Six-monthly ultrasound and alpha-fetoprotein is the guideline-recommended surveillance for cirrhosis and selected chronic hepatitis B. This expert opinion begins from an uncomfortable premise: the strategy has poor sensitivity for early-stage disease, poor real-world adherence, and no randomised evidence that it reduces hepatocellular carcinoma-related or all-cause mortality in cirrhosis.
Two alternatives are reviewed. Abbreviated MRI protocols aim to keep the accuracy of a full liver protocol while cutting time and cost — dynamic contrast-enhanced abbreviated MRI can make a definitive diagnosis without recall imaging, and non-contrast abbreviated MRI has shown better sensitivity and specificity than ultrasound in recent trials. Blood-based panels, both protein-based (GALAD, HES 2.0) and circulating tumour DNA assays, may improve adherence but need prospective longitudinal evaluation before their detection advantage can be assumed to translate into outcomes.
The argument that carries beyond the tests themselves is about selection. Guidelines target surveillance by risk of developing cancer; the authors argue it should be targeted by probability of benefit — that earlier detection leads to effective treatment and better outcome. Those are not the same patients. Someone with decompensated cirrhosis and poor performance status may have the highest cancer risk and the least to gain from finding it early.
- Keep six-monthly ultrasound and AFP as the standard; none of the alternatives is ready to replace it.
- Ask whether an individual patient would actually be a candidate for curative treatment before intensifying surveillance.
- Consider MRI-based surveillance where ultrasound is repeatedly inadequate — obesity and severe steatosis are the common reasons.
- Treat GALAD and circulating tumour DNA panels as research tools; adherence gains have not been shown to become outcome gains.
- Adherence is the biggest modifiable factor in most settings, including Indian practice — a test the patient returns for beats a better test they do not.
Why it matters
Surveillance is targeted at who is most likely to get cancer, which is not the same as who benefits most from finding it.
Don't overread it
This is an expert opinion proposing a framework, not evidence that benefit-stratified surveillance improves outcomes.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for gastroenterology & hepatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free