/w=3840,quality=90,fit=scale-down)
Raphael ThysGenetic Engineering & Biotechnology News
[Credit: Cybrain/Getty Images]
Insilico Medicine has been granted the FDA’s first Orphan Drug Designation for a drug discovered and designed using artificial intelligence (AI)—the company’s lead pipeline candidate INS018_055, a small molecule inhibitor treatment for idiopathic pulmonary fibrosis (IPF).
INS018_055 is Insilico’s first wholly-owned program in which AI was used to identify a novel target—which the company has identified only as “Target X”—and generate novel small molecules through the company’s Pharma.AI platform.
INS018_055 is set to enter a global, randomized multi-site Phase II trial “early” this year.
“Study endpoints include safety, tolerability, pharmacokinetics, and efficacy to include change in FVC [forced vital capacity] in patients with idiopathic pulmonary fibrosis,” Alex Zhavoronkov, PhD, Insilico’s founder and CEO, told GEN.
According to Insilico, INS018_055 represents a potential first-in-class treatment against a novel target with potential relevance in a broad range of fibrotic indications—including kidney fibrosis (now in the IND-enabling phase) and skin fibrosis (now in discovery phase).
“Insilico used this newly discovered target as the basis for the structure-based design of a potentially first-in-class novel small molecule inhibitor,” Zhavoronkov told GEN. “Simultaneous success in target discovery and drug candidate generation for a broad indication such as fibrosis is a precedent among AI systems, and it provides proof of concept for the ability of deep learning to link biology and chemistry in an integrated workflow.”
“We plan to disclose additional details about the unique mechanism of action of the drug in an upcoming publication,” Zhavoronkov added.
Just last month, Insilico announced positive topline safety, tolerability, and pharmacokinetics (PK) results from a Phase I trial (NCT05154240) assessing INS018_055 in 78 healthy volunteers in New Zealand.
The topline results—which were in line with Insilico’s preclinical modeling—showed INS018_055 was generally safe and well tolerated, with a favorable PK profile and no significant accumulation after seven days. No deaths or serious adverse events were reported during the trial though one participant who received INS018_055 in the 30 mg once daily cohort of the study’s multiple ascending dose portion of the study discontinued treatment due to a moderate influenza-like illness that was deemed unrelated to the study treatment.
“We are progressing the global clinical development of the program at top speed to allow patients with fibrotic diseases to benefit from this novel therapeutic as soon as possible,” Zhavoronkov said.
Insilico is among a number of companies that have recently claimed development and clinical milestones related to AI-based drugs. Exscientia, an Oxford-based pioneer in using AI to design and develop small molecule drugs, was the first company to have an AI-designed molecule entering clinical trials in 2020—EXS-21546, a majority-owned A2A receptor antagonist which was co-invented and is being developed with Evotec as an anti-cancer immunotherapy.
Exscientia recently announced its latest clinical candidate EXS4318, a small molecule, potentially first-in-class selective Protein kinase C (PKC) theta inhibitor designed by the company for immunology and inflammation indications and licensed to Bristol Myers Squibb (BMS).
Last month Absci, which has declared its aim of growing into the Google search engine of protein-based drug discovery and biomanufacturing, announced an AI drug milestone of its own, saying that its AI platform had succeeded in creating and validating de novo antibodies in silico.
Insilico began development of INS018_055 in February 2021 using Pharma.AI. The platform incorporates a pair of specific-function platforms.
One of those is PandaOmics, which is designed to enable multi-omics discovery of novel targets through a proprietary pathway analysis approach called iPanda that infers pathway activation or inhibition, finding connections between seemingly unrelated genes based on dysregulated molecular processes.
PandaOmics scores potential targets by generating a ranked list of potential targets or “target hypotheses” for a given disease (or disease subtype), then filters those target hypotheses based on their novelty, accessibility by small molecules, biologics, and safety. The panda name reflects the company’s Chinese origins (Insilico is based in Hong Kong).
The other specific-function platform, Chemistry42, is an automated, machine-learning de-novo drug design and scalable engineering platform which, according to the company, enables users to find novel lead-like molecules in as little as a week. Drawing on large numbers of compounds and molecular fragments, Chemistry42 uses generative AI to create new drug-like molecules optimized to have specific properties.
Chemistry42 sets rules for molecule shape, chemical complexity, synthetic accessibility, metabolic stability, and other properties that novel molecules must satisfy. Once generated, a new compound is annotated with all properties—including physico-chemical parameters, binding scores, drug-likeness features—then mapped on vendors’ catalogs and proprietary libraries for any similarity and novelty.
“The FDA’s orphan drug designation for the IPF indication is an important milestone in the development of INS018_055,” Feng Ren, PhD, Insilico’s co-CEO and CSO, said in a statement. “Insilico scientists are now further advancing clinical validation and accelerating the project to meet clinical needs and benefit patients worldwide.”