Hepatocellular carcinoma and other liver masses

• Hepatocellular carcinoma (HCC) is a cancer that starts in hepatocytes, the main working cells of the liver. It is the commonest type of primary liver cancer and usually arises in a liver that is already scarred by cirrhosis.
• Cirrhosis from any cause is the strongest risk factor, but chronic hepatitis B can cause HCC even without cirrhosis, and hepatitis C, alcohol, MASLD, aflatoxin, hemochromatosis, PBC and tobacco add to the risk.
• People with cirrhosis are offered ultrasound of the liver, with or without alpha fetoprotein (AFP), about every 6 months, because finding a tumour early is what makes cure possible.
• In someone with cirrhosis or chronic hepatitis B, a liver nodule showing arterial phase hyperenhancement and washout on multiphase CT or MRI is HCC on imaging alone, without a biopsy; an AFP above 400 ng/mL is suggestive but no longer diagnostic.
• Curative treatment (resection, transplantation, ablation) is for early-stage disease. Intermediate disease is treated mainly with transarterial chemoembolisation (TACE), and advanced disease with systemic therapies such as atezolizumab plus bevacizumab or durvalumab plus tremelimumab.
• Metastases to the liver are far commoner than primary liver cancer, most often from colorectal cancer. Benign masses such as hemangioma and focal nodular hyperplasia must be distinguished from cancer and from adenoma, which can rupture or become malignant.

By the end of this page you should be able to:

1. Explain why most HCC develops in cirrhosis, and state the main regional and sex differences in incidence.
2. List the at-risk groups who need HCC surveillance and the tests and interval used.
3. Interpret a CT or MRI of a liver nodule in an at-risk patient: arterial phase hyperenhancement, washout, and the LI-RADS categories.
4. Stage HCC with the Barcelona Clinic Liver Cancer (BCLC) system and match each stage to its first-choice treatment.
5. State the Milan criteria and the principles of downstaging and bridging to liver transplantation.
6. Compare the first-line systemic therapies for advanced HCC and name the cautions with atezolizumab plus bevacizumab.
7. Distinguish the common benign and malignant liver masses from HCC.

Primary liver cancer is the sixth most common cancer worldwide and the third leading cause of cancer-related death worldwide and in the United States as of 2020 ^[1], and HCC is its most common type, accounting for between 75% and 86% of cases ^[1].

HCC is commonly staged with the BCLC system, which combines tumour burden, liver function and performance status. The categories are:

• BCLC 0 (very early): a single tumour 2 cm or less.
• BCLC A (early): a single tumour of any size, or up to 3 nodules none larger than 3 cm, without vascular invasion or extrahepatic spread, in a patient with preserved liver function.
• BCLC B (intermediate): multifocal disease beyond BCLC A, preserved liver function, no vascular invasion or extrahepatic spread.
• BCLC C (advanced): vascular invasion or extrahepatic spread, in a patient who is still relatively fit with preserved liver function.
• BCLC D (end stage): major cancer symptoms or severe liver dysfunction without the option of transplantation.

• Global burden. Primary liver cancer is the sixth most common cancer worldwide and the third leading cause of cancer-related deaths both worldwide and in the United States as of 2020 ^[1].
• Sex. Men are affected approximately two to three times more than women, with higher incidence and mortality across most countries ^[1].
• Regional pattern. The highest age-standardised incidence is in East Asia and sub-Saharan Africa, where chronic hepatitis B is endemic and aflatoxin exposure persists; sub-Saharan Africa carries a particularly young profile, with a median age at diagnosis of 46 years and more than one third of affected people from Africa diagnosed before age 40 ^[1]. Projections suggest incidence in that region will double by 2040 ^[1].
• Underlying disease. Cirrhosis from any cause is present in over 80% of patients with HCC, and patients with cirrhosis from any aetiology typically have about a 2% annual risk of developing HCC ^[1].
• United States. Incidence and mortality rose from 1970 to 2010, then incidence began to fall from 2011 and mortality plateaued in 2013, with one study showing a subsequent decrease of roughly 3% per year ^[1].
• Changing causes. In North America and Europe, viral-hepatitis HCC has plateaued where elimination programmes work, while alcohol-related and MASLD-related (metabolic dysfunction-associated steatotic liver disease, formerly non-alcoholic fatty liver disease) HCC are increasing; NAFLD is now the fastest growing cause of HCC in liver transplant candidates ^[1]. About one quarter to one third of NAFLD-related HCC occurs without cirrhosis, although surveillance is not cost-effective in non-cirrhotic NAFLD ^[1].

The strongest risk factor is cirrhosis from any liver disease aetiology ^[1]. The following modify risk in people with chronic liver disease; several combine synergistically rather than additively ^[1].

• Chronic hepatitis B (HBV), even without cirrhosis, especially in endemic regions; vertical (mother-to-child) transmission explains the young age at diagnosis in sub-Saharan Africa ^[1].
• Chronic hepatitis C (HCV), including after sustained virological response when cirrhosis is present ^[1].
• Alcohol: alcohol-associated cirrhosis is a known risk factor, and alcohol as a cofactor with another aetiology can increase HCC risk as much as 5-fold ^[1].
• MASLD/NAFLD: a rising cause, linked to obesity and metabolic syndrome; the components of metabolic syndrome, including diabetes, nearly double HCC risk even without overweight or obesity ^[1].
• Aflatoxin B1 and aristolochic acid: cofactors for HCC in people with HBV infection through dietary exposure ^[1].
• Hemochromatosis, primary biliary cholangitis (PBC) and other chronic cholestatic and metabolic liver diseases through their progression to cirrhosis.
• Tobacco: smoking is associated with a 20% to 86% increased risk of HCC, and this risk can return nearly to baseline after 30 years of smoking cessation ^[1].
• Diabetes mellitus and metabolic syndrome.
• Obesity: associated with 1.5 to 4.5 times higher risk of HCC and contributes to nearly 10% of HCC worldwide ^[1].
• Lifestyle cofactors such as physical inactivity and excess body weight; a healthy weight, balanced diet, avoidance of tobacco and alcohol and control of metabolic risk factors may reduce HCC risk ^[1].

HCC usually develops along a predictable sequence: chronic necroinflammation, fibrosis, cirrhosis and, finally, malignant transformation of hepatocytes. Each risk factor drives this sequence in its own way.

• Chronic HBV integrates into the host genome and expresses the HBx protein, promoting genomic instability and carcinogenesis even before cirrhosis is present.
• HCV drives chronic inflammation and oxidative stress; eradicating HCV reduces but does not abolish HCC risk once cirrhosis has developed.
• Alcohol and MASLD act through steatosis, insulin resistance, inflammation and fibrosis.
• Aflatoxin B1 is metabolised to an epoxide that binds DNA, producing characteristic TP53 mutations and acting synergistically with HBV.
• In cirrhosis, regenerating nodules accumulate somatic mutations; key driver changes include TERT promoter mutations, CTNNB1 (beta-catenin) and TP53 alterations, which explain the molecular heterogeneity of HCC.

The result is a tumour with a rich arterial blood supply (unlike the surrounding cirrhotic liver, which depends mainly on portal venous flow). This vascular difference is the basis of the imaging hallmarks and of transarterial therapy.

HCC is often found incidentally during surveillance ultrasound in a patient with known cirrhosis. When it presents symptomatically, features reflect tumour burden, location and decompensation of the underlying cirrhosis.

Symptoms

• Right upper quadrant pain or heaviness, abdominal distension, early satiety.
• Weight loss, anorexia, fatigue.
• Decompensation of previously stable cirrhosis: new ascites, jaundice, variceal bleeding or hepatic encephalopathy.
• Fever and night sweats are possible.
• Sudden severe pain with shock from tumour rupture and intraperitoneal bleeding, which is a surgical emergency.

Signs

• Hepatomegaly, sometimes with an irregular or nodular liver edge; an arterial bruit over the liver is occasionally audible.
• Signs of cirrhosis: spider naevi, palmar erythema, gynaecomastia, ascites, splenomegaly.
• Jaundice and encephalopathy if decompensated.

Paraneoplastic features are characteristic of HCC and arise from ectopic hormone or factor production by the tumour:

• Hypoglycaemia: fasting hypoglycaemia from a large tumour consuming glucose or producing insulin-like growth factor.
• Erythrocytosis: excess erythropoietin production, raising the haematocrit.
• Hypercalcaemia: parathyroid hormone-related peptide.
• Dyslipidaemia, skin changes and thrombocytosis may also occur.

Vascular invasion and metastasis. Portal vein tumour thrombus is common in advanced disease and produces rapidly worsening liver function and portal hypertension. Extrahepatic spread most often goes to the lungs, bones and lymph nodes, and is assessed during staging.

The first distinction is between HCC and metastatic cancer in the liver (metastases are far more common). The second is between HCC and benign masses. Use the table of other liver masses below and the imaging characteristics (arterial hypervascularity with washout favours HCC).

First-line evaluation

• Liver function tests, bilirubin, albumin, prothrombin time/INR, platelet count (thrombocytopenia signals portal hypertension), renal function.
• AFP: elevated in some HCCs and used in surveillance and prognosis; it should not be used alone to diagnose.
• Imaging: multiphase CT or contrast-enhanced MRI of the liver, acquired in at least arterial and portal venous phases, to demonstrate arterial phase hyperenhancement and washout.
• Chest CT (non-contrast) for staging in tumours beyond BCLC 0; PET CT is not recommended routinely because of low sensitivity of 50% to 65% ^[1].
• Upper endoscopy to assess varices before systemic therapy with anti-VEGF agents.

What to look for. The diagnosis in at-risk patients rests on imaging: a nodule showing arterial phase hyperenhancement with washout on portal venous or delayed phases is HCC ^[1]. MRI has higher sensitivity (82% versus 66%) with similar specificity (92% versus 91%) than CT in meta-analysis, but the two are equally recommended, chosen by local expertise and availability ^[1].

In at-risk patients (cirrhosis, or chronic HBV infection with intermediate or high risk), HCC can be diagnosed on imaging without biopsy. The AASLD supports the Liver Imaging Reporting and Data System (LI-RADS) ^[1].

• LR-5 lesion: 95% to 99% probability of HCC; accepted as the diagnosis in an at-risk patient.
• LR-4 lesion: roughly 75% probability of HCC; multidisciplinary discussion, with biopsy or repeat imaging in 3 months ^[1].
• LR-3 observation: roughly 30% probability; repeat cross-sectional imaging in 3 to 6 months ^[1].
• LR-M: features suggestive of malignancy but not specifically HCC (93% to 100% malignant on tissue, but only 29% to 44% are HCC); biopsy is recommended ^[1].
• Size matters: for nodules of at least 2 cm, arterial phase hyperenhancement plus one major ancillary feature (washout, enhancing capsule or threshold growth) is enough; for 10 to 19 mm nodules, arterial phase hyperenhancement plus washout or threshold growth, or two major features, is required ^[1].
• LI-RADS does not apply outside its defined populations (no cirrhosis and no HBV); these patients need a biopsy.

Biopsy is required for nodules in patients without cirrhosis or without HBV, for indeterminate (LR-4) findings when a diagnosis would change management, for LR-M lesions, and to confirm diagnosis in clinical trials ^[1]. AASLD also encourages multidisciplinary consideration of biopsy of LR-4 and LR-5 lesions to confirm diagnosis or enable molecular analysis ^[1]. Staining for two of glypican 3, glutamine synthetase, heat shock protein 70 and clathrin heavy chain is highly specific for HCC ^[1]. Sensitivity of biopsy ranges from 70% to 93% for most tumours and can be as low as about 60% for tumours under 2 cm; a negative biopsy does not exclude HCC, and a second biopsy is recommended if the lesion behaves like cancer ^[1]. Risk of complications such as bleeding or seeding is about 3%, reduced by coaxial needle technique ^[1].

AFP at a threshold of 400 ng/mL was previously used as a diagnostic criterion, but over 40% of HCC have normal AFP and AFP can be raised in other cancers, so AFP is no longer recommended for diagnosis ^[1]. AFP 400 ng/mL or higher remains a threshold for systemic therapy decisions (ramucirumab in the second line) and for transplantation eligibility and downstaging ^[1].

The BCLC system is recommended because it incorporates tumour burden, liver dysfunction and performance status, and provides treatment allocation ^[1][2]. The BCLC 2022 update refined patient characterisation using the albumin-bilirubin (ALBI) score and AFP, recognised heterogeneity within BCLC B, and incorporated downstaging and stage migration ^[1][2]. Expected median survival in the BCLC model is more than 5 years for stage 0/A, 2.5 years for stage B, 2 years for stage C and 3 months for stage D ^[2].

All patients need multiphase CT or contrast-enhanced MRI for tumour extent; those with tumours 2 cm or larger should have a non-contrast chest CT to look for lung metastases ^[1]. Bone or pelvic CT is not routine but may be considered when AFP exceeds 1000 ng/mL, with macrovascular invasion or multifocal bilobar disease ^[1].

Risk stratification for transplant. RETREAT and similar scores stratify 5-year post-transplant recurrence risk from under 3% in the lowest-risk group up to 75% in the highest risk; even within Milan criteria, HCC recurs after transplantation in 10% to 15%, with a median survival of about 1 year from recurrence and under 20% eligible for resection ^[1].

Resection is the curative treatment of choice for localised HCC without cirrhosis ^[1]. In cirrhosis, it is the treatment of choice for patients with limited tumour burden, compensated (Child-Pugh A) cirrhosis, no clinically significant portal hypertension, and an adequate future liver remnant (FLR), typically more than 30% without cirrhosis and more than 40% with cirrhosis ^[1]. In such patients, 5-year survival exceeds 70% and perioperative mortality is under 3% ^[1].

Clinically significant portal hypertension means a hepatic venous pressure gradient of 10 mmHg or more, or surrogates such as varices, splenomegaly or platelets under 100,000 per microlitre ^[1]. Portal vein embolisation (and increasingly radioembolisation) can increase the FLR. Extended indications (multifocal disease, or macrovascular portal vein tumour thrombus in selected patients) are confined to high-volume centres and multidisciplinary decision-making ^[1]. Laparoscopic and robotic resection are options in selected patients and may extend resection to those with mild portal hypertension ^[1].

After resection, the risk of recurrence is 50% to 70% at 5 years, highest in the first year ^[1]. Surveillance with contrast-enhanced CT or MRI of abdomen and chest plus AFP is advised every 3 to 6 months, indefinitely ^[1].

For early-stage HCC with significant portal hypertension or decompensated cirrhosis, transplantation is the treatment of choice because it cures both the tumour and the underlying liver disease ^[1]. Median survival is about 10 years, and the 5-year recurrence risk after transplant (about 10%) is far lower than after resection or ablation (50% to 60%) ^[1].

Milan criteria: one lesion between 1 and 5 cm, or two to three lesions each between 1 and 3 cm, without vascular invasion or metastases ^[1]. Expanded criteria (UCSF, up-to-seven, total tumour volume, Extended Toronto, Kyoto) achieve post-transplant 5-year survival of roughly 65% to 81% in selected patients ^[1]. In the United States, downstaging to within Milan criteria from the UNOS downstaging criteria (single lesion 5.1 to 8 cm; two to three lesions each 5 cm or less with total diameter 8 cm or less; four to five lesions each 3 cm or less with total diameter 8 cm or less) allows MELD exception points after a period of observation and 6 months within Milan criteria ^[1]. Patients with an AFP of 1000 ng/mL or higher must be downstaged to under 500 ng/mL to qualify ^[1].

Bridging therapy (ablation, TACE or radioembolisation) is used while waiting for an organ, with systemic therapy not recommended routinely as bridging but not an absolute exclusion ^[1]. Atezolizumab or other immune checkpoint inhibitors must be stopped at least 3 months before transplant because of rejection risk ^[1]. Salvage transplantation after recurrence of a resected tumour within Milan criteria is an option, and living-donor transplantation is used in many centres ^[1].

After transplant, surveillance uses contrast-enhanced abdominal CT or MRI plus chest CT, because the lungs and liver are the commonest sites of recurrence; immunosuppression with mTOR inhibitors may reduce recurrence, though the SiLVER trial did not show overall long-term benefit ^[1].

Thermal ablation (radiofrequency or microwave) is the treatment of choice for early-stage HCC of 3 cm or less in patients who are not surgical candidates ^[1]. An ablation-first strategy may also be used for centrally located tumours requiring major resection, and for tumours under 2 cm ablation gives survival comparable with resection in randomised data ^[1]. Thermal ablation yields 3-year overall survival of 76% and recurrence-free survival of about 46% for unifocal HCC of 3 cm or less ^[1]. Tumours larger than 3 cm or next to critical structures are better treated with other locoregional modalities.

• Radiation segmentectomy (selective TARE) controls solitary unresectable disease and can bridge to transplant ^[1].
• Stereotactic body radiotherapy (SBRT) and proton beam therapy provide durable local control in up to 5 sessions, including for central tumours, but are avoided in significant liver dysfunction (Child-Pugh score 8 or more, uncontrolled ascites or encephalopathy) ^[1].

TACE is the primary treatment for BCLC stage B disease ^[1], selectively catheterising feeding arteries. A systematic review of conventional TACE found an objective response rate of 52.5% (95% confidence interval 43.6% to 61.5%) and median survival of 19.4 months ^[1]. Common adverse effects include liver enzyme abnormalities, fever, marrow suppression, pain and vomiting; mortality was 0.6% ^[1]. Patients achieving an objective response survive longer than non-responders ^[1].

TACE suitability falls with poor liver function, large tumour burden, vascular invasion or markedly elevated AFP, and patients considered unsuitable are better treated with systemic therapy ^[1]. Conventional and drug-eluting bead TACE have similar efficacy ^[1]. Combining TACE with multikinase inhibitors has not shown benefit and is not recommended outside trials ^[1].

Radioembolisation (TARE, yttrium-90) is an accepted alternative for intermediate-stage disease and for radiation segmentectomy of solitary unresectable disease ^[1]. The LEGACY study (solitary HCC up to 8 cm, Child-Pugh A, good performance status) reported an objective response rate of 88.3% and a duration of response of at least 6 months in 76.1% ^[1]. Personalised dosimetry improved outcomes over standard dosimetry in the DOSISPHERE-01 trial ^[1].

Radiologic response is assessed with mRECIST (modified response evaluation criteria in solid tumours) about 6 weeks after TACE and about 12 weeks after TARE or radiotherapy; patients with no response or progression after one to two sessions are TACE/TARE-refractory and should move to systemic therapy ^[1].

Systemic therapy is for unresectable HCC not amenable to locoregional therapy, including BCLC C and suitable BCLC B. It requires preserved liver function (Child-Pugh A, or selected Child-Pugh B7) and ECOG performance status 0 to 1 ^[1].

First line

• Atezolizumab plus bevacizumab (IMbrave150): median overall survival 19.2 months versus 13.4 with sorafenib, progression-free survival 6.8 versus 4.3 months, objective response rate 29.8% versus 11.3% ^[1]. It is the preferred first-line option for Child-Pugh A patients who are candidates for VEGF-targeted therapy, but requires an upper endoscopy beforehand because of bleeding risk; large varices should be banded and carvedilol considered, with at least 2 weeks between banding and starting treatment ^[1].
• Durvalumab plus tremelimumab (HIMALAYA): median overall survival 16.4 versus 13.8 months with sorafenib, and 3-year survival 30.7% versus 20.2% ^[1]. Another preferred first-line option, particularly when VEGF therapy is contraindicated (recent gastrointestinal bleeding, severe proteinuria, or after transplantation).
• Lenvatinib and sorafenib remain first-line alternatives when immune-based combinations are contraindicated (for example severe autoimmune disease or post-transplant). Sorafenib improved survival by about 3 months over placebo in the SHARP trial (10.7 versus 7.9 months) ^[1]; lenvatinib was non-inferior to sorafenib in REFLECT ^[1].
• Pembrolizumab plus lenvatinib, and cabozantinib plus atezolizumab, are not recommended as first-line by AASLD ^[1].

Second line and beyond. After progression on atezolizumab plus bevacizumab, sorafenib or lenvatinib are preferred; after durvalumab plus tremelimumab, sorafenib or lenvatinib. Following sorafenib or lenvatinib, cabozantinib, regorafenib or ramucirumab (if AFP is 400 ng/mL or higher) are preferred; pembrolizumab may be used in patients with no prior immunotherapy ^[1]. In high-risk patients after resection or ablation, adjuvant atezolizumab plus bevacizumab initially improved recurrence-free survival at interim analysis (hazard ratio 0.72, 95% confidence interval 0.56 to 0.93, with 12-month recurrence-free survival of 78% versus 65%), but the benefit was not sustained with longer follow-up (hazard ratio 0.90, 95% confidence interval 0.72 to 1.12), and the 2025 AASLD critical update advises against adjuvant systemic therapy, with surveillance as the standard of care ^[3][4].

For BCLC D disease (major cancer symptoms or severe liver dysfunction without transplant option), treatment does not improve survival and the focus is on symptom control, management of ascites and encephalopathy, and palliative care coordination ^[2].

• After resection or ablation: CT or MRI every 3 to 6 months indefinitely ^[1].
• After transplant: abdominal CT or MRI plus chest CT, with interval guided by risk scores such as RETREAT ^[1].
• After TACE: reimaging at 6 weeks, then every 3 to 6 months; after TARE or SBRT: about 12 weeks ^[1].

• Tumour rupture with haemoperitoneum: emergency surgery or embolisation.
• Progressive liver failure from tumour replacement or treatment.
• Portal vein tumour thrombus: worsening portal hypertension and liver function.
• Post-treatment: post-hepatectomy liver failure, biliary injury, variceal bleeding, post-embolisation syndrome (fever, pain), immune-related adverse events with combination immunotherapy, and hand-foot skin reaction, diarrhoea and hypertension with multikinase inhibitors.
• Recurrence after curative therapy is the dominant late complication.

Outcome depends entirely on stage at detection and treatment access. BCLC 0/A patients have median survival of more than 5 years with curative therapy; stage B about 2.5 years; stage C about 2 years; stage D about 3 months ^[2]. Even with treatment, advanced (BCLC C) disease has median survival of about 2 years. Liver transplantation gives the best long-term cancer-free survival (median about 10 years, 5-year recurrence about 10%). Recurrence after resection is 50% to 70% at 5 years ^[1]. AFP of 400 ng/mL or higher and rising AFP signal aggressive biology and worse outcomes ^[1].

Primary prevention

• HBV vaccination for all newborns and high-risk adults who were not vaccinated at birth; universal newborn vaccination in Asia is associated with large falls in HCC incidence ^[1].
• Antiviral therapy for HBV and HCV in everyone who meets treatment criteria significantly reduces HCC risk, even after cirrhosis has developed, although risk persists with advanced fibrosis ^[1].
• Control of cofactors: avoid alcohol and tobacco, maintain healthy weight, control diabetes and metabolic syndrome; coffee consumption may be recommended in chronic liver disease ^[1].
• Control of aflatoxin through grain storage, food handling and agricultural policy in endemic regions; aflatoxin B1 acts synergistically with HBV ^[1].
• Statins, aspirin and metformin are not recommended solely to prevent HCC, though they may be used for their usual indications ^[1].

Surveillance ^[1]

• Test of choice: liver ultrasound, with or without AFP, at semi-annual (about 6-month) intervals. AASLD recommends ultrasound plus AFP ^[1].
• Target population: Child-Pugh A or B cirrhosis from any aetiology (with HCC incidence around 1% or more per year); Child-Pugh C only if the patient is a transplant candidate; non-cirrhotic chronic hepatitis B at increased risk (men older than 40 from endemic countries, women older than 50, people from Africa at earlier ages, family history of HCC, PAGE-B score above 10).
• Not for: patients with life expectancy under 1 to 2 years from non-HCC causes, non-cirrhotic NAFLD (incidence too low, about 0.008 per 100 person-years), and post-SVR HCV without cirrhosis.
• Quarter-annual surveillance did not improve detection or survival versus semi-annual surveillance in a randomised trial ^[1].
• Recall: subcentimetre lesions are followed with ultrasound at 3 to 6 months; a new or enlarging lesion 1 cm or larger, or AFP of 20 ng/mL or higher or rising, triggers multiphase CT or MRI ^[1]. Markedly elevated AFP of 200 or 400 ng/mL without a mass needs alternative imaging and possibly PET ^[1].

• Children. HCC in children is rare and usually occurs in the setting of inherited metabolic disease, HBV infection or tyrosinaemia; hepatoblastoma is the more common childhood liver cancer.
• Pregnancy. Management requires multidisciplinary teams; HCC during pregnancy is uncommon and treatment is individualised.
• Older adults. Resection and systemic therapy decisions balance tumour biology and comorbidity; many trials included only well-performing patients with Child-Pugh A disease.
• HBV-endemic populations and migrants. Patients from Africa may develop HCC at a young age; start surveillance earlier and maintain it even after migration ^[1].
• Post-transplant patients should not receive immune checkpoint inhibitors because of graft loss and death; sorafenib or lenvatinib is first-line instead ^[1].

• Aflatoxin control and HBV vaccination remain the highest-impact public health interventions; many regions without robust viral hepatitis programmes have a disproportionately high burden of HBV-related HCC and younger onset ^[1].
• Ultrasound is the single test that makes screening feasible; learn to recognise a mobile or hypervascular liver mass, rising AFP and atypical cirrhosis in any clinic with ultrasound.
• Where CT and MRI are unavailable, diagnostic decisions must rely on ultrasound, clinical context and AFP, with referral to a centre that can perform contrast imaging.
• Antiviral therapy for HBV and curative treatments such as resection, transplantation and TACE depend on regional capacity; earlier referral improves outcomes. TACE and radioembolisation may be unavailable, leaving systemic therapy as the practical option where accessible.
• Tumour markers and pathology capacity may be limited; store tissue and send for staining where possible, because mixed HCC-cholangiocarcinoma changes management.

• HCC is the commonest primary liver cancer and arises most often in cirrhosis, which carries about a 2% annual risk of HCC.
• Men are affected about two to three times more than women; incidence is highest in East Asia and sub-Saharan Africa.
• Chronic HBV can cause HCC without cirrhosis; HCV, alcohol (up to 5-fold cofactor effect), MASLD, aflatoxin, hemochromatosis, PBC, tobacco and diabetes also raise risk.
• Screening the at-risk (cirrhosis Child-Pugh A or B, or selected HBV carriers) with ultrasound plus AFP every 6 months improves early detection and survival.
• In at-risk patients, arterial phase hyperenhancement with washout on CT or MRI is HCC without biopsy; LI-RADS categories quantify the probability.
• AFP above 400 ng/mL is suggestive but no longer diagnostic; over 40% of HCC have normal AFP.
• BCLC staging combines tumour burden, liver function and performance status and dictates treatment: resection for favourable single lesions, transplantation within Milan criteria, ablation up to 3 cm, TACE for intermediate disease, systemic therapy (atezolizumab plus bevacizumab or durvalumab plus tremelimumab first-line) for advanced disease.
• Milan criteria: one lesion 1 to 5 cm, or up to three lesions each up to 3 cm.
• Systemic therapy requires preserved liver function and good performance status; atezolizumab plus bevacizumab needs endoscopy beforehand because of bleeding risk.
• Metastases are far commoner than primary liver cancer; hemangioma and focal nodular hyperplasia are benign and harmless, while adenoma can rupture or become malignant.

• Cirrhosis and its complications, surveillance ultrasound, hepatitis B and C management, liver transplantation, alcohol-related liver disease, metabolic dysfunction-associated steatotic liver disease, colorectal cancer and liver metastases, amoebic and pyogenic liver abscess.