In the past week, two announcements from the drug discovery world have had Insilico Medicine’s generative artificial intelligence (AI) target discovery engine, PandaOmics, front and center.

The first involved the use of PandaOmics to identify actionable drug targets for the lysosomal storage disease cystinosis and to validate them in preclinical models of the disease. This was work done by the Mechanisms of Inherited Kidney Disorders (MIKADO) group at the University of Zurich. The research collaboration between Insilico and MIKADO was first announced in March 2022.

These results, which open new therapeutic possibilities for this disease, were published in Nature Communications in the article, “Lysosomal cystine export regulates mTORC1 signaling to guide kidney epithelial cell fate specialization.

Cystinosis is a rare genetic disease (affecting 1 in 100,000 to 200,000 newborns worldwide) characterized by inappropriate storage of cystine in cells. Nephropathic cystinosis is the most common and severe form of the disease, typically emerging in infancy. The relentless accumulation of cystine slowly destroys the body’s organs, causing kidney failure, diabetes, hypothyroidism, myopathy, and central nervous system deterioration. There are currently no curative treatments for children with cystinosis.

Amino acid transporters play a vital role in facilitating the movement of essential nutrients across the membrane of lysosomes. Cystinosis is caused by mutations that invalidate the cystinosin (CTNS) transporter, causing cystine to accumulate, and driving lysosomal storage disease. Scientists suspected that the mechanism responsible for the cell damage was connected to the regulation of rapamycin complex 1 (mTORC1).

In the study, the researchers used the PandaOmics platform as a novel way to prioritize disease-target associations and prioritize actionable (drug) targets in cystinosis cells. The AI-based analysis predicted that mTOR leads a ranked list of actionable drug targets.

The study’s findings revealed a causal association between the regulation of mTORC1 and the disease. They indicate that hyperactive mTOR signaling drives the dysfunction of kidney tubular cells and is a targetable pathway in cystinosis.

Alessandro Luciani, PhD external principal investigator in Drug Discovery Strategy for Rare Diseases at University of Zurich, explains: “Our research showed that cystine storage stimulates the activation of the mTORC1 protein, leading to the impairment of kidney tubular cell differentiation and function.”

Tested on cellular systems and animal models, the researchers found that treatment with the Food and Drug Administration (FDA)-approved therapeutic rapamycin restores the degradative activities of the lysosomes and ameliorates the dysfunction of kidney tubular cells—the earliest telltale manifestation of the disease. These results identify mechanisms and therapeutic targets for dysregulated homeostasis in cystinosis.

“Cystinosis is a commonly neglected disease with a large unmet need. With the power of artificial intelligence-driven, systems biology-based drug discovery, we have unlocked new understanding of cystinosis disease and accelerated the discovery of actionable drug targets, with the goal of bringing novel breakthrough medicines to patients,” said Olivier Devuyst, MD, head of MIKADO group at the UZH.

Inhaled AI drugs

Just a few days after the publication about cystinosis, Insilico Medicine announced an inhalable formula, and new device, for its AI-designed drug for idiopathic pulmonary fibrosis (IPF.)

The company is nominating the preclinical candidate of an inhalation solution for INS018_055, the first AI-powered anti-fibrotic small molecule inhibitor developed leveraging its proprietary AI drug discovery platform Pharma.AI for the treatment of IPF.

In February 2021, Insilico nominated INS018_055 as a preclinical candidate for the treatment of IPF, a chronic lung disease that results in progressive and irreversible decline in lung function, and initiated a first-in-human study in November 2021. The FDA granted Orphan Drug Designation to INS018_055 for the treatment of IPF in February 2023. INS018_055 is currently being assessed in international multicenter Phase II trials, with the first patients dosed in June 2023.

Inhalation offers advantages including rapid action onset, high bioavailability, lowered effective dose, reduced side effects, and direct delivery to the lungs.

The new solution has proven to be well tolerated in preclinical studies with good anti-fibrotic and anti-inflammatory efficacy. 

“We continue to be committed to providing new treatment options for patients with this devastating lung disease,” said Alex Zhavoronkov, PhD, founder and CEO of Insilico Medicine. “We hope that this new formulation will provide another path for patients in need and further demonstrate the capabilities of our AI drug discovery platform.” 

Inhalation solutions deliver treatment as vapor or aerosol through special devices into the respiratory tract, which is considered one of the most efficient routes of drug delivery as it may provide advantages including rapid action onset, high bioavailability, lowered effective dose. and reduced side effects with noninvasive and targeted route of administration.

In preclinical studies, the inhalation solution of INS018_055 was able to achieve higher lung exposure with low systemic exposure and demonstrated antifibrotic and anti-inflammatory efficacy in animal models. Moreover, it showed good pharmacokinetic (PK) and safety profiles, as well as good stability and solubility.

“We’re encouraged by the initial studies of this new formulation and look forward to advancing it,” said Feng Ren, PhD, Co-CEO and CSO of Insilico Medicine. “We believe this lead AI-discovered and AI-designed drug can be explored in different formulations to offer as many treatment options as possible to patients with this terrible condition.”

Previous articleAI Helps to Unveil the Genetic Basis of Human Skeletal Proportions
Next articleTackling the “Don’t Eat Me” Signal in Brain Cancer