Endometrial Hyperplasia and PCOS: The Risk Nobody Explains Properly

What is endometrial hyperplasia, exactly?

Endometrial hyperplasia is a measurable overgrowth of the uterine lining, and it comes in two clinically distinct grades — only one of which carries meaningful cancer risk. Hyperplasia without atypia is an overgrown but structurally normal-looking lining, usually reversible with treatment. Atypical hyperplasia has abnormal-looking cells within that overgrowth and is the form that carries a real chance of already containing, or progressing to, cancer. In a study of 57 anovulatory, amenorrhoeic women biopsied specifically because of their symptom pattern, 43 of whom had PCOS, almost all of the hyperplasia found — 45.6% of the whole group — was the non-atypical kind; only one case, 1.75%, showed atypia (Tingthanatikul et al., 2006). That distinction — common but usually reversible, versus rare but more serious — is the one this whole page is built around.

How does PCOS actually cause this?

Every missed ovulation is a missed dose of progesterone, and it is progesterone, not oestrogen, that normally triggers the endometrium to shed and reset on a schedule. PCOS’s chronic anovulation leaves the lining exposed to oestrogen’s growth signal for months at a time with nothing to organise or stop it — the same “unopposed oestrogen” mechanism behind irregular and skipped periods generally, just carried further, for longer, in the small number of cases where a thickened endometrial lining crosses from overgrown into abnormal (Teede et al., 2023). This is also why the underlying driver is the missed ovulation, not PCOS as a label — anything else that causes chronic anovulation produces the same risk through the same pathway.

How much does PCOS actually raise endometrial cancer risk?

Women with PCOS have close to three times the risk of endometrial cancer compared with women without it, in a Danish nationwide cohort study that followed 1,719,121 women for an average of 23.7 years and recorded 7,862 endometrial cancer cases (HR 3.02, 95% CI 2.03–4.49) — rising to nearly 6-fold specifically in premenopausal women (HR 5.82, 95% CI 3.64–9.30) (Frandsen et al., 2024). That estimate is not an outlier: a separate 2025 meta-analysis pooling 13 studies found a closely matching relative risk of 2.91 (95% CI 2.29–3.70) (Ghaempanah et al., 2025). Two independent bodies of evidence, built from different populations, landing in the same range — roughly threefold — is what “genuinely elevated” means here, as distinct from a single study’s number that might not replicate.

What is the absolute risk — the number that actually matters?

The absolute risk stays low even with a tripled relative risk: about 1.7% of women with PCOS develop endometrial cancer, compared with about 0.5% of women overall, based on U.S. prevalence data pooling nine studies (pooled RR 3.46, 95% CI 2.28–5.23) (Pace et al., 2024). Put plainly: out of 100 women with PCOS, roughly 98 will not develop endometrial cancer. A threefold relative increase sounds alarming in isolation; multiplied against a baseline risk that is already small, it produces an absolute chance that stays in the low single digits over a lifetime.

Who is actually at higher risk within PCOS?

Risk stacks with specific, named factors rather than applying evenly to everyone with a PCOS diagnosis. Long-standing untreated amenorrhoea, higher body weight, type 2 diabetes, and a persistently thickened endometrium seen on ultrasound each add measurable risk on top of PCOS itself, and the population most concentrated for risk is exactly the one presenting with abnormal bleeding: in a 2026 meta-analysis of 6,064 women with abnormal uterine bleeding, having PCOS raised the odds of hyperplasia 3.09-fold and of endometrial cancer specifically 6.98-fold, compared with women who had abnormal bleeding without PCOS (Chu et al., 2026).

What monitoring and protection actually exist?

Routine endometrial screening is not recommended for women with PCOS who have no symptoms, under the 2023 international guideline, precisely because absolute risk stays low across the group as a whole (Teede et al., 2023). Screening — transvaginal ultrasound to check endometrial thickness, and biopsy when indicated — is reserved for people with symptoms or with the additional risk factors in the table above, not applied as a blanket test. For protection rather than diagnosis, the guideline’s recommended approach is cycle regulation: a combined hormonal contraceptive or a cyclic or continuous progestin restores regular shedding for anyone not actively trying to conceive, which is the same reason cycle regulation gets recommended even to people with no current interest in pregnancy.

Once hyperplasia is actually diagnosed on biopsy, progestin therapy is the first-line treatment, and delivery method matters: a levonorgestrel intrauterine system produced a higher rate of complete regression than oral cyclic medroxyprogesterone acetate in non-obese women with non-atypical hyperplasia specifically (RR 1.36, 95% CI 1.07–1.73), across a meta-analysis of randomised trials, though the two performed similarly in women with obesity (RR 1.03, 95% CI 0.94–1.13) (Yuk et al., 2017). A repeat biopsy after a course of treatment — typically several months — is what actually confirms regression; symptoms improving is a good sign but not a substitute for that confirmation.

What symptoms actually warrant an assessment?

Specific bleeding patterns, not a PCOS diagnosis on its own, are what should prompt an assessment. Any bleeding at all once you are 12 months past your last period is never explained by PCOS by default and needs evaluation that week. Heavy bleeding with clots that is new or has clearly worsened, bleeding between periods that persists for more than two weeks rather than resolving, and cycles that have shifted from occasionally irregular to almost constant spotting are the patterns worth raising specifically — what each individual bleeding pattern usually means on its own is covered in more detail separately for heavy bleeding and for spotting and discharge. The blood panel that typically accompanies a bleeding work-up is covered here.

Note: in May 2026, PCOS was renamed polyendocrine metabolic ovarian syndrome, or PMOS, by a global consensus of more than 50 organisations. Same anovulatory mechanism, same risk figures, same monitoring approach — only the name on the diagnosis changed. This article uses PCOS, since that is still the term most readers search.

Who this does not apply to, and what this doesn’t mean

This does not mean a single delayed or skipped period is dangerous — an isolated missed cycle is expected in PCOS and is not what the risk figures above describe. It does not mean everyone with irregular periods will develop hyperplasia; most people with chronically infrequent cycles never do, which is exactly what “elevated relative risk, low absolute risk” means in practice. It does not mean someone on a regularly shedding regimen — a combined pill, a cyclic progestin, or genuinely ovulatory natural cycles — carries the same exposure as someone with untreated, long-standing amenorrhoea; the mechanism this whole risk is built on requires the unopposed exposure to actually be happening. And it does not mean hyperplasia, if it is found, is the same thing as cancer — most of it, especially in a population with PCOS rather than older age as the driver, is the non-atypical, reversible kind.

Your next step

If your cycles have been chronically infrequent — fewer than 8 a year, or gaps over 90 days — for longer than a year without any form of cycle regulation, that specific history, not a general sense of worry, is what to bring to a clinician: it is the exposure this whole risk is built on, and it is the piece that decides whether protection now is worth discussing before any symptom shows up at all.