Up to one in ten women have endometriosis, according to the US National Institutes of Health. Medicine still cannot fully explain why. A study published on 20 November in Nature Genetics does not solve that puzzle either. What it does offer is one of the broadest genetic maps of the condition so far, drawn from more than a million women whose data sit in biobanks across North America, Europe and Asia.
If you have ever watched Benoit Blanc work a room in Knives Out, you know the method. Nobody gets arrested in the first act. The detective narrows the list of suspects. This study works the same way: it names no single cause but points to the parts of the genome that deserve a much closer look, along with the biological processes behind them.
What the researchers actually did
The paper, titled “Multiancestry genome-wide association and integrative multiomics analyses implicate key pathways in endometriosis”, comes from a team led by Shefali Setia-Verma at the Institute for Biomedical Informatics of the University of Pennsylvania’s Perelman School of Medicine, with Lindsay Guare as first author. Funding came from the NIH’s child health institute, NICHD.
The team ran a genome-wide association study. Put simply, they scanned the genetic data of very large groups of people for spots where a variant shows up more often in women with the condition than in women without it. They pooled summary data from 14 biobanks, including UK Biobank, FinnGen, BioBank Japan, the Taiwan Biobank, the US Million Veteran Program and All of Us. In total the analysis covered 1,063,657 women, 50,265 of them with endometriosis. Nearly a third of participants were of non-European ancestry, which matters more than it might sound. The authors themselves point out that most genetic reference data still comes from people of European descent.
On top of the genome scan, the researchers layered further analyses of gene activity, proteins, metabolites, gut bacteria and cell types. They also used Mendelian randomization, a statistical method that uses genetic variants to probe whether one condition might influence another.
58 regions, 27 of them new
The headline result: 58 regions of the genome are linked to endometriosis, 27 of which had not been reported before. Some signals appeared only in people of African or admixed American ancestry, which is exactly the kind of finding that disappears when a study relies on one population. The team flagged 314 genes as possible drug targets. Among the genes they highlight are ESR1, which carries the blueprint for an estrogen receptor. Another is FSHR, the receptor for follicle-stimulating hormone. The authors also single out focal adhesion, the machinery cells use to anchor themselves to surrounding tissue, as a potential treatment target.
The study also found genetic overlap with 135 other conditions or traits. The genetic tendencies that raise endometriosis risk tend to go together with those linked to antidepressant prescriptions, headache and insomnia. They run in the opposite direction to age at first birth. These are correlations at the level of genes, not diagnoses. They do not show that one condition causes another.
The Mendelian randomization results suggest that endometriosis may influence raised blood lipids, antacid prescriptions and lower levels of the gut bacteria family Bifidobacteriaceae. The authors describe these as hypotheses for future research rather than established effects.
How much is in your genes?
The team estimates that common genetic variants explain roughly 11 to 13 percent of the variation in endometriosis risk across the ancestry groups studied. That is higher than the roughly 7 percent from earlier genetic studies, yet far below the 47 percent suggested by twin research. The gap is a familiar one in genetics. It suggests that rare variants, interactions between genes and environment, or factors not captured in health records still hide a large part of the story.
“Our findings will help guide future research that could eventually lead to better diagnosis,” Setia-Verma said in the NIH press release. John Gaitanis, director of NICHD, said a greater understanding “will help us take steps toward developing diagnostic tools and treatments.” Both statements come from people involved in or funding the work. At the time of writing we found no published comment from independent experts.
What the study cannot tell you
The authors are refreshingly candid about the limits. In their own words, “we did not perform any experimental functional validation”. Everything rests on statistics and computer models, not on lab experiments. The team had no access to individual-level data, only summary statistics. Diagnosis codes in health records are an imprecise way to identify endometriosis. The stricter, surgically confirmed definitions produced much smaller samples. Rare and structural genetic variants were not captured. Despite the size of the study, the authors write that they still lacked the power to draw many conclusions in understudied populations.
One more detail for the record: earlier versions of the work have circulated as preprints since late 2024. The peer-reviewed paper is new. The underlying research is not entirely fresh. In a news cycle that loves the word “breakthrough”, that distinction is worth keeping in mind.
What it means for you
In practical terms, nothing changes in the doctor’s office tomorrow. There is no genetic test for endometriosis coming out of this study. There is no new treatment either. NIH stresses that the findings are associations. If you live with pelvic pain or suspect endometriosis, the route still runs through a gynecologist and a proper diagnostic workup.
The longer view is more hopeful. A map built from more than a million women, with real diversity in ancestry, gives researchers a far better place to start. The detective has a shortlist now. The next acts belong to the lab.
This article is for general information only and does not replace advice from a doctor or other qualified health professional.
Key facts at a glance
- What: A genome-wide study in Nature Genetics, published on 20 November 2026, linked 58 regions of the genome to endometriosis, 27 of them newly identified.
- Who: The analysis was led by researchers at the University of Pennsylvania Perelman School of Medicine with funding from the NIH child health institute NICHD.
- Data: The study pooled data from 14 biobanks covering 1,063,657 women, including 50,265 with endometriosis.
- Findings: The authors flagged 314 genes as possible drug targets plus genetic overlap with 135 other conditions or traits, including headache and insomnia.
- Note: for general information only, not medical advice.
FAQ
What did the new endometriosis genetics study find?
It linked 58 regions of the genome to endometriosis, 27 of them new, using data from more than one million women. It also found genetic overlap with conditions such as headache and insomnia.
Is there now a genetic test for endometriosis?
No. The study identifies statistical associations and research targets, not a diagnostic test or a treatment.
Why does ancestry matter in the endometriosis study?
Nearly a third of participants had non-European ancestry, which revealed some risk regions seen only in people of African or admixed American ancestry. According to the authors, most genetic reference data still comes from people of European descent.
Sources
- Multiancestry genome-wide association and integrative multiomics analyses implicate key pathways in endometriosis (Nature Genetics)
- Large NIH-supported study identifies key underlying pathways in endometriosis (NIH)
- Penn Medicine press release on the endometriosis genetics study (EurekAlert)
- Report on the new endometriosis study (LiveNOW from FOX)
- Preprint version of the multiancestry endometriosis study (PubMed Central)
This article was created with AI assistance and is based on the sources listed.





