Insilico Medicine's AI-Discovered Drug Rentosertib Enters Phase III Trials for IPF
# Insilico Medicine Advances AI Drug for IPF to Phase III Trials
## Breakthrough in AI-driven drug testing moves forward as Insilico Medicine targets severe lung disease
Insilico Medicine has progressed to Phase III human trials for its AI-discovered drug, rentosertib, aimed at treating idiopathic pulmonary fibrosis (IPF). This advancement represents a significant milestone within the computational drug discovery arena, marking the transition from initial safety tests to late-stage efficacy validation.
IPF is a devastating condition that severely scars lung tissue, ultimately jeopardizing patients' ability to breathe. Patients diagnosed with this illness generally have a median survival rate of only two to four years. Rentosertib works by inhibiting TRAF2- and NCK-interacting kinase, addressing critical mechanisms underlying the disease when taken orally.
In a randomized trial, 71 participants were recruited from 22 clinical sites across China. These patients were divided into two groups: one receiving a placebo and the other receiving varying doses of the drug at 30 mg or 60 mg daily over a 12-week observation period.
Those on the 60 mg daily dose showcased significant improvement, achieving a mean forced vital capacity gain of +98.4 mL, while the placebo group experienced an average loss of 20.3 mL in lung capacity. The safety profile was satisfactory, with adverse events aligning closely with anticipated baseline rates across all trial groups. The U.S. Food and Drug Administration (FDA) granted the drug ‘Orphan Drug Designation’ in February 2023.
### Computational Discovery and Target Prioritization
The drug's development hinges on Insilico Medicine’s proprietary platform, Pharma.AI. This computational system employs distinct engines for various biological and chemical engineering tasks.
PandaOmics, the engine for initial target identification, processes extensive biological datasets, including genomics, clinical trial data, academic research, and patent information, to create comprehensive biological network models. Through causal inference algorithms, it exposes novel disease connections hidden within vast data frameworks.
PandaOmics identified TNIK as a key biological target for IPF. Notably, it avoided existing antifibrotic medication pathways involving receptor tyrosine kinases.
The software mapped TNIK as a pivotal regulator of fibrosis and inflammation, influenced by signaling channels such as Wnt, TGF-?, Hippo/YAP-TAZ, JNK, and NF-?B. The target selection process utilized a framework focusing on the hallmarks of aging, scoring targets based on their involvement in aging mechanisms, chronic inflammation, and extracellular matrix remodelling.
Feng Ren, PhD, Co-CEO and Chief Scientific Officer of Insilico Medicine, stated: “IPF is one of the clearest clinical examples of an age-related disease in which fibrosis, chronic inflammation, extracellular matrix remodeling, and cellular senescence intersect.”
### Innovative Molecular Engineering Approach
Upon selecting a target, the Chemistry42 engine employs generative molecular design, marking a departure from traditional high-throughput screening methods. Instead of searching through existing compound libraries, Chemistry42 utilizes Generative Tensorial Reinforcement Learning to develop molecules compatible with the target protein pocket. This process ensures a balance between structural fit and necessary pharmacological properties.
During this phase, 79 molecules were synthesized, and the 55th iteration was chosen for preclinical testing. This focused approach enabled the timeline from initial project to preclinical candidate selection to be condensed to just 18 months.
The methodology has its roots in a 2019 publication of Insilico Medicine’s GENTRL approach in Nature Biotechnology, which created reproducible systems that streamline molecular generation, sidestepping the costly trial-and-error methods typical in conventional pharmaceutical chemistry.
### Assessing Biological Impact Through Proteomics
The clinical evaluation incorporates complex proteomic analysis to confirm the biological interactions predicted by AI algorithms. Insilico Medicine utilizes its internal proteomic aging-clock frameworks in the IPF trial to capture exploratory geroscience data.
Several chronological-age proteomic clocks, such as ProtAge and OrganAgechrono, measure biological-age alterations due to treatment. The research team employs UK Biobank age-associated trajectories as external comparison datasets to contextualize treatment-responsive proteins within broader population data.
Additionally, mortality-risk-related proteomic clocks like PAC offer alternative analytical streams alongside standard clinical endpoints. The teams conduct analyses such as SenMayo and CellAge signature assessments to explore senescence and senescence-associated secretory phenotype biology in cellular models.
Peer-reviewed research published in Aging and Disease validated that pharmacological inhibition of TNIK leads to senomorphic activity, resulting in clear reductions in indicators of extracellular matrix remodelling.
### Comprehensive Documentation of the Computational Pipeline
Rentosertib’s clinical journey demonstrates a documented and peer-reviewed data trail vital for substantiating AI capabilities within life sciences. Nature Biotechnology published a comprehensive overview of the entire discovery-to-clinic trajectory. This publication details the algorithmic prioritization of the TNIK target, the generative chemistry outcomes, preclinical efficacy data, and details on Phase I pharmacokinetics.
The Journal of Medicinal Chemistry published findings on structural biology validation, highlighting the discovery of novel TNIK inhibitor chemotypes and the supporting structural evidence via the TNIK kinase domain co-crystal structure. Nature Medicine presented data on Phase IIa safety and lung-function outcomes, confirming the computational predictions.
Alex Zhavoronkov, PhD, Founder and CEO of Insilico Medicine, remarked: “Rentosertib is a defining program for Insilico because it represents the full arc of our mission: using AI not only to move faster, but to originate new biology, new chemistry, and new therapeutic opportunities.”
The advancement through Phase III trial underscores the necessity for verifiable data regarding human outcomes in the field of AI biopharma.
