CA 19-9 Unit Converter
CA 19-9 Unit Converter
U/mL and kU/L are the same number for CA 19-9 — and around one person in ten is Lewis-antigen-negative and cannot produce it at all, which makes a normal result uninformative in them.
CA 19-9 converter
U/mL ⇄ kU/LCA 19-9 22 U/mL
Formula and conversion factors
kIU/L = U/mL
U/L = U/mL × 1000
- U/mL = kU/L
- a unit per millilitre is a kilounit per litre, so the two are the same number
- arbitrary units
- defined by the assay calibrator against the 1116-NS-19-9 antibody, not by mass, so there is no molar conversion
- Lewis phenotype
- CA 19-9 is the sialylated Lewis a antigen; people who lack a functional fucosyltransferase cannot make it, and no unit conversion rescues a result from that
- bilirubin
- biliary obstruction raises CA 19-9 independently of cancer, so a result taken while a patient is jaundiced should be repeated after drainage
Worked example
CA 19-9 22 U/mL
22 U/mL = 22 kU/L = 22 kIU/L — all the same number
22 × 1000 = 22,000 U/L
Below the conventional cut-off of 37 U/mL
In a Lewis-antigen-negative person this result would be uninformative rather than reassuring
The four units
| Unit | Relationship | Example |
|---|---|---|
| U/mL | The usual reporting convention | 22 |
| kU/L | Identical to U/mL | 22 |
| kIU/L | Identical to U/mL | 22 |
| U/L | U/mL × 1000 | 22,000 |
Reading a CA 19-9
| Question | Answer |
|---|---|
| What is the conventional cut-off? | 37 U/mL, with roughly 81% sensitivity and 90% specificity for pancreatic cancer in symptomatic patients |
| Who cannot produce CA 19-9 at all? | Lewis-antigen-negative individuals, around 5 to 10 per cent of people — a normal result in them is uninformative, not reassuring |
| What raises it without cancer? | Biliary obstruction and cholangitis above all, then pancreatitis, cirrhosis, cystic fibrosis and thyroid disease |
| Is it a screening test? | No. It is not recommended for detecting pancreatic cancer in people without symptoms |
| What is it actually good for? | Prognosis, assessing resectability and monitoring response or recurrence in known pancreatic cancer |
| When should a raised result be repeated? | After biliary drainage, since jaundice alone can push it into the hundreds or thousands |
Why some people can never have a raised CA 19-9
The arithmetic is trivial. U/mL, kU/L and kIU/L are the same number, because a unit per millilitre is a kilounit per litre, and only U/L differs, at a thousand times larger. There is no molar unit: CA 19-9 is measured in arbitrary units defined by the assay calibrator against the 1116-NS-19-9 antibody, and the target is a carbohydrate epitope rather than a molecule with a fixed mass. Everything difficult about CA 19-9 lies in interpretation.
The most important limitation is genetic. CA 19-9 is the sialylated Lewis a antigen, and its synthesis requires a functional fucosyltransferase. Around 5 to 10 per cent of people are Lewis-antigen-negative and cannot make it at all. In those individuals the assay reads low no matter what disease is present, so a normal CA 19-9 is uninformative rather than reassuring, and the test should not be used to follow them. If a patient with known pancreatic cancer has a persistently undetectable CA 19-9, this is the likely explanation.
The second limitation is biliary. Obstruction and cholangitis raise CA 19-9 powerfully, often into the hundreds or thousands, with no malignancy present, and the level usually falls once the duct is drained. A high value measured in a jaundiced patient therefore says little about tumour burden and should be repeated after drainage. Pancreatitis, cirrhosis, cystic fibrosis and thyroid disease raise it more modestly.
What CA 19-9 is genuinely useful for is prognosis and monitoring in a patient already known to have pancreatic cancer: a high preoperative level predicts a worse outcome and can suggest occult metastatic disease, and serial measurements track response to treatment and recurrence. It is not a screening test and is not recommended for detecting pancreatic cancer in people without symptoms, where the low prevalence of the disease means most raised results are false alarms. Any CA 19-9 result needs a clinician who has the imaging, the bilirubin and, where it matters, the Lewis phenotype.
Frequently asked questions
Is U/mL the same as kU/L for CA 19-9?
Yes, and kIU/L too — a unit per millilitre is a kilounit per litre. Only U/L is different, being a thousand times larger, so 22 U/mL is 22 kU/L and 22,000 U/L. There is no molar unit for CA 19-9.
Why can some people not produce CA 19-9?
CA 19-9 is the sialylated Lewis a antigen, and making it requires a functional fucosyltransferase. Around 5 to 10 per cent of people are Lewis-antigen-negative and cannot synthesise it, so their result stays low whatever disease is present and a normal value tells you nothing.
What is the normal range for CA 19-9?
The conventional cut-off is 37 U/mL, at which sensitivity for pancreatic cancer in symptomatic patients is around 81 per cent and specificity around 90 per cent. Laboratories vary, and the cut-off cannot be applied to Lewis-negative individuals at all.
Can a blocked bile duct raise CA 19-9?
Yes, markedly — biliary obstruction and cholangitis are the commonest causes of a CA 19-9 in the hundreds or thousands without cancer, and the level usually falls after drainage. A result taken while a patient is jaundiced should be repeated once the obstruction is relieved.
Can CA 19-9 be used to screen for pancreatic cancer?
No. It is not recommended for detecting pancreatic cancer in people without symptoms: the disease is uncommon enough that most raised results would be false alarms, and Lewis-negative patients would be missed entirely. Its role is prognosis and monitoring in known disease.
Related calculators
References
- Ballehaninna UK, Chamberlain RS. The clinical utility of serum CA 19-9 in the diagnosis, prognosis and management of pancreatic adenocarcinoma: an evidence based appraisal. J Gastrointest Oncol. 2012;3(2):105–119.
- Parra-Robert M, Santos VM, Canis SM, et al. Relationship between CA 19.9 and the Lewis phenotype: options to improve diagnostic efficiency. Anticancer Res. 2018;38(10):5883–5888.
- Locker GY, Hamilton S, Harris J, et al. ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol. 2006;24(33):5313–5327.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.