New semaglutide research spans mood, kidney, and fat regain

The clearest new finding is that semaglutide changed how people with major depressive disorder made choices in a laboratory task, and the change was not explained by weight loss. Researchers randomized 72 people with major depressive disorder to oral semaglutide or placebo for 16 weeks ¹. They played a four-armed restless bandit task before and after treatment. In this task, each of four options pays off at different rates that shift over time. To do well, a player has to decide when to stay with the option that seems best and when to try another option that might be better.

After 16 weeks, the semaglutide group made more choices that maximized expected reward, a pattern called exploitation. They also made more directed exploration toward uncertain options. They repeated their previous choice more often, a pattern called perseveration. The average treatment effects were 0.0242 for exploitation, 0.0619 for directed exploration, and 0.3017 for perseveration. The differences were statistically significant, meaning they were unlikely to be due to chance ¹. The research team found these effects were moderated by, but ultimately independent of, baseline characteristics and weight loss during the trial ¹.

This is a small secondary analysis, so it does not establish that semaglutide treats depression. It does suggest that a metabolic change can shift a decision-making process that is often impaired in depression.

The FLOW trial already showed that semaglutide slowed kidney decline in people with type 2 diabetes and chronic kidney disease ⁴. The new analysis asked a practical safety question: did semaglutide cause more problems as kidney function worsened? The data did not suggest added harm from semaglutide as kidney function fell.

FLOW included 3,533 people with type 2 diabetes and chronic kidney disease, randomized to semaglutide 1.0 mg once weekly or placebo ². Kidney function was measured with eGFR, a number that estimates how well the kidneys filter waste. A lower eGFR means worse kidney function. Among people with an eGFR of 60 or above, 45.1% of those on semaglutide had a serious adverse event, compared with 50.4% on placebo. Among those with an eGFR under 30, the rates were 56.4% on semaglutide and 60.4% on placebo ². In every kidney function subgroup, the semaglutide group reported fewer serious adverse events than the placebo group.

Treatment discontinuations due to any adverse event rose as kidney function fell in both groups, but the semaglutide and placebo rates stayed nearly identical at each level. At eGFR 60 or above, discontinuations were 11.2% with semaglutide and 11.3% with placebo. At eGFR under 30, they were 18.8% with semaglutide and 18.7% with placebo ². This suggests the higher discontinuation rate comes from advanced kidney disease itself, not from the drug.

A mouse experiment followed what happens to fat tissue after semaglutide stops and body weight returns. Male mice with diet-induced obesity received high-dose semaglutide or vehicle for three weeks, then the drug was stopped while a high-fat diet continued ³. During treatment, semaglutide produced rapid fat loss while preserving lean mass. Fasting blood glucose, insulin levels, insulin sensitivity, and glucose tolerance all improved ³.

In visceral fat, the kind packed around internal organs, the researchers saw a clear anti-inflammatory change. Pro-inflammatory macrophages, immune cells that drive tissue damage, fell sharply. Crown-like structures, clusters of immune cells around stressed fat cells, also decreased. Inflammatory gene expression dropped. Subcutaneous fat, the layer under the skin, was minimally affected ³.

After the drug was withdrawn, the mice regained weight and fat quickly. Visceral fat inflammation returned to levels seen in chronically obese mice. Macrophage infiltration and inflammatory gene expression matched the obese controls. Fat buildup in the liver and liver inflammatory markers also returned to obese levels. Despite that reversal, fasting blood glucose and insulin levels remained partly improved ³.

This experiment shows that semaglutide's effect on fat tissue inflammation is tied to active treatment and does not survive regain, at least under these conditions in mice.

These three reports sit beside a much larger body of semaglutide research. The depression trial is a secondary analysis of 72 people, so it is not strong enough on its own to change conclusions. It is consistent with earlier laboratory work showing semaglutide activates appetite-suppressing neurons in the brain ⁵ and with observational data linking GLP-1 drugs to lower Parkinson's disease diagnosis ⁶. But it is one of the first randomized experiments to measure a decision-making pattern in major depressive disorder ¹.

The FLOW safety analysis reinforces the main FLOW result, which found reduced sustained kidney decline, end-stage kidney disease, and cardiovascular death ⁴. The new safety data by kidney function subgroup add confidence that the benefit-risk pattern does not flip as kidney disease advances ².

The mouse study fits with known weight regain after stopping semaglutide. In the STEP 4 trial, participants switched to placebo regained 6.9% of their body weight within a year ⁷. The mouse data add a tissue-level detail: the anti-inflammatory effect in visceral fat is reversible during regain, even though some blood sugar improvements persist ³.

Together, the new evidence does not establish a new use. It adds texture to what the drug does in the brain, in advanced kidney disease, and after treatment stops.

The depression trial used a small sample and a short 16-week treatment period, so it cannot say whether semaglutide improves mood or daily functioning in depression. Larger trials with clinical outcomes are needed ¹.

The FLOW safety analysis reported events descriptively, so the pattern is informative but not a formal test ². The mouse study used one sex, one high dose, and a short treatment window, and mouse fat biology does not always translate to humans ³. What happens to fat tissue inflammation during longer treatment, slower regain, or partial maintenance dosing remains open.