Insilico Medicine reaches Phase 2a clinical milestone with generative AI-designed TNIK inhibitor for IPF in under 30 months
“Insilico Medicine reaches Phase 2a clinical milestone with generative AI-designed TNIK inhibitor for IPF in under 30 months” documents a Drug Discovery & Development deployment in Pharmaceutical & Life Science at Insilico Medicine. communities.springernature.com reports time to phase 1 clinical success: Under 30 months; this directory has not independently verified that result.
Evidence at a glance
- Evidence status:
- Automated evidence gate passed
- Deployment timeframe:
- Not reported by source
- Reported outcome metrics:
- 3 cited below
- Directory entry published:
- Source link checked:
The source-link check confirms reachability, not independent re-verification of every claim.
Source-reported figures — cited source: communities.springernature.com
The Challenge
Traditional drug discovery takes more than 10 years and costs over $2 billion, making it slow and financially risky. Idiopathic pulmonary fibrosis (IPF) is a debilitating, age-related lung disease with no curative treatment and significant unmet clinical need. Identifying novel, druggable targets and designing effective small molecules through conventional methods remained prohibitively time-intensive.
The Solution
Insilico Medicine deployed its proprietary generative AI platform, Pharma.AI, to accelerate the full drug discovery pipeline. The Biology AI module identified TNIK as a novel IPF target in 2019; the Chemistry AI module then aided medicinal chemists in designing ISM001-055, a small-molecule TNIK inhibitor. The compound achieved preclinical candidate nomination 18 months after initial AI-driven target identification and advanced through Phase 0 and two independent Phase 1 trials before entering Phase 2a.
Results
ISM001-055 completed the journey from AI-driven hypothesis to Phase 1 clinical success in under 30 months—a fraction of the industry average. In a randomized, double-blind, placebo-controlled Phase 2a trial enrolling 71 IPF patients across 21 sites in China, the drug met its primary endpoint of safety and tolerability at all dose levels. Secondary efficacy endpoints were also met, with a dose-dependent improvement in forced vital capacity (FVC) observed, with the 60 mg QD cohort showing the largest lung function improvement over 12 weeks.
Key Takeaways
- Generative AI can compress the target identification-to-clinical-candidate timeline from years to ~18 months when biology and chemistry AI modules are tightly integrated.
- A dose-dependent FVC response in IPF after only 12 weeks of treatment is clinically meaningful and validates AI-designed molecular mechanisms in a notoriously heterogeneous disease.
- End-to-end AI-native drug discovery (target ID → molecule design → clinical validation) is now demonstrably feasible, setting a precedent for the broader pharmaceutical industry.
Explore Related
Details
- Industry
- Pharmaceutical & Life Science
- Use Case
- Drug Discovery & Development
- AI Technology
- Large Language Models & Generative AI
- Company Size
- Startup
- Company
- Insilico Medicine
- Evidence status
- Automated evidence gate passed
- Deployment timeframe
- Not reported by source
- Directory entry published
- Source link checked
Cited source
communities.springernature.comHave a similar implementation?
Share your customer's AI results and link it to your vendor profile.
Submit a case study →