Diazoxide Targets High Insulin to Tackle Fatty Liver Disease
A Phase 1 trial tests whether lowering compensatory hyperinsulinemia with diazoxide improves glucose and lipid metabolism in NAFLD patients.
Summary
Researchers at Columbia University ran a completed Phase 1 trial testing diazoxide — a drug that suppresses insulin secretion — in adults with overweight or obesity, insulin resistance, and non-alcoholic fatty liver disease (NAFLD). The trial compared two doses of diazoxide (1 mg/kg and 2 mg/kg) against placebo over 14 days. The core hypothesis is that chronically high insulin — not just high blood sugar — drives fat accumulation in the liver, and that temporarily lowering insulin could improve how the liver handles fats and glucose. Participants wore continuous glucose monitors and consumed deuterated water to measure liver fat production in real time. Results from this trial could reshape thinking about NAFLD treatment by shifting focus from blood sugar to insulin levels as the primary therapeutic target.
Detailed Summary
Non-alcoholic fatty liver disease affects roughly one in four adults globally and sits at the intersection of obesity, insulin resistance, and metabolic dysfunction. Standard approaches focus on lowering blood glucose, but a growing body of evidence suggests that compensatory hyperinsulinemia — chronically elevated insulin driven by insulin resistance — may itself be a primary driver of hepatic fat accumulation. This trial directly tests that idea.
Researchers at Columbia University conducted a completed, randomized, placebo-controlled Phase 1 trial in adults with overweight or obesity who had insulin resistance with or at high risk for NAFLD. Participants were randomized to oral diazoxide at 1 mg/kg per dose, 2 mg/kg per dose, or placebo, taken 27 times over 14 days — roughly three times daily. Diazoxide is a well-known potassium channel opener that suppresses pancreatic insulin secretion, making it a precise pharmacological tool to reduce hyperinsulinemia without directly targeting glucose.
To measure outcomes, the trial used fasting blood draws on four mornings, tracking plasma glucose, serum insulin, triglycerides, free fatty acids, and apolipoprotein B. Critically, participants also consumed deuterated (heavy) water across the study period — a validated metabolic tracer technique that allows researchers to quantify de novo lipogenesis, the liver's synthesis of new fat from carbohydrates. Continuous glucose monitors provided real-time glycemic data throughout.
The results have not yet been published in full, but the mechanistic design is elegant. If lowering insulin reduces hepatic de novo lipogenesis and improves lipid profiles without worsening glucose control, it would provide strong human evidence that hyperinsulinemia — not hyperglycemia alone — is a causal driver of NAFLD.
Caveats are significant: this is a short-term Phase 1 trial focused on safety and metabolic signals, not clinical endpoints. The summary is based on the trial registration abstract only, and full results are pending.
Key Findings
- Trial tests whether suppressing high insulin with diazoxide reduces liver fat production in NAFLD patients.
- Deuterated water tracing allows direct measurement of de novo lipogenesis — the liver's conversion of carbs to fat.
- Two doses (1 mg/kg and 2 mg/kg) compared against placebo over 14 days in insulin-resistant adults.
- Continuous glucose monitoring tracks real-time glycemic safety while insulin is pharmacologically lowered.
- Findings could reframe NAFLD treatment strategy from glucose-lowering to insulin-lowering.
Methodology
Randomized, double-blind, placebo-controlled Phase 1 trial at Columbia University enrolling adults with overweight/obesity and insulin resistance with or at risk for NAFLD. Participants took 27 doses of diazoxide or placebo over 14 days alongside 32 doses of deuterated water for de novo lipogenesis assessment. Fasting labs and continuous glucose monitoring provided metabolic endpoints.
Study Limitations
This is a short-term Phase 1 trial designed primarily for safety and metabolic signal detection, not clinical outcomes such as liver fibrosis or steatosis scores. The full dataset and results have not yet been published; this summary is based on the trial registration abstract only. The 14-day window is insufficient to draw conclusions about long-term efficacy or durability of any metabolic changes observed.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: