Can Endometriosis Cause Cancer?

Can endometriosis cause cancer?

Endometriosis can rarely contribute to a pathway toward cancer, but it is not cancer, and most people with it never develop cancer. The clearest association is with two ovarian cancer subtypes: clear cell and endometrioid ovarian cancers. That association does not mean endometriosis inevitably, or generally directly, causes cancer.

If you’d like a second opinion on your imaging or a recommended treatment, our team at Lotus Endometriosis Institute can review it.

A 2025 review of published clinical and molecular findings in people with endometriosis and gynecologic cancers reported malignant transformation rates of approximately 0.7–2.5%. Malignant transformation means a lesion becomes cancerous. That range depends on the populations examined and how transformation was defined; it is not a personal risk estimate for everyone with endometriosis.

The distinction matters: having a higher risk of a particular cancer and having an endometriosis lesion turn into cancer are related but different questions. Neither means cancer is the expected outcome.

Why endometriosis and cancer can share biological features

Endometriosis can spread, invade nearby tissue, and behave aggressively without being cancer. Some lesions share biological features associated with cancer, but those features alone do not establish malignancy or predict that a lesion will become malignant.

Inflammation, iron and oxidative stress

Endometriosis lesions can create a persistent inflammatory environment. Long-standing inflammation is strongly associated with cancer and may contribute causally to it, although the pathway from endometriosis to cancer remains uncommon.

The local pelvic fluid environment can be iron-rich, with higher levels of reactive oxygen species, molecules that cause oxidative stress. More free iron has been linked with more oxidative stress, which can damage cells over time. This environment is particularly relevant to ovarian endometriomas.

Immune behavior that lets lesions persist

Macrophages are immune cells that can attack threats or support tissue repair. In endometriosis, pelvic macrophages may lean toward an M2-like profile, meaning their behavior favors tissue repair and tolerance rather than clearing the lesions.

This helps explain how lesions can persist despite an immune response. Over time, immune activity that should help protect or repair tissue can instead sustain an inflammatory environment that favors abnormal cell growth.

Mutations in a lesion that still appears benign

A 2019 review of molecular findings in uterine tissue, pelvic endometriosis, and ovarian cancers described driver mutations in genes including PIK3CA, KRAS, and ARID1A. These changes were found in the gland-like epithelial cells of some endometriosis lesions that still appeared benign, as well as in cancers. A driver mutation can encourage cell growth, but finding one does not by itself mean cancer is present.

The supporting cells, called stromal cells, may be free of those mutations while having changes in gene regulation that promote inflammation and resistance to progesterone. Estrogen-driven signals from this supporting tissue can encourage neighboring mutated cells to multiply.

Some lesions therefore combine mutations with an inflammatory, oxidative environment that can rarely promote malignant transformation, particularly in the ovary. The same biological features do not produce the same outcome in every location.

Risk depends on the lesion, pathology and personal history

Risk is not evenly distributed across everyone with endometriosis. Lesion location, microscopic findings, age, family history, and genetics all matter more than a single general percentage.

Atypical pathology is not the same as an unusual-looking implant

Atypical endometriosis diagnosed by a pathologist under the microscope can carry precancerous implications. This describes concerning cellular or tissue changes, not simply an implant’s appearance during surgery.

Dr. Vasilev emphasizes that this is different from a surgeon describing visually atypical implants. White or vesicular, bubble-like implants may look unlike classic brown implants, but they do not carry precancerous implications merely because they look unusual.

If the word “atypical” appears in your records, clarify whether it comes from the pathology report or the surgeon’s visual description. Those meanings should not be treated as interchangeable.

Ovarian, deep and bowel lesions

Ovarian endometriomas receive particular attention because the strongest cancer association concerns ovarian clear cell and endometrioid cancers. Deep infiltrating lesions can also contain driver mutations and substantial inflammation, but those findings do not establish that they will become cancerous.

ARID1A alterations have been found most often in patients with deep infiltrating implants and are being explored as potential tools for sorting patients into different risk groups. Such molecular risk assessment is not yet routine clinical care for most people.

The ovarian-subtype association does not establish that bowel endometriosis causes bowel cancer. Likewise, a stage 4 endometriosis diagnosis is not a cancer stage or a prediction of malignant transformation. The specific lesion and its pathology matter more than the stage label alone.

Family history, genetics and prior surgery

A high-risk family history or inherited cancer predisposition deserves attention, particularly as age increases into the late forties and beyond. Dr. Vasilev stresses that a small percentage still matters across a very large population.

As a hypothetical illustration, a 1% risk applied to one million people would represent 10,000 cases. This is arithmetic to show why a small risk matters, not a measured risk estimate or a count of people known to be at high risk.

Genetic risk also changes how findings should be interpreted. In a 2025 descriptive multicenter cohort study of nearly 17,000 women carrying pathogenic BRCA1 or BRCA2 variants, 2.4% reported an endometriosis diagnosis. Ovarian cancer prevalence was not higher among those reporting endometriosis; in simple comparisons, it was lower, at 10% versus 15%.

That BRCA-carrier study does not show that endometriosis is protective. Self-reported diagnoses, risk-reducing surgery, and other differences between groups could influence the result. It does suggest that the endometriosis-ovarian cancer relationship may differ across genetic risk groups.

A history of hysterectomy alone also does not settle questions about a known endometriosis lesion or ovarian finding. Which organs were removed, whether ovaries or lesions remain, and what the surgical pathology showed all affect the individual assessment.

Distinguishing a familiar endometrioma from a concerning change

An endometrioma is a benign ovarian endometriosis cyst, not automatically a cancerous mass. The practical question is whether a presumed endometrioma still has reassuring features or has changed enough to require a different explanation.

Symptoms and imaging findings can overlap between benign disease and cancer. A previous endometriosis diagnosis should not automatically explain every later change.

Changes in symptoms, bleeding or a cyst

Meaningful changes to include in an assessment are:

• New or rapidly worsening pelvic pain, especially after a period of stability.
• A growing ovarian cyst or an endometrioma that changes in appearance.
• Unexplained weight loss, persistent bloating, early satiety, meaning feeling full unusually quickly, or other ongoing abdominal symptoms.
• Changing abnormal uterine bleeding, particularly heavy bleeding that changes over time or bleeding after menopause.

None of these is a diagnosis of cancer. Fibroids, polyps, and other benign conditions may also explain symptoms, but a persistent change should not simply be attributed to familiar endometriosis.

A clinician’s assessment may include updated imaging, interpretation of tumor markers, and consideration of a specialist referral. The choice depends on the combination of symptoms, imaging, personal history, and prior pathology.

What ultrasound, MRI and PET-CT can add

Ultrasound and MRI help assess a cyst’s appearance and whether it has changed. Some centers can also use quantitative MRI techniques to obtain information beyond a general description of a mass as simple or complex.

A 2025 retrospective, single-center MRI study examined 156 patients with pathology-confirmed benign endometriosis or endometriosis-associated ovarian cancer. It combined intravoxel incoherent motion, or IVIM, with T2 mapping, techniques that measure aspects of water movement, blood flow, and tissue structure. Cancerous lesions tended to have lower diffusion-related and T2 measurements.

The combined model performed better than individual measurements, with an area under the curve of about 0.87, a measure of how well a test separates two groups. Reported sensitivity and specificity were in the low 80% range. This is promising, but it does not make MRI definitive, and the numerical cutoffs are not universal across scanners and protocols.

Dr. Vasilev’s view is that PET-CT is generally not accurate for determining whether cancer is present in endometriosis. PET-CT is often used in ovarian cancer assessment, however, and it may be reasonable to consider in selected cases because cancer can produce a markedly stronger tracer signal than endometriosis.

Imaging can guide how a mass is assessed and managed. Pathology remains the final diagnosis when cancer is in question.

Normal blood tests cannot settle the question

Endometriosis can be very local in its biological effects. Blood inflammatory markers can be lower than those seen in advanced ovarian cancer even while the pelvic fluid shows substantial inflammatory and oxidative activity.

This helps explain why severe symptoms and significant disease can coexist with normal or nonspecific bloodwork. Normal blood tests do not establish that a lesion is harmless, and abnormal results do not automatically establish cancer.

Markers such as CA-125, CA19-9, and HE4 require context. The 2025 gynecologic cancer review emphasizes the limited specificity of commonly used markers and standard imaging. A marker result needs to be interpreted alongside symptoms, imaging findings, and personal history rather than used as a stand-alone answer.

Adenomyosis and uterine cancer are separate questions

The strongest association remains between endometriosis and specific ovarian cancer subtypes. Claims about adenomyosis or uterine cancer are much less certain and should not be inferred from that ovarian cancer connection.

Associations between endometriosis and endometrial cancer have been reported, but they do not establish that endometriosis causes endometrial cancer. Cervical cancer is generally not considered to have a true causal connection with endometriosis.

What adenomyosis tissue studies do and do not show

A 2025 tissue study comparing 21 women with adenomyosis and 14 controls found higher levels of the protein Numb in the uterine lining and muscle of women with adenomyosis. Numb helps regulate cell behavior, and its presence may relate to pathways involved in invasive growth, including Notch signaling.

The same adenomyosis tissue study also analyzed separate cancer datasets, where NUMB alterations were associated with different survival patterns in endometrial cancer. These findings help explain potentially shared cell-growth biology. They do not demonstrate that adenomyosis becomes cancer or provide a reason for new cancer screening.

Coexisting fibroids, polyps and endometrial findings

A 2026 retrospective surgical pathology study examined 1,692 women with endometriosis, adenomyosis, or both who underwent hysterectomy with removal of one or both ovaries and fallopian tubes. Fibroids and polyps commonly coexisted with these conditions. Fibroids were particularly frequent in women with both endometriosis and adenomyosis.

The hysterectomy pathology study found more endometrial cancer diagnoses in the adenomyosis-only group. However, that association did not remain independent after statistical adjustment. It does not support a conclusion that adenomyosis causes cancer, and findings from this selected surgical population should not be treated as everyone’s risk.

For someone with changing pain or bleeding, coexisting conditions remain part of the discussion. Symptoms attributed to adenomyosis or endometriosis may also reflect fibroids, polyps, or other endometrial findings.

Questions to bring to your clinician

Use the appointment to clarify what your own history and findings mean rather than relying on a general statistic. These questions can help make that discussion specific:

• Based on my history and imaging, am I in a higher-risk group because of a long-standing endometrioma, atypical features, or concerning growth?
• If I have an ovarian endometrioma, what changes on ultrasound or MRI would make you concerned?
• Would an MRI add useful information in my case, and does your center use quantitative methods?
• How should we interpret CA-125, CA19-9, or HE4 for me, given that they are not specific?
• Could my symptoms be explained by coexisting issues such as fibroids or polyps, and should we check for them?
• Given the change in my symptoms or cyst, do we need updated imaging?
• Would tumor markers add useful information about this change?
• Is a specialist referral indicated?

What a risk estimate cannot tell you

Finding mutations or immune changes in a lesion helps explain possible pathways, but it does not identify who will develop cancer. Many available datasets come from surgical or referral populations, which can overrepresent more severe or complicated cases.

Biomarkers and imaging models still need validation in other populations and clinical settings, including across different MRI scanners and protocols. Genetics, lesion location and depth, and microscopic atypia can also alter risk, limiting what any one number can tell an individual.

A personal risk discussion should establish what prior pathology showed, whether a lesion is stable, how symptoms have changed, and what follow-up fits those findings. A headline or a single percentage cannot provide that plan.