Liver Cancer Screening in Cirrhosis: When to Scan, How HCC Is Staged, and What Treatment Looks Like

If you have cirrhosis — from any cause — you need an ultrasound of the liver every 6 months, usually with an AFP blood test, for life. A 2022 meta-analysis of 59 studies and 145,396 patients found that this surveillance nearly doubled early-stage detection and cut the risk of death by about a third. Liver cancer found early can be cured; found late, it usually cannot.
— Dr Chetan Kalal, DM Hepatology, Gleneagles Hospital Mumbai

Take a composite of a patient I meet far too often: a 58-year-old man from Thane whose hepatitis B was diagnosed twelve years ago, treated for a while, then forgotten. He felt well. Nobody scanned him. He arrives now with a 9 cm tumour that has grown into the portal vein, and the first question his family asks — “Can it be removed?” — has an answer none of us want to give.

The tragedy is not that he had liver cancer. It is that an inexpensive 6-monthly ultrasound would very probably have found it at 2 cm, when it could have been burnt away in an afternoon. Hepatocellular carcinoma — HCC, the commonest form of liver cancer — grows quietly in a liver that has already been damaged, and it produces no symptoms until it is large. Surveillance is the only reliable way to catch it while it is still curable — which is why “when do I get scanned?” is the right first question for anyone with cirrhosis.

Who needs liver cancer screening?

Two groups. First, anyone with cirrhosis, whatever caused it — hepatitis B, hepatitis C, alcohol, fatty liver, autoimmune disease or anything else. Second, a subset of people with chronic hepatitis B who do not yet have cirrhosis, because the virus can cause cancer directly, without scarring first.

Both the 2025 EASL guidelines and the 2023 AASLD guidance put cirrhosis of any cause at the centre of surveillance. On hepatitis B without cirrhosis, the AASLD lists those who qualify: men over 40 and women over 50 from regions where hepatitis B is common (India is one), people of African origin at a younger age, anyone with a family history of liver cancer, and anyone whose score on a validated risk calculator such as PAGE-B is intermediate or high. If you have hepatitis B and are unsure whether you fit, ask your hepatologist to run the numbers; it takes two minutes.

Equally useful is knowing who does not need it. The AASLD advises against routine surveillance in people with advanced fibrosis short of cirrhosis — whether from cured hepatitis C or from fatty liver — and in anyone too unwell from other illnesses to benefit from treatment. Surveillance only makes sense if a positive scan would change what we do.

One point I stress to every patient whose hepatitis C has been cured or whose hepatitis B is suppressed on tablets: if you had cirrhosis before treatment, you still need surveillance after it. Clearing the virus lowers the cancer risk. It does not remove it. The same holds for alcohol-related cirrhosis after years of sobriety — my cirrhosis page covers this.

What does screening involve, and does it actually work?

An ultrasound of the liver every 6 months, combined with a blood test for alpha-fetoprotein (AFP). That is the whole programme. The AASLD guidance grades this a strong recommendation on level 2 evidence and specifically prefers ultrasound plus AFP over ultrasound alone. Six-monthly, not yearly: the guidance notes that semi-annual surveillance finds tumours at an earlier stage and improves survival compared with annual scans, even after correcting for lead-time bias. CT and MRI are reserved for people in whom ultrasound is technically poor — severe obesity, a very nodular liver — and for working up anything the ultrasound finds.

Does it save lives? The best evidence comes from a 2022 meta-analysis in the Journal of Hepatology by Singal and colleagues, who pooled 59 studies covering 145,396 patients with HCC. Those whose cancer was found by surveillance were 1.86 times more likely to be diagnosed at an early stage, 1.83 times more likely to receive curative treatment, and had a 33% lower risk of death (hazard ratio 0.67) after adjusting for lead-time bias. The authors are honest that the studies varied considerably, and that between 8.8% and 27.5% of patients experienced some harm from false alarms — an unnecessary CT, a biopsy that found nothing — though most such harms were mild. For hepatitis B specifically, the AASLD guidance cites a large randomised trial in which surveillance reduced liver cancer deaths with a hazard ratio of 0.63.

Put plainly: of every three people with cirrhosis who would otherwise die of liver cancer, surveillance saves roughly one. Few things in hepatology offer that return for a scan and a blood test twice a year.

The failure is rarely the test; it is the missed appointment. If you have cirrhosis and your last scan was more than 6 months ago, book the next one before you finish reading this.

What the AFP test can and cannot tell you

AFP is a protein that many — not all — liver cancers release into the blood. On its own it is a poor screening test, and that is exactly why we pair it with ultrasound rather than rely on it.

A 2018 meta-analysis in Gastroenterology by Tzartzeva and colleagues, pooling 32 studies and 13,367 patients with cirrhosis, put numbers on this. Ultrasound alone detected cancer at any stage with 84% sensitivity, but early-stage cancer — the kind we are actually looking for — with only 47% sensitivity. In the studies that compared the two strategies head to head, ultrasound plus AFP found early-stage cancer 63% of the time versus 45% for ultrasound alone. The price is more false positives: adding AFP slightly lowers specificity. The AASLD guidance states that at the usual cut-off of 20 ng/mL, AFP has a sensitivity of about 60% and a specificity of about 90%.

So three rules. A normal AFP does not mean you have no cancer; roughly four in ten early tumours never raise it. A mildly raised AFP does not mean you have cancer either — active hepatitis can push it up. And a steadily rising AFP across successive tests, even within the “normal” range, is a reason for a contrast CT or MRI, which is how HCC is actually diagnosed; in cirrhosis a typical appearance on imaging is enough, and biopsy is seldom required.

BCLC staging explained in plain language

Once a cancer is confirmed, the next question is: how far along is it? Liver cancer is unusual because the answer depends on three things at once — the tumour (number, size, whether it has entered blood vessels or spread outside the liver), how well the underlying liver still works, and how fit the patient is. The Barcelona Clinic Liver Cancer (BCLC) system, updated in 2022 and used by both EASL and AASLD, combines all three into five stages. It is the framework behind every liver cancer treatment decision I make in Mumbai; the table below is a simplified version.

Two things the table cannot show. The stage is not a sentence — a tumour shrunk by TACE can turn an intermediate case into a transplant candidate. And the survival figures are what the BCLC authors expect for a well-treated group; an individual may do far better or worse. Staging belongs to a team — hepatologist, liver surgeon, interventional radiologist, oncologist — which is how we work at Gleneagles.

Early-stage liver cancer: three ways to cure it

At stages 0 and A the goal is cure, and there are three routes.

Ablation destroys the tumour from inside with heat, delivered through a needle guided by ultrasound or CT — radiofrequency ablation or its faster cousin, microwave ablation. It needs no surgery and a one- or two-night stay, and for tumours of 2 cm or less the BCLC 2022 update states plainly that ablation and resection offer the same survival benefit.

Resection — surgically removing the part of the liver containing the tumour — suits a single tumour, of any size, in a patient whose liver function is preserved and whose portal pressure is not raised, because the scarred liver left behind must be able to cope. The AASLD guidance expects 5-year survival above 70% after resection in well-selected patients, with surgical mortality under 3%.

Liver transplant is the one treatment that removes both the cancer and the cirrhosis that caused it. The rule that decides eligibility is the Milan criteria, published by Mazzaferro and colleagues in the New England Journal of Medicine in 1996: one tumour of 5 cm or less, or up to three tumours each 3 cm or less, with no vessel invasion or spread. In that original study of 48 transplanted patients, 4-year survival was 75% overall and 85% in those whose explanted liver confirmed they were within the criteria; the cancer came back in only 8%. Thirty years on, Milan remains the benchmark, though both AASLD and EASL now accept that patients slightly beyond it can be “downstaged” with TACE or TARE and transplanted if they respond — after a 3-to-6-month observation period, in the AASLD’s wording.

In India most transplants are living-donor transplants, which removes the long wait for a deceased-donor organ but demands a healthy, willing family donor; the transplant page explains how we assess both. Our programme has performed over 1,000 transplants, around 300 of them complex, with 92% survival.

Intermediate and advanced liver cancer: control rather than cure

When the cancer is too extensive to cut out or burn away, the aim shifts to controlling it for as long as possible, and often to shrinking it back into transplant territory.

TACE — transarterial chemoembolisation — is the standard treatment for stage B. An interventional radiologist threads a catheter into the artery feeding the tumour and delivers chemotherapy beads that both poison the tumour and block its blood supply. It requires reasonably good liver function; in a failing liver it does more harm than good.

TARE, or Y-90 radioembolisation, delivers microscopic radioactive beads through the same arterial route — an alternative to TACE in selected patients, and useful for downstaging larger tumours before transplant.

Systemic therapy is the treatment for stage C and for stage B patients who are not suited to TACE. For over a decade the only option was sorafenib, a targeted tablet which in the 2008 SHARP trial extended median survival from 7.9 to 10.7 months — modest, but real. Lenvatinib, another targeted tablet, followed. The larger shift came with immunotherapy: in the IMbrave150 trial, published in the New England Journal of Medicine in 2020, atezolizumab with bevacizumab reduced the risk of death by 42% compared with sorafenib (hazard ratio 0.58), with 67.2% of patients alive at 12 months versus 54.6%. A second immunotherapy combination, durvalumab with tremelimumab, is also used first-line internationally. These drugs are available in India, though cost and access remain real constraints, and the choice between them depends on bleeding risk, varices, liver function and other illnesses — a decision for your treating team, never from an article.

Whatever the stage, the cirrhosis underneath still needs treating — the virus, the alcohol, the ascites, the nutrition. A well-managed liver tolerates cancer treatment better and lives longer with it.

What I tell every patient with cirrhosis

Put two dates in your calendar each year for a liver ultrasound and AFP, and keep them even when you feel perfectly well — especially then. If an ultrasound shows a nodule, do not panic and do not wait: a contrast CT or MRI within a few weeks settles the question. And if HCC is confirmed, ask for the BCLC stage and insist on a multidisciplinary opinion before any treatment starts, because the difference between the right first treatment and the wrong one is measured in years.

For the man in the composite above, the honest answer to his family’s question was no. For his younger brother, also hepatitis B positive, the answer to “when do I get scanned?” was: this week, and every 6 months after that.

References

1. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatocellular carcinoma. J Hepatol. 2025;82(2):315–374. PMID 39690085. doi:10.1016/j.jhep.2024.08.028
2. Singal AG, Llovet JM, Yarchoan M, et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78(6):1922–1965. PMID 37199193. doi:10.1097/HEP.0000000000000466
3. Reig M, Forner A, Rimola J, et al. BCLC strategy for prognosis prediction and treatment recommendation: The 2022 update. J Hepatol. 2022;76(3):681–693. PMID 34801630. doi:10.1016/j.jhep.2021.11.018
4. Singal AG, Zhang E, Narasimman M, et al. HCC surveillance improves early detection, curative treatment receipt, and survival in patients with cirrhosis: A meta-analysis. J Hepatol. 2022;77(1):128–139. PMID 35139400. doi:10.1016/j.jhep.2022.01.023
5. Tzartzeva K, Obi J, Rich NE, et al. Surveillance imaging and alpha fetoprotein for early detection of hepatocellular carcinoma in patients with cirrhosis: A meta-analysis. Gastroenterology. 2018;154(6):1706–1718.e1. PMID 29425931. doi:10.1053/j.gastro.2018.01.064
6. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996;334(11):693–699. PMID 8594428. doi:10.1056/NEJM199603143341104
7. Finn RS, Qin S, Ikeda M, et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med. 2020;382(20):1894–1905. PMID 32402160. doi:10.1056/NEJMoa1915745
8. Llovet JM, Ricci S, Mazzaferro V, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359(4):378–390. PMID 18650514. doi:10.1056/NEJMoa0708857