Insulin, body fat, and endometriosis — the fuel the lesion runs on

What turns that local adaptation into a systemic vulnerability is what feeds it from the outside. Two signals circulating in the blood reinforce the lesion's metabolic program directly. The first is insulin, and the second is insulin-like growth factor 1, IGF-1. When insulin runs chronically high — hyperinsulinemia, the metabolic state shared by insulin resistance, prediabetes, and a great deal of excess weight — and when IGF-1 is elevated alongside it, both signals converge on the IGF-1 receptor and switch on one of the cell's most important growth pathways: PI3K–AKT–mTOR. This is the axis that tells a cell to grow, to build protein, to divide, and crucially, not to die. In the endometriotic cell, its activation lowers the apoptotic threshold — making the cells harder to kill off — and amplifies their responsiveness to the growth factors already saturating the lesional environment. The blunt way to put it is that insulin and IGF-1 are fertilizer. They do not start the lesion, but they feed its proliferative machinery, and the more of them in circulation, the more fuel reaches the implant.

Body fat compounds the problem along a second, parallel route that has nothing to do with insulin and everything to do with estrogen. Adipose tissue is the dominant site of extragonadal aromatization — the conversion of androgens into estrogens by the enzyme aromatase, happening out in the fat of the body rather than in the ovaries. That makes fat mass an estrogen reservoir, and because endometriosis is an estrogen-driven disease, the link is mechanically simple: more adipose tissue means more peripheral estradiol production and a higher systemic estrogenic tone, the very tone the lesion grows in response to. So body fat acts on endometriosis twice over — once through the insulin and IGF-1 it tends to accompany, and once through the estrogen it manufactures directly.

There is a further wrinkle that ties insulin and estrogen together more tightly than they first appear. High insulin lowers sex hormone-binding globulin, the protein that keeps estrogen bound and inactive in the bloodstream; with less of that globulin in circulation, a larger fraction of whatever estrogen is present becomes free and biologically available to act on tissue. So hyperinsulinemia does not only feed the lesion's growth machinery through IGF-1 — it also raises the free, active estrogen fraction the lesion responds to. The two systemic problems, the metabolic and the estrogenic, are not running on separate tracks; they reinforce each other. This is part of why the source review treats the metabolic axis as a genuine amplifier rather than a bystander, and why so many affected patients carry a parallel diagnosis of insulin resistance or PCOS. The same internal state that defines those conditions is the state that most efficiently supplies the endometriotic lesion.

It is worth dwelling on how unusual this makes the lesion's position. Most tissues in the body are buffered against swings in systemic metabolism; they take what they need and ignore the rest. The endometriotic cell, by contrast, has wired its survival to exactly the signals that a metabolically stressed body produces in excess. Its Warburg-like glycolysis makes it dependent on glucose availability. Its reliance on PI3K–AKT–mTOR makes it responsive to insulin and IGF-1. And its estrogen-driven growth makes it sensitive to the aromatase activity that body fat carries out. Three separate features of the lesion each point back to the same systemic state — the state of carrying excess adiposity and running chronically high insulin. That convergence is what makes the metabolic axis worth treating as a driver rather than a coincidence, because a lesion this finely tuned to a particular internal environment is, by the same token, vulnerable to that environment changing.

The therapeutic implication of all this is unusually clean. If insulin, IGF-1, and adipose aromatization are fuel lines, then improving metabolism withdraws the fuel. Lower the circulating insulin and you quiet the IGF-1R/PI3K–AKT–mTOR drive; reduce fat mass and you lower both the IGF-1 it tends to raise and the estrogen it aromatizes. This is not a speculative leap. It is the same principle that makes metabolic intervention work in polycystic ovary syndrome, a condition that is metabolically and inflammatorily analogous to endometriosis and that shares the same insulin/IGF-1 and adipose-estrogen axis. The PCOS precedent is the strongest read-across the field has, and it is the reason the metabolic levers below are taken seriously. They are, with one notable exception, still investigational in endometriosis — but the mechanism connecting them to the disease is well drawn, and the evidence in the neighboring condition is real.

The incretin agonists are the most prominent of these levers, and they are the GLP-1 drugs that have become household names. The class is differentiated by how many receptors each agent engages: semaglutide activates the GLP-1 receptor; tirzepatide is a dual GLP-1/GIP agonist; retatrutide adds glucagon-receptor agonism as a tri-agonist, producing progressively greater weight loss. Three of their actions matter to endometriosis. The core one is metabolic: by enhancing glucose-dependent insulin secretion, slowing gastric emptying, and reducing appetite, these drugs lower circulating insulin and reduce adiposity, which withdraws insulin and IGF-1 drive from the PI3K–AKT–mTOR axis so that less fuel reaches the implant's proliferative machinery. The second is estrogenic: because adipose tissue is the dominant site of extragonadal aromatization, reducing fat mass directly lowers peripheral estradiol — a systemic counterpart to the local aromatase blockade that prescription aromatase inhibitors achieve. The third is independent of weight loss altogether. GLP-1 receptor signaling raises intracellular cAMP in immune and endothelial cells, and that cAMP rise attenuates assembly of the NLRP3 inflammasome and the maturation of IL-1β, repolarizes macrophages away from the permissive M2-like phenotype, and lowers systemic inflammatory tone. So an incretin agonist plausibly touches the metabolic, the estrogenic, and the immune arms of the disease at once.

The third action — the immunometabolic one, independent of weight loss — deserves emphasis because it connects the metabolic story back to the inflammatory one. The cAMP rise that GLP-1 signaling produces in immune and endothelial cells does more than dampen inflammation in the abstract; it attenuates the NLRP3 inflammasome and the maturation of IL-1β, and it repolarizes macrophages away from the permissive M2-like phenotype that, in endometriosis, tolerates and nourishes ectopic tissue rather than clearing it. That is the same macrophage problem that drives the disease's fibrosis and angiogenesis. So an incretin agonist is not only a metabolic lever; it plausibly reaches the immune milieu of the lesion as well, and it does so through a mechanism that would still operate even in a patient who lost no weight at all. This is why the class is interesting beyond its obvious effect on body fat: it touches the metabolic amplifier, the adipose estrogen reservoir, and the inflammatory environment, three of the disease's reinforcing circuits, through a single intervention. Whether that translates into lesion-level benefit in endometriosis is exactly what has not yet been established.

On evidence, the honest labeling is layered. In PCOS, GLP-1 and dual GIP/GLP-1 agonists produce well-documented controlled-trial benefit — weight reduction, improved insulin sensitivity, lower androgens, restored ovulation — proving the class can favorably reset the same axis implicated in endometriosis. In endometriosis specifically, the data are early: a 2025 patient-reported survey found that roughly two-thirds of incretin users with confirmed or suspected disease reported improvement in at least one symptom, with about a third reporting complete resolution of at least one, alongside reductions in pelvic pain, bloating and low-back pain; semaglutide shows synergy with progestins in hormone-sensitive tissue models; and a controlled trial pairing an incretin agonist with a levonorgestrel intrauterine device is in progress. That is a measurable early signal, not a controlled result. The endometriosis indication remains investigational, and retatrutide is not yet approved. Among the metabolic levers, this class simply has the strongest combination of coherent mechanism and early human signal.

Berberine reaches the same metabolic territory by a different switch. The plant alkaloid inhibits mitochondrial complex I, raising the AMP-to-ATP ratio and thereby activating AMP-activated protein kinase, AMPK — the cell's master energy sensor. AMPK activation suppresses mTORC1, reducing protein synthesis and proliferation, and improves insulin sensitivity, lowering the insulin/IGF-1 amplifier; berberine also independently inhibits NF-κB signaling, so it engages the inflammatory hub and the metabolic amplifier together. In endometriotic cell models this translates to reduced proliferation, migration and invasion, lower inflammatory cytokine output, and pro-apoptotic effects. The qualifiers are real: human endometriosis data are lacking, so its use in the disease is preclinical and investigational, and berberine carries CYP3A4 and P-glycoprotein interactions along with gastrointestinal effects that make drug-interaction review with a prescribing provider non-optional.

It is worth noting that berberine and the incretins converge on the same downstream destination, mTOR, from opposite directions — the incretins by lowering the insulin and IGF-1 that drive the PI3K–AKT–mTOR axis from outside the cell, berberine by activating AMPK to suppress mTORC1 from inside it. mTOR is, in a sense, the metabolic counterpart to the inflammatory hub: the place where the lesion's growth signals are integrated into a decision to build and divide. Quieting it from either end has the same intended consequence, which is a cell less able to sustain the proliferative tempo the disease depends on. That two mechanistically distinct agents aim at the same integration point is, as with the inflammatory node, a sign that the point is real and central rather than incidental.

MOTS-c sits one tier further out, in the realm of plausible support rather than defined target. It is a 16-amino-acid mitochondrial-derived peptide encoded within the mitochondrial 12S ribosomal RNA gene, and under metabolic stress it translocates to the nucleus to help regulate adaptive antioxidant and metabolic gene programs. Its principal cytoplasmic action is activation of AMPK, which increases glucose uptake and insulin sensitivity, promotes fatty-acid oxidation, and restrains mTOR-driven proliferation — in aggregate reproducing several of the metabolic adaptations that exercise normally induces. The endometriosis-relevant consequence would be withdrawal of the insulin/IGF-1 proliferative amplifier, paralleling the diet, berberine, and incretin arms. But human therapeutic data for MOTS-c are preliminary, and no endometriosis data exist at all; it belongs to a research agenda, not a treatment plan.

A word on how these metabolic levels rank against one another, because the evidence is genuinely uneven and the ranking matters. The incretins sit at the front not because they are proven in endometriosis — they are not, and the indication is investigational — but because they pair a coherent triple mechanism with a strong controlled-trial record in the analogous condition and an early, structured human signal in endometriosis itself. Berberine sits behind them: mechanistically sound, supported by endometriotic cell-model work, but lacking human endometriosis data and carrying real drug-interaction caveats. MOTS-c sits further back again, with preliminary human data of any kind and none at all in this disease. And nutrition sits beneath all of them as the foundation rather than a competitor — the lever with no regulatory ambiguity, available to anyone, acting on the same insulin and IGF-1 amplifier the drugs target. The ordering is not a ranking of mechanism, which is broadly similar across the group, but a ranking of evidence, and keeping those two scales separate is the whole discipline of reading this field honestly. A strong mechanism is a reason to test something, not a reason to assume it works.

Underneath all of the pharmacology sits the simplest lever, which is what is eaten. Low-glycemic, anti-inflammatory nutrition reduces the insulin and IGF-1 amplifier through diet alone, lowering the same PI3K–AKT–mTOR drive the drugs target, and the source review frames it as structured rather than prescriptive — a reduction of the metabolic load rather than a rigid elimination regimen. It is the foundation the rest of the metabolic strategy is built on, and it carries no regulatory ambiguity. What the whole arm shares is a single underlying idea, and it is worth ending on, because it reframes the disease. Endometriosis is usually approached as tissue to be excised and estrogen to be suppressed. The metabolic view adds a third reading: the lesion is also a cell that has to be fed, drawing on circulating insulin, IGF-1, and the estrogen that body fat manufactures. If those supply lines can be narrowed — and in the analogous setting of PCOS they demonstrably can — then improving metabolism is not merely good general health advice layered onto a gynecologic problem. It is acting on a driver of the disease itself, which is a different and more consequential thing, even as the agents that do it most powerfully remain, for endometriosis, investigational and best navigated with a prescribing provider.