Mifepristone reduces HbA1c in adults with T2D, hypercortisolism

Treatment with mifepristone significantly reduced HbA_1c levels in patients with difficult-to-control type 2 diabetes (T2D) and hypercortisolism compared with placebo, according to the final results of the CATALYST trial presented at ADA 2025.

In the part 2 treatment phase, 136 patients with inadequately controlled T2D and hypercortisolism (defined as a cortisol level of >1.8 µg/dL after dexamethasone suppression test) were randomized in a 2:1 ratio to receive either mifepristone* (n=91) or placebo (n=45) for 24 weeks. The mean HbA_1c levels were 8.62 and 8.41 percent, respectively, at baseline.

By week 24, patients receiving mifepristone had a statistically significant reduction in HbA_1c compared with those receiving placebo (-1.47 percent vs -0.15 percent), corresponding to a difference of -1.32 percent between the groups (p<0.001). [ADA 2025, abstract CT-SY28-1]

Prof Vivian Fonseca from Tulane University, School of Medicine in New Orleans, Louisiana, US, who presented the results of the study, highlighted that the HbA_1c reductions were achieved despite that more patients in the mifepristone group had reduced or discontinued their glucose-lowering medications, such as long-acting (49.2 percent vs 13 percent) and fast-acting (30.3 percent vs 10.5 percent) insulins and sulfonylureas (22.2 percent vs 10.5 percent), by week 12 than those in the placebo group.

In addition, treatment with mifepristone showed similar HbA1c reductions in patients with (-1.26 percent vs 0.15 percent; difference of -1.41 percent; p=0.005) and without (-1.58 percent vs -0.29 percent; difference of -1.29 percent; p<0.001) adrenal imaging abnormalities, compared with placebo.

“This is the first randomized, placebo-controlled study to show that hypercortisolism without an adrenal imaging abnormality responds to cortisol-directed pharmacological therapy. I think this is a very new paradigm in treatment,” said Fonseca.

Overall, the CATALYST trial met its primary endpoint, demonstrating a significant HbA_1c reduction at week 24 with mifepristone compared with placebo in patients with difficult-to-control T2D and hypercortisolism, according to the researchers. [Diabetes Care 2025;doi:10.2337/dc25-1055]

Other key secondary endpoints

At week 24, patients treated with mifepristone lost 4.4 kg in body weight, whereas those treated with placebo gained 0.7 kg (difference of -5.1 kg; p=0.001).

As a result, those taking mifepristone achieved significantly reduced waist circumference (-5.2 vs -0.1 cm; difference of -5.1 cm; p=0.002) and body mass index ([BMI]; -1.5 vs 0.3 kg/m²; difference of -1.7 kg/m²; p=0.001) than those taking placebo.

“Waist circumference decreased along with body weight, suggesting loss of primarily visceral fat mass … and BMI decreases continued at study end,” Fonseca noted.

“These findings demonstrate a potentially promising treatment solution for patients with difficult-to-treat T2D and hypercortisolism, a patient population frustrated with poor results of their diabetes care,” said Dr John Buse from the University of North Carolina School of Medicine, North Carolina, US, one of the authors of the study. “By treating the root cause—excess cortisol—we saw meaningful improvements in both blood glucose and weight which could be game-changing when it comes to managing T2D.”

Safety endpoints

Serious treatment-emergent adverse events (TEAEs) occurred more frequently with mifepristone than with placebo (31.9 percent vs 4.7 percent), which led to higher treatment discontinuation rates (28.6 percent vs 2.3 percent).

The most common TEAEs reported with mifepristone were hypokalaemia (29.7 percent), fatigue (20.9 percent), nausea (20.9 percent), and vomiting (15.4 percent).

These TEAEs were mostly mild-to-moderate in severity, manageable, and consistent with mifepristone’s known safety profile, noted Fonseca.

Takeaways

“Overall, in a cohort with difficult-to-control T2D and hypercortisolism, mifepristone resulted in clinically and statistically significant improvements in HbA_1c and other comorbidities, with comparable reductions in HbA_1c observed in participants with and without adrenal imaging abnormalities,” said Fonseca.