What Is the CA-125 Test? Which Cancers Cause High Levels, and What Is Normal?
In this article
What is CA-125, and what is it used for?
The key to understanding CA-125 correctly lies in one sentence: this test is a marker of an irritated membrane, not of cancer. When the lining of the abdomen, lungs or uterus is stretched, inflamed or in contact with fluid, it releases this protein, whether cancer is present or not. This single fact explains the test's behavior: why it rises during menstruation, why it increases in heart failure, why it fails as a screening tool, and why it remains valuable in a patient with an established diagnosis.
- The screening question has been settled. In UKCTOCS, which followed 202562 women for 16 years, screening did detect cancer at an earlier stage (stage 1 diagnoses increased by 47.2%), but it did not reduce deaths from ovarian cancer.
- And its harms have been measured. In the PLCO study in the United States, 9.6% of screened women received a false alarm, 1080 underwent surgery, and for approximately every 100 operations, 15 serious complications occurred, all in women who did not have cancer.
- The common “20% increase” rule is wrong. The internationally accepted criterion is that the value rises to at least twice the upper limit of normal and that this is confirmed with a second sample at least one week later.
- Detecting recurrence early did not prolong survival. When patients who began chemotherapy as soon as CA-125 rose were compared with those who waited until symptoms appeared, survival was the same (25.7 versus 27.1 months). Moreover, quality of life deteriorated earlier in the early-treatment group.
- The test's best-established use is an area absent from the old text: deciding whether a woman with a pelvic mass should be referred to a gynecologic oncologist.
- 35 U/mL is not a universal “normal.” The same number does not mean the same thing before and after menopause.
CA-125 (cancer antigen 125) is a large glycoprotein called MUC16 whose main source is a membrane tissue called mesothelium: the peritoneum lining the abdomen, the pleura lining the lungs and the pericardium surrounding the heart. The uterine lining, fallopian tubes and ovarian surface also produce it.
When does this tissue release CA-125? When it is stretched, inflamed or in contact with fluid. In other words, the stimulus is mechanical and inflammatory irritation, not cancer.
This single fact explains all the test's peculiarities: why it rises during menstruation (shedding of the endometrium), in endometriosis and pelvic inflammation, in cirrhosis and heart failure (fluid and stretching of the abdominal lining), why it can be very high in tuberculous peritonitis, and why it rises in ovarian cancer—because that cancer also irritates the same membrane. The test detects the membrane's distress, not cancer.
The Screening Question: Two Large Studies, Two Clear Answers
This is the study that settled the issue. 202562 postmenopausal women were divided into three groups: annual CA-125 testing, interpreted not against a fixed threshold but using the ROCA algorithm, which follows each person's own value trajectory, plus ultrasound when necessary; ultrasound alone; and no screening. Median follow-up was 16.3 years.
Screening did its job: in the screened group, stage 1 diagnoses increased by 47.2% (95% CI 19.7-81.1), while stage 4 diagnoses decreased by 24.5%. In other words, the disease truly was detected earlier.
But the outcome did not change: 1206 deaths in total: 296 in the multimodal screening arm, 291 in the ultrasound arm and 619 in the no-screening arm. No significant reduction (p=0.58 and p=0.36). In the authors' words: “Screening did not significantly reduce deaths from ovarian and fallopian tube cancer.”
It is important to understand why this result matters so much: detecting cancer early and saving lives are not the same thing. Some ovarian tumors detected at an early stage were slow-growing anyway, while fast-growing tumors slipped through between annual screens.
In the study conducted in the United States, groups of approximately 39000 women each were compared. In the screening group, CA-125 was measured annually for six years, this time using a fixed threshold of 35 U/mL, and ultrasound was performed for four years.
- Result: ovarian cancer deaths were 4.4 versus 3.8 per 10000 person-years, a relative risk of 1.18 (0.98-1.4). No benefit.
- Harm: 3285 women received a false alarm (9.6%); of these, 1080 underwent surgery and 163 experienced a serious complication, approximately 15 for every 100 operations. None of these women had cancer.
The two studies must be considered together. PLCO used a crude threshold and failed. One might say, “Then the threshold was wrong.” But UKCTOCS used a much smarter algorithm, ROCA, which follows a person's own trajectory, and it failed too. The problem therefore lies not in the threshold, but in what the test measures.
CA-125 testing to screen for ovarian cancer in women without symptoms is not recommended. The United States Preventive Services Task Force reaches the same conclusion: the harms of screening outweigh the benefits, and screening does not reduce ovarian cancer deaths.
For carriers of BRCA1/2 gene alterations, the situation is no different: the effectiveness of screening with CA-125 and vaginal ultrasound has not been established. The intervention that genuinely reduces risk in these people is preventive surgery (removal of the ovaries and fallopian tubes), which reduces risk by approximately 90%. The timing of surgery should be planned with genetic counseling, considering individual risk and the effects of early menopause.
The Test's Actual Performance
In one study, CA-125 was elevated in approximately 6% of benign diseases and in 28.5% of nongynecologic cancers. An elevated result therefore tells you neither the source nor whether it is malignant.
One Sentence to Remember: a normal CA-125 does not rule out ovarian cancer, and an elevated CA-125 does not prove cancer.
35 U/mL Is Not a Universal “Normal”
35 U/mL is a statistical threshold below which approximately 99% of the healthy population falls, not a “health boundary.” And its meaning changes with menopausal status:
| Before menopause | After menopause | |
|---|---|---|
| Why the difference? | Menstruation, endometriosis, fibroids, ovarian cysts and pelvic inflammation all increase CA-125 and are common in this age group | Most of these benign causes are no longer present |
| Specificity | The specificity of CA-125 alone is only ~55%, meaning that a large proportion of elevated results occur without cancer | Substantially higher |
| In Practice | The referral thresholds used are higher (200 U/mL in some sources) | Values above 35 U/mL are taken more seriously |
During menstruation, CA-125 can rise approximately threefold. If a premenopausal woman is to have the test, it should be performed outside menstruation; otherwise, a completely physiological elevation may trigger a series of unnecessary investigations.
Best-Established Use: Assessing a Pelvic Mass
This is CA-125's most firmly established use, although it is almost absent from older texts. The question is: should a woman with a pelvic mass be operated on by a general gynecologist or referred to a gynecologic oncology center? This distinction matters because ovarian cancer outcomes are better when surgery is performed at an experienced center.
| Method | How does it work? |
|---|---|
| RMI (Risk of Malignancy Index) | Multiplies three pieces of information: ultrasound findings × menopausal status × CA-125. Menopausal status enters the formula directly (a coefficient of 1 before menopause and 3 after menopause), so the same CA-125 value is weighted three times more heavily after menopause. |
| ROMA score | CA-125 + HE4 + menopausal status. The two are used together because HE4 is less misleading where CA-125 is weakest (endometriosis and benign cysts). Different thresholds apply before and after menopause. |
| Guideline Approach | The guideline in England: in a woman with symptoms, ultrasound is performed if CA-125 is ≥35; if RMI is above 250, the patient is referred to a specialist multidisciplinary team. American gynecologic societies use referral thresholds of 200 before menopause and 35 U/mL after menopause, together with findings such as residual fluid, a fixed nodular mass and family history. |
Notice that none of these approaches uses CA-125 on its own. Each puts the test in context: ultrasound findings, menopausal status and the presence of symptoms. In every approach, the test is ordered for a woman with symptoms, never for screening. This is precisely where CA-125 is useful: the presence of a mass is already known, and the test only helps decide where the patient with that mass should be referred.
Treatment and Follow-up: The Correct Criteria
The statement “a CA-125 increase of more than 20% is significant” circulates in Turkish sources. This is not a recognized criterion. The criteria accepted by the international gynecologic cancer group are:
- Disease progression: CA-125 rising to at least twice the upper limit of normal or, if it has never returned to normal, reaching twice its lowest value (nadir). This must be confirmed with a second sample taken at least one week later.
- Treatment response: In a patient whose baseline value is at least twice the upper limit of normal, a decrease of at least 50%, maintained for at least 28 days.
Confirmation is as important as the number itself. Decisions are not made on a single high measurement, because CA-125 may rise with infection, inflammation or other temporary causes.
If CA-125 is monitored in a patient who has completed ovarian cancer treatment and achieved a complete response, and it begins to rise, should chemotherapy start immediately or should treatment wait until symptoms appear?
This question was tested in a randomized study. When CA-125 was confirmed to have risen to twice the upper limit of normal, 529 patients were assigned either to immediate treatment or to waiting until clinical symptoms appeared:
- Overall survival: HR 0.98 (0.80-1.20), p=0.85: no difference.
- Median survival in the early-treatment arm: 25.7 months, compared with 27.1 months in the waiting arm, numerically slightly worse with early treatment.
- Quality of life deteriorated earlier in the early-treatment arm because patients started chemotherapy while they still felt well.
The authors' conclusion: the benefit of routine CA-125 monitoring during follow-up after a complete response has not been established.
What This Means for the Patient: a rise in your CA-125 during follow-up is not automatically an emergency. Decisions consider your symptoms, imaging findings and preferences together. Some patients prefer to know and start early; others choose to preserve their time without symptoms. Both approaches can be defended, and knowing this allows patients to participate in their own decisions.
Noncancerous Causes of Elevated CA-125
| Cause | Explanation |
|---|---|
| Menstruation | May increase the value approximately threefold. The test should not be performed during this period. |
| Pregnancy (the first three months) and the postpartum period | Physiological. |
| Endometriosis | One of the most common and misleading causes. A large or leaking endometrioma may produce very high values. |
| Fibroids and benign ovarian cysts | Common causes. |
| Pelvic inflammatory disease | Inflammation directly irritates the membrane. |
| Liver cirrhosis and fluid in the abdomen (ascites) | Contact with fluid and stretching of the abdominal lining. |
| Heart failure | See below: this is an instructive section that most sources omit. |
| Tuberculous peritonitis | ⚠ Particularly important in Turkey. Tuberculosis of the abdominal lining can, in its clinical presentation, imaging findings and CA-125 levels, closely mimic advanced ovarian cancer, with fluid in the abdomen, thickening of the abdominal lining and very high CA-125 (values above 1000 have been reported). This diagnosis should be considered before proceeding to surgery. |
| Diverticulitis | Intra-abdominal inflammation. |
The clearest indication that CA-125 is not a “cancer marker” is its place in the cardiology literature. Fluid accumulating in heart failure stretches the abdominal and lung linings; inflammatory signals (IL-6, TNF-α) also contribute, and mesothelial cells release CA-125. In these patients, therefore, CA-125 acts as a marker of congestion (fluid overload).
This is so consistent that it has been tested: in a study of 380 patients, CA-125-guided treatment improved a composite endpoint compared with standard care. This improvement was mainly due to fewer hospital readmissions; there was no effect on mortality. This use is not yet included in guidelines.
The Main Message for the Reader: in a patient with heart failure, high CA-125 may mean fluid overload rather than cancer. This single example explains why “CA-125 = cancer” is wrong better than all the other explanations.
CA-125 in Other Cancers
Besides ovarian cancer, cancers in which CA-125 may rise include fallopian tube, uterine (endometrial), colorectal, breast, pancreatic, stomach and lung cancers. However, “it can rise” and “it is used in that cancer” are different things.
In Uterine (Endometrial) Cancer CA-125 elevations have been reported, and very high values may be associated with spread beyond the uterus. However, we could not access current guideline texts on the extent to which it should be used to predict spread before surgery or during follow-up; we therefore make no definite recommendation here.
In other cancers, CA-125 is not recommended as a diagnostic or follow-up tool; these cancers have their own markers (such as CEA in colorectal cancer and CA 19-9 in pancreatic cancer).
Multicancer early detection tests designed to identify several cancers early with a single blood test are being developed. Some studies have reported high sensitivity for ovarian cancer, but these results come from studies comparing people with cancer and healthy people, not from a real screening setting. This distinction is critical: it is precisely what UKCTOCS taught us.
The large screening study conducted in England was completed in 2024, and results on its effect on mortality are awaited. The consensus in the field is that these tests are not yet ready for routine screening.
Frequently Asked Questions
- Should I Have CA-125 Tested During a Checkup?
- No. In the study following 202562 women for 16 years, screening detected cancer at an earlier stage but did not reduce deaths. Its harms, however, were measured: in the study in the United States, 9.6% of screened women received a false alarm, 1080 underwent surgery, and approximately 15 serious complications occurred for every 100 operations, all in women without cancer. This test is not recommended for women without symptoms.
- My CA-125 Is 60. Do I Have Ovarian Cancer?
- Most likely not, but context is decisive. Before menopause, this value may result from endometriosis, fibroids, an ovarian cyst, pelvic inflammation or menstruation; in this age group, the specificity of CA-125 alone is only ~55%. The first steps are to repeat the test outside menstruation and have an ultrasound. Decisions are based not on CA-125 alone, but on ultrasound findings, menopausal status and symptoms together.
- My CA-125 Is Normal. Does That Definitely Mean I Do Not Have Ovarian Cancer?
- No. CA-125 is normal in approximately half of early-stage ovarian cancers and generally does not rise at all in mucinous cancers. A normal result is not a reason to stop evaluating a woman with persistent symptoms such as bloating, abdominal or pelvic pain, early fullness or frequent urination.
- My Mother/Sister Has Ovarian Cancer. Can I Be Monitored With CA-125?
- No, and this is one of the most frequently misunderstood points. Even in BRCA1/2 carriers, the effectiveness of screening with CA-125 and ultrasound has not been established. If there is a familial risk, the correct step is not a blood test but genetic counseling; the intervention that genuinely reduces risk is preventive surgery (reducing risk by approximately 90%). These decisions should be made together, including discussion of the effects of early menopause.
- My CA-125 Has Risen During Follow-up After Treatment. Should I Start Chemotherapy Immediately?
- This is not automatically an emergency, and the question has been tested in a randomized study. Between patients who began treatment as soon as CA-125 rose and those who waited until symptoms appeared, no survival difference was found (25.7 versus 27.1 months); moreover, in the early-treatment group, quality of life deteriorated earlier. Nor are decisions made on a single high measurement: the accepted criterion is a rise to twice the upper limit of normal, confirmed with a second sample at least one week later. The decision should consider your symptoms, imaging findings and preferences together.
- I Have Endometriosis, and My CA-125 Is Always High. What Should I Do?
- Endometriosis is one of the most common and misleading causes of elevated CA-125; large or leaking endometriomas can produce very high values. The test has limited value for follow-up in this situation: your disease is monitored with imaging and clinical assessment. Your doctor assesses whether a new rise is significant by comparing it with your own known baseline.
- Do I Need to Fast Before the Test?
- No fasting or other preparation is needed. The only practical warning is not to have the test during menstruation, when the value may rise approximately threefold and trigger unnecessary investigations.
Glossary
- CA-125 (MUC16)
- A large glycoprotein produced by membrane (mesothelial) cells. It is not specific to cancer; it responds to irritation.
- Mesothelium
- The layer of cells lining the abdominal cavity, lungs and heart. It is the main source of CA-125.
- ROCA algorithm
- A method that follows changes in a person's own CA-125 trajectory rather than using a fixed threshold. Used in UKCTOCS.
- RMI (Risk of Malignancy Index)
- A score combining ultrasound findings, menopausal status and CA-125. Used to guide referral decisions for pelvic masses.
- ROMA score
- A risk score combining CA-125 and HE4 with menopausal status.
- HE4
- A second marker used alongside CA-125; less misleading in benign conditions such as endometriosis.
- Nadir
- The lowest value reached after treatment. Subsequent rises are interpreted against it.
- Upper limit of normal (ULN)
- The upper value of the reference range. The recurrence criterion is twice this value.
- Specificity
- The proportion of people without the disease in whom the test remains normal. CA-125 has low specificity in premenopausal women.
- False positive
- An elevated test result despite the absence of disease. This is the main harm of screening.
- Stage shift
- Detecting disease at an earlier stage through screening. It does not in itself mean that lives were saved, which is precisely what UKCTOCS showed.
DROZDOGAN Academy Commentary
CA-125 is among the tumor markers with the most supporting data, yet it is one of the most misused tests. Interestingly, its limitations were not guessed: they were measured in a study involving two hundred thousand women, and the result was entirely counterintuitive.
⚡ Why Does This Matter?
Because ovarian cancer causes symptoms late, it is among the cancers women fear most. This fear fuels the search for a “simple blood test to detect it early.” CA-125 is not that test, and this is no longer an opinion but a measured result. Yet calling it a bad test would also be wrong: it works really well when the right question is asked. The problem is not the quality of the tool, but the job it is used for.
✅ Strengths / What Is Improving
There is a clarity rarely seen in this field. UKCTOCS settled the question with 16 years of follow-up; PLCO quantified the harms; OV05 tested the monitoring strategy. All three were randomized, so we have evidence rather than opinion. In pelvic mass assessment, tools such as RMI and ROMA make CA-125 genuinely useful by placing it in context rather than leaving it on its own. GCIG criteria also require a confirmatory sample when defining response and progression, preventing decisions based on a single measurement.
⚠️ Limitations / What Should Not Be Overstated
The test is normal in approximately half of early-stage cancers and does not rise in the mucinous type. Its specificity before menopause is 55%, almost a coin toss. The most unsettling UKCTOCS finding is that screening did its job (stage 1 diagnoses increased by 47%), but the outcome did not change. This shows that the assumption “early diagnosis always saves lives” does not hold for this disease. The most frequently repeated error in the field—“a 20% increase is significant”—appears in no accepted criterion.
🩺 Our Position in Practice
- We do not order CA-125 for women without symptoms, including BRCA carriers. If there is familial risk, we refer them to genetic counseling, not the laboratory.
- We do not test premenopausal women during menstruation and do not make decisions on a single high value.
- We do not interpret CA-125 alone in a pelvic mass; we use it alongside ultrasound findings and menopausal status to decide on referral.
- We use a confirmed doubling as the criterion for recurrence, not percentage increases.
- We do not treat a rising CA-125 during follow-up as an automatic reason to start treatment. We share the OV05 result with the patient, because this is one of the rare decisions in which patients can exercise their own preference.
- We do not turn away a woman with persistent symptoms on the basis of a normal CA-125.
- When we see abdominal fluid and high CA-125, we consider tuberculous peritonitis; in Turkey, this is a question that should be asked before surgery.
❓ Open Questions
UKCTOCS said “this test does not work,” not “no test can work.” Multicancer early detection tests and circulating tumor DNA will reopen this question, but the data so far come from comparisons of patients and healthy people rather than real screening settings. UKCTOCS taught us exactly how important this distinction is. We could not access current guideline texts on the role of CA-125 before surgery and during follow-up in uterine cancer. And a question specific to Turkey: how many women in this country undergo unnecessary oncologic surgery because of tuberculous peritonitis?
- Corrections: The statement “a CA-125 increase of more than 20% is considered significant” has been removed. The accepted criterion is that the value rises to twice the upper limit of normal (or the nadir) and is confirmed with a second sample at least one week later. The treatment response criterion has also been added (a ≥50% reduction maintained for at least 28 days). It is noted that 35 U/mL is not a universal normal and that interpretation varies with menopausal status. “More than 80% in advanced stages” is presented alongside the ~82% figure for all epithelial ovarian cancers. It is also noted that CA-125 generally does not rise in the mucinous type.
- Additions: UKCTOCS and PLCO results and measured harms of screening; the OV05/EORTC 55955 trial and follow-up debate; pelvic mass assessment (RMI, ROMA, guideline thresholds); low specificity before menopause; avoiding testing during menstruation; tuberculous peritonitis; CA-125 in heart failure; and preventive surgery rather than screening in BRCA carriers.
- Figures Not Given Because They Could Not Be Verified: The average early-detection lead time reported in OV05 and numerical quality-of-life results (the direction was verified, but not the magnitude). The behavior of CA-125 in clear cell ovarian cancer. The highest values reached in tuberculous peritonitis and endometrioma (no numbers given because these were at case-report level). Recommended age ranges for preventive surgery in BRCA carriers, because the guideline text could not be accessed. The role of CA-125 before surgery and during follow-up in uterine cancer. Exact ROMA score thresholds, which vary by the test used.
- Nature of the Sources: Numerical PLCO data were taken from the National Cancer Institute's summary because the primary publication could not be accessed. Referral thresholds from American gynecologic societies were cited from a clinical review rather than the primary guideline document.
- Turkey: No verifiable data, guideline or national screening program text specific to Turkey could be found; therefore, no numerical claims about Turkey are made. No information is provided on SGK reimbursement.
- Internal links are partly retained from the existing article and partly tentative; they should be checked before publication.
This article is for information only and does not replace personalized medical advice. Your doctor should interpret test results together with your symptoms and imaging findings.
References
- Menon U, Gentry-Maharaj A, Burnell M, et al. Ovarian cancer population screening and mortality after long-term follow-up in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS): a randomised controlled trial. Lancet. 2021;397(10290):2182-2193.
- Buys SS, Partridge E, Black A, et al. Effect of screening on ovarian cancer mortality: the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Randomized Controlled Trial. JAMA. 2011. (Data cited from the National Cancer Institute PDQ summary.)
- Rustin GJS, van der Burg MEL, Griffin CL, et al. Early versus delayed treatment of relapsed ovarian cancer (MRC OV05/EORTC 55955): a randomised trial. Lancet. 2010;376(9747):1155-1163.
- Rustin GJS, Vergote I, Eisenhauer E, et al. Definitions for response and progression in ovarian cancer clinical trials incorporating RECIST 1.1 and CA 125 agreed by the Gynecological Cancer Intergroup (GCIG). Int J Gynecol Cancer.
- US Preventive Services Task Force. Screening for ovarian cancer: recommendation statement.
- National Institute for Health and Care Excellence. Ovarian cancer: recognition and initial management. NICE CG122.
- Jacobs I, Oram D, Fairbanks J, et al. A risk of malignancy index incorporating CA 125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer. Br J Obstet Gynaecol.
- Moore RG, McMeekin DS, Brown AK, et al. A novel multiple marker bioassay utilizing HE4 and CA125 for the prediction of ovarian cancer in patients with a pelvic mass (ROMA). Gynecol Oncol.
- CA-125 (Cancer Antigen 125). StatPearls [Internet].
- National Cancer Institute. PDQ — Genetics of Breast and Gynecologic Cancers; Ovarian, Fallopian Tube, and Primary Peritoneal Cancer Screening.
- Núñez J, et al. CA125-guided therapy in acute heart failure (CHANCE-HF): a randomized study.
Sources
2. Screening for ovarian cancer: US Preventive Services Task Force Recommendation Statement. (2018). https://jamanetwork.com/journals/jama/fullarticle/2672638
Was this article helpful?
Your feedback helps us improve our content.
Reader comments
Share your comment
Your name and comment are published only after administrator approval. Your email address is never shown to visitors.
Be the first to comment.