Stage 4 Colon Cancer and CEA Levels: What You Should Know

Carcinoembryonic antigen, or CEA, is one of the most commonly measured blood markers in people with stage 4 (metastatic) colorectal cancer, and elevated levels at diagnosis are tied to worse survival outcomes. In one large population-based study, patients with stage IV colorectal cancer who had elevated baseline CEA faced a 56% higher risk of dying from the disease compared with those whose CEA was not elevated.¹ But the relationship between CEA and advanced colon cancer is not as straightforward as “high number equals bad news.” Some patients with widespread metastases have perfectly normal CEA, and CEA can spike temporarily during effective treatment, confusing the picture for patients and clinicians alike.

What CEA Actually Is and Why Colorectal Tumors Produce So Much of It

CEA is a protein found on the surface of cells, where it plays a role in how cells stick to one another. It is produced abundantly during fetal development but becomes limited to a few cell types in healthy adults, with the strongest expression in the lining of the colon and rectum.² That background explains why colorectal cancers are the biggest CEA producers of any tumor type. In a tissue microarray study of more than 13,700 tumors spanning 120 cancer types, colorectal adenocarcinomas were positive for CEA roughly 99% of the time, the highest rate observed.²

CEA is not just an innocent bystander leaking off the tumor surface. Research suggests it actively helps cancer cells survive and spread. When scientists knocked down CEA production in colon cancer cells in the lab, metastasis dropped by more than half, and the cells became less able to migrate.³ Separate work showed that reducing CEA inhibited tumor cell clumping by about 70% and increased the rate at which cancer cells self-destructed, suggesting CEA protects tumors from their own programmed cell-death machinery.⁴ CEA also activates molecules involved in cell adhesion, which may help circulating tumor cells latch onto distant organs, particularly the liver.⁵

How CEA Levels Relate to Prognosis in Stage 4 Disease

The standard cutoff used in most labs is 5 ng/mL. A level above that threshold is considered “elevated,” though in stage 4 disease the numbers are often dramatically higher. In a study of colorectal cancer patients stratified by stage, most people with CEA above 5 ng/mL were in stage IV, while those below the cutoff were more often in earlier stages.⁶ That pattern makes intuitive sense: more tumor cells making CEA means more CEA in the bloodstream.

The prognostic hit from an elevated CEA at diagnosis is substantial. One analysis of two large cohorts found that patients categorized as having an abnormal pretreatment CEA had roughly a 70% higher risk of death compared with those whose CEA was normal, regardless of the treatment they received.⁷ In patients specifically with stage IV colon cancer, elevated preoperative CEA has been confirmed as an independent predictor of poor survival alongside factors like tumor grade, primary tumor location, and whether the patient received chemotherapy.⁸ In the population-based study mentioned in the introduction, median cancer-specific survival was 14 months for patients with elevated CEA versus 24 months for those with normal levels.¹

Importantly, CEA is one piece of a larger puzzle. It does not override staging, imaging findings, or molecular testing. A high CEA adds information but does not, on its own, dictate a treatment plan. And about half of all colorectal cancer patients across all stages actually have CEA levels within the normal range.⁶

Using CEA to Track Treatment Response

For patients undergoing chemotherapy for metastatic disease, CEA is typically drawn before each cycle or at regular intervals. The idea is simple: if the treatment is working, the tumor shrinks, and CEA should fall. If the disease progresses, CEA should rise. One study tracking CEA changes during chemotherapy found that shifts in CEA between roughly week 12 and week 18 of treatment reliably mirrored what CT scans showed, and a CEA increase of about 3% or more during that window was independently linked to worse survival.⁹

The catch is that CEA monitoring works well mainly in patients whose CEA was elevated to begin with. When researchers divided patients into those with initially high and initially normal CEA, the sensitivity of CEA for detecting disease progression was 67% in the elevated group but only 20% in those who started with normal levels.¹⁰ If your CEA was never high, tracking it during treatment adds little useful information.

The CEA Surge That Mimics Failure

One of the most anxiety-inducing scenarios for patients is seeing CEA jump in the weeks after starting chemotherapy. It is natural to assume the treatment is not working. But a phenomenon known as a “CEA surge” can cause a temporary spike even when the cancer is responding. The leading explanation is that as chemotherapy kills tumor cells, they release their stored CEA into the blood before the body clears it.

In a study of patients receiving oxaliplatin-based chemotherapy, about 15% experienced a surge, with the median rise reaching roughly 260% above baseline before levels dropped.¹¹ Separate work looking at a larger group found that CEA surges lasted less than four months in every case and were associated with a clinical benefit from treatment, not failure.¹² The same pattern has been documented with irinotecan-based regimens.¹³

The practical takeaway: a CEA rise in the first few months of chemotherapy should not, by itself, trigger a change in treatment. Guidelines generally advise confirming progression with imaging rather than reacting to CEA alone, and a rise lasting less than four months does not meet the formal definition of CEA-based progression.

When CEA Stays Normal Despite Advanced Disease

Not every stage 4 colon cancer pumps out CEA at high levels, and understanding why matters for how much weight you and your medical team should place on the test. About half of colorectal cancer patients across all stages have normal CEA, and even in stage IV, a meaningful fraction will not have elevated levels.

Tumor genetics play a role. Tumors carrying RAS mutations, one of the most common mutation types in colorectal cancer, were associated with a higher likelihood of having a normal CEA at presentation in one study.¹⁴ That finding does not mean the cancer is less aggressive; it means that particular tumor simply makes less CEA protein. In contrast, other work on KRAS-mutant metastatic colorectal cancer found an association between KRAS mutations and higher initial CEA, suggesting the relationship between mutation status and CEA production is not yet fully settled.¹⁵ What is clear is that mutation testing now forms a standard part of the workup for metastatic colorectal cancer, and the tumor’s molecular profile influences both treatment options and how much to trust CEA as a surveillance tool.

Poorly differentiated or mucinous tumors can also behave differently. These subtypes tend to spread in patterns that involve peritoneal metastases rather than liver-only disease, and their CEA production can be unpredictable.¹⁶

CEA Together With CA 19-9

Oncologists frequently order CA 19-9 alongside CEA. CA 19-9 is another blood-based tumor marker, originally associated with pancreatic cancer but also elevated in a portion of colorectal cancer patients. In a study of colorectal cancer patients at various stages, most people diagnosed at stage IV had either CEA alone elevated (about 32%) or both markers elevated (about 42%), whereas those at early stages more commonly had both markers in the normal range.¹⁷

The combination adds value in specific clinical situations. For stage IV patients who undergo surgery to completely remove all visible disease, having both postoperative CEA and CA 19-9 measured together was an independent predictor of recurrence, performing better than either marker alone.¹⁸ Changes in CA 19-9 over time during first-line chemotherapy also correlate with what scans show, adding another data point when imaging results are ambiguous.¹⁹

Where Metastases Spread in Stage 4 and What That Means for CEA

Stage 4 colorectal cancer is not one disease in terms of spread pattern. The liver is the most common site, but where exactly the cancer goes depends partly on where the primary tumor sits. In a large registry analysis, liver-only metastases were found in about 43% of right-sided colon cancer patients, 54% of left-sided colon cancer patients, and 52% of rectal cancer patients.²⁰ Peritoneal spread was far more common in right-sided tumors (about 33%), while lung metastases were most frequent in rectal cancer (about 28%).²⁰

This matters for CEA interpretation because liver metastases tend to be the most reliable driver of elevated blood CEA. When cancer spreads primarily to the peritoneum, standard serum CEA can be less sensitive. Interestingly, measuring CEA directly in peritoneal fluid (rather than blood) appears far more accurate in that specific scenario. One study found that peritoneal CEA had a sensitivity of about 92% and a specificity of about 92% for detecting peritoneal metastases, outperforming both serum CEA and peritoneal CA 19-9.²¹ This is a niche test, done during surgery, but it illustrates how the usefulness of CEA depends on context.

Non-Cancer Reasons CEA Can Be Elevated

If you are being monitored for colorectal cancer and your CEA creeps up, it is worth remembering that CEA is not cancer-specific. Smoking is a well-established cause of elevated CEA in people without cancer. One study found that cigarette smokers had an average CEA of about 9 ng/mL, well above the standard 5 ng/mL cutoff, and levels climbed with the number of cigarettes smoked per day.²² Beyond smoking, conditions like chronic liver disease, kidney failure, inflammatory bowel disease, and other chronic inflammations can all push CEA above normal.²³ For cancer patients who also smoke or have liver or kidney issues, a single elevated CEA reading is harder to interpret, and trends over time matter more than any one measurement.

CEA After Surgery for Stage 4 Disease

A growing number of stage 4 patients undergo surgery with the goal of removing all visible cancer, particularly when metastases are limited to the liver. In these cases, the rate at which CEA drops after surgery carries prognostic meaning. A study looking at patients with elevated preoperative CEA found that those whose CEA fell by roughly 50% or more after curative resection had dramatically better outcomes: estimated five-year overall survival was about 77% in the group with a large drop versus about 31% in those whose CEA declined more slowly.²⁴ The speed and magnitude of the postoperative CEA drop can therefore help guide decisions about how aggressively to pursue additional therapy.

Circulating Tumor DNA as an Alternative to CEA

One of the more promising developments in cancer monitoring is circulating tumor DNA, or ctDNA. These are tiny fragments of DNA shed by the tumor into the bloodstream, and they carry the tumor’s specific genetic mutations, making them potentially more specific than CEA. In a study of patients with metastatic colorectal cancer monitored during chemotherapy, ctDNA was detectable in 97% of patients, compared with 83% who had CEA above 5 ng/mL. More strikingly, ctDNA predicted 80% of progression events detected on CT scans, versus only 30% caught by CEA, and when both markers detected progression, the ctDNA rise came earlier.²⁵

That sounds like ctDNA should replace CEA immediately, but the picture is more nuanced in practice. In patients who had undergone surgery with curative intent, a head-to-head comparison found that ctDNA did not outperform standard imaging in catching recurrences, and the combination of imaging plus CEA measurement actually had a slightly higher sensitivity (about 73%) than ctDNA alone (about 53%).²⁶ The technology is improving rapidly, but for now, ctDNA and CEA are best understood as complementary rather than as replacements for each other. CEA remains cheaper and more widely available, while ctDNA offers higher specificity and the potential to catch certain relapses that CEA misses.

CEA as a Therapeutic Target

Because CEA is produced so heavily and specifically by colorectal tumors, researchers have tried to use it as a homing signal for treatment. One approach is CAR-T therapy, which engineers a patient’s own immune cells to recognize and attack cells displaying CEA on their surface. A phase I dose-escalation trial enrolled 10 patients with CEA-positive metastatic colorectal cancer who had already progressed on prior treatment. Seven of the 10 achieved stable disease after the infusion, two remained stable for more than 30 weeks, and two showed tumor shrinkage on imaging. Most patients saw a decline in serum CEA, and severe adverse events related to the CAR-T cells were not observed.²⁷ This is still very early-stage work, far from standard care, but it illustrates how CEA could eventually shift from passive marker to active therapeutic target.

The Emotional Weight of CEA Numbers

For patients living with stage 4 colorectal cancer, few moments in the treatment cycle carry as much emotional charge as waiting for CEA results. A survey of colorectal cancer patients found a strong link between concern about cancer recurrence and anxiety specifically about CEA test results, and this anxiety was particularly acute in patients with younger children.²⁸ It is worth knowing, though, that more frequent testing does not appear to make anxiety worse. A trial comparing intensified CEA monitoring with standard follow-up schedules found no significant differences in depression, anxiety, or fear of recurrence between the two groups.²⁹ Patients who understand what CEA can and cannot tell them, particularly the surge phenomenon and the limitations in non-secreting tumors, tend to be better equipped to put individual readings in context rather than riding a roller coaster with every blood draw.