A blood test caught more early liver cancers than ultrasound

Design

Prospective multicentre diagnostic study, head-to-head with ultrasound, CT/MRI reference

Population

2,293 evaluable adults with cirrhosis or chronic hepatitis B eligible for HCC screening (US)

Primary outcome

Early-stage HCC sensitivity (non-inferiority) and specificity

Effect

Early-stage sensitivity 66.7% vs 22.2% (difference 44.4 points); specificity 81.8% vs 98.6%

The ALTUS study prospectively compared a multi-target blood test (mt-HBT) head to head with surveillance ultrasound in 2,293 evaluable adults with cirrhosis or chronic hepatitis B eligible for hepatocellular carcinoma screening, using CT or MRI as the reference.

Among 37 cancers (27 early stage), the blood test detected 66.7% of early-stage cancers versus 22.2% for ultrasound, a 44-percentage-point absolute gain, and more than doubled any-stage detection. The trade-off was specificity: 81.8% for the blood test (just missing its 82% target) against 98.6% for ultrasound, meaning more false positives and downstream imaging.

This matters because ultrasound misses early tumours, especially in obesity and steatotic liver, and adherence is poor. A sensitive blood test could broaden surveillance, but the lower specificity means more benign scares and confirmatory scans. It is cross-sectional diagnostic performance, not evidence that the test lowers liver-cancer death, and it is not yet a guideline-endorsed substitute for imaging.

• The blood test detected 66.7% of early-stage liver cancers versus 22.2% for ultrasound.
• Any-stage detection was also more than doubled (73.0% vs 29.7%).
• Specificity was lower (81.8% vs 98.6%), so expect more false positives.
• It could help where ultrasound is limited, as in obesity or steatotic liver.
• This is diagnostic accuracy, not proof of reduced liver-cancer mortality.

Why it matters

It offers a route around ultrasound's poor early sensitivity and low uptake, the main weaknesses of current HCC surveillance.

Don't overread it

Cross-sectional diagnostic study; the lower specificity drives false positives, and it has not been shown to reduce liver-cancer mortality or endorsed to replace imaging.

The statistics, in plain English

A 44-point sensitivity gain is large, but specificity of 81.8% versus 98.6% means roughly one in five without cancer screens positive, generating confirmatory imaging; cross-sectional accuracy does not show the test saves lives.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.