Appetite Hormones Stay Stubbornly Out of Sync as Anorexia Patients Gain Weight in Treatment

Recovery from anorexia nervosa has long been measured on a scale: kilograms regained, body mass index restored, meals completed. But a new study from Norway suggests that the story of recovery is written just as much in the blood as on the scale. In one of the most detailed longitudinal portraits of the gut–brain axis in anorexia nervosa to date, researchers tracked a panel of appetite-related hormones in patients undergoing inpatient treatment and found that even as weight climbed steadily over twelve weeks, several key molecular signals of hunger and satiety refused to fall back into line with those of healthy people.

The work, published in the Journal of Eating Disorders, emerges from the Norwegian Microbiota Study in Anorexia Nervosa, known as NORMA, a project funded by the Norwegian Research Council that examines how gut biology and eating disorders intersect. The team, led by Ida Heir Hovland and Siv Kjølsrud Bøhn of the Norwegian University of Life Sciences together with collaborators at Oslo University Hospital and other Norwegian centers, and including eating disorders researcher Cynthia M. Bulik of the Karolinska Institutet and the University of North Carolina at Chapel Hill, enrolled 42 patients with anorexia nervosa admitted for inpatient treatment and 42 age-matched healthy controls. Blood samples were drawn from patients at baseline, after six weeks, and after twelve weeks of treatment, while the healthy controls provided a single baseline sample for comparison.

The analytical approach was deliberately broad. Using Luminex technology, a bead-based immunoassay platform that can quantify many proteins simultaneously in a small volume of serum, the researchers measured an array of hormones and signaling molecules central to appetite regulation: ghrelin, the stomach-derived hormone that stirs hunger; leptin, the fat-cell hormone that signals long-term energy stores; insulin and C-peptide, markers of pancreatic insulin secretion; adipsin and resistin, adipose-derived factors; adiponectin, another fat-cell hormone with metabolic and anti-inflammatory roles; the incretin hormones GLP-1 and GIP, which are released by the gut after eating and modulate insulin secretion and satiety; glucagon; plasminogen activator inhibitor-1; and visfatin. Dietary intake was captured with a three-day digital food diary and interview, allowing the team to connect hormone levels not just to weight but to what patients actually ate.

The baseline differences were striking. Compared with healthy controls, patients with anorexia nervosa showed significantly lower levels of adipsin, leptin, insulin, and resistin, and significantly higher levels of adiponectin and glucose-dependent insulinotropic polypeptide, or GIP. Ghrelin, the hunger hormone, was also elevated in patients, though the difference fell just short of statistical significance with a p-value of 0.08. No baseline differences emerged for C-peptide, plasminogen activator inhibitor-1, or visfatin. Intriguingly, the size of the group difference in GLP-1 and glucagon depended on age, with discrepancies between patients and controls being greater in older participants, a hint that the hormonal signature of the illness may shift across the lifespan.

These patterns make physiological sense in the context of severe starvation. Leptin is produced by adipose tissue in proportion to fat stores, so its near-absence in emaciated patients is expected. Low insulin reflects reduced nutrient intake and reduced body fat. Elevated ghrelin is the body’s classic cry for food. But the elevated adiponectin and GIP are less intuitive, and their persistence raises questions about whether the gut–brain axis in anorexia nervosa is merely responding to starvation or has been fundamentally rewired by it. Adiponectin typically rises as fat mass falls, yet its other roles in inflammation and energy metabolism suggest it may be doing more than simply tracking body composition.

Then came treatment. Over twelve weeks of structured inpatient care, designed around nutritional rehabilitation, the patients’ mean body mass index rose significantly from 15.5 to 18.2 kilograms per square meter, a substantial gain that nonetheless left most patients below the healthy BMI range. The hormonal response to this refeeding was a study in partial recovery. Adipsin, leptin, insulin, and C-peptide all increased, moving in the direction of healthy control values. Ghrelin decreased and approached the levels seen in controls. In other words, the hormones most tightly coupled to fat mass and acute nutrient flux responded to treatment much as one would predict from basic endocrinology.

But four signals stood apart. Glucagon, adiponectin, GIP, and GLP-1 were largely resistant to change across the twelve weeks of treatment. These are precisely the hormones most entangled with the gut side of the gut–brain axis: GLP-1 and GIP are secreted by intestinal enteroendocrine cells in response to food, and glucagon, produced by the pancreatic alpha cells, governs glucose mobilization during fasting. Their stubbornness suggests that refeeding and weight gain, at least over this timeframe, are not sufficient to normalize every arm of the appetite-signaling network. The authors interpret this as evidence of continued gut–brain axis dysregulation, a molecular echo of the illness that persists even as the body is refed.

Perhaps the most consequential finding of the study concerns diet. When the researchers examined associations between the biomarkers and what patients ate, they found that body mass index alone did not fully account for the hormonal differences between patients and controls. Dietary composition, particularly the intake of fiber and fat, appeared to matter independently for hormone regulation in anorexia nervosa. This is a technically important observation because it implies that the quality and macronutrient profile of the diet, not merely the total energy delivered, may shape how the appetite-signaling system recovers. Fiber is well known to influence gut hormone secretion through fermentation by the gut microbiota, producing short-chain fatty acids that stimulate enteroendocrine L-cells, the very cells that produce GLP-1. Fat intake, meanwhile, influences both incretin release and the absorption kinetics of meals. The NORMA study’s microbiota focus makes this connection especially tantalizing, since the gut bacteria that ferment fiber are themselves altered in anorexia nervosa.

The clinical implications are considerable but deliberately framed as open questions by the authors. Nutritional treatment is a cornerstone of anorexia nervosa care, and weight restoration remains essential, yet the persistence of abnormal levels of glucagon, adiponectin, GIP, and GLP-1 after twelve weeks of intensive treatment suggests that weight gain alone may not fully repair the biological machinery of appetite. Whether these unresponsive hormones can be targeted through tailored nutritional strategies, such as manipulating fiber or fat intake, or through pharmacological therapies aimed at the gut–brain axis, remains to be investigated. The GLP-1 system is already a major drug target in obesity and diabetes, and understanding why it behaves anomalously in anorexia nervosa could open new therapeutic avenues for a disorder where restoring the drive to eat is the central clinical challenge.

The study’s design carries caveats that the authors acknowledge through its framing as exploratory. With 42 patients and 42 controls, the sample is modest, and the healthy controls were only measured once, limiting the ability to model within-person change in that group. Twelve weeks is a short window in the arc of recovery from an illness that often runs a chronic course, and the patients’ BMIs at the end of treatment were still below the healthy range, so it remains possible that some hormones would normalize with further weight restoration. Still, the trial was prospectively registered with ClinicalTrials.gov under identifier NCT06144905, and the longitudinal sampling within treatment gives the findings a temporal texture that cross-sectional studies cannot match. What emerges is a picture of anorexia nervosa as not only a psychological disorder of eating but a systemic endocrine condition, one in which the molecular conversation between gut and brain is disrupted at baseline, partially restored by refeeding, and in key respects strikingly resistant to change. For clinicians and researchers alike, the message is that the scale tells only part of the story, and that the hormones of hunger may hold the rest.

Subject of Research: Appetite-related biomarker changes during inpatient treatment of anorexia nervosa and their association with diet and body mass index

Article Title: Exploring appetite-related biomarkers in anorexia nervosa during treatment: associations with diet and comparison with healthy controls

Article References: Hovland, I. H., Otterdal, K. L., Bang, L., Vinje, H., Herfindal, A. M., Strømland, S. S., Aspholm, T. E., Arsenovic, D., Reistad, H. T., Dahl, J., Ones, M. L., Weider, S., Bulik, C. M., Rø, Ø., & Bøhn, S. K. (2026). Exploring appetite-related biomarkers in anorexia nervosa during treatment: associations with diet and comparison with healthy controls. Journal of Eating Disorders. https://doi.org/10.1186/s40337-026-01752-1

Image Credits: AI Generated

DOI: 10.1186/s40337-026-01752-1

Keywords: anorexia nervosa, appetite hormones, gut-brain axis, ghrelin, leptin, GLP-1, GIP, adiponectin, dietary fiber, nutritional rehabilitation, biomarkers, eating disorders