Skip to content

Hepatocellular carcinoma — guidelines

TL;DR — Major guidelines agree on surveillance of sufficiently high-risk, treatment-eligible populations; multiphase imaging diagnosis in defined at-risk livers; joint tumor/liver staging; curative local options for early disease; locoregional therapy for selected intermediate disease; and immunotherapy-based first-line treatment for eligible advanced HCC (Singal 2023, PMID 37199193; EASL 2025, PMID 39690085; Vogel 2025, PMID 39986353). Disagreement clusters where evidence is weakest or resources differ: AFP in surveillance, contrast-enhanced ultrasound, non-cirrhotic HBV/MASLD eligibility, resection with portal hypertension, transplant expansion, TACE versus systemic migration, and availability of regional regimens. APASL, Japanese, Korean, Chinese, European, and US documents are not interchangeable because disease etiology, donor systems, technology, and regulatory access differ (Omata 2017, PMID 28620797; Kudo 2021, PMID 34239808; KLCA-NCC 2022, PMID 36263666). Guideline age matters acutely in HCC because systemic and TACE-combination evidence changes within months. Recommendations should therefore be recorded with year, jurisdiction, evidence level, and supersession status.

Core documents in this synthesis

Body/document Year Region Scope PubMed record
AASLD Practice Guidance 2023 USA Prevention, surveillance, diagnosis, staging, treatment Singal et al., PMID 37199193
EASL Clinical Practice Guidelines 2025 Europe/international Full HCC management EASL, PMID 39690085
ESMO Clinical Practice Guideline 2025 Europe/international oncology Diagnosis, treatment, follow-up Vogel et al., PMID 39986353
BCLC strategy update 2022 International framework Prognosis and treatment allocation Reig et al., PMID 34801630
APASL guideline update 2017 Asia-Pacific Full HCC management Omata et al., PMID 28620797
JSH consensus update 2021 Japan Prevention through systemic therapy Kudo et al., PMID 34239808
KLCA-NCC practice guideline 2022 Korea Full HCC management KLCA-NCC, PMID 36263666
Chinese guideline review 2022 guideline summarized 2023 China Updated Chinese recommendations Xie et al., PMID 37124695

The guideline registry records document-level status and supersession; this page synthesizes recommendations.

Surveillance eligibility

Population Broad consensus Main disagreement
Cirrhosis, treatment eligible Six-month surveillance Modality, AFP, exceptions near end-stage liver disease
Chronic HBV without cirrhosis Selected demographic/clinical risk groups Age/sex/region thresholds and score use
HCV cirrhosis after cure Continue surveillance Whether/when risk falls low enough to stop
Advanced fibrosis without cirrhosis Individualized in some systems Evidence and cost-effectiveness
Non-cirrhotic MASLD No blanket surveillance How to enrich high-risk subgroup
Child-Pugh C not transplant eligible Generally no surveillance benefit Reversibility and candidacy reassessment

AASLD conditions surveillance on meaningful treatment eligibility and uses ultrasound plus AFP at approximately six-month intervals (Singal 2023, PMID 37199193). EASL 2025 emphasizes personalized, risk-based surveillance and new imaging/biomarker strategies while retaining ultrasound-based practice (PMID 39690085).

AFP disagreement

Historic guideline differences over AFP reflect a familiar tradeoff:

  • adding AFP increases early-stage sensitivity;
  • specificity falls and downstream imaging rises;
  • assay availability and underlying inflammatory activity vary;
  • longitudinal algorithms are not standardized.

Meta-analysis found early-stage sensitivity of 47% for ultrasound alone and 63% with AFP, supporting the contemporary AASLD combined approach while leaving room for different value judgments (Tzartzeva 2018, PMID 29425931).

Imaging diagnosis

Major guidelines permit non-invasive HCC diagnosis in defined high-risk populations using multiphase CT/MRI arterial enhancement and washout-based criteria. Differences include:

  • which high-risk populations qualify;
  • minimum lesion size;
  • extracellular versus hepatobiliary MRI contrast interpretation;
  • contrast-enhanced ultrasound as first- or second-line diagnosis;
  • handling of indeterminate lesions;
  • whether pathology is required before systemic therapy.
Issue US/European tendency Asia-Pacific variation
CT/MRI Central non-invasive pathway Central, with region-specific criteria
CEUS Often second-line/problem-solving More prominent in some guidelines
Hepatobiliary phase Ancillary/structured use Sometimes integrated more strongly
Biopsy Indeterminate/non-at-risk/LR-M Varies with imaging algorithm and treatment plan

LI-RADS provides a standardized US framework; regional criteria should not be mixed within one lesion report.

Staging and allocation

BCLC 2022 is the main AASLD/EASL-linked strategy, combining tumor burden, liver function, and performance status with treatment options (Reig 2022, PMID 34801630). Asian guidelines may use regional algorithms that recommend more aggressive resection or locoregional therapy for selected disease beyond typical BCLC boundaries (Omata 2017, PMID 28620797; Kudo 2021, PMID 34239808).

Disagreement does not always mean contradictory evidence. It can reflect:

  • HBV-predominant versus HCV/MASLD populations;
  • living- versus deceased-donor availability;
  • surgical and ablation expertise;
  • access to TARE, proton therapy, or hepatic arterial infusion;
  • national drug approvals;
  • different acceptable tradeoffs between treatment intensity and certainty.

Resection

Question Shared principle Variation
Liver reserve Must be adequate Measurement and permissible impairment
Portal hypertension Raises risk Absolute versus relative contraindication
Multifocal disease Usually not standard Western early-stage resection More expansive selection in some Asian algorithms
Major vascular invasion Usually systemic/advanced pathway Selected surgical approaches in expert Asian centres
Future liver remnant Essential Threshold depends on liver quality and technique

Guidelines are converging toward multidisciplinary selection rather than a single diameter rule, but comparative evidence for expanded surgery is often observational.

Ablation and radiation

Thermal ablation is a curative option for very small early tumors, particularly when surgery is unsuitable. Modality and size cutoffs vary. External-beam radiation has moved from rescue therapy toward a recognized alternative or combination in AASLD/EASL-era practice, but strength and placement differ with local expertise (Singal 2023, PMID 37199193; EASL 2025, PMID 39690085).

Transplantation

Milan remains the common benchmark, but guidelines and allocation systems differ in:

  • expanded morphology;
  • AFP thresholds;
  • downstaging entry limits;
  • required stability/waiting time;
  • living-donor exceptions;
  • post-transplant surveillance.

Metroticket 2.0 and AFP-based models move selection toward continuous biology, while public allocation still needs transparent categorical rules (Mazzaferro 2018, PMID 28989060).

Intermediate-stage disease

All modern guidelines recognize BCLC B heterogeneity. TACE remains central for selective embolization candidates, but treatment migration is increasingly explicit.

Feature TACE direction Systemic direction
Discrete, catheter-selectable nodules Favored Optional/after failure
Preserved portal flow/liver function Favored Still eligible
Diffuse/infiltrative burden Low expected benefit Favored
Repeated incomplete response Stop/reassess Migrate
Decompensation Often unsafe Systemic evidence also weak; supportive/transplant review

EMERALD-1 and LEAP-012 were published after several core guidelines and will drive updates on TACE-systemic combinations (Sangro 2025, PMID 39798579; Kudo 2025, PMID 39798578).

Advanced systemic therapy

Regimen Guideline position driver Selection concern
Atezolizumab-bevacizumab IMbrave150 OS/PFS superiority Varices/bleeding, hypertension, proteinuria
Tremelimumab-durvalumab HIMALAYA OS superiority Immune toxicity; transplant/autoimmune context
Nivolumab-ipilimumab CheckMate 9DW OS 23.7 vs 20.6 months; HR 0.79 (95% CI 0.65–0.96) Early crossing of survival curves; immune toxicity (Yau 2025, PMID 40349714)
Lenvatinib or sorafenib Phase 3 TKI evidence When IO unsuitable; oral toxicity
Regional IO/TKI regimens CARES-310 and national approvals Jurisdiction and access

No guideline can infer head-to-head superiority from separate sorafenib-controlled trials.

Liver-function evidence boundary

Guidelines appropriately inherit trial eligibility limits. Most systemic RCT evidence is Child-Pugh A; recommendations for Child-Pugh B are conditional and selective. Locoregional recommendations also narrow as bilirubin, ascites, encephalopathy, and portal flow worsen.

A guideline table should never be read without the liver-function footnote.

Areas of genuine disagreement

  1. AFP inclusion and interpretation.
  2. CEUS placement in diagnosis.
  3. Surveillance of advanced fibrosis and selected non-cirrhotic HBV.
  4. Resection with portal hypertension or limited multifocal disease.
  5. Transplant expansion/downstaging boundaries.
  6. TARE, radiation, and hepatic arterial infusion positioning.
  7. When BCLC B should start systemic therapy.
  8. Choice and sequencing among first-line IO combinations.
  9. Regional availability of drugs and technologies.

How to use guidelines safely

  • Confirm publication year and supersession.
  • Identify region and resource assumptions.
  • Separate recommendation strength from evidence certainty.
  • Check whether the patient population matches trial eligibility.
  • Use multidisciplinary judgment where recommendations diverge.
  • Document why a non-default strategy was chosen.
  • Recheck rapidly changing systemic and combination sections.

Guideline positions are partly evidence and partly system design

Decision AASLD/EASL tendency Asia-Pacific/Korea/China tendency Evidence beneath disagreement
Surveillance biomarker Ultrasound plus AFP in AASLD; EASL historically more cautious about AFP AFP and additional markers have wider roles in several Asian pathways US+AFP early sensitivity ~63% versus ~47% for US alone, with more false positives (Tzartzeva 2018, PMID 29425931)
Surveillance interval Six months Usually six months, with risk-adapted variants in some systems Six versus twelve months increased early detection, RR 1.17 (95% CI 1.08–1.26), in pooled evidence (Yang 2023, PMID 36921104)
Noninvasive diagnosis High specificity in a defined at-risk population; CT/MRI central Wider use of hepatobiliary-phase MRI, CEUS, and region-specific agents/criteria Different contrast agents and transplant allocation consequences change acceptable sensitivity/specificity (Kim 2019, PMID 30759967; Joo 2023, PMID 36606612)
Resection with portal hypertension More restrictive in classic Western algorithms Selected resection more often considered in high-volume Asian practice Observational selection dominates; risk is increased but not binary (Cho 2023, PMID 36907570)
TACE Default for suitable BCLC B, with early migration for unsuitable/refractory disease More granular transarterial algorithms and HAIC use in parts of Asia BCLC B heterogeneity and regional expertise drive divergence (Lee 2023, PMID 36606613)
Systemic therapy IO combinations first line; agent choice conditioned by bleeding/autoimmune risk Similar global backbone plus jurisdiction-specific agents and sequences APASL 2024 incorporates Asian phase-3 agents unavailable elsewhere (Lau 2024, PMID 39570557)
Transplant diagnosis/allocation OPTN rules require allocation-specific imaging categories National allocation rules differ A diagnostic guideline and an organ-allocation rule serve different error tolerances (Kierans 2023, PMID 36097856)

Imaging disagreement has downstream consequences

International comparisons show that lesion size threshold, arterial-phase hyperenhancement definition, washout phase, capsule, hepatobiliary-phase hypointensity, and CEUS eligibility differ among systems (Kim 2019, PMID 30759967). In transplant candidates, gadoxetate MRI and CT can classify the same lesion differently against explant pathology, affecting both diagnosis and allocation (Jeon 2020, PMID 32333148; Odedra 2022, PMID 36195953). OPTN therefore cannot simply import every LI-RADS research refinement without considering allocation specificity, auditability, and equity (Kierans 2023, PMID 36097856).

Korean 2022 updates explicitly separate imaging, local ablation, and transarterial recommendations, reflecting modalities such as Sonazoid CEUS and high-volume regional practice (Joo 2023, PMID 36606612; Lee 2023, PMID 36606614; Lee 2023, PMID 36606613). China’s 2024 guideline similarly integrates surgical, interventional, radiotherapy, HAIC, and systemic strategies within its national resource and evidence context (Zhou 2025, PMID 41063733). These are not merely translations of BCLC.

New evidence creates timing mismatches

Guidelines freeze evidence at different search dates. The 2025 EASL and ESMO documents can incorporate newer first-line immunotherapy evidence than APASL 2017; APASL’s separate 2024 systemic guideline updates only one domain (EASL 2025, PMID 39690085; Vogel 2025, PMID 39986353; Omata 2017, PMID 28620797; Lau 2024, PMID 39570557). A “current” transplant section and a “current” systemic section may therefore come from different documents within the same region.

Community evidence can also challenge the evidence hierarchy. Annual HBV screening in 14,426 Chinese participants was associated with corrected mortality HR 0.74 for prevalence-round and 0.52 for incident-round cancers, supporting feasibility but not settling six- versus twelve-month guidance for cirrhosis (Zeng 2023, PMID 37667043). Patient-centered imaging reviews argue for switching away from repeatedly limited ultrasound, while capacity and cost prevent a universal CT/MRI recommendation (Heald 2025, PMID 39527256).

How to use conflicting recommendations

  1. Identify the patient population to which the recommendation applies: cirrhosis versus HBV without cirrhosis, transplant candidate versus general diagnosis, Child-Pugh A versus decompensated disease.
  2. Record the evidence date and jurisdictional approval/access.
  3. Separate the society’s evidence judgment from its health-system constraints.
  4. State the class/strength and evidence level when the document provides them.
  5. For disagreement, present both thresholds and the tradeoff rather than silently selecting one.
  6. Reassess after each treatment because stage migration and liver-function change can move the patient into a different recommendation branch.

Unresolved disagreements

  • AFP. Added sensitivity is measurable, but false-positive workup and assay interpretation produce different recommendations (Tzartzeva 2018, PMID 29425931).
  • Hepatobiliary-phase MRI. Greater lesion sensitivity can lower specificity and alter transplant listing; the appropriate balance differs for clinical care and organ allocation (Jeon 2020, PMID 32333148; Kierans 2023, PMID 36097856).
  • Resection boundaries. Asian algorithms more often extend resection to selected portal-hypertensive, multifocal, or vascular-invasion contexts; randomized comparative evidence is sparse (Cho 2023, PMID 36907570; Wen 2022, PMID 35197399).
  • TACE combinations. PFS-positive phase-3 trials arrived after many guidance searches; whether to recommend them universally before mature OS depends on society methodology and resource context (Sangro 2025, PMID 39798579; Kudo 2025, PMID 39798578).
  • Regional systemic agents. Donafenib, camrelizumab-rivoceranib, tislelizumab, and other agents have phase-3 evidence but uneven approval, reimbursement, and comparator relevance (Lau 2024, PMID 39570557).

Open questions

  • Can one risk-based surveillance framework transport across HBV-, HCV-, alcohol-, and MASLD-dominant regions?
  • Which portal-hypertension measure should determine resection eligibility?
  • How should EMERALD-1 and LEAP-012 reshape intermediate-stage algorithms before mature OS (Sangro 2025, PMID 39798579; Kudo 2025, PMID 39798578)?
  • Can transplant expansion rules optimize both recurrence and geographic equity?
  • How should living-donor availability influence guideline boundaries without weakening donor protection?

References

  1. Singal AG, et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78:1922-1965. PMID 37199193
  2. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatocellular carcinoma. J Hepatol. 2025;82:315-374. PMID 39690085
  3. Reig M, et al. BCLC strategy for prognosis prediction and treatment recommendation: the 2022 update. J Hepatol. 2022;76:681-693. PMID 34801630
  4. Omata M, et al. Asia-Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int. 2017;11:317-370. PMID 28620797
  5. Kudo M, et al. Management of hepatocellular carcinoma in Japan: JSH consensus statements and recommendations 2021 update. Liver Cancer. 2021;10:181-223. PMID 34239808
  6. Korean Liver Cancer Association and National Cancer Center Korea. 2022 KLCA-NCC Korea practice guidelines for the management of hepatocellular carcinoma. Clin Mol Hepatol. 2022;28:583-705. PMID 36263666
  7. Xie DY, et al. A review of 2022 Chinese clinical guidelines on the management of hepatocellular carcinoma: updates and insights. Hepatobiliary Surg Nutr. 2023. PMID 37124695
  8. Tzartzeva K, et al. Surveillance imaging and AFP for early detection of HCC in cirrhosis: a meta-analysis. Gastroenterology. 2018;154:1706-1718.e1. PMID 29425931
  9. Mazzaferro V, et al. Metroticket 2.0 model after liver transplantation for HCC. Gastroenterology. 2018;154:128-139. PMID 28989060
  10. Sangro B, et al. Durvalumab with or without bevacizumab with TACE in HCC (EMERALD-1). Lancet. 2025;405:216-232. PMID 39798579
  11. Kudo M, et al. TACE with lenvatinib-pembrolizumab (LEAP-012). Lancet. 2025;405:203-215. PMID 39798578
  12. Lau G, et al. APASL clinical practice guidelines on systemic therapy for HCC—2024. Hepatol Int. 2024;18:1661-1683. PMID 39570557
  13. Kim TH, et al. Comparison of international guidelines for noninvasive HCC diagnosis: 2018 update. Clin Mol Hepatol. 2019;25:245-263. PMID 30759967
  14. Jeon SK, et al. Guideline comparison for HCC diagnosis using gadoxetic MRI in transplant candidates. Eur Radiol. 2020;30:4762-4771. PMID 32333148
  15. Lee MW, et al. 2022 KLCA-NCC guidelines for local ablation: what is new? Korean J Radiol. 2023;24:10-14. PMID 36606614
  16. Lee IJ, et al. 2022 KLCA-NCC guidelines for transarterial therapy: what is new? Korean J Radiol. 2023;24:6-9. PMID 36606613
  17. Joo I, et al. 2022 KLCA-NCC guidelines for imaging diagnosis: what is new? Korean J Radiol. 2023;24:1-5. PMID 36606612
  18. Zhou J, et al. China Liver Cancer Guidelines for diagnosis and treatment of HCC, 2024 edition. Liver Cancer. 2025;14:779-835. PMID 41063733
  19. Wen N, et al. Clinical management of HCC worldwide: guideline comparison, 2022 update. Biosci Trends. 2022;16:20-30. PMID 35197399
  20. Cho Y, et al. Overview of Asian HCC clinical practice guidelines. Clin Mol Hepatol. 2023;29:252-262. PMID 36907570
  21. Odedra D, et al. International diagnostic guidelines and transplant allocation: gadoxetic MRI versus CT with explant correlation. Cancer Imaging. 2022;22:55. PMID 36195953
  22. Heald J, et al. Patient-centered HCC surveillance: complementary roles of ultrasound and CT/MRI. Abdom Radiol (NY). 2025;50:2088-2096. PMID 39527256
  23. Kierans AS, et al. OPTN HCC classification, LI-RADS alignment, and gaps. Liver Transpl. 2023;29:206-216. PMID 36097856
  24. Zeng H, et al. HCC screening in HBsAg-positive individuals in China: prospective multicenter study. Nat Cancer. 2023;4:1382-1394. PMID 37667043
  25. Yang J, et al. Comparative effectiveness of HCC screening intervals and modalities: a meta-analysis. Chin Med J (Engl). 2023;136:1322-1330. PMID 36921104
  26. Vogel A, et al. Hepatocellular carcinoma: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2025;36:491-506. PMID 39986353
  27. Yau T, et al. Nivolumab plus ipilimumab versus lenvatinib or sorafenib as first-line treatment for unresectable hepatocellular carcinoma (CheckMate 9DW). Lancet. 2025;405:1851-1864. PMID 40349714