer the expanded BioNeMo models as a series of cloud-based enterprise services—with predictors for drug molecule binding successes and changes in protein structures—hosted through its AWS HealthOmics platform. The announcement was made during Nvidia’s GTC conference this week in San Jose, California. The new AI programs include DNABERT, a foundation model trained on genomic sequences that aims to predict the function of different regions of genetic code and the effects of mutations. Another upcoming model is scBERT, trained on single-cell RNA sequencing data for exploring gene knockout hypotheses. Meanwhile, EquiDock looks to forecast how two proteins will interact with each other in 3D.  Nvidia said that more than 100 biopharma and AI drug discovery companies already employ its BioNeMo platform, including Amgen, Astellas Pharma, Cadence, Iambic, Insilico Medicine, Recursion and Terray Therapeutics.  "Generative AI is transforming drug discovery by allowing us to build sophisticated models and seamlessly integrate AI into the antibody design process,” said Amgen’s chief technology officer David Reese. “Our team is harnessing this technology to create the next generation of medicines that will bring the most value to patients.”

The latest around the BIOSECURE Act, Takeda's plan to shut down a gene therapy facility, and an early death issue in Carvykti's clinical trial made our news this week. WuXi AppTec left the Biotechnology Industry Organization amid increasing pressure on the trade group stemming from a draft U.S. bill targeting certain Chinese biotechs. Takeda will shut down a facility in Austria as the Japanese pharma pivots away from research in adeno-associated virus gene therapies. The FDA questioned whether Carvykti should be approved in earlier multiple myeloma after noticing an early detriment to patients' survival in a clinical trial. And more. 1. Biotech trade group pledges support for BIOSECURE Act, plans to boot member WuXi AppTec The Biotechnology Industry Organization (BIO) is severing ties with WuXi AppTec as one of the “important steps” that the trade group is taking to support U.S. national security. In a revised statement, BIO said WuXi “proactively ended its membership.” BIO said it supports the BIOSECURE Act, which looks to block “adversary biotech companies,” including WuXi, from getting U.S. government contracts. BIO’s move came after its lobbying activity against the draft bill drew the ire of a lawmaker. 2. Takeda winds down manufacturing and R&D operations in Austria after gene therapy pivot, plots 190 layoffs Takeda is shutting down operations at its adeno-associated virus gene therapy plant in Orth, Austria, and nearly 200 employees will lose their jobs. Takeda said it would divest the site about a year ago as part of its decision to end early R&D activities around AAV gene therapies. But after a “thorough and lengthy effort,” the company couldn’t find a buyer, a Takeda spokesperson said. 3. FDA flags early deaths in CAR-T myeloma trials for J&J and Legend’s Carvykti, Bristol Myers’ Abecma The FDA has raised concerns about an increased risk of death for Johnson & Johnson and Legend Biotech’s Carvykti compared with standard combinations in the first few months of its phase 3 trial in earlier lines of treatment for multiple myeloma. More patients in the Carvykti arm died than those in the control arm in the first 10 months of the CARTITUDE-4 trial. That overall survival data set is up for discussion during an FDA advisory committee meeting. 4. Astellas puts the cork back in $350M Xork deal after Cartesian-Selecta merger Astellas has decided to end a licensing deal with Cartesian Therapeutics about a year after signing it originally with Selecta Biosciences. Under the deal, Astellas paid $10 million upfront to use Cartesian’s Xork, an immunoglobulin protease, with its gene therapy candidate AT845. The hope was that the asset would help more patients become eligible for adeno-associated virus-based gene therapies. 5. Insilico 'pulls back the curtain' on AI drug discovery process in hope of convincing skeptics In a Nature Biotechnology paper, Insilico Medicine described how it used artificial intelligence to discover, develop and test its lead candidate, INS018_055, a TNIK inhibitor that’s in phase 2 development for the lung disease idiopathic pulmonary fibrosis. The company’s AI platform, PandaOmics, was used to both identify a therapeutic target and to design a drug candidate. 6. Biocon continues to divest with $150M sale of insulin and other brands to Eris Biocon has agreed to sell its Indian branded formulations business to Eris for about $150 million. The deal includes a portfolio of metabolic, oncology and critical care products, including Insugen, which was the first recombinant human insulin product developed domestically in India. The divested franchise accounted for 3.7% of Biocon’s revenue in its fiscal year 2023. 7. Innovent shutters ADC-focused Maryland R&D site, moves lab to California (BioSpace, Yicai) Innovent Biologics will shut down its Rockville, Maryland, site, which serves as the Chinese firm’s U.S. headquarters and is conducting research in antibody-drug conjugates. A BioSpace article, citing former employees, reported that the R&D work will move to Chin and that the company has axed the entire research team at the Maryland location. But Innovent told Yicai that the lab will be relocated to California. Other News of Note: 8. Chinese authorities rumored to be working on policies to support innovative drugs (Bloomberg) 9. Aurobindo unit begins shipping drugs from India plant after FDA inspection triggered production halt 10. BeiGene bumps AbbVie, J&J from top lymphoma event sponsor spot 11. Sun Pharma issues recall of gout drug due to microbial contamination 12. AstraZeneca taps Mankind to distribute Symbicort in India (The Economic Times)

Along with the paper, Insilico Medicine also released an interactive GPT that answers questions about TNIK and the company’s AI process.  A new paper in Nature Biotechnology published March 8 offers the first in-depth look at how AI-driven biotech Insilico Medicine used artificial intelligence to discover, develop and test its lead drug candidate.  Insilico founder Alex Zhavoronkov, Ph.D., called the paper an “essential blueprint for the pharma and biotech industry” on how to use AI in drug development.  “We hope this will serve as a tipping point and motivation to accelerate the pace of AI drug discovery and increase pharma partnerships on our AI-designed therapeutics, particularly in the preclinical stages,” Zhavoronkov said in an email to Fierce Biotech Research. In the paper, Insilico describes how AI was used to find a potent target for fibrosis, then a therapeutic and then a drug to match, using 13 different experiments and three clinical trials. “In order to advance a new era of generative AI-driven drug discovery, it’s important to set the benchmark for what that means and to prove those results in a peer-reviewed scientific journal,” Zhavoronkov said. The article goes step-by-step through the process that Insilico undertook to develop INS018_055, a small-molecule TNIK inhibitor that is currently in phase 2 trials for the treatment of the lung disease idiopathic pulmonary fibrosis (IPF). That includes how Insilico used its proprietary AI platform, PandaOmics, to both identify a target and come up with a drug candidate to treat the disease. The paper also discusses additional potential indications for INS018_055 beyond IPF, including skin and kidney fibrosis as well as diseases of aging.  The team discovered that INS018_055 might have potential as a treatment for age-related disease by combining disease analysis with the company’s “Hallmarks of Aging Assessment,” a complex algorithm that takes into account the various biological mechanisms involved in aging to identify drug targets.  “TNIK emerged as one of the most promising dual-purpose targets related to anti-fibrosis and aging in this assessment, and we are very interested in further research in this direction,” Zhavoronkov said. He added that metabolic regulation may be another direction to explore, as a recent study published in Science Advances shows that the enzyme regulates lipid metabolism in obesity.  At the same time the paper was published, Insilico also released an interactive GPT that answers questions about TNIK and the company’s AI process.  By pulling back the curtain on AI-driven drug development at Insilico, Zhavoronkov hopes the paper will give any remaining skeptics of the company’s technology and process “a clear understanding of Insilico’s unique capabilities, and the quality of our AI-discovered and designed drugs, backed by scientific data,” he said.