# Hepatocellular Carcinoma - Screening and Management

## Epidemiology

Hepatocellular carcinoma is the third leading cause of cancer-related death worldwide and the sixth most common cancer overall. Its incidence is rising in Western countries, driven primarily by the growing burden of metabolic dysfunction-associated steatotic liver disease and the aging cohort of patients with prior hepatitis C virus infection. There is a male predominance of 2 to 4:1, and incidence increases with age. The overwhelming majority of hepatocellular carcinoma, 80 to 90%, arises within a cirrhotic liver, though 10 to 20% develops in patients without cirrhosis, particularly in the settings of chronic hepatitis B and MASLD.

## Risk Factors

Cirrhosis of any etiology is the strongest risk factor for hepatocellular carcinoma, with an annual incidence of 1 to 6% in cirrhotic patients. Hepatitis B virus is unique among hepatocellular carcinoma risk factors in its ability to cause cancer without cirrhosis, through direct integration of HBV DNA into the host genome, promoting hepatocarcinogenesis. The risk is proportional to the HBV DNA level and the duration of infection.

Hepatitis C virus-associated hepatocellular carcinoma risk persists even after sustained virologic response in patients with advanced fibrosis or cirrhosis, though the risk is reduced by 68 to 76%. MASLD and MASH represent the fastest growing etiologic category, and the development of hepatocellular carcinoma in non-cirrhotic MASLD patients (10 to 20% of cases) creates a particularly challenging screening paradigm. Alcohol use is synergistic with viral hepatitis and obesity in promoting hepatocarcinogenesis.

Aflatoxin B1 exposure, a mycotoxin produced by Aspergillus species endemic in Sub-Saharan Africa and Southeast Asia, causes a characteristic TP53 R249S hotspot mutation and substantially increases hepatocellular carcinoma risk. Genetic conditions including hemochromatosis, alpha-1 antitrypsin deficiency, Wilson disease, and glycogen storage diseases also confer increased risk.

## Screening

### Who to Screen (AASLD)

All patients with cirrhosis, regardless of etiology, should undergo hepatocellular carcinoma screening, unless their expected survival from comorbid conditions is limited. Patients with chronic hepatitis B without cirrhosis should be screened based on demographic and virologic risk: Asian males over age 40, Asian females over age 50, African and African American individuals over age 20, those with a family history of hepatocellular carcinoma, and those with HBV DNA exceeding 2000 IU/mL with significant fibrosis. For non-cirrhotic MASLD, no established screening protocol currently exists, though risk stratification tools are under active development.

### How to Screen

The AASLD recommends ultrasound with or without alpha-fetoprotein every 6 months as the standard screening approach. Ultrasound sensitivity for early hepatocellular carcinoma ranges from 47 to 84% and is operator-dependent, with limitations in patients with obesity and nodular cirrhosis. Alpha-fetoprotein alone has insufficient sensitivity (40 to 60%) to serve as a standalone screening test, though specificity improves at a cutoff above 20 ng/mL.

The GALAD score, incorporating gender, age, AFP-L3 fraction, AFP, and des-gamma-carboxy prothrombin, is a combined biomarker score with sensitivity of 85 to 95% for early hepatocellular carcinoma and may supplement or eventually replace ultrasound. Abbreviated MRI is emerging as an alternative to ultrasound, offering higher sensitivity, with cost-effectiveness studies ongoing. Computed tomography is not recommended as a screening tool due to radiation exposure and cost.

## Diagnostic Imaging

### LI-RADS (Liver Imaging Reporting and Data System)

The Liver Imaging Reporting and Data System provides a standardized framework for the diagnosis of hepatocellular carcinoma in at-risk patients undergoing contrast-enhanced CT or MRI. The system categorizes observations as LR-1 (definitely benign), LR-2 (probably benign), LR-3 (intermediate probability), LR-4 (probably HCC), LR-5 (definitely HCC), LR-M (probably or definitely malignant but not HCC-specific, which may represent intrahepatic cholangiocarcinoma or combined HCC-cholangiocarcinoma), and LR-TIV (tumor in vein, indicating portal vein tumor thrombus).

### Key Imaging Features for LR-5

The hallmark imaging feature of hepatocellular carcinoma is arterial phase hyperenhancement, reflecting the hypervascular nature of the tumor fed predominantly by the hepatic artery. Additional major features include non-peripheral washout in the portal venous or delayed phase, an enhancing capsule, and threshold growth defined as a 50% or greater increase in diameter within 6 months or less.

The LR-5 criteria for definitive hepatocellular carcinoma diagnosis require arterial phase hyperenhancement plus washout and/or an enhancing capsule in a lesion measuring 10 mm or greater. When these criteria are met, the diagnosis of hepatocellular carcinoma is established without the need for biopsy.

### When to Biopsy

Biopsy should be considered for LR-3 or LR-4 lesions when histologic confirmation would change clinical management. LR-M lesions should be biopsied to determine the specific tumor type, as intrahepatic cholangiocarcinoma carries different treatment and prognostic implications than hepatocellular carcinoma. Biopsy is generally recommended for lesions arising in non-cirrhotic livers, where the specificity of imaging criteria is lower. The risk of tumor seeding with biopsy is less than 2%, which is acceptable when the diagnosis will meaningfully impact management decisions.

<image>A LI-RADS classification diagram showing the diagnostic imaging algorithm for liver observations in at-risk patients. Start with "Liver observation detected in at-risk patient (cirrhosis or chronic HBV) on contrast-enhanced CT or MRI." Decision tree based on size and imaging features. First: "Arterial phase hyperenhancement (APHE) present?" If no APHE: categorize as LR-1 to LR-3 based on other features (size, stability). If APHE present and size < 10 mm: "LR-4 (probably HCC) — follow-up in 3 months." If APHE present and size >= 10 mm: check for "Non-peripheral washout AND/OR enhancing capsule?" If both present: "LR-5 (definitely HCC — diagnosis confirmed, no biopsy needed)." If only one present: "LR-4." If neither: "LR-4." Separate pathway for "Targetoid appearance (peripheral washout, rim enhancement)": "LR-M (malignant, not HCC-specific — biopsy recommended)." Include representative CT/MRI images for each LI-RADS category: arterial phase with bright enhancing nodule (APHE), portal venous phase showing washout (darker than surrounding liver), and capsule enhancement. Label phases clearly. Use green for benign, yellow for intermediate, orange for probable HCC, red for definite HCC/malignant.</image>

## Staging

### Barcelona Clinic Liver Cancer (BCLC) Staging System

| BCLC Stage | Tumor | Liver Function | PS | Treatment | Survival |
|---|---|---|---|---|---|
| 0 (Very early) | Single ≤2 cm | CTP-A | 0 | Resection, ablation, or transplant | >5 years |
| A (Early) | Single or ≤3 nodules ≤3 cm | CTP-A/B | 0 | Resection, transplant (Milan), or ablation | >5 years (transplant) |
| B (Intermediate) | Multinodular, no vascular invasion | CTP-A/B | 0 | TACE or TARE | >2.5 years |
| C (Advanced) | Vascular invasion and/or extrahepatic spread | CTP-A/B | 1-2 | Atezo+bev or durva+treme (systemic) | ~19 months |
| D (Terminal) | Any | CTP-C (not transplant candidate) | 3-4 | Best supportive care | <3 months |

The Barcelona Clinic Liver Cancer staging system integrates tumor stage, liver function as assessed by the Child-Turcotte-Pugh classification, and performance status to guide treatment decisions. BCLC stage 0 (very early) applies to patients with a single nodule of 2 cm or less with Child-Pugh A liver function and performance status 0; these patients are candidates for resection, ablation, or transplantation. BCLC stage A (early) encompasses patients with a single tumor or up to 3 nodules each 3 cm or less, with Child-Pugh A or B function and performance status 0; curative options include resection, ablation, or transplantation.

BCLC stage B (intermediate) describes patients with multinodular disease without vascular invasion or extrahepatic spread, Child-Pugh A or B function, and performance status 0; transarterial chemoembolization is the recommended treatment. BCLC stage C (advanced) includes patients with vascular invasion and/or extrahepatic spread, Child-Pugh A or B function, and performance status 1 to 2; systemic therapy is indicated. BCLC stage D (terminal) applies to patients with Child-Pugh C function who are not transplant candidates or who have performance status 3 to 4; best supportive care is appropriate.

### Milan Criteria (for Transplant Eligibility)

The Milan criteria define transplant eligibility as a single tumor 5 cm or less, or up to 3 tumors each 3 cm or less, with no macrovascular invasion and no extrahepatic disease. Patients transplanted within Milan criteria achieve 5-year post-transplant survival of 70 to 75% with a recurrence rate below 10 to 15%.

### Beyond Milan

The UCSF criteria expand eligibility to a single tumor 6.5 cm or less, or 2 to 3 tumors each 4.5 cm or less with a total tumor diameter not exceeding 8 cm. The AFP score, developed in France, incorporates alpha-fetoprotein level into transplant selection criteria. AFP levels exceeding 1000 ng/mL are associated with high recurrence risk and may exclude patients from transplant eligibility at some centers. Downstaging through locoregional therapy (TACE, Y-90, ablation) to bring tumors within Milan criteria has been validated, with successful downstaging yielding transplant outcomes comparable to patients who were within Milan criteria from the outset.

## Treatment

### Curative Therapies

#### Surgical Resection

Surgical resection is the preferred curative approach for non-cirrhotic hepatocellular carcinoma and for patients with compensated cirrhosis (Child-Pugh A) with a single tumor and preserved liver function. Preoperative assessment includes evaluation of hepatic reserve through indocyanine green retention testing and future liver remnant volumetry, as well as assessment of portal hypertension, since an HVPG exceeding 10 mmHg is associated with worse postoperative outcomes. Platelet count and MELD score provide additional prognostic information.

Five-year survival following resection ranges from 50 to 70%, though 5-year recurrence rates are 50 to 70%, reflecting the high rate of de novo tumor development in the remaining cirrhotic liver. Laparoscopic and robotic approaches offer equivalent oncologic outcomes with reduced perioperative morbidity.

#### Liver Transplantation

Liver transplantation is the only curative option that simultaneously addresses both the tumor and the underlying cirrhotic liver disease. The Milan criteria remain the standard, though center-specific expanded criteria are increasingly applied. Locoregional bridging therapy, using ablation or TACE while on the transplant waiting list, prevents tumor progression and reduces dropout. Living donor liver transplantation expands the donor pool and may permit more liberal selection criteria, though comprehensive donor risk assessment is essential given a mortality risk of 0.1 to 0.5% for right lobe donation.

#### Ablation

Radiofrequency ablation is effective for tumors 3 cm or less and has been shown to be equivalent to resection for tumors 2 cm or less in randomized controlled trials. Microwave ablation provides faster treatment times, larger ablation zones, and less susceptibility to the heat-sink effect from adjacent blood vessels, and is increasingly preferred over RFA. Cryoablation and irreversible electroporation serve as alternatives for lesions adjacent to critical structures. A minimum ablation margin of 1 cm is recommended for optimal local control. Local recurrence occurs in 10 to 20%, and new tumors develop in the cirrhotic liver at rates comparable to those observed after resection (50 to 70% at 5 years).

### Locoregional Therapies

#### Transarterial Chemoembolization (TACE)

TACE is the standard of care for BCLC stage B (intermediate) hepatocellular carcinoma in patients with multinodular disease and compensated cirrhosis without vascular invasion. Conventional TACE involves emulsification of lipiodol with a chemotherapeutic agent, typically doxorubicin, followed by injection of embolic particles such as gelfoam. Drug-eluting bead TACE uses calibrated microspheres loaded with doxorubicin, providing more standardized drug delivery with possibly fewer systemic side effects, as demonstrated in the PRECISION V trial, though overall efficacy is comparable to conventional TACE.

Treatment sessions are repeated every 6 to 12 weeks based on imaging response assessed by modified RECIST criteria. Post-embolization syndrome, consisting of fever, abdominal pain, and nausea, occurs in 50 to 60% of patients and is self-limited. TACE is contraindicated in decompensated cirrhosis (Child-Pugh C), main trunk portal vein thrombosis, and severe hepatic dysfunction.

#### Transarterial Radioembolization (TARE/Y-90)

Transarterial radioembolization delivers selective internal radiation therapy using yttrium-90-labeled microspheres (glass or resin). Unlike TACE, it can be safely used in patients with portal vein invasion. It can be delivered in a lobar or segmental fashion, with the radiation segmentectomy concept providing high local doses to achieve ablative radiation effects.

The DOSISPHERE-01 trial demonstrated that personalized dosimetry improved response rates and survival outcomes. Y-90 has an expanding role as a bridge to transplantation or resection, as a downstaging strategy, as an alternative to TACE for BCLC stage B disease, and for treatment in the setting of portal vein tumor thrombus. Side effects include fatigue and post-radioembolization syndrome, with radiation-induced liver disease being a risk if delivery is non-selective.

### Systemic Therapy

#### First-Line

| Line | Regimen | Key Trial | OS (months) | Key Notes |
|---|---|---|---|---|
| 1st | Atezolizumab + bevacizumab | IMbrave150 | 19.2 vs 13.4 (sorafenib) | Screen/treat varices before bevacizumab |
| 1st | Durvalumab + tremelimumab (STRIDE) | HIMALAYA | Superior to sorafenib | No variceal bleeding risk; single priming treme dose |
| 1st (alt) | Lenvatinib | REFLECT | Non-inferior to sorafenib | Higher response rate; for non-immunotherapy candidates |
| 2nd | Cabozantinib | CELESTIAL | 10.2 vs 8.0 | After sorafenib/lenvatinib progression |
| 2nd | Ramucirumab | REACH-2 | Benefit only if AFP ≥400 | Anti-VEGFR2; biomarker-selected |
| 2nd | Regorafenib | RESORCE | 10.6 vs 7.8 | After sorafenib progression |
| Adjuvant | Atezolizumab + bevacizumab | IMbrave050 | Improved RFS | First positive adjuvant therapy (FDA 2023) |

The combination of atezolizumab (anti-PD-L1) plus bevacizumab (anti-VEGF) established a new standard of care based on the IMbrave150 trial, which demonstrated an overall survival of 19.2 months compared with 13.4 months for sorafenib. Before initiating bevacizumab, patients must be screened for and treated for esophageal varices to mitigate the bleeding risk associated with anti-VEGF therapy. Adequate liver function, preferably Child-Pugh A, is recommended.

The STRIDE regimen of durvalumab (anti-PD-L1) plus tremelimumab (anti-CTLA-4) provides an alternative first-line option based on the HIMALAYA trial. This regimen uses a single priming dose of tremelimumab 300 mg followed by durvalumab 1500 mg every 4 weeks, and notably carries no variceal bleeding risk, making it suitable for patients with untreated or high-risk varices.

Sorafenib, the multi-kinase inhibitor that established the first systemic therapy for hepatocellular carcinoma in the SHARP trial (overall survival 10.7 versus 7.9 months compared with placebo), is now relegated to second-line use or reserved for patients who are not candidates for immunotherapy. Lenvatinib, another multi-kinase inhibitor, demonstrated non-inferiority to sorafenib in the REFLECT trial with higher response rates, and serves as an alternative when immunotherapy-based regimens are not suitable.

#### Second-Line

Cabozantinib demonstrated an overall survival of 10.2 months versus 8.0 months after sorafenib progression in the CELESTIAL trial. Ramucirumab, an anti-VEGFR2 antibody, showed benefit in the REACH-2 trial but only in patients with AFP levels of 400 ng/mL or greater. Regorafenib was validated in the RESORCE trial for progression after sorafenib. Pembrolizumab has received conditional approval based on the KEYNOTE-240 and KEYNOTE-394 trials.

<image>A BCLC staging and treatment algorithm for hepatocellular carcinoma. Create a horizontal staging diagram with five columns for each BCLC stage (0, A, B, C, D). For each stage, show: Patient criteria (tumor characteristics, CTP class, performance status), Recommended treatment, and Expected survival. BCLC 0 (very early): single <=2 cm, CTP-A, PS 0; treatment: resection/ablation/transplant; survival: >5 years. BCLC A (early): single or <=3 nodules <=3 cm, CTP-A/B, PS 0; treatment: resection, transplant (within Milan), or ablation; survival: >5 years with transplant. BCLC B (intermediate): multinodular, no vascular invasion, CTP-A/B, PS 0; treatment: TACE or TARE; survival: >2.5 years. BCLC C (advanced): portal invasion and/or extrahepatic spread, CTP-A/B, PS 1-2; treatment: atezolizumab+bevacizumab or durvalumab+tremelimumab (first-line systemic); survival: ~19 months. BCLC D (terminal): CTP-C not for transplant, PS 3-4; treatment: best supportive care; survival: <3 months. Include arrows showing stage migration (downstaging from B to A, upstaging from A to C). Show treatment modalities with small icons: scalpel for surgery, coil for ablation, catheter for TACE/TARE, pills/IV for systemic, liver icon for transplant. Use color gradient from green (early/curative) to red (terminal). Include Milan criteria box and beyond-Milan downstaging pathway.</image>

## Adjuvant Therapy

The IMbrave050 trial demonstrated that adjuvant atezolizumab plus bevacizumab following curative resection or ablation reduced recurrence-free survival events, leading to FDA approval in 2023 as the first positive adjuvant therapy in hepatocellular carcinoma. This represents a landmark advance in post-curative treatment.

## Key Clinical Pearls

- Hepatocellular carcinoma can be diagnosed by imaging alone (LI-RADS 5) in at-risk patients. Biopsy is not required if classic features are present (arterial phase hyperenhancement plus washout and/or capsule in a lesion 10 mm or greater).
- All cirrhotic patients should be screened with ultrasound with or without AFP every 6 months. The GALAD score and abbreviated MRI may improve sensitivity.
- Atezolizumab plus bevacizumab is the current first-line systemic therapy for unresectable hepatocellular carcinoma (IMbrave150). Patients must be screened for varices before bevacizumab.
- Durvalumab plus tremelimumab (STRIDE) is an alternative first-line option without variceal bleeding risk.
- Successful downstaging to within Milan criteria yields post-transplant outcomes equivalent to those within Milan from the outset.
- MASLD-related hepatocellular carcinoma can occur without cirrhosis in 10 to 20% of cases, representing a major screening challenge.
- Adjuvant atezolizumab plus bevacizumab (IMbrave050) is the first proven adjuvant therapy after curative hepatocellular carcinoma treatment.
- AFP exceeding 1000 ng/mL is associated with poor post-transplant outcomes, and some programs use AFP-based scoring to refine transplant eligibility.

## References
1. Reig M, et al. BCLC strategy for prognosis prediction and treatment recommendation: The 2022 update. *J Hepatol*. 2022;76(3):681-693.
2. Finn RS, et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma (IMbrave150). *N Engl J Med*. 2020;382(20):1894-1905.
3. Abou-Alfa GK, et al. Tremelimumab plus durvalumab in unresectable hepatocellular carcinoma (HIMALAYA). *NEJM Evid*. 2022;1(8):EVIDoa2100070.
4. Marrero JA, et al. Diagnosis, staging, and management of hepatocellular carcinoma: 2018 Practice Guidance by the AASLD. *Hepatology*. 2018;68(2):723-750.
5. Qin S, et al. Atezolizumab plus bevacizumab versus active surveillance in patients with resected or ablated high-risk hepatocellular carcinoma (IMbrave050). *Lancet*. 2023;402(10415):1835-1847.