Tumour Marker Blood Tests (AFP, CEA, CA-125, CA 19-9): What They Can and Can't Tell You

It's one of the most tempting boxes to tick on a health-check menu: a "cancer blood test." A single tube, a few markers — AFP, CEA, CA-125, CA 19-9 — and the reassuring promise of catching cancer early. These tests are real and genuinely valuable, but not for the job they're usually sold for. Using them to screen healthy, symptom-free people is something the major cancer authorities specifically advise against, because it tends to cause more harm than good. This is a plain-language, evidence-based guide to what tumour markers actually do — general education, not a diagnosis, and certainly not a substitute for a conversation with a doctor about your own risk.

What tumour markers are actually for

A tumour marker is a substance — usually a protein — that can be higher when a cancer is present. The crucial point is when they're useful. The US National Cancer Institute is explicit that, for screening, studies have "generally found that these markers do not work well": they miss people who have cancer (not sensitive enough) and flag people who don't (not specific enough) ¹.

Their established, valuable roles are different ¹:

• Monitoring a known cancer — tracking whether treatment is working.
• Watching for recurrence after treatment.
• Helping with diagnosis in someone who already has symptoms or an abnormal scan — alongside imaging and biopsy, never alone.

In other words, they shine after a cancer is on the radar — not as a first-line "do I have cancer?" check in a well person.

Why screening healthy people backfires

The problem is in the two kinds of error, and both do damage:

• False positives. Many of these markers rise in completely benign situations. A raised result in a healthy person triggers anxiety, then a cascade of scans and invasive follow-up — and for ovarian screening, sometimes unnecessary surgery to remove ovaries, with real surgical risks ⁴.
• False negatives. A normal marker can give false reassurance to someone who does have a cancer, potentially delaying care.

The clearest evidence comes from ovarian cancer. In UKCTOCS, a trial of over 200,000 women, annual CA-125-based screening did not reduce ovarian-cancer deaths — mortality was the same whether women were screened or not ³. On that evidence, the US Preventive Services Task Force gives ovarian screening of average-risk women a Grade D — recommend against — citing no mortality benefit and real harms ⁴. A test can find more early cancers and still not save lives, if it also sets off enough false alarms.

The four markers, briefly

• AFP (alpha-fetoprotein) — used with ultrasound for 6-monthly liver-cancer surveillance in high-risk people, notably those with cirrhosis — not the general public ¹. Also a germ-cell (testicular) tumour marker.
• CEA — a colorectal-cancer monitoring test, not a screen; it's elevated in only about 30% of early-stage disease, and runs higher in smokers ⁶.
• CA-125 — an ovarian-cancer monitoring marker that rises in many benign conditions — menstruation, endometriosis, fibroids, pregnancy — so its value for finding early cancer in healthy women is poor ³.
• CA 19-9 — a pancreatic-cancer monitoring marker with a striking blind spot: 5–10% of people can't produce it at all (they lack the Lewis antigen), so it can read normal even with advanced cancer, and it also rises in benign biliary and pancreatic conditions ⁵.

So what should you do?

The useful message isn't "never test" — it's "test for the right reason." If you have a specific, genuine risk, there are proper, guideline-based pathways: someone with cirrhosis or chronic hepatitis B or C has a real liver-cancer risk and benefits from structured surveillance (ultrasound plus AFP), and someone with a strong family history or a hereditary cancer syndrome has tailored screening options. The right move is to ask a doctor whether you fall into one of those groups — not to buy a marker panel off a menu and try to interpret it yourself.

It's also worth knowing that the much-publicised multi-cancer early-detection blood tests (such as Galleri) are a different, still-investigational category — none is approved for routine screening, and they're being tested in large trials now ¹. They aren't the same as the single tumour markers above, and aren't a reason to change what you do today.

What we see at the clinic

People arrive in Pattaya fairly often with a health-package printout showing a slightly raised CEA or CA 19-9 and understandable fear that it means cancer. In the great majority, once it's read in context — a smoker's CEA, a benign reason for a CA 19-9, a premenopausal woman's CA-125 — it resolves into reassurance, sometimes after a sensible confirmatory step. That's the quiet harm of selling these as a "cancer check": they generate a lot of worry, scans and follow-up for very few true cancers found. We try to do two honest things: calm the false alarms by interpreting the result properly, and make sure anyone who is genuinely high-risk gets the real, guideline-based surveillance they should have. We don't diagnose or rule out cancer from a marker panel — that's not what it can do — and any worrying result is a conversation with a doctor, not a verdict.

Common questions

Can a tumour marker blood test tell me if I have cancer? Not reliably in a healthy person. These markers miss real cancers and flag harmless conditions, which is why authorities don't recommend them as screens for symptom-free people ¹. They're built to monitor a known cancer, not to find one.

My CEA / CA 19-9 came back high — do I have cancer? Usually not. Both rise in benign situations — CEA in smokers, CA 19-9 in biliary and pancreatic conditions — so a mildly raised value most often has an innocent explanation that's confirmed with proper follow-up ⁵. It needs interpreting in context, not panic.

Why isn't CA-125 a good ovarian-cancer screen? Because a 200,000-woman trial showed CA-125-based screening didn't reduce ovarian-cancer deaths, and it rises in many benign conditions — so it produces false alarms (and sometimes unnecessary surgery) without saving lives ³. It's a monitoring tool, not a screen.

Is there anyone who should have these tests? Yes — people with a specific risk. For example, AFP with ultrasound is recommended for liver-cancer surveillance in those with cirrhosis or chronic hepatitis ¹. The way to know is to ask a doctor whether you're in a genuine high-risk group.

Should I buy a 'cancer marker package' for peace of mind? On the evidence, it's more likely to cause worry than provide it — false positives lead to scans and procedures, false negatives give false reassurance ¹. Your money and energy are better spent on the screening that is proven for your age and risk, decided with a doctor.

Key takeaway

Tumour markers — AFP, CEA, CA-125, CA 19-9 — are valuable tools for monitoring a diagnosed cancer and watching for recurrence, and for targeted surveillance in genuinely high-risk people ¹. What they are not is a good "cancer check" for healthy, symptom-free people: they miss real cancers and flag benign ones, and the clearest trial evidence — over 200,000 women screened with CA-125 — showed no lives saved and real harms ³. If you have a specific risk, like cirrhosis or chronic hepatitis, there's a proper, guideline-based pathway worth following. Otherwise, the honest advice mirrors what we say about PSA: don't buy a marker panel as a cancer screen — talk to a doctor about the screening that actually fits your risk.