Tumor Marker Result of 33: How to Read CEA, CA125, CA19-9, and CA15-3 Reference Ranges, Non-Cancer Fluctuation, and Serial Trends With Imaging

From a cancer support group: today’s blood test shows a tumor marker of 33—is that high?

A member shared that a surveillance blood test showed a tumor marker of 33 and asked whether that counts as high.
This worry is common after treatment or during follow-up, because a number on a lab slip is often read as a verdict on recurrence.

The short answer is that 33 cannot be interpreted without the marker name, and a single result should not decide whether disease has returned.
If the focus is colorectal cancer and CEA after treatment, see
Stage 3–4 colorectal cancer with a normal CEA after treatment

1. Is 33 high? Name the marker first

Direct answer: 33 does not mean the same thing for every test.
CA125 often uses an upper limit near 35 units, while CEA in most people stays below about 5 units, so a CEA of 33 is clearly above the usual cutoff.
A CA125 or CA15-3 near 33 may still sit near that lab’s reference range.

What each marker is used for, and why ranges differ

National Academy of Clinical Biochemistry guidance
(Sturgeon 2008)
supports CEA for colorectal cancer follow-up, CA15-3 or CEA for monitoring advanced breast cancer, and CA125 for ovarian cancer monitoring.
ASCO gastrointestinal guidance (Locker 2006)
recommends postoperative CEA every 3 months for at least 3 years in stage II–III disease if further treatment is an option, and states that rising serial CA19-9 in pancreatic cancer needs confirmation with other studies.

Numbers from the NORIP reference study

In 498 healthy people from the NORIP project
(Bjerner 2008),
the CA125 upper limit was 35.8 kU/L, CA19-9 in secretors was 28.3 kU/L, and MUC1/CA15-3 rose with age and BMI, for example 31.7 kU/L at age 40 and BMI 24, and 37.5 kU/L at age 70 and BMI 24.
CEA rose with age and smoking; nonsmokers had upper limits of 3.59 µg/L at age 50 and 4.12 at age 70.
So a CEA of 33 is not a mild rise, while a CA125 of 33 may sit near the upper limit.

Lab reference ranges can differ

Quality guidance on tumor marker use (Sturgeon 2008, quality)
stresses standardized methods, the lab’s own reference interval, and caution when switching assays, because numbers can shift even when disease does not.
For long-term follow-up, use the same lab and method when possible.

2. Non-cancer reasons a marker can move

Direct answer: Tumor markers are not exclusive to cancer cells.
Inflammation, cysts, liver disease, biliary obstruction, postoperative healing and smoking can raise levels without recurrence.

CA125

European Group on Tumor Markers guidance
(Sölétormos 2016)
advises against screening asymptomatic women with CA125 alone, because early-stage sensitivity is about 50–62% and specificity about 94–98.5%.
Endometriosis, pelvic inflammation and other peritoneal processes can raise CA125, so results need symptoms and imaging context.

CEA

NORIP confirmed that smoking and age raise CEA. Duffy concluded that circulating markers are not useful for detecting early breast cancer because sensitivity and specificity are too limited
(Duffy 2006).
Bowel inflammation, liver disease and smoking should be asked about whenever CEA moves.

CA19-9

A patient with gallstones and acute cholangitis had a CA19-9 of 9,586 U/mL that fell quickly after treating the obstruction
(Korkmaz 2010).
Even very high values can come from biliary blockage or pancreatitis. ASCO therefore advises confirming rising serial CA19-9 with other studies.

CA15-3

Duffy notes that CA15-3 is most useful for monitoring advanced breast cancer that cannot be followed easily with imaging,
and that a rise may precede imaging by about 2–9 months, though the clinical value of that lead time is not always clear. Starting or changing treatment still depends on the oncologist and the whole clinical picture.

3. Why serial trends matter more than one number

Direct answer: A single value has a high false-alarm rate. Looking at direction across draws separates recurrence better,
and still needs imaging and symptoms.

Half-life and rise rate

Bidart and colleagues (Bidart 1999)
summarized that marker half-life after therapy and doubling or rise rate help distinguish benign fluctuation from metastatic disease better than one point,
while kidney or liver disease and sampling timing still matter.

Evidence from the FACS CEA analyses

The FACS randomized trial in 1,202 people after curative-intent colorectal cancer treatment
(Primrose 2014)
found that CEA or CT follow-up found more recurrences treatable with curative intent than minimum follow-up.
Secondary analysis of 582 people in the CEA arms (Shinkins 2017)
found a single CEA test had an AUC of 0.74; at 5 µg/L, sensitivity was about 50%, and about 4 in 10 people without recurrence had at least one false alarm.
Decisions based on a serial trend performed better (AUC 0.85).

Bar length reflects AUC from Shinkins 2017, not disease-resolution rates.

Questions to ask when you see 33

• Which marker is this: CEA, CA125, CA19-9, CA15-3 or another
• What was my previous value, and was the same lab or assay used
• Has it risen across several draws, or is this one bump
• Do I need CT, MRI or PET now, and when is the next visit

4. Imaging and the role of blood numbers

Direct answer: Tumor markers are alerts and monitoring tools, not scans.
Confirming recurrence or treatment response usually needs CT, MRI or sometimes PET together with symptoms and examination.

Why imaging still matters

ASCO advises confirmation when CA19-9 rises in series. Sturgeon supports CEA within a treatment plan, and FACS showed roles for both CEA and CT in finding operable recurrence.
Extra scans without an indication do not always mean more safety and can add radiation and worry. Let the treating team choose the timing.

What a blood number cannot do

It does not replace diagnosis, locate a mass, or alone justify starting a new drug.
For overall timing of progression, see
How fast cancer moves from stage 1 to 2 to 3

5. Easing worry while waiting for results

Direct answer: Fear of recurrence is understandable.
What helps is knowing the marker name, comparing your own baseline, noting symptoms and asking about the next test plan—not searching lab numbers online every night.

While waiting

Limit late-night reopening of results, schedule a talk with the oncology team and use the question list above.
If worry disrupts sleep, eating or daily life, tell the oncology team or psycho-oncology staff; that is part of care, not something to endure alone.
For supplements in remission, see
Dietary supplements after cancer remission

Table: reference ranges and reading a value of 33

Ranges are examples from studies and guidelines. Use the reference interval printed on your own lab report.

Table: temporary fluctuation versus needing more tests

Use this to prepare for a clinic visit, not to diagnose yourself.

Table: comparing follow-up options

Choices depend on cancer type, stage, prior treatment and the treating team.

Common beliefs versus the evidence

CA125, CA19-9 and CA15-3 all use a cutoff of 35

They are in a similar neighborhood, but not identical across labs. CA125 is often near 35; CA19-9 and CA15-3 ranges differ and can depend on assay, age or BMI.

A CEA of 33 is only a mild to moderate rise

Usual CEA cutoffs are near 5 units, so 33 is clearly elevated and needs clinical review, not a pass.

One result can prove recurrence

It cannot. CEA analyses show serial trends discriminate better than a single value, and imaging plus symptoms still matter.

Any small rise means an immediate PET scan

Not for everyone. Clinicians may repeat the blood test or choose CT or MRI based on the cancer type and risk.

Still within the reference range means you are safe

A value still inside the printed range can matter if it keeps rising from your own baseline. That is why clinicians watch personal trends.

Frequently asked questions (FAQ)

Is a tumor marker of 33 very high?

It depends on the marker. A CA125 near 35 may sit near the upper limit. A CEA with a usual cutoff near 5 units is clearly elevated at 33. In every case, compare with your prior values and ask your clinician; do not decide from one number.

How do CA125, CA19-9, CA15-3 and CEA reference ranges differ?

CA125 is often near 35 units. CA19-9 is often near 37 or lab-specific. CA15-3 is often in the low-to-mid 30s by age and BMI. CEA is often under about 5 units and higher in smokers.

What non-cancer factors can raise a marker to around 33?

Inflammation, cysts, endometriosis, biliary stones or cholangitis, liver disease, postoperative healing, smoking and assay differences can move values without recurrence.

Why do clinicians look at serial trends?

Single values false-alarm often. FACS-linked CEA analyses found serial trends discriminate better than one result, and imaging is still used when indicated.

When are CT, MRI or PET needed?

When values rise across draws, new symptoms appear, or a planned surveillance scan is due. PET is not required for every small one-time bump.

Should I stay with the same laboratory?

Yes when possible. Different assays can give different numbers even if disease is unchanged.

My result is still inside the reference range but higher than before. Should I worry?

It can matter, especially if the rise continues. Bring the report so your clinician can compare it with your own baseline and symptoms.

How can I ease worry while waiting?

Write down the marker name, prior values and questions for clinic, limit late-night rechecking of results, and use the oncology clinic contact channel. If worry disrupts sleep or eating, tell the team.

Scientific mechanism

Glycoproteins and oncofetal antigens: CA125, CA19-9 and CA15-3 are glycoproteins that many epithelial tissues can release during inflammation or cell turnover.
CEA is an oncofetal antigen that can reappear in gut mucosa with inflammation or cancer, so it is not a cancer-only on/off switch
(Sturgeon 2008,
Duffy 2006).

Kinetics in blood: After treatment, levels fall according to half-life. Persistent producing tissue slows the fall or drives a later rise.
Rate of rise over time carries more information than one point, but kidney or liver disease and sampling intervals still matter
(Bidart 1999).

Laboratory error modes: Heterophilic antibodies, hook effects and assay differences can distort numbers.
Using the same method, the same lab and cumulative reports supports safer long-term monitoring
(Sturgeon 2008, quality).

Compiled and reviewed by Asst. Prof. Dr. Norawit Raatpiboon

Related articles