Insilico Medicine AI Drug Discovery: Full Timeline

Insilico Medicine used its own AI to pick a new fibrosis target and design a drug against it, and that drug, rentosertib, has now been through a randomized trial. Here is the full record, graded by how far each result has been proven.

Presented By Our Partners

The short answer: Insilico Medicine builds its own AI platform, Pharma.AI, with PandaOmics for target discovery and Chemistry42 for molecule design. Its lead drug, rentosertib, hits TNIK, a fibrosis target its AI nominated, and a 71-patient randomized Phase 2a trial in Nature Medicine found it safe enough to keep testing, with a lung function signal at the top dose. That makes it one of the most advanced tests of a drug whose target and molecule both came from AI, but the trial was small, lasted 12 weeks and had safety, not efficacy, as its primary endpoint. Commercially, Insilico listed in Hong Kong in 2025 and has signed deals with Eli Lilly.

Tier: 2, AI science lab
Own models: Yes. PandaOmics, Chemistry42, inClinico and the nach0 foundation model (PMID 38404138, PMID 38846388)
Own wet lab: Yes. It runs a robotics lab for AI-driven drug discovery, as reported by Medical Design and Development
First health move: May 2016, deep neural networks that estimate age from a routine blood test (PMID 27191382)
Furthest result so far: Tested in people, Phase 2 (rentosertib Phase 2a, peer-reviewed; primary endpoint safety)
Tier 1 partners: None. We found no funding or partnership with OpenAI, Anthropic, Google DeepMind, Microsoft or Meta. NVIDIA researchers co-authored its nach0 model paper (2024), and it has used Google DeepMind’s public AlphaFold structures in published work

What Is Insilico Medicine?

Insilico Medicine is an AI drug discovery company that both develops its own medicines and licenses its software and drug candidates. Author affiliations on its papers place teams in Hong Kong, Shanghai, Cambridge in Massachusetts, Montreal and Abu Dhabi (PMID 38459338, PMID 40461817). Insilico appointed Feng Ren, a former GSK executive, as co-CEO (Insilico via PR Newswire, Endpoints News), and Alex Zhavoronkov is the corresponding author on most of its papers.

Its platform, Pharma.AI, includes PandaOmics for target and biomarker discovery, Chemistry42 for generating new small molecules and inClinico for forecasting Phase 2 trial outcomes (PMID 38404138). The company’s best-known output is rentosertib, a TNIK inhibitor for idiopathic pulmonary fibrosis whose target and molecule both came from these tools (PMID 38459338).

Insilico raised a $110 million Series E in March 2025 (HLTH) and then listed on the Hong Kong Stock Exchange in what Longevity.Technology called 2025’s largest biotech IPO. PharmCube reported that the offering was backed by Eli Lilly and Tencent, and Nasdaq described it as the first AI-driven biotech to list under Chapter 8.05 of the exchange’s rules. Funding and a listing show investor demand; they are not evidence that its drugs work.

Insilico Medicine Healthcare Timeline: Every Move Since 2016

Entries are newest first, back to Insilico’s earliest peer-reviewed AI health work we could find. Result rows carry an evidence level from our fixed scale; deals, funding and appointments are marked Not a result.

Date	Move	Type	Evidence level	Source
Sep 2026	Nature Biotechnology analysis of serum from the rentosertib Phase 2a trial: six proteomic aging clocks all predicted lower biological age in treated arms.	Result	Peer-reviewed, Phase 2 (exploratory biomarker analysis)	PMID 42706338
Apr 2026	Formed a Longevity Board chaired by Eli Lilly’s Andrew Adams, with Nobel laureate Michael Levitt among its members.	Hire	Not a result	Insilico via AlphaGalileo
Nov 2025	Entered a research and licensing collaboration with Eli Lilly, reported as a $100 million deal.	Partnership	Not a result	Drug Discovery Online
Jun 2025	Rentosertib Phase 2a results published in Nature Medicine: 71 patients with idiopathic pulmonary fibrosis, 12 weeks, safety primary endpoint similar across arms, lung function signal at the highest dose.	Result	Tested in people, Phase 2	PMID 40461817
May 2025	ISM5939, an oral ENPP1 inhibitor designed with generative AI, slowed tumor growth in mice alongside PD-1/PD-L1 drugs and chemotherapy.	Result	Peer-reviewed, Preclinical	PMID 40410143
Mar 2025	Raised a $110 million Series E.	Funding	Not a result	HLTH
May 2024	nach0, a foundation model for chemistry and biomedical language tasks, published with NVIDIA co-authors.	Model	Peer-reviewed, Research tool	PMID 38846388
Mar 2024	Nature Biotechnology paper: AI nominated TNIK as a fibrosis target and generated INS018_055 (now rentosertib); Phase 1 in 78 healthy participants; about 18 months from target discovery to preclinical candidate.	Result	Tested in people, Phase 1	PMID 38459338
Feb 2024	PandaOmics, its AI platform for target and biomarker discovery and part of the Pharma.AI suite with Chemistry42, described in a peer-reviewed paper.	Model	Peer-reviewed, Target discovery	PMID 38404138
Sep 2023	Licensed exclusive global rights to ISM3091, a USP1 inhibitor, to Exelixis, reported as an $80 million deal.	Partnership	Not a result	Exelixis via Business Wire
Aug 2023	inClinico, a transformer-based platform to predict Phase 2 trial outcomes, described with validation run by the company, partly with drug companies and financial institutions.	Model	Peer-reviewed, Research tool	PMID 37483175
Jan 2023	Used AlphaFold-predicted protein structures with PandaOmics and Chemistry42 to find a CDK20 inhibitor hit for liver cancer within 30 days after synthesizing 7 compounds.	Result	Peer-reviewed, Molecule design	PMID 36794205
Sep 2019	GENTRL generative model designed DDR1 kinase inhibitors in 21 days; one lead showed favorable pharmacokinetics in mice.	Result	Peer-reviewed, Molecule design	PMID 31477924
Jun 2016	Deep neural networks classified drugs into therapeutic categories from their transcriptional profiles.	Result	Peer-reviewed, Research tool	PMID 27200455
May 2016	Deep neural networks predicted a person’s age from a routine blood test, an early AI aging biomarker.	Result	Peer-reviewed, Research tool	PMID 27191382

Insilico Medicine’s Biggest Results, and How Far Each Has Been Proven

Insilico has published more peer-reviewed results than most AI drug discovery companies, nearly all written by its own staff. The rentosertib trial is the one that matters most, so we grade it in detail.

Rentosertib Phase 2a in idiopathic pulmonary fibrosis. Tested in people, Phase 2 (Phase 2a, peer-reviewed). In a double-blind, placebo-controlled trial, 71 patients received 30 mg once daily (18), 30 mg twice daily (18), 60 mg once daily (18) or placebo (17) for 12 weeks (PMID 40461817). The primary endpoint was safety: the share of patients with at least one treatment-emergent adverse event was 72.2%, 83.3% and 83.3% on the three doses versus 70.6% on placebo, and the most common reasons for stopping treatment were liver toxicity or diarrhea. On the secondary endpoint of forced vital capacity, the 60 mg group gained a mean 98.4 mL (95% confidence interval 10.9 to 185.9) while the placebo group lost 20.3 mL (95% confidence interval minus 116.1 to 75.6). Those intervals overlap, the arms were tiny, every investigator site listed is in China, and Insilico staff co-authored the paper. The authors themselves say the drug warrants larger and longer trials. What would prove it next: a larger, longer, multinational trial with lung function as the primary endpoint, then independent replication.

TNIK target and Phase 1 safety. Tested in people, Phase 1 (peer-reviewed). The 2024 Nature Biotechnology paper reports that AI nominated TNIK as an anti-fibrotic target, that INS018_055 showed anti-fibrotic activity in several animal models, and that a randomized Phase 1 trial in 78 healthy participants supported its safety and pharmacokinetics, with a separate Phase 1 in China showing a comparable profile (PMID 38459338). The roughly 18 months from target discovery to preclinical candidate is a company-reported timeline. What would prove it next: other groups confirming TNIK as a fibrosis driver.

Aging clocks in the rentosertib trial. Peer-reviewed, Phase 2 (exploratory biomarker analysis). A September 2026 Nature Biotechnology paper applied six proteomic aging clocks to serum from the Phase 2a trial and found all six predicted lower biological age in treated arms (PMID 42706338). The authors state that the clocks alone cannot separate effects on aging from effects on the disease. What would prove it next: prespecified aging endpoints in a larger trial.

Featured Partner

Invest in the Infrastructure Behind Modern Medicine

As healthcare expands beyond hospital walls, the buildings and campuses supporting that shift are generating compelling returns for investors who move early. The Healthcare Real Estate Fund offers qualified investors direct access to a curated portfolio of medical office, outpatient, and specialty care facilities.

Learn More →

GENTRL and DDR1 inhibitors. Peer-reviewed, Molecule design. In 2019 Insilico reported that its GENTRL generative model produced DDR1 kinase inhibitors in 21 days; four compounds were active in biochemical assays, two worked in cells and one lead had favorable pharmacokinetics in mice (PMID 31477924). It showed speed, not a medicine. What would prove it next: a GENTRL-derived compound entering human trials.

ISM5939 for solid tumors. Peer-reviewed, Preclinical. ISM5939, an oral ENPP1 inhibitor designed with Insilico’s generative chemistry, worked with PD-1/PD-L1 drugs and chemotherapy to suppress tumor growth in mouse models with good tolerance (PMID 40410143). What would prove it next: published human trial data.

Who Insilico Medicine Works With

Eli Lilly is Insilico’s deepest pharma relationship. Lilly entered a research and licensing collaboration with Insilico in November 2025 (Drug Discovery Online), reported as a $100 million deal (The Pharma Letter). Lilly has also struck a deal with Insilico reported at up to $2.75 billion (Wealth Professional). Lilly also backed the Hong Kong listing, and Lilly’s Andrew Adams chairs the Longevity Board that Insilico formed in April 2026, which includes Nobel laureate Michael Levitt (Insilico via AlphaGalileo, R&D World).

Insilico has also licensed out drug candidates. Exelixis took exclusive global rights to ISM3091, a USP1 inhibitor, in September 2023 (Exelixis) in what The Fly reported as an $80 million deal, and the Menarini Group licensed a KAT6 inhibitor for breast cancer and other cancers (Menarini and Insilico) in a deal BioSpectrum Asia valued at $500 million. Most of the value in deals like these is milestone payments that depend on future trial success. On the science side, the rentosertib Phase 2a was run with investigators at more than a dozen hospitals in China (PMID 40461817), and its aging clock analysis added co-authors from the Broad Institute and Massachusetts General Hospital (PMID 42706338).

Tier 1 ties: none. We found no funding or partnership with OpenAI, Anthropic, Google DeepMind, Microsoft or Meta. NVIDIA researchers co-authored the nach0 foundation model paper (PMID 38846388), which shows a research collaboration, not an investment. Insilico also used Google DeepMind’s publicly released AlphaFold structures to find a CDK20 inhibitor hit (PMID 36794205), which is use of a public tool, not a partnership.

What to Watch Next

These are the measurable events that would move Insilico’s evidence level.

A larger, longer trial of rentosertib in idiopathic pulmonary fibrosis with lung function as a primary endpoint, including sites outside China.
Peer-reviewed patient data for a second Insilico drug.
Independent replication of TNIK as a fibrosis target by a group with no tie to Insilico.
Named candidates, milestone payments or trial starts under the Lilly, Exelixis and Menarini agreements.
First human trial data for ISM5939.

Related: see how every lab is graded in our AI healthcare breakthroughs tracker, our full analysis of the rentosertib clinical verdict, the Insilico Medicine Longevity Board, where AI-designed drugs stand in our status report on AI-designed drugs in human trials and the contrasting record of Recursion.

What This Means for You

If you are a patient with idiopathic pulmonary fibrosis, rentosertib is not approved and cannot be prescribed. The Phase 2a result is encouraging but early, and approved antifibrotic drugs remain the standard of care. If you are interested in trials, ask your pulmonologist and check ClinicalTrials.gov.

If you work in research or care, read the rentosertib paper for what it is: a well-run safety trial with a promising secondary signal, not proof of efficacy. Insilico’s published tools are useful to study, but almost every result so far comes from company authors, so independent replication is the next bar to watch. For investors, the Lilly deals and the Hong Kong listing reflect expectations that one 71-patient trial cannot yet support on its own.

Compiled by Healthcare Discovery from company announcements, named news reports and PubMed. Every PMID on this page was checked against PubMed on October 28, 2026. Evidence levels follow the scale explained in our AI healthcare breakthroughs tracker. Healthcare Discovery uses AI research tools from several companies on this page, including OpenAI, Anthropic and Google models; no company listed reviewed or paid for this page.

What is Insilico Medicine?

Insilico Medicine is an AI drug discovery company that builds its own platform, Pharma.AI, including PandaOmics for finding drug targets and Chemistry42 for designing molecules. It develops its own drugs, licenses software and drug candidates to partners such as Eli Lilly, Exelixis and Menarini, and listed on the Hong Kong Stock Exchange in 2025.

What is rentosertib?

Rentosertib, formerly ISM001-055 and INS018_055, is a small-molecule TNIK inhibitor for idiopathic pulmonary fibrosis. Insilico’s AI nominated TNIK as the target and generated the molecule. It is one of the most advanced tests of a drug whose target and structure both came from AI.

Did rentosertib work in its Phase 2a trial?

It passed a safety test and showed an early efficacy signal, but it has not been proven to work. In a 71-patient, 12-week randomized trial published in Nature Medicine in 2025, the primary endpoint was safety, and adverse event rates were similar across arms. Forced vital capacity rose by 98.4 mL at the highest dose versus a 20.3 mL fall on placebo, a secondary endpoint that the authors said warrants larger, longer trials.

What are PandaOmics and Chemistry42?

They are Insilico’s in-house AI tools. PandaOmics analyzes omics and biomedical text data to propose drug targets and biomarkers. Chemistry42 is its generative chemistry platform that designs new small molecules. Both are part of the Pharma.AI suite and are available through licensing or collaboration.

Who is on Insilico’s Longevity Board?

Insilico formed a Longevity Board in April 2026, chaired by Andrew Adams of Eli Lilly, with Nobel laureate Michael Levitt among its members. The board is an advisory and strategy move, not a clinical result.

Has rentosertib been approved?

No. As of October 2026 its most advanced published data come from a 71-patient Phase 2a trial, and the authors called for larger, longer trials before any conclusion on efficacy.

Sources

Xu Z, et al. A generative AI-discovered TNIK inhibitor for idiopathic pulmonary fibrosis: a randomized phase 2a trial. Nature Medicine, 2025. PMID 40461817
Zhavoronkov A, et al. Integration of proteomic aging clocks in a phase 2a clinical trial supports simultaneous geroprotective assessment. Nature Biotechnology, 2026. PMID 42706338
Ren F, et al. A small-molecule TNIK inhibitor targets fibrosis in preclinical and clinical models. Nature Biotechnology, 2024. PMID 38459338
Pu C, et al. Oral ENPP1 inhibitor designed using generative AI as next generation STING modulator for solid tumors. Nature Communications, 2025. PMID 40410143
Livne M, et al. nach0: multimodal natural and chemical languages foundation model. Chemical Science, 2024. PMID 38846388
Kamya P, et al. PandaOmics: An AI-Driven Platform for Therapeutic Target and Biomarker Discovery. Journal of Chemical Information and Modeling, 2024. PMID 38404138
Aliper A, et al. Prediction of Clinical Trials Outcomes Based on Target Choice and Clinical Trial Design with Multi-Modal Artificial Intelligence. Clinical Pharmacology and Therapeutics, 2023. PMID 37483175
Ren F, et al. AlphaFold accelerates artificial intelligence powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor. Chemical Science, 2023. PMID 36794205
Zhavoronkov A, et al. Deep learning enables rapid identification of potent DDR1 kinase inhibitors. Nature Biotechnology, 2019. PMID 31477924
Aliper A, et al. Deep Learning Applications for Predicting Pharmacological Properties of Drugs and Drug Repurposing Using Transcriptomic Data. Molecular Pharmaceutics, 2016. PMID 27200455
Lane E, et al. Deep biomarkers of human aging: Application of deep neural networks to biomarker development. Aging, 2016. PMID 27191382
Insilico via AlphaGalileo. Insilico Medicine announces industry’s first longevity board to accelerate AI-driven aging research for drug discovery, April 2026. AlphaGalileo
R&D World. Insilico launches industry’s first longevity board with Lilly’s Andrew Adams as chair, Nobel laureate Michael Levitt on roster. R&D World
Drug Discovery Online. Insilico, Lilly enter research and licensing collaboration to advance AI-driven drug discovery, November 2025. Drug Discovery Online
The Pharma Letter. Insilico enters $100 million AI deal with Lilly, November 2025. The Pharma Letter
Wealth Professional. Eli Lilly strikes up to $2.75bn deal with Insilico to accelerate push into AI-developed medicines. Wealth Professional
Longevity.Technology. Insilico Medicine lists on Hong Kong Stock Exchange in 2025’s largest biotech IPO, 2025. Longevity.Technology
PharmCube. Insilico Medicine Launches HKEX IPO Backed by Lilly, Tencent. PharmCube
Nasdaq. Insilico Medicine Debuts On HKEX As First AI-Driven Biotech Under Chapter 8.05. Nasdaq
HLTH. Insilico Medicine raises $110M in Series E funding to advance AI-driven drug discovery, March 2025. HLTH
Exelixis via Business Wire. Exelixis and Insilico Medicine Enter into Exclusive Global License Agreement for ISM3091, a Potentially Best-in-Class USP1 Inhibitor, September 2023. Business Wire
The Fly via TipRanks. Exelixis in $80M deal with Insilico for global rights to develop ISM3091, September 2023. TipRanks
Menarini Group and Insilico via BioSpace. Menarini Group and Insilico Medicine enter global exclusive license agreement for novel KAT6 inhibitor for potential breast cancer treatment and other oncology indications. BioSpace
BioSpectrum Asia. Menarini Group and Insilico Medicine ink oncology deal worth $500 M. BioSpectrum Asia
Insilico via PR Newswire. Insilico Medicine appoints Dr. Feng Ren as co-CEO. PR Newswire
Endpoints News. As their first clinical data test looms, global AI player Insilico bumps up a GSK vet to co-CEO in charge of development. Endpoints News
Medical Design and Development. Insilico Launches Robotics Lab to Accelerate Its AI-Driven Drug Discovery. Medical Design and Development

Free Daily Briefing

The Latest Longevity Science.
Delivered Every Morning.

Join researchers, physicians, and health professionals getting daily breakthroughs in AI-driven medicine, epigenetics, and longevity research.

Support the research that powers this editorial

No spam. Unsubscribe anytime. We respect your inbox.