PCOS is 50–70% heritable. Genetic markers explain less than 10% of cases. What that gap means.

Photo by Adam Winger on Unsplash.

PCOS has a heritability of 50–70%. Identified genetic markers explain less than 10% of cases.

Those two numbers are not contradictory. The gap between them is where the most clinically useful questions live.

Endometriosis sits at 45–55% heritable. Adenomyosis has a well-documented familial pattern with deeply intriguing epigenetic underpinnings. All three conditions run in families. All three are described to women as genetic. And all three are then handed back to them as if the investigation is complete.

It isn't.

What heritability actually measures

Heritability is a population-level statistic. It measures how much of the variation in a trait, across a given population, can be attributed to genetic differences between individuals rather than environmental differences.

It does not tell you what percentage of any individual's disease is determined by their genes.

A heritability of 70% does not mean that 70% of your PCOS is genetic. It means that in the populations studied, 70% of the variation in who develops PCOS can be statistically explained by genetic differences. The remaining 30% is accounted for by environment: diet, stress load, toxin exposure, metabolic history, reproductive history, the conditions you were developing in before you were even born.

This distinction matters enormously in practice, because "it's genetic" is frequently used to mean "there's nothing modifiable to investigate." Heritability, correctly understood, says nothing of the sort. It describes where variation comes from in populations. It says very little about what shapes severity, timing, or expression in an individual.

The 10% figure for PCOS makes this concrete. If 50–70% of the variation in PCOS is heritable, but identified genetic markers explain less than 10% of cases, that means the vast majority of the genetic contribution is still unexplained. We haven't found the genes that account for it. Which suggests that what we're dealing with is not a straightforward genetic condition at all, but a complex interaction between genetic vulnerability and environment: everything from intrauterine conditions and early hormonal exposures to adult stress physiology, nutritional status, and endocrine disruption.

There is also a methodological caveat worth naming honestly. Most heritability estimates for these conditions come from twin studies, and twin studies have real limitations. Identical twins share more than DNA: in many cases they share the same intrauterine environment, the same maternal hormonal milieu, the same placenta (in monochorionic pairs), and often the same early childhood exposures. When we measure concordance in identical versus fraternal twins, we are trying to isolate genetic from environmental contribution, but if the intrauterine environment itself varies systematically between identical and fraternal twins, the genetic estimate is inflated. Diagnosis is also frequently self-reported in these studies, and there is a known ascertainment bias: if one identical twin is diagnosed, the other is more likely to seek evaluation. These are not reasons to dismiss the heritability data. They are reasons to treat the upper end of the estimates with some caution, and to hold the possibility that the environmental contribution, broadly defined to include the womb environment, is larger than the headline figures suggest.

PCOS and PMOS: what the genetics actually show

The most robust heritability data for PCOS comes from a 2006 Dutch twin study by Vink and colleagues, which found concordance rates suggesting heritability in the 50–70% range. A 2023 international evidence-based guideline from the International PCOS Network confirmed that lifestyle and environmental factors significantly modify how severely PCOS presents: even in genetically predisposed individuals, supportive conditions meaningfully reduce disease burden.

A 2019 study by Risal and colleagues in Nature Medicine added a layer that is still not widely discussed in clinical settings. Using a mouse model, they demonstrated transgenerational transmission of PCOS-like traits through prenatal androgen exposure. The female offspring and granddaughters of mice exposed to excess androgens during pregnancy showed PCOS-like presentations, even without direct androgen exposure themselves. The implication is significant: some of what looks like a genetic predisposition to PMOS may actually be an epigenetically transmitted hormonal environment, shaped by the conditions that existed during your mother's pregnancy.

Before going further, a word on terminology. The condition is almost universally called PCOS — polycystic ovary syndrome — but the name is, in many ways, a misnomer. Research consistently shows that most women with this diagnosis do not actually have cysts on their ovaries. What they have are multiple small follicles in an arrested state, a pattern more accurately described as multifollicular ovaries. PMOS — polycystic and multifollicular ovary syndrome — is a more precise and more honest term, and one that better reflects the actual biology. I'll use PMOS from here.

The phenotypic variability within PMOS is itself instructive. The insulin-resistant presentation (often associated with weight changes, hyperandrogenism, and metabolic markers) is the one that dominates clinical conversation. But the adrenal presentation looks very different: characterised by excess DHEA-S, often in women who are insulin-sensitive rather than resistant, and who frequently find that high exercise loads and caloric restriction worsen rather than improve their picture. This is not a trivial distinction. It changes every practical recommendation. And yet the genetic framing tends to flatten it into a single condition with a single explanation.

What drives predisposition into disease:Insulin resistance and hyperinsulinaemia (even without overt diabetes), chronic psychological and physiological stress, low energy availability, micronutrient depletion (particularly zinc, magnesium, and inositol), and environmental endocrine disruptors (plastics, pesticides, certain personal care products) are consistently implicated in the literature as the factors that determine whether a genetic predisposition becomes a clinical condition, and how severe that condition is.

Endometriosis: a condition the body grows from its environment

Endometriosis has a heritability of approximately 45–55%, based on a 2018 meta-analysis of genome-wide association studies by Rahmioglu and colleagues. Twin concordance rates in the research sit at 50–60% for identical twins and 20–30% for fraternal twins, confirming a meaningful but far from deterministic genetic contribution.

The GWAS work has identified susceptibility loci, particularly involving the genes WNT4, VEZT, and ESR1. These are involved in oestrogen signalling, reproductive development, and cell adhesion processes. Sisters of women with endometriosis have a prevalence approximately five to six times higher than the general population, and familial clustering is well established.

But notice what this tells us. Even in identical twins (who share 100% of their DNA), concordance is only 50–60%. That means in nearly half of identical twin pairs where one twin has endometriosis, the other doesn't. The same genetic blueprint, expressed differently, depending on what the body's environment has been.

It is also worth noting that endometriosis, once assumed to be a condition of older reproductive-age women, is increasingly being identified in adolescents. A 2025 review in Biomedicines examined the genetic and epigenetic components in the pathogenesis of both endometriosis and adenomyosis specifically in younger populations, and found that the same susceptibility genes (WNT4, VEZT, ESR1) appear to influence early lesion development, alongside aberrant DNA methylation and chromatin remodelling as epigenetic drivers. This matters for two reasons: it suggests that the biological vulnerabilities are established earlier than the diagnosis often is, and it argues for earlier investigation in adolescents with significant period pain rather than the default of oral contraceptives as a first (and often only) response.

The research strongly implicates inflammatory load, immune dysregulation, oestrogen clearance pathways, and environmental toxin exposure as the drivers that determine how a genetic vulnerability expresses. Women with endometriosis show consistently elevated inflammatory markers, altered immune surveillance in the peritoneal cavity(which appears to be less effective at clearing ectopic endometrial tissue in those who develop the condition), and impaired oestrogen clearance through liver and gut pathways.

The toxin connection is one that rarely receives adequate clinical attention. Dioxins, PCBs, and phthalates have documented associations with endometriosis in both animal models and epidemiological research. These are oestrogen-mimicking compounds that accumulate in the body over time and actively suppress immune function. The relationship between endometriosis severity and environmental toxin load is an area where the evidence is more established than the clinical conversation suggests.

Adenomyosis: the most epigenetically driven of the three

Adenomyosis is the most understudied of these three conditions, and the hardest to characterise genetically, partly because it was historically only diagnosable via hysterectomy and therefore identified later in women's reproductive lives. This has limited the quality of epidemiological data significantly, and there is currently no reliable heritability percentage from twin or population studies for adenomyosis specifically. The absence of that figure is itself informative: we don't have it not because the condition isn't heritable, but because the research simply hasn't been possible to conduct with the same rigour as PMOS and endometriosis. There are also no comparative data from hunter-gatherer or traditional populations of the kind that exist for some other conditions, which would allow us to assess how much of the disease burden is a product of contemporary environments rather than genetic architecture. This gap points toward environment as a likely major driver, even if we can't yet quantify it precisely.

What we do know is that familial clustering is clearly present. If your mother or sister has adenomyosis, your risk is meaningfully elevated. Emerging GWAS data has identified 171 distinct genetic risk signatures specific to adenomyosis, with some overlap with endometriosis: nine genes appear in both conditions, but in distinct SNP (single nucleotide polymorphism) combinations and through mechanistically different pathways. Adenomyosis-specific genetic signatures are particularly enriched for pathways involved in cell-cell adhesion: the mechanisms by which cells hold their correct positional identity in tissue.

The shared genes (WNT4 and ESR1 appear again here, alongside VEZT) suggest common ground in oestrogen signalling and reproductive tissue development, while the distinct signatures suggest that the two conditions, though they frequently co-occur, are not simply the same disease expressed differently.

What is particularly striking in the adenomyosis literature is the epigenetic picture. Research points to pre-existing epigenetic defects in the basalis layer of the endometrium as a likely precondition for adenomyosis development: specifically, aromatase excess (the enzyme that converts androgens to oestrogens, producing a locally oestrogen-rich environment), progesterone resistance in stromal cells (which means the normal hormonal signalling that should protect against lesion development doesn't function correctly), and a hyperactive inflammasome. KRAS mutations have been found in the epithelial cells of adenomyosis lesions.

Critically, the literature suggests these epigenetic abnormalities may be acquired during embryonic or perinatal life and may even be transmitted to the next generation. This is the same transgenerational epigenetic transmission mechanism found in PMOS research. It means that what is inherited may not be a gene directly causing the condition, but an epigenetically programmed cellular environment, established before birth, shaped by the conditions of pregnancy and potentially even by the grandmother's environment.

The modifiable drivers in adenomyosis overlap substantially with endometriosis: inflammatory load, oestrogen excess or impaired clearance, immune dysregulation, and environmental toxin accumulation. Aromatase excess (which may itself be epigenetically determined) creates a self-sustaining oestrogen environment that drives lesion development. Supporting oestrogen clearance pathways, reducing inflammatory substrate, and addressing the immune environment are clinically meaningful areas of investigation.

Why these three conditions so often co-occur

PMOS, endometriosis, and adenomyosis overlap at a rate that is not explained by chance. Many women carry more than one of these diagnoses, and the biology tells us why.

They share overlapping genetic architecture: WNT4, ESR1, and the oestrogen signalling pathways appear in the GWAS data for all three. They share immune dysfunction as a common pathway: altered immune surveillance and an elevated inflammatory substrate appear across all three presentations. They share a relationship with oestrogen excess or dysregulated clearance. And they share sensitivity to environmental oestrogen-mimicking compounds.

The condition that manifests most prominently is often a function of which tissue environment has accumulated the most load. Adenomyosis develops in the myometrium; endometriosis in ectopic sites outside the uterus; PMOS in the ovarian and hormonal architecture. But the underlying physiology creating vulnerability in all three is, to a significant degree, the same.

This is why treating them as entirely separate conditions, each with its own specialist and its own intervention, often misses the shared substrate underneath.

What epigenetics adds to this picture

Epigenetics is the study of how gene expression is regulated without changes to the underlying DNA sequence. Genes can be switched on or off, up- or down-regulated, by a range of mechanisms: DNA methylation, histone modification, the action of microRNAs. These mechanisms are sensitive to environment in ways that fixed DNA sequence is not.

This is why the environment you developed in, the nutritional status your mother carried during pregnancy, the stress load and toxin exposure of your early life, the conditions of your childhood, all of these can shape how your genetic predisposition expresses. Not by changing the genes themselves, but by altering which ones are active, and how active they are.

Aberrant DNA methylation and chromatin remodelling are specifically documented in adenomyosis and endometriosis. In PMOS, prenatal androgen exposure appears to epigenetically programme later ovarian and hormonal function. These are not theoretical mechanisms. They are observable, measurable, and, in some respects, modifiable.

The implication is that when we say these conditions are genetic, we are describing a system where the gene is one actor in a complex environment, not a pre-written script. Changing the environment changes the expression. Not always completely, and not always immediately. But meaningfully, and in ways that matter clinically.

Understanding your SNPs: genetics as a map, not a verdict

For women who want to go further than the population-level data and understand their own genetic picture, SNP (single nucleotide polymorphism) testing offers something genuinely useful: not a prediction of disease, but a map of vulnerability.

The Eat For Your Genes (EFYG) report includes dedicated sections on PCOS and endometriosis genetics, covering the most replicated GWAS findings for both conditions. Because adenomyosis shares a significant portion of its genetic risk architecture with endometriosis (the EFYG report itself notes they "share much of the same biology"), this information is relevant across all three conditions.

In the PCOS section, the key SNPs assessed include DENND1A (the most replicated PCOS GWAS gene, involved in androgen production in theca cells), SHBG (which regulates how much free testosterone is circulating and is also a strong marker of insulin resistance), FTO (a significant metabolic phenotype modifier), and others affecting FSH receptor sensitivity and aromatase activity. In the endometriosis section, the most notable finding is CDKN2B-AS1/ANRIL, the single most replicated genetic risk allele for endometriosis, which is also a strong risk allele for cardiovascular disease. This is not a coincidence: both conditions involve the same underlying biology of chronic inflammation and tissue remodelling. The inflammatory substrate that drives endometriosis lesion development is the same substrate that drives cardiovascular risk.

Beyond the dedicated reproductive health section, the detoxification genes are directly relevant. CYP2C19, an ultra-fast phase 1 metaboliser variant, carries an explicitly flagged increased risk for endometriosis in the EFYG report. The CYP1B1 variants (involved in catechol oestrogen metabolism) shape how oestrogen is processed and whether it produces inflammatory metabolites. The UGT1A6 variants (glucuronidation pathway) affect the liver's ability to package oestrogen for excretion. In other words, the oestrogen environment that drives all three of these conditions is shaped by genetic variants scattered across multiple systems, not just the reproductive section.

These SNPs are not diagnostic. Carrying DENND1A or CDKN2B-AS1 does not mean PCOS or endometriosis is inevitable. What it means is that specific biological pathways may be operating with less resilience than average, and knowing which ones changes where effort is directed. Someone with impaired oestrogen clearance variants needs a different intervention focus than someone whose primary vulnerability sits in inflammatory signalling. The genetics become a guide to where to look, and what to prioritise, rather than a verdict about what is fixed.

This is a fundamentally different relationship with genetic information than the clinical norm. And it is one that, in my experience, women find genuinely empowering rather than frightening.

What a more useful investigation looks like

When a woman comes to me with any of these diagnoses, I am not dismissing the genetic component. I'm beginning with it, and then asking: what in this person's biology, history, and environment is determining how that predisposition is expressing?

The questions that matter:

What is the inflammatory load? This includes gut health (the gut is the primary source of the systemic inflammatory substrate in most chronic presentations), dietary factors, immune activation patterns, and environmental exposures.

What is the oestrogen environment? Is oestrogen being cleared efficiently through liver phase 1 and phase 2 pathways? Are the gut microbiome and bile flow supporting enterohepatic oestrogen clearance? Is there external oestrogen exposure (through plastics, cosmetics, food packaging) adding to the load?

What is the immune environment? All three conditions have immune dysfunction as a significant feature. In endometriosis especially, peritoneal immune surveillance appears compromised. What is driving immune dysregulation, and what is it being asked to manage?

What is the stress and nervous system context? The HPA axis (our stress response system) directly modulates both immune function and inflammatory activity. Chronic stress physiology is not a background variable here. It is a mechanistic driver.

What is the nutritional status? Zinc, magnesium, B vitamins, vitamin D, omega-3 fatty acids, and iron all appear in the literature as relevant to these conditions. Nutrient depletion is both a consequence of the inflammatory burden and a factor that maintains it.

These questions do not change the diagnosis. They change what's possible.

Predisposition is not a verdict

The heritability data for these three conditions is real and meaningful. If your mother or sister has PMOS, endometriosis, or adenomyosis, your genetic risk is genuinely elevated. That is worth knowing, because knowing it changes what you look for and how early you look for it.

But heritability describes vulnerability, not certainty. And the evidence across all three conditions is consistent: supportive conditions reduce severity, sometimes substantially. The genetic predisposition sets the stage. The environment determines much of what plays out on it.

"It's genetic" is a starting point for the investigation, not the end of it. The questions that follow are the ones that change the outcome.

If you're navigating PMOS, endometriosis, or adenomyosis, whether that's period pain that disrupts your life, cycles that feel out of your control, or fertility challenges that haven't been given a satisfying explanation, I work with women one-on-one to look at the whole picture. Not just the diagnosis, but what in your particular history, biology, and environment is shaping how it's expressing, and what can actually be done about it.

Many of the women I work with have been told that pain is expected, that their fertility options are limited by genetics, or that management is the best that can be offered. In most cases, there is considerably more that can be investigated, and considerably more that can change.

As always, thank you for reading.

Lou