AI Platform Delivers Eye Drop Candidate That Triples Efficacy Against Leading Cause of Vision Loss

Artificial intelligence continues to accelerate drug discovery timelines while potentially improving therapeutic outcomes, as evidenced by Insilico Medicine's latest preclinical candidate nomination. The company has designated ISM9077, an AI-designed small molecule targeting inflammatory and degenerative ocular diseases, as its 32nd preclinical candidate since 2021—a milestone that underscores the scalability of generative chemistry platforms in pharmaceutical development.

Superior Efficacy in Preclinical Models

ISM9077 demonstrated approximately three times the efficacy of current therapies in preclinical models of dry age-related macular degeneration (dry AMD), a leading cause of irreversible central vision loss in older adults. According to the company, the compound improved retinal structure, visual function, and histopathology across key endpoints in these models.

The candidate also showed promising results in uveitis models, where it reduced inflammation and inflammatory cytokine release while improving retinal function. In dry eye disease models, ISM9077 increased tear production and reduced corneal inflammation, outperforming cyclosporine A, an established treatment. Validation studies in Cynomolgus monkey models—which more closely approximate human biology—yielded positive results supporting the mechanism of action.

The compound's pharmacokinetic profile revealed retinal exposure levels 2-5.5 times higher than plasma concentrations, with favorable oral bioavailability, low-to-moderate clearance, and a wide safety margin. These properties support formulation as an eye-drop medication, potentially offering a non-invasive alternative to injectable therapies currently used for some ocular conditions.

Generative AI Cuts Development Timeline in Half

Insilico Medicine developed ISM9077 using its Chemistry42 generative chemistry platform, employing a structure-based drug design approach. The company reports achieving preclinical candidate nomination in 12-18 months, compared to an estimated 2.5-4 years for traditional drug discovery programs—representing a 40-60% reduction in early-stage development timelines.

"ISM9077 exemplifies the power of AI-driven drug discovery to address unmet medical needs with unprecedented speed and precision," said Alex Zhavoronkov, Founder, CEO, and Chief Business Officer of Insilico Medicine, in the company's announcement. Feng Ren, Co-CEO and Chief Scientific Officer, noted that the candidate targets a protein designated as "Target Y" involved in pathological inflammation, with a mechanism potentially applicable beyond ophthalmology to neurodegenerative diseases, inflammatory bowel disease, cardiovascular disease, metabolic dysfunction-associated steatohepatitis (MASH), and obesity.

Pipeline Productivity Signals AI Maturation

Insilico's track record provides benchmark data for AI-driven drug discovery productivity: of the 32 preclinical candidates nominated since 2021, 13 have received Investigational New Drug (IND) approval or clearance from regulatory authorities. This translation rate offers evidence of generative AI's capacity to produce clinically viable therapeutic candidates at scale.

For patients with dry AMD, uveitis, and dry eye disease, ISM9077 represents a potential treatment option that combines superior efficacy with convenient eye-drop administration. If successfully developed through clinical trials, the compound could improve quality of life for millions affected by these blinding ocular conditions while addressing current limitations in toxicity, efficacy, and dosing adherence that characterize existing therapies.

Sources

1. Insilico Medicine nominates AI-designed candidate for ocular diseases — Drug Target Review