Recovery is slippery.
Doctors fix the nutrition. Psychologists untangle the thoughts. It sounds like a solid plan. But forty percent of those discharged end up back in a hospital bed within six months. Why? We didn’t know. Until now.
Researchers think the answer hides in a hormone battle. Specifically, the tug-of-war between ghrelin—the stuff that screams eat now —and its antagonist, LEAP2.
Virginie Tolle, a neuroscientist at INSERM, presented findings last year that point directly at this imbalance. She notes that anorexia carries the highest mortality rate of any psychiatric disorder.
“[Anorexia nervosa]… has the highest mortality rate amongst all psychiatry disorders,” Tolle said.
Current treatment relies on feeding and talking. It works, slowly. Relapse remains stubbornly high.
The team tracked thirty women undergoing four months of refeeding therapy. They took blood at the start, at the end, and again six months later.
Here is the pattern they found.
At the height of illness, these women had significantly higher levels of LEAP2. By twenty percent. This protein blocks ghrelin. It shuts down hunger signals. Even when the body screams for energy, LEAP2 whispers silence.
As weight returned during treatment, LEAP2 dropped. Ghrelin got a voice back.
But not everyone stayed recovered.
The relapsers saw their LEAP2 spike again. It crept back up, silencing the hunger cues that should have stabilized their weight.
The data got clearer. The ratio of ghrelin to LEAP2 linked directly to impulse control. Patients who maintained stable weight had different ratios than those who didn’t.
They tested this on mice, too.
Starve a mouse for a bit—just lose twenty-five percent of its weight. Offer a choice. Eat a small treat now. Or wait. And eat a feast later.
The starved mice chose the immediate sugar. High LEAP2 locked them into this impulsivity. Even after refeeding, the behavior didn’t fully vanish. The biology remembered the famine.
This suggests the brain’s decision-making gets hijacked by metabolism. Not just psychology. Physiology.
Can a blood test save lives?
If larger studies confirm these markers, clinicians might spot the slide before the crash. High LEAP2 could flag a patient for early intervention. Not when the weight is gone, but when the hormone screams it will be.
It changes the game. We aren’t just watching numbers on a scale. We’re watching chemistry.
“Metabolic signals that normally regulate hunger adapt differently… These signals also influence the brain.”
Maybe we finally have a handle on the unseen hand guiding the relapse. Maybe the next treatment isn’t just food. Or talk therapy.
But a drug that tweaks the ratio.
We don’t have that yet. The data is promising. It’s just one study.
But look at the mouse. It kept choosing the small reward, starving its future for the comfort of now. Until the biology shifted.
Perhaps that shift is possible. For us, too.
At least, the science suggests the door isn’t closed.
