Insilico Medicine has dosed the first patient in a Phase II trial of its lead pipeline candidate INS018_055, a small molecule inhibitor treatment for idiopathic pulmonary fibrosis (IPF) that was discovered and designed using generative artificial intelligence (AI).
The Phase II trial consists of parallel studies—one launched in China in April, the other being launched in the U.S. after Insilico received FDA approval last month. Both are randomized, double-blind, placebo-controlled studies designed to assess the safety, tolerability, pharmacokinetics, and preliminary efficacy of oral INS018_055 12 weeks after treatment in participants with IPF who are divided into four parallel cohorts.
Following the cohort studies, Insilico plans to assess INS018_055 in larger populations by recruiting 60 participants with IPF at about 40 sites in both China and the U.S. The company expects to have topline results from the studies available next year.
Citing competitive reasons, Insilico has not disclosed the drug’s target, which the company has identified only as “Target X.” Alex Zhavoronkov, PhD, Insilico’s founder and CEO, shed some light on the target’s function, though not its identity, in an interview with GEN.
“We can say that this target is a regulator of at least three pathways implicated in fibrosis,” Zhavoronkov said.
The three, he said, are the yes-associated protein and transcriptional coactivator with PDZ-binding motif, known collectively as the YAP/TAZ pathway; the Wnt pathway (derived from wingless/integrated); and the transforming growth factor beta pathway (TGF-β).
Insilico won the FDA’s first Orphan Drug Designation for an AI drug last year. Based in Hong Kong, privately-held Insilico supports every one of its development programs through an autonomous robotics laboratory in Suzhou, China, that Zhavoronkov recently showcased for an exclusive report by GEN Edge.
Insilico reasons that by focusing on fibrosis-related pathways, it can develop INS018_055 for a broad range of fibrotic indications beyond IPF. One is skin fibrosis, a program now in the discovery phase.
Further along is the development of INS018_055 for kidney fibrosis, which is in the IND-enabling phase. Kidney fibrosis is the final manifestation of chronic kidney disease, with >10% of the world’s population (843.6 million people in 2017) living with a kidney disease that is driven by or associated with kidney fibrosis.
That makes kidney fibrosis a much larger and thus more lucrative market for Insilico than IPF, a rare chronic scarring lung disease characterized by progressive and irreversible decline in lung function, which affects approximately five million people worldwide.
However, IPF is also a potentially lucrative indication for Insilico since the top two current standard-of-care treatments are both blockbuster drugs that slow the progression of disease but do not halt or reverse lung fibrosis—and face patent-cliff loss of exclusivity. Last year, Boehringer Ingelheim’s OFEV® (nintedanib) generated net sales of €$3.227 billion ($3.5 billion), while Roche’s Esbriet® (pirfenidone) racked up CHF 718 million ($800 million).
While OFEV sales climbed 29.5% from €2.491 billion ($2.7 billion) in 2021, the drug will lose exclusivity next year. Esbriet is already facing generic competition, explaining why its sales dropped 31% from CHF 1.039 billion (about $1.2 billion) in 2021. The first generic versions reached the U.S. market in Q2 2022, aided by a U.S. District Court for the District of Delaware decision favoring generic drugmaker Sandoz.
“Both drugs are poorly tolerated by a substantial number of patients, and use is further limited by high cost and lack of perceived benefit on symptoms and quality of life,” observed last month in a commentary published in JAMA.
Insilico reasons that losses of exclusivity for nintedanib and pirfenidone will facilitate treatment combinations with INS018_055 as the main drug, to be followed by either nintedanib or pirfenidone. While Insilico evaluated potential combinations in preclinical studies, its clinical program is assessing INS018_055 individually, seeking to establish its own efficacy.
“We shall see”
“We have benchmarked against both nintedanib and pirfenidone, and demonstrated that our compound either outperformed or demonstrated synergetic effect. So together our drug with pirfenidone, they have helped IPF models in a pretty dramatic way,” Zhavoronkov said. “In patients, we shall see.”
“In the clinical trial, we are not combining it with anything,” he added. “We are confident that it’s going to outperform the standards of care. Otherwise, we wouldn’t be doing this. We have a completely different mechanism. Also, this mechanism is likely to be implicated in aging, so we actually want to demonstrate that in one way or another, it benefits older patients.”
Insilico faces a headwind in the likelihood of slow enrollment for its studies since IPF is a rare disorder, patients have to satisfy detailed criteria to enter the studies, and because several other drug developers are studying IPF treatments. [See IPF Candidates Crowd the Clinic, below]
Among them is Roche itself, which in February halted development of its zinpentraxin alfa (RG6354) in IPF after the recombinant form of pentraxin-2 (PTX-2), a naturally circulating human protein, failed the Phase III STARSCAPE trial (NCT04552899) in that indication. Roche inherited zinpentraxin alfa when it acquired Promedior in 2019 for up to $1.4 billion ($390 million upfront), and had planned to submit approval applications in 2024. Roche continues to develop zinpentraxin alfa for myelofibrosis, where it is in Phase II development and applications envisioned for 2025 and beyond.
However, Boehringer Ingelheim has fared better with its next-gen IPF candidate BI1015550, advancing it into the global Phase III FIBRONEER™ -IPF trial (NCT05321069) late last year after the phosphodiesterase 4B (PDE4B) inhibitor was shown to slow down the rate of lung function decline in IPF patients 12 weeks after treatment in a Phase II trial (NCT04419506), according to a study published in The New England Journal of Medicine.
INS018_055 is Insilico’s first wholly owned program in which AI was used to identify a novel target and generate novel small molecules through the company’s Pharma.AI platform.
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.
Three clinical candidates
INS018_055 is one of three Insilico candidates to have advanced into clinical trials. The others are ISM3312, an oral small molecule 3CLPro inhibitor designed to treat COVID-19; and ISM3091, a small molecule ubiquitin specific protease 1 (USP1) inhibitor being developed to treat BRCA-mutant breast cancer. ISM3312 is now in a Phase I trial and actively dosing patients, Zhavoronkov said.
Last month, the FDA approved Insilico’s IND application for ISM3091, making it the company’s first oncology program to advance into the clinic. Insilico expects in July to launch the U.S, portion of an open-label, multicenter Phase I trial of ISM3091 that will be conducted simultaneously in China. The trial will study the safety, tolerability, pharmacokinetics, and preliminary efficacy of ISM3091, and will include dose escalation and dose optimization studies designed to identify a recommended dose level for Phase II monotherapy.
Insilico reasons that USP1 is a suitable target because it plays a crucial role in DNA damage response and repair, which has led to its emergence as a new synthetic lethal target for cancer treatment. The first drugs approved to leverage synthetic lethality are poly (ADP-ribose) polymerase (PARP) inhibitors, which have shown positive clinical effects in many patients. However, between 40% and 70% of patients have developed resistance over time to PARP inhibitors, which has fueled interest in alternatives by Insilico and other drug developers.
One such developer, KSQ Therapeutics, launched the Phase I KSQ-4279-1101 trial last December, designed to assess the company’s USP1-targeting small molecule KSQ-4279 in patients with advanced solid tumors across multiple indications. KSQ-4279 is being studied both in combination with a PARP inhibitor, and in combination with chemotherapy.
Zhavorionkov estimates that KSQ is between nine months and a year ahead of Insilico—though its not the only USP1 developer: Debiopharm in March acquired for an undisclosed price global rights from Novo Nordisk to FT-3171 (renamed Debio 0432), which targets an undisclosed novel DNA damage repair pathway and is in late preclinical phases.
“KSQ Therapeutics is probably 9 months ahead of us, maybe up to a year, but our chemistry is absolutely novel,” Zhavoronkov said. He said that after Insilico applied PandaOMICs to prioritize the USP1 target, only to find competitors also looking at USP1, “We actually thought that it would be beneficial to have somebody de-risk the target and invest a lot of effort into biology. We’ve focused on chemistry, which is much, much more selective, much, much more potent than KSQ.”
ISM3091 is one of nine Insilico pipeline programs for oncology indications, making cancer the company’s largest therapeutic area.
The rest of Insilico’s preclinical oncology pipeline includes two discovery phase programs—a solid tumor program targeting cyclin dependent kinase 12 (CDK12); and a hepatocellular carcinoma (HCC) program targeting the transcription factor cellular MYC (cMYC)—and six programs in the IND-enabling phase.
Furthest along of the IND-enabling phase programs is a novel cancer immunotherapy for cold tumors that inhibits glutaminyl-peptide cyclotransferase-like protein (QPCTL), being co-developed with Fosun Pharma.
The IND-enabling programs also include a methionine adenosyltransferase 2α (MAT2A) inhibitor designed to treat MTAP-deficient cancer using a synthetic lethality mechanism; a KAT6A inhibitor to treat ER+/HER2- breast cancer; and three solid tumor programs—one each targeting ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1), diacylglycerol kinase alpha (DGKA), and the TEA domain (TEAD) transcription factor.
The TEAD program is of keen interest to Insilico because it combines a known target in TEAD with a pathway that has drawn increasing attention from researchers—YAP/TAZ, named for a pair of transcriptional coactivators, members of the Hippo signaling pathway which play a critical role in cell growth regulation, embryonic development, regeneration, proliferation, and cancer origin and progression.
“The level of novelty is lower, but the ability to partner this with another company is much higher,” Zhavoronkov said. “Our PandaOmics predicts this target to be extremely hot in the next five years. That means that the level of interest is likely to increase. It’s a preclinical candidate in the IND-enabling phase, which means that soon we will be ready to go clinical.”
“In the ideal world it should be this year,” Zhavoronko said. “In the not so ideal world, it would probably be early next year.”