CA 15-3 Blood Test: What It Measures and Why Levels Rise

CA 15-3 is a blood-based tumor marker most closely associated with breast cancer, used primarily to track how advanced disease responds to treatment and to watch for recurrence after initial therapy. It is not a screening test and performs poorly at catching breast cancer in its early stages. The marker’s real clinical value emerges once cancer has spread, where rising or falling levels can signal whether treatment is working or the disease is progressing. Understanding what CA 15-3 can and cannot tell you matters, because misinterpreting a single result leads to unnecessary anxiety or, worse, premature changes in treatment.

What CA 15-3 Actually Measures

CA 15-3 stands for Cancer Antigen 15-3. It is a fragment of a protein called MUC1, which sits on the surface of many types of cells, including those lining the breast ducts. When breast cancer cells multiply and break down, they shed pieces of MUC1 into the bloodstream, where a standard blood test can detect them. The more tumor activity occurring in the body, the more of this protein tends to circulate, which is why the test becomes useful when cancer is advanced and producing large amounts of the antigen.

Most labs report CA 15-3 results in units per milliliter (U/mL). The commonly used upper limit of normal is around 25 to 30 U/mL, though this can vary slightly between laboratories depending on the assay they use. One large study defined normal as 0 to 30 U/mL, while another noted that patients with stage 0 through stage II breast cancer still had values within the normal range of 0 to 25 U/mL.¹² This slight disagreement on the cutoff is worth knowing about: if your lab report says “normal up to 25” and you read elsewhere that it should be 30, neither is wrong. Different assays and reference populations explain the gap.

Why CA 15-3 Is Not a Screening Tool

One of the most persistent misconceptions about CA 15-3 is that it can be used to detect breast cancer early. It cannot. The marker’s fundamental limitation is that blood levels are rarely elevated in patients with early or localized disease.³ In one study of breast cancer patients without metastases, the sensitivity of CA 15-3 was just about 6%, meaning roughly 94 out of every 100 early-stage patients had normal values.⁴ A test that misses that many cases is simply not useful for finding cancer before it has spread.

Researchers have tried to improve on this. One approach involves looking not just at the protein backbone of CA 15-3 (which is what the standard test detects) but at its sugar-chain patterns, since cancer cells often attach unusual sugar structures to the protein. An experimental assay using a lectin called Concanavalin A to detect these abnormal sugar chains showed promise in distinguishing breast cancer patients from healthy controls more effectively than the standard test.⁵ This work is still in early stages, but it illustrates why the standard CA 15-3 test falls short: it measures the quantity of the protein without accounting for how cancer changes its molecular appearance.

Where It Becomes Useful: Monitoring Advanced Disease

The story changes once breast cancer has metastasized. In patients with confirmed spread to distant sites, CA 15-3 sensitivity jumps dramatically, reaching roughly 83% in one study.⁶ At that stage, the test becomes a practical way to gauge whether chemotherapy, hormonal therapy, or other treatments are having an effect. A steady decline in CA 15-3 over the course of treatment generally indicates the tumor burden is shrinking. A persistent rise after the initial cycles of therapy, on the other hand, suggests the disease is not responding well.

In advanced breast cancer, one study found that an increase in CA 15-3 of more than 15 U/mL after the second cycle of treatment correlated with shorter progression-free survival and poorer clinical outcomes.⁷ This makes the trend over time far more informative than any single reading. Oncologists almost never make decisions based on one CA 15-3 value in isolation; they look at a series of measurements over weeks or months.

The Spike That Fools Everyone

If you are being treated for breast cancer and your CA 15-3 suddenly shoots up in the first few weeks of a new therapy, you might understandably panic. But a temporary spike shortly after starting treatment is a well-documented phenomenon, and it does not necessarily mean the cancer is getting worse. Dying tumor cells release their contents into the blood, and that burst of cellular debris can transiently push CA 15-3 levels up before they come back down.

Research on patients with bone metastases, for example, showed that CA 15-3 displayed a significant but transient elevation around 30 days after beginning systemic treatment, returning to pretreatment levels by day 60.⁸ Another analysis described this “flaring” as a rise of 125% or more over baseline in the first four to six weeks of treatment, and warned that this phenomenon has been linked to premature, unnecessary changes in treatment regimen when clinicians mistakenly interpret the spike as disease progression.⁹ The practical takeaway: if your oncologist sees a spike in the first month or so, the standard approach is to wait and retest rather than abandon a potentially effective therapy.

Tracking Recurrence After Early-Stage Treatment

After surgery and adjuvant therapy for early-stage breast cancer, patients naturally worry about the cancer coming back. Serial CA 15-3 testing during follow-up visits is one way to look for recurrence, and there is real data suggesting it can provide a useful early signal. In a large retrospective study of over 700 patients, a rising CA 15-3 was the first indicator leading to the diagnosis of metastatic recurrence in about 37% of cases. Patients whose relapse was discovered through a CA 15-3 rise had a longer median overall survival compared to those whose recurrence was detected by other means: roughly 35 months versus 22 months.¹⁰

That survival difference sounds encouraging, and it is, but it comes with an important caveat. Detecting recurrence earlier through a blood test does not automatically mean the patient will live longer. It could reflect lead-time bias, meaning the test finds the recurrence sooner, so the clock starts ticking earlier, without actually changing the outcome. A combined analysis of seven international trials concluded that whether routine use of CA 15-3 and similar biomarkers genuinely improves overall survival remains an open question.¹¹

This uncertainty is reflected in major clinical guidelines. Both the European Society for Medical Oncology (ESMO) and the American Society of Clinical Oncology (ASCO) discourage routine serial CA 15-3 measurement in the follow-up of early breast cancer patients, citing insufficient evidence that it produces a survival benefit.¹² Many oncologists still order the test periodically, often because patients request it, but it is not part of the standard recommended follow-up protocol for early-stage disease. If your doctor does order it, know that a normal result is reassuring but not a guarantee, and a mildly elevated result does not necessarily mean the cancer is back.

How CA 15-3 Compares With Other Breast Cancer Markers

CA 15-3 is not the only blood-based tumor marker used in breast cancer. CEA (carcinoembryonic antigen) and CA 27.29 are two others you may encounter on a lab report. CA 27.29 detects a slightly different region of the same MUC1 protein that CA 15-3 targets, so the two tests tend to perform similarly. In one head-to-head comparison, CA 27.29 and CA 15-3 had nearly identical sensitivity for metastatic breast cancer (about 86% and 83%, respectively), and both outperformed CEA, which came in at roughly 63%.¹³ In practice, most labs offer one or the other rather than both, and your oncologist will interpret whichever one is available.

CEA is more commonly associated with colon cancer but is sometimes ordered alongside CA 15-3 for breast cancer monitoring, particularly in advanced disease. Using the two markers together can provide a slightly fuller picture, since a rise in either one can signal trouble. However, neither CEA nor CA 27.29 solves the fundamental problem of low sensitivity in early-stage disease. All three markers share the same basic limitation: they only become reliably elevated when there is a substantial tumor burden.

CA 15-3 and Tumor Subtypes

Breast cancer is not one disease, and CA 15-3 does not behave identically across all its molecular subtypes. A study examining the relationship between tumor markers and breast cancer subtypes found that elevated CA 15-3 was significantly more common in estrogen-receptor-negative tumors compared to estrogen-receptor-positive ones (about 22% versus 10%). Elevated CEA, by contrast, was more strongly associated with HER2-positive tumors.¹⁴ Both markers correlated with larger tumor size and more advanced lymph node involvement, which fits with the general pattern that CA 15-3 rises as tumor burden increases. But the subtype-specific differences matter: if you have a hormone-receptor-positive breast cancer, your CA 15-3 may be less likely to be elevated even if the disease is progressing, and your oncologist would lean more heavily on imaging and clinical exams.

Not Just Breast Cancer: Other Reasons CA 15-3 Can Be Elevated

One of the trickiest aspects of interpreting CA 15-3 is that breast cancer is far from the only thing that raises it. Other cancers, benign conditions, and even normal physiological states can push levels above the standard cutoff.

Other Cancers

Because MUC1 is expressed on many types of epithelial cells throughout the body, cancers in other organs can produce elevated CA 15-3. Early research found that levels above 40 U/mL occurred in about 46% of ovarian cancers, 30% of liver cancers, and 26% of lung cancers.¹⁵ A separate study using a lower cutoff of 22 U/mL reported even higher rates: about 71% of bronchogenic (lung) carcinoma patients, 66% of those with epithelial ovarian cancer, and 44% of gastrointestinal cancer patients had elevated CA 15-3.¹⁶ This cross-reactivity means an unexpectedly high CA 15-3 in someone without known breast cancer warrants a broad evaluation, not an automatic assumption of breast cancer.

Benign and Autoimmune Conditions

You do not even need to have cancer for CA 15-3 to be elevated. Among patients with benign diseases, about 3% had levels above 40 U/mL in one large series. The conditions most likely to cause these false positives included chronic hepatitis (over 40% of those patients had elevated levels), liver cirrhosis, sarcoidosis, tuberculosis, and systemic lupus erythematosus.¹⁷ Rheumatoid arthritis patients with interstitial lung disease were another notable group: about 30% had CA 15-3 above the upper limit, with no evidence of any malignancy on thorough workup.¹⁸ If you have a known autoimmune or liver condition, an elevated CA 15-3 is far less alarming than it might otherwise be.

Pregnancy

Pregnancy is another common cause of elevated CA 15-3 that can catch people off guard, especially breast cancer survivors who are watching their markers closely. CA 15-3 levels rise naturally during pregnancy, particularly in the third trimester. In one study, the median level in the third trimester was 26.0 U/mL compared to 14.0 U/mL in the first trimester, and over 37% of pregnant women had a value above the standard normal cutoff at some point during their pregnancy. These levels typically normalized after delivery, and the elevations did not correlate with pregnancy complications.¹⁹ For breast cancer survivors who become pregnant, this is critical information. A rising CA 15-3 in the third trimester does not automatically signal recurrence; it may simply be a normal pregnancy effect.

Race, Ethnicity, and Individual Variation

Baseline CA 15-3 levels are not the same across all populations, and this has practical implications for how the test is interpreted. A study of postmenopausal women in the United States found that non-Hispanic Black women had higher average CA 15-3 levels than non-Hispanic White women. The researchers concluded that race and ethnicity should be considered when assigning thresholds for diagnostic or screening purposes.²⁰

Other factors affect baseline levels too. An analysis of women without breast cancer found that perimenopausal age, a history of salpingectomy, and longer oral contraceptive use were associated with lower CA 15-3, while Black ethnicity, higher ferritin levels, and a history of endometriosis were associated with higher levels.²¹ These associations suggest that using a single fixed cutoff for everyone may lead to more false positives in some groups and more false negatives in others. There is no widely adopted race-adjusted or age-adjusted reference range yet, but awareness that the “normal” range has real variability from person to person helps explain why a result of 32 U/mL in one individual might mean nothing at all while the same number in another could warrant follow-up.

How CA 15-3 Fits Into the Bigger Diagnostic Picture

A prospective study tracking patients with localized breast cancer after treatment found that while CA 15-3 was fairly specific (meaning an elevated result usually pointed to something real), its sensitivity for detecting first relapse was poor, and its positive predictive value was especially low in patients with a relatively good prognosis. The researchers concluded that the test is not suitable as a standalone tool for breast cancer follow-up.²² This is the consistent theme across the research: CA 15-3 is a supporting player, not a lead actor.

In practice, oncologists use CA 15-3 as one piece of a larger puzzle that includes imaging (CT scans, PET scans, bone scans, MRI), physical examination, and symptom evaluation. A rising CA 15-3 on its own may prompt further investigation with imaging, but it would not typically trigger a change in treatment without confirmation. And a normal CA 15-3 does not rule out active disease, especially in early stages.

For patients living with metastatic breast cancer, the test takes on a more prominent role. When imaging is difficult to interpret, such as with bone-only metastases that do not always change in appearance on scans even when responding to treatment, CA 15-3 trends can provide a clearer signal of whether therapy is working. In that context, the trend line over multiple blood draws becomes one of the most practical tools available to the oncologist and patient together.

Pregnancy Monitoring After Breast Cancer

Breast cancer survivors who become pregnant face a unique monitoring challenge. Standard surveillance includes periodic tumor marker checks, and a climbing CA 15-3 during pregnancy can create real clinical dilemmas. Since more than a third of pregnant women exceed the normal threshold without any malignancy, clinicians generally advise caution in interpreting isolated elevations during pregnancy. The recommended approach is to track the trend, note whether the values return to normal within a few weeks after delivery, and reserve aggressive workup for cases where levels continue rising postpartum or where other clinical findings are concerning. This is one of those areas where the test can cause more harm than good if taken at face value without context.