Tumour Markers & Oncology Labs

Convert alpha-fetoprotein between ng/mL, µg/L, pmol/L and WHO IU/mL, with the adult reference limit, the 400 ng/mL hepatocellular carcinoma threshold, and why a normal AFP excludes very little.

Calculate BSA by the Mosteller or Du Bois formula, the basis of most chemotherapy dosing and of eGFR indexing.

U/mL and kU/L are the same number, so the work on this page is the interpretation: the 35 U/mL convention, the long list of benign causes, and why CA 125 is not a screening test.

U/mL and kU/L are the same number for CA 15-3. The conventional cut-off is about 30 U/mL, and the test is for monitoring known breast cancer — never for diagnosing it.

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.

Convert CEA between ng/mL, µg/L and pmol/L against separate smoker and non-smoker reference limits, and see why CEA is a monitoring test rather than a screening one.

Convert free PSA between ng/mL, µg/L and pmol/L — but a free PSA on its own has no interpretation. It means something only as a percentage of the total PSA, so no reference band is shown here.

Calculate the percentage of free PSA, validated in the 4–10 ng/mL diagnostic grey zone to help avoid unnecessary biopsy.

Convert total PSA between ng/mL, µg/L and pmol/L, with age-specific limits and an honest account of the 4.0 ng/mL convention — a referral threshold, not a diagnosis.

Correct total PSA for prostate size to separate benign enlargement from a result that warrants biopsy.

Estimate the exponential doubling time between two PSA values, a strong predictor of metastasis after biochemical recurrence.

Calculate the absolute annual rise in PSA — informative at higher PSA values, though no longer a standalone biopsy trigger.

Combine HE4, CA 125 and menopausal status into a single risk estimate that guides referral for an adnexal mass.