Can a tumor marker test alone diagnose cancer?

No. A raised tumor marker does not mean someone has cancer, and the National Cancer Institute says so plainly. Marker results are combined with other tests, such as biopsies or imaging, to make a diagnosis.

A tumor marker is anything made by cancer cells, or by other cells reacting to cancer, that tells you something about a cancer. Traditionally these are proteins produced in higher amounts by cancer cells and measurable in blood, urine, stool, tumor tissue, or other fluids. Increasingly they are genomic markers too, including tumor gene mutations and changes in tumor DNA.

The two ways a single number misleads

The test can be wrong in both directions, which is the whole reason it cannot stand alone.

A high result without cancer. Noncancerous conditions can raise marker levels. A number outside the reference range does not identify what pushed it there.

A normal result with cancer. NCI notes that not everyone with a given cancer has a raised level of the marker linked to that cancer. A reassuring number is not proof of anything.

What markers are genuinely good at

Diagnosis is the weakest use. The strong uses come after a diagnosis exists.

Markers help identify the type and stage of a cancer, and they contribute to an estimate of prognosis. They indicate which treatment may work: markers that show whether you are a candidate for a particular targeted therapy are often called biomarkers for cancer treatment, and they are usually measured in tumor tissue.

Repeat measurements taken during treatment, sometimes called serial measurements, show whether the tumor is responding. Measurements taken after treatment ends can check for recurrence. A trend across several draws carries far more information than any single value.

Tumors also shed cells and biological material into the blood, which is what a liquid biopsy measures.

Why they are poor screening tests

This is the part worth understanding before paying for a test out of pocket.

Researchers hoped circulating tumor markers could screen people without symptoms. Studies have generally found they do not work well for that. They are often not sensitive enough, meaning they miss people who do have the disease. And they are often not specific enough, meaning they flag people who do not.

Multi-cancer detection tests, or MCDs, are the current attempt at solving this. Most look at DNA that tumor cells release into the blood, and some also analyze proteins. Several are already being marketed.

NCI is careful here, and you should be too. Much remains to be learned about how best to use these tests and about their harms and benefits. The critical unanswered question is whether treating the cancers an MCD finds would actually reduce deaths from those cancers. NCI is launching a clinical trial to find out.

Being marketed is not the same as being proven.

What to do with a result you already have

If a marker came back high, ask three things: what noncancerous conditions can raise this one, what the next test would be, and whether a single reading or a trend is being used.

If a marker is being used to track treatment, ask what your baseline was and what size of change would prompt a scan or a change in plan.

All tumor marker tests, including those developed inside a single clinical laboratory, are run in labs meeting standards set by the Clinical Laboratory Improvement Amendments program. Our guide to tumor markers lists the common ones by cancer type, and the piece on how cancer is diagnosed explains where biopsies fit.

Sources

• https://www.cancer.gov/about-cancer/diagnosis-staging/diagnosis/tumor-markers-fact-sheet
• https://www.cancer.gov/about-cancer/diagnosis-staging/diagnosis