Liver cancer in cirrhosis: the six-monthly ultrasound that finds it in time

Liver cancer is one of the tumours usually found too late. In Germany about 9,680 people are diagnosed with it each year and almost 8,400 die of it; relative five-year survival is only around 15%. The most common type, hepatocellular carcinoma (HCC), almost always arises in a liver that is already diseased. That is the opportunity: the risk group is known, and the test is simple – an ultrasound scan every six months. The problem is that many of the people concerned do not know they belong to that group.

Disclosure: the author heads the liver cancer centre of the CCC München and works as a liver and transplant surgeon at LMU Klinikum München.

Who is at risk

The single most important risk factor is cirrhosis – the scarring of the liver at the end of years of damage. Any cause counts: alcohol, fatty liver disease linked to the metabolic syndrome (MASLD), hepatitis B or C, iron overload and others. The other group is chronic hepatitis B, which can cause liver cancer even without cirrhosis.

For hepatitis B without cirrhosis, the PAGE-B score (age, sex, platelet count) helps. In the European study of 1,815 treated patients that developed it, nobody with fewer than 10 points developed liver cancer within five years; at 10–17 points it was 3–4%, and at 18 points or more 16–17%.

What "surveillance" means

Surveillance means regular checks for people with a known risk; Germany has no statutory liver cancer screening for the general population. The standard is a liver ultrasound every six months, often combined with a blood test for alpha-fetoprotein (AFP), a tumour marker.

Ultrasound alone misses many early tumours. In a meta-analysis of 32 studies with 13,367 patients it detected only 45% of early-stage cancers; with AFP added, 63%. In people with obesity or a very nodular liver the view is often limited. MRI is then an option. A shorter, faster protocol – abbreviated MRI – is being studied intensively but is not yet standard.

The evidence – and its limits

There is only one large randomised trial. In Shanghai in the 1990s, 18,816 people aged 35 to 59 with hepatitis B or chronic hepatitis were randomly assigned to ultrasound plus AFP every six months or to no screening. In the screened group almost half the tumours found were small, and 46.5% could be removed surgically (control group: 7.5%). Deaths from liver cancer fell by 37%. In absolute terms, 32 of 9,373 people died of liver cancer in the screened group versus 54 of 9,443 in the control group – roughly 3.4 instead of 5.7 per 1,000 over the study period – even though the screened group attended only 58% of the offered examinations. The results cannot simply be transferred to people with cirrhosis in Europe today.

For people with cirrhosis there are no randomised trials, only observational data. A meta-analysis of 59 studies with more than 145,000 HCC patients found that under surveillance the cancer was found at an early stage almost twice as often (relative risk 1.86), treated with curative intent almost twice as often (1.83), and the risk of death was about a third lower (hazard ratio 0.67).

Such comparisons have pitfalls. Earlier diagnosis makes survival look longer even if the date of death does not change – lead-time bias (see Why longer survival after screening does not prove it works). The authors adjusted for it and the benefit remained. Still, people who attend regular checks tend to be healthier and better cared for overall. There are harms too: in four studies, 8.8% to 27.5% of people under surveillance experienced physical harms from false-positive or indeterminate results, mostly mild ones such as extra imaging.

What the guidelines say

• German S3 guideline (German Guideline Programme in Oncology; current version 5.2, version 5.0 from 2024): everyone with cirrhosis has an increased risk regardless of cause; ultrasound every six months is recommended. People with hepatitis B and a PAGE-B score of 10 or more, and people with hepatitis C and advanced fibrosis, should be offered surveillance. For fatty liver hepatitis (MASH) without cirrhosis it is an optional ("may") recommendation, because the benefit is not established.
• EASL 2025 (European Association for the Study of the Liver): patients with cirrhosis should undergo surveillance if they could be treated. Ultrasound every six months remains the standard; ultrasound plus AFP is a valid option. Data on abbreviated MRI are still limited. Surveillance is not recommended for advanced fibrosis without cirrhosis.
• AASLD 2023 (US): ultrasound plus AFP about every six months (strong recommendation); CT or MRI if the ultrasound view is limited. In end-stage cirrhosis (Child-Pugh C) only if transplantation is possible; no routine surveillance for fatty liver disease with advanced fibrosis but no cirrhosis.

All three share one condition: surveillance only makes sense if a tumour that is found could actually be treated.

The real problem: unrecognised cirrhosis

Cirrhosis causes no symptoms for a long time, and liver enzymes can be normal (see Fatty liver disease: the silent epidemic). In a nationwide Swedish study, 39% of HCC patients with cirrhosis had not known about their cirrhosis. Their median survival after the cancer diagnosis was 0.9 years, compared with 3.8 years for patients under surveillance. In German health-insurance data (BARMER, 2016–2020), only 4% of people later diagnosed with HCC had received the recommended twice-yearly ultrasound over several years; 55% never had it in any year. The gap lies less in the test than in finding the people who need it.

When a tumour is found early

A small HCC can be cured. Depending on its position, size and liver function there are three options: surgical removal (resection), ablation, in which the tumour is destroyed with heat through a needle probe, and liver transplantation. Transplantation treats tumour and diseased liver at once and is an option when the cancer lies within the Milan criteria: one tumour up to 5 cm or up to three tumours each up to 3 cm, with no vascular invasion or spread. In the original Milan study of 1996, 75% of transplanted patients were alive after four years. The German S3 guideline recommends that patients be seen at a transplant centre before any bridging treatment begins.

What this means for you

• Know your liver status. If you drink heavily, have diabetes or marked obesity, ask for an assessment of liver scarring – for example the FIB-4 score from routine blood tests, or elastography (a scan that measures liver stiffness).
• Use the hepatitis test. Since October 2021, people with statutory health insurance (GKV) aged 35 and over are entitled to a one-time test for hepatitis B and C as part of the "Check-up 35" health examination.
• If you have cirrhosis: ask who coordinates the six-monthly ultrasound, whether AFP is measured, and whether the liver could be seen well on the scan.
• If you have hepatitis B: ask for your PAGE-B score. From 10 points, surveillance is recommended.
• If something is found: work-up belongs in a centre whose tumour board covers liver surgery, ablation and transplantation.

This article is general information and not individual medical advice. Decisions about tests and treatment belong in a conversation with your own physician.