AI-designed molecule may provide safer Cushing’s treatment

Insilico's ISM6200 targets gene, counteracts excess cortisol

Written by Margarida Maia, PhD |

A dropper hovers next to four vials of blood.

An artificial intelligence (AI)-designed small molecule that targets the NR3C1 gene may offer a safer, more effective treatment for Cushing’s syndrome by counteracting excess cortisol.

Insilico Medicine said it will focus on its molecule, ISM6200, as a promising new drug that may become a leading treatment for ovarian cancer, Cushing’s syndrome, obesity linked to high cortisol levels, and glaucoma.

“We started working on a portfolio of small molecule drugs targeting NR3C1 to help unlock the full potential of this target,” Alex Zhavoronkov, PhD, Insilico’s founder and co-CEO, said in a company press release. “The nomination of ISM6200 highlights the transformative power of generative AI in addressing complex medicinal chemistry challenges while reducing traditional risks.”

Cushing’s syndrome occurs when the body is exposed to high levels of cortisol, a hormone that helps control the response to stress. Cushing’s disease, a type of Cushing’s syndrome, is caused by a tumor in the brain’s pituitary gland. When cortisol remains elevated for too long, it can cause a range of symptoms, including weight gain, high blood pressure, and muscle weakness.

Researchers are looking for better ways to control cortisol’s effects in the body. One potential target is the NR3C1 gene, which encodes the glucocorticoid receptor, the switch that cortisol uses to affect cells. When cortisol binds to this receptor, it changes how the body responds to stress. However, in Cushing’s syndrome, this signaling pathway is overactive.

Recommended Reading
An illustration depicting a person with curly hair writing at a desk, with papers whirling in the air, as the banner image of

The Cushing’s disease symptoms that creep in when I’m off my meds

Engineered to reduce risk

ISM6200 is designed to block the receptor, reducing the effects of excess cortisol. It was developed using AI tools that help researchers design better compounds faster. Early studies suggest that ISM6200 is selective, targeting the glucocorticoid receptor without affecting other signaling pathways. This is important to avoid unwanted side effects and interactions with other medications.

“ISM6200 was specifically engineered with outstanding properties, including lower [drug–drug interaction] risks, which allow broader combination options,” said Feng Ren, PhD, Insilico’s co-CEO and chief scientific officer. “More importantly, the molecule can deliver strong efficacy at a much lower projected human dose, offering a safer and more effective treatment path for patients in need.”

ISM6200 demonstrated promising efficacy in treating Cushing’s syndrome in animal studies. It lowered insulin resistance, which occurs when the body does not respond well to insulin (the hormone that helps sugar enter cells from the bloodstream), leading to high blood sugar. After a short treatment period, insulin resistance was reduced, and blood pressure improved toward normal levels.

ISM6200 may also be useful to treat other diseases. Relacorilant, which also blocks NR3C1, can make cancer cells more sensitive to chemotherapy, potentially helping standard treatments work better. It may also be used to treat other conditions associated with excess cortisol, such as obesity or glaucoma, a group of eye diseases.

“I am especially encouraged by multi-purpose candidates like this one, as it is indicated in systems across the body, even exceeding our expectation of targeting both aging and certain diseases,” Zhavoronkov said. “By using AI, we can identify high-quality molecules with broad potential, in record time, and we are advancing more pipelines with AI-driven efficiency.”