Skip to content
Prof. Dr. Mustafa Özdoğan · Medical Oncologist

What Is the CA 15-3 Test? Its Importance in Breast Cancer

What Is the CA 15-3 Test? Its Importance in Breast Cancer

What is the CA 15-3 test? Its true role in breast cancer

At the heart of this article is a sentence that older texts almost never include: CA 15-3 measurement is not recommended in an asymptomatic woman who has completed breast cancer treatment. This is not an opinion; it is a finding tested in two studies that randomized 2,563 women in 1994 and reaffirmed in the 2026 guideline. The reason is instructive: finding recurrence earlier is not the same as treating it better.

In brief
  • The guideline's exact wording in 2026 is: "ASCO does not recommend routine complete blood counts, chemistry panels or tumor markers such as CEA, CA 15-3 and CA 27.29 for follow-up in asymptomatic patients after breast cancer treatment."
  • This recommendation is evidence-based. In two Italian randomized studies published in 1994, intensive follow-up did indeed find recurrences earlier (112 versus 71 isolated metastases)— but did not change 5-year mortality (18.6% versus 19.5%). Moreover, quality of life did not improve either.
  • The test does not work well in early-stage disease anyway: in early-stage breast cancer, fewer than half of cases have elevated CA 15-3. Even in advanced disease, in 20-25% of patients, the tumor does not secrete this protein at all.
  • A point that needs correcting: CA 15-3 and CA 27.29 measure the same protein (MUC1), but examine different regions and are not numerically interchangeable.
  • And an important warning: unlike other markers, an early CA 15-3 rise after treatment begins should not be considered an innocent fluctuation —a study associated this rise with the risk of disease progression.
  • New-generation tests face the same obstacle: circulating tumor DNA (ctDNA) can detect recurrence, but the benefit of starting early treatment on this basis has not yet been demonstrated —and the guideline does not recommend it outside clinical research.
What is CA 15-3? And how does it differ from CA 27.29?

CA 15-3 measures fragments in blood of MUC1, a large glycoprotein produced by breast tissue cells. Healthy breast cells also produce it; cancer cells secrete more, and more enters the bloodstream.

Its relationship to CA 27.29 is often confused. Both measure the same protein (MUC1), but use antibodies binding to different but overlapping regions of the protein—the sequence examined by CA 15-3 and the one examined by CA 27.29 are shifted by three amino acids. Their measurement methods also differ.

The practical consequence: "CA 15-3 and CA 27.29 are the same, so either can be measured instead of the other" is only half true. They measure the same molecule, but their numbers cannot be compared directly. The same applies to different laboratories: a patient under follow-up should, if possible, always have measurements taken at the same laboratory using the same method. The difference is mostly due to calibration, making it unclear whether a "rise" in a series is real or reflects a change of laboratory.

Early-stage follow-up: what does the evidence say?

The wording of the 2026 guideline

The American Society of Clinical Oncology's current breast cancer follow-up guideline published in 2026 states:

"ASCO does not recommend routine complete blood counts, chemistry panels or tumor markers such as CEA, CA 15-3 and CA 27.29 for follow-up in asymptomatic patients after breast cancer treatment."

The same guideline also does not recommend routine CT, PET, bone scintigraphy, liver ultrasound or chest X-rays in asymptomatic patients (97% of the panel agreed on this). Circulating tumor DNA is also not recommended for recurrence monitoring outside clinical research.

What follow-up is recommended? Medical history, physical examination and mammography. In other words, follow-up after a woman's surgery uses discussion, examination and mammography, not blood tests.

Why? The two studies of 1994

This recommendation is often questioned because it is counterintuitive: "Why would finding it earlier not be better?" Two large Italian randomized studies tested this question in 1994, with a clear answer.

The GIVIO study The Rosselli Del Turco study
Number of patients 1,320 (26 centers) 1,243
What was tested? Intensive follow-up (bone scintigraphy, liver ultrasound, chest X-rays and blood tests) versus clinical follow-up Intensive follow-up (chest X-rays and bone scintigraphy) versus clinical follow-up
Was recurrence found earlier? — Yes: 112 versus 71 isolated metastases
Survival 132 deaths (20%) versus 122 (18%)— no difference 5-year mortality 18.6% versus 19.5%—no difference
Quality of life No difference at months 6, 12, 24 and 60 —
Two lessons from these two tables

1. "It found recurrence earlier but did not change the outcome." The second study is the clearest example: in the intensive follow-up group, more isolated metastases were found (112 versus 71)—but mortality remained unchanged. Patients therefore lived longer knowing about their disease; they did not live longer. In medicine, this is called lead-time bias (lead-time bias).

2. The "but it provides reassurance" argument was also tested and not confirmed. The GIVIO study measured quality of life at four separate time points and found no difference at any of them. Intensive follow-up neither prolonged life nor provided peace of mind.

A current systematic review including these two studies (5 randomized studies, 4,023 women) reached the same conclusion: no difference in overall survival, disease-free survival or quality of life.

What is new in 2026: risk-based follow-up rather than the same follow-up for everyone

The most important change in the current guideline is the replacement of a uniform follow-up schedule with a three-level follow-up model whose intensity varies with risk (low, intermediate and high intensity). Patients with low and high recurrence risk will therefore no longer be seen at the same frequency.

However, the guideline itself gives an honest warning: for comparisons of follow-up intensity by risk level, there are no randomized studies; this three-level model was developed by expert consensus. It is therefore a sensible recommendation, not a proven superiority.

So is the test never used? The situation in metastatic disease

A defensible use for CA 15-3 is in patients with widespread (metastatic) breast cancer whose marker was already elevated at diagnosis, to monitor their treatment response.

The guideline's measured wording here

The guideline on biomarker use in metastatic breast cancer states that CA 15-3 and CA 27.29 "may be used as adjunct assessments"; however, it says there are no data supporting their use alone and treatment-change decisions must be based on clinical assessment and imaging.

A noteworthy detail: the guideline rates even this relatively permissive recommendation as "insufficient evidence quality, informal consensus". Even the statement "CA 15-3 is measured in metastatic disease" therefore rests on practical consensus, rather than strong evidence.

What this means for the patient: a single rise in CA 15-3 is not a reason to change treatment. Decisions consider symptoms, examination and imaging together.

Test performance

Rate of elevated CA 15-3 in breast cancer
These are approximate values; they are given as approximations because the primary source texts could not be accessed.
Metastatic (widespread) disease
~80%
Early-stage disease
below 50%
In metastatic disease, tumors that do not secrete the protein at all
20-25%
The bottom row is the test's most important limitation: in one of every four or five patients with advanced breast cancer, the tumor does not secrete this protein at all. CA 15-3 monitoring is meaningless from the outset in these patients. The test can therefore be used for monitoring only in patients in whom it was shown to be elevated at diagnosis.

The normal level is generally considered below 30 U/mL, but this limit varies by method and laboratory—a threshold of 40 U/mL was used in some studies.

An important warning: a rise at the start of treatment

A "temporary rise" in this marker should not be considered innocent

In some tumor markers (CEA, CA 19-9), a temporary rise at treatment initiation is attributed to tumor-cell breakdown and is generally unimportant. It is not correct to say the same for CA 15-3.

In a study of 122 patients, the proportion with a CA 15-3 rise after starting treatment was approximately 25%; however, this rise was found to be associated with a higher risk of disease progression (OR 2.62; 95% CI 1.45-4.72).

The practical consequence: a CA 15-3 rise at treatment initiation should not be dismissed as "temporary anyway." The measurement should be repeated and assessed together with imaging. We could not find verified data on the typical timing and duration of this rise.

Noncancerous causes of elevation

In a series of 1,220 people, CA 15-3 was elevated in only 3.1% of benign diseases (at a threshold of 40 U/mL), so this is not common. However, the rate rises substantially in certain diseases:

Condition Rate of elevation
Chronic hepatitis 42.9%
Sarcoidosis 16.7%
Liver cirrhosis 13.3%
Tuberculosis 9.7%
Systemic lupus erythematosus 6.7%
All benign diseases (average) 3.1%
Particularly important for Turkey: liver diseases

The table's most striking row is the 42.9% rate in chronic hepatitis—meaning CA 15-3 may be elevated in two of every five people with chronic hepatitis. The rate in cirrhosis is also 13.3%.

Given that chronic viral hepatitis is relatively common in Turkey, a liver-related cause must be considered when CA 15-3 rises in a patient undergoing breast cancer follow-up. This simple question may prevent unnecessary chains of imaging and anxiety.

Note: We could not find verified data on how much these benign conditions raise CA 15-3—for example, whether it exceeds 100 U/mL—so we do not give an upper limit.

CA 15-3 may also rise in cancers outside the breast (lung, ovarian, pancreatic, colorectal), but it is not used as a diagnostic or monitoring tool in these cancers.

The next generation: circulating tumor DNA faces the same obstacle

Circulating tumor DNA (ctDNA) is expected to replace tumor markers. It has been shown to detect recurrence months before it becomes visible on imaging. But does this change the outcome?

The ZEST study: the field's most instructive failure

In a study testing early treatment initiation in ctDNA-positive patients, 1,901 patients were screened. The result:

  • Only 147 (7.7%) were ctDNA-positive.
  • The study was terminated early because too few patients could be recruited— only 40 patients could be randomized.
  • And the most striking finding: among 98 patients positive on the first test, 51 (55%) already had recurrence visible on imaging.

Thus, more than half of the patients whom ctDNA was presumed to have "detected before imaging" already had visible disease.

In another study of triple-negative breast cancer, the 12-month detection rate ranged from 23.7-29.0%; the researchers note that interventional studies are still needed. Other studies continue.

The same lesson, a second time

Placed alongside 1994, this table reveals a clear pattern. Thirty-two years ago, intensive follow-up found more recurrences but did not change outcomes. Today, ctDNA detects earlier, but the benefit of using it to guide treatment has not yet been demonstrated.

The common reason is the same: finding an untreatable disease earlier is not the same as treating it better. The field has learned this lesson twice—and that does not mean ctDNA will not work; it only shows that the difference between detection and benefit must be measured. That measurement is still being made.

Frequently asked questions

My breast cancer treatment is finished. Should I have CA 15-3 checked regularly?
No—the current guideline does not recommend this. Routine tumor markers (CEA, CA 15-3, CA 27.29) are not recommended in asymptomatic patients, nor are routine CT, PET or bone scintigraphy. Recommended follow-up is medical history, physical examination and mammography. This is not an arbitrary recommendation: two randomized studies in 1994 tested intensive follow-up—recurrences were found earlier, but mortality did not change and quality of life did not improve.
But would finding it earlier not be better?
This is the most counterintuitive but best-tested point. In a study of 1,243 patients, the intensive follow-up group found more isolated metastases (112 versus 71)— yet 5-year mortality remained unchanged (18.6% versus 19.5%). Patients lived knowing about their disease for longer; they did not live longer. The "it reassures me" argument was also measured: quality-of-life assessments at four separate time points found no difference.
My CA 15-3 is normal. Does that mean there is no cancer?
No. In early-stage breast cancer, fewer than half of cases have elevated CA 15-3. Moreover, even in advanced disease, in 20-25% of patients, the tumor does not secrete this protein at all. A normal result neither excludes cancer nor makes follow-up unnecessary—and cannot justify stopping evaluation in a patient with symptoms.
My CA 15-3 rose. Has the cancer recurred?
A single elevated value does not show this. First, repeat the measurement, then exclude noncancerous causes —particularly liver diseases: CA 15-3 may be elevated in 42.9% of people with chronic hepatitis and 13.3% of those with cirrhosis. Measurement at a different laboratory can also create a false "rise." Decisions consider symptoms, examination and imaging together.
Can I have CA 27.29 measured instead of CA 15-3?
Both measure the same protein (MUC1), but examine different regions and their numbers are not interchangeable. For a patient under follow-up, the main rule is: always the same test, always at the same laboratory. Otherwise, it is impossible to distinguish a real "rise" in a series from a change in method.
I started treatment and my CA 15-3 rose. Is it temporary?
Assuming that "a temporary rise is innocent" with this marker is not safe. In a study of 122 patients, a rise at treatment initiation occurred in approximately one-quarter, but was found to be associated with a higher progression risk (OR 2.62). It should therefore not be dismissed; the measurement should be repeated and assessed with imaging.
Should I have a circulating tumor DNA (ctDNA) test?
It is not recommended for recurrence monitoring outside clinical research. The test can detect recurrence, but a benefit from starting early treatment on this basis has not yet been demonstrated. A large study testing this was terminated early because too few patients could be found, and revealed a striking finding: among ctDNA-positive patients, 55% already had recurrence visible on imaging. The field is still investigating this question.

Glossary

CA 15-3 / MUC1
A test measuring fragments in blood of a large glycoprotein (MUC1) produced by breast cells. It is not cancer-specific.
CA 27.29
A test measuring a different region of the same protein. It is not numerically interchangeable with CA 15-3.
Lead-time bias
An apparent increase in survival because a disease is detected earlier—although lifespan has not changed, only the date of diagnosis has moved earlier.
Intensive follow-up
Regular blood tests and imaging in an asymptomatic patient. Randomized studies in breast cancer have tested it and failed to demonstrate a survival benefit.
ctDNA (circulating tumor DNA)
DNA fragments entering the blood from tumor cells. It may detect recurrence early; the benefit of using it to guide treatment is not yet proven.
Molecular residual disease (MRD)
A very small amount of disease remaining in the body after treatment, undetectable by imaging.
OR (odds ratio)
A measure of how many times a factor changes risk. If the confidence interval includes 1, the result is not significant.
Calibration
Adjusting a laboratory test against reference values. The main source of differences between laboratories.

DROZDOGAN Academy Commentary

This article is really more about a habit of thinking than patient information. The story of CA 15-3 is the best-documented example of the most common logical mistake in oncology: assuming detection and improvement are the same thing.

⚡ Why does this subject matter?

What a woman who has finished breast cancer treatment most wants is a checkup that will detect recurrence early. This wish is entirely understandable—and that is precisely why it was tested. The results showed that it could not be fulfilled: intensive follow-up found recurrences earlier, did not change mortality, and did not even provide peace of mind. Nevertheless, routine CA 15-3 is still requested in practice. The cost is more than an unnecessary test: each elevated value can trigger a chain of imaging and weeks of anxiety.

✅ Strengths / positive developments

There is rare clarity in this area: the question was asked in randomized studies in 1994, confirmed by systematic review in 2016, and reaffirmed in the 2026 guideline. The answer has not changed in thirty-two years. The 2026 guideline's three-level, risk-based follow-up model is also a sensible advance—and explicitly stating that it has "no randomized evidence, expert consensus" enhances the guideline's credibility. The test's role in metastatic disease is also positioned realistically.

⚠️ Limitations / things not to overstate

We could not obtain the test-performance figures (approximately 80%, below 50%, 20-25%) from primary sources, so they had to be given as approximations. The upper limit of CA 15-3 elevations in benign disease is unknown. The timing and duration of a rise at treatment initiation are also uncertain. And an important point: imaging and treatment available in the 1994 studies differed from today. This does not invalidate the findings but does not entirely close the question of whether the same results would occur with today's treatments. Indeed, ctDNA studies are asking that question again.

🩺 Our position in practice

  1. We do not request routine tumor markers in early-stage follow-up —and explain why to patients. Explaining the difference between "we are not requesting it" and "we do not care" is the most critical part of this conversation.
  2. We put symptoms ahead of blood tests. New, persistent bone pain, shortness of breath or neurological findings are investigated despite normal CA 15-3.
  3. In metastatic patients, we monitor the test only if elevated at diagnosis —because in one-quarter of patients, the tumor does not secrete this protein at all.
  4. We do not change treatment on the basis of a single value. The guideline itself says the same.
  5. Always the same laboratory, always the same test. We do not use CA 15-3 and CA 27.29 interchangeably.
  6. When a value rises, we ask about the liver first. The 42.9% rate in chronic hepatitis represents a possibility that cannot be neglected in Turkey.
  7. We do not dismiss a rise at the start of treatment as "temporary" —the data do not support that for this marker.

❓ Open questions

Would the 1994 studies produce the same results if repeated with today's treatments? With improving interventions for oligometastatic disease, this question has become relevant again—and ctDNA studies are testing precisely that. Whether the three-level, risk-based follow-up model genuinely makes a difference has not yet been tested in randomized studies either. And a Turkey-specific question remains: how often are unnecessary tumor markers requested in early-stage follow-up? No study has measured this.

Our brief verdict: CA 15-3 becomes reliable only when disease is widespread—so it cannot do precisely what patients expect of it. This expectation was tested in 2,563 women in 1994: intensive follow-up found more recurrences, did not change mortality and did not improve quality of life. Thirty-two years later, the guideline still says the same thing. And ctDNA is now facing the same obstacle. The lesson is that finding an untreatable disease earlier is not treating it better —and this lesson has been learned twice in thirty-two years.
Transparency note
  • Additions (the article's main rationale): The recommendation against tumor markers in early-stage follow-up and the guideline's exact wording. The two randomized studies and systematic review underlying this recommendation. The guideline's measured wording in metastatic disease and the fact that even this recommendation is rated as "insufficient evidence quality." The 2026 risk-based follow-up model. Why an early treatment-related rise in this marker should not be considered innocent. Rates for benign causes. The current status of ctDNA.
  • Corrections: The statement "CA 15-3 and CA 27.29 have the same function and measure the same protein" was completed—they measure different regions of the same protein and are not numerically interchangeable. The older text's statement that "TPS antigen is the best candidate for predicting recurrence" was not used: it comes from a single-center study covering a narrow time period and included in no guideline. Statements about differences related to ethnicity, fibroid history and oral contraceptive use were also removed: the confidence interval of the relevant comparison includes zero (so it is not significant), and the other claims rely not on a peer-reviewed publication but on a conference abstract.
  • Items given as approximations or omitted because they could not be verified: Stage-specific sensitivity rates (the primary source text could not be accessed, so approximations were given). The highest values reached in benign diseases. The timing and duration of a rise at treatment initiation. CA 15-3 behavior in pregnancy, breastfeeding and kidney failure (found only in secondary sources). The texts of ESMO and NCCN follow-up recommendations could not be accessed, so no opinions were attributed to these organizations.
  • Guideline version: The recommendations reported are based on the current follow-up guideline published in 2026 and the guideline on biomarker use in metastatic disease.
  • Turkey: No verifiable Turkey-specific clinical data were found. No information about SGK reimbursement has been provided.
  • Internal links were partly retained from the existing article and are partly estimated; they should be checked before publication.

This article is for information and does not replace individualized medical advice. Your oncologist should determine your follow-up plan, considering your disease stage and risk level.

References

  1. Nahleh Z, Lustberg M, et al. Follow-Up Care and Surveillance for Survivors of Breast Cancer: ASCO Guideline. J Clin Oncol. 2026. doi:10.1200/JCO-26-01700
  2. The GIVIO Investigators. Impact of follow-up testing on survival and health-related quality of life in breast cancer patients: a multicenter randomized controlled trial. JAMA. 1994;271(20):1587-1592.
  3. Rosselli Del Turco M, Palli D, Cariddi A, et al. Intensive diagnostic follow-up after treatment of primary breast cancer: a randomized trial. JAMA. 1994;271(20):1593-1597.
  4. Moschetti I, Cinquini M, Lambertini M, et al. Follow-up strategies for women treated for early breast cancer. Cochrane Database Syst Rev. 2016.
  5. Van Poznak C, Somerfield MR, Bast RC, et al. Use of biomarkers to guide decisions on systemic therapy for women with metastatic breast cancer: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2015;33(24):2695-2704.
  6. MUC1 gene-derived glycoprotein assays for monitoring breast cancer (CA 15-3, CA 27.29, BR): are they measuring the same antigen? Arch Pathol Lab Med. 2004;128(10):1131-1135.
  7. Transient early increase in CA 15-3 during systemic therapy for metastatic breast cancer. Breast Cancer Res Treat. 2009.
  8. CA 15-3 levels in benign diseases. (Series of 1,220 people.)
  9. ZEST study: ctDNA-guided niraparib in patients with molecular residual disease. (SABCS 2024 presentation and related assessments.)
  10. Turner NC, et al. c-TRAK TN: ctDNA surveillance and intervention in early-stage triple-negative breast cancer.

Sources

Selchick F. and Sissons B. (08.10.2021). What to know about the CA 15-3 test and breast cancer.
Medical News Today: www.medicalnewstoday.com

Appointment request

DROZDOGAN AI

Explore this article together

Ask about the article and read answers alongside published sources from our site.

DROZDOGAN AI

Drozdogan Assistant

What Is the CA 15-3 Test? Its Importance in Breast Cancer

Ask about the article and read answers alongside published sources from our site.

Conversation and privacy information

Drozdogan Assistant answers this conversation. Authorized site administrators may review it for quality. Do not share personal information.

Replies use OpenAI infrastructure. Questions and replies are processed by that service; conversation records are stored on our site. Do not share identifying, contact or personal health information.

0 / 1000

For information only; no personal diagnosis or treatment advice.

Related articles

Explore other articles on this topic.

Stay informed

Receive new articles and scientific updates by email.

Google Analytics runs on public pages independently of this window and may use analytics cookies. Your choice here controls external services only.

Cookie policy