Skip to content

Hepatocellular carcinoma — overview

TL;DR — HCC accounts for approximately 90% of primary liver cancer (EASL 2025, PMID 39690085); liver cancer is the third leading cause of cancer death globally (Bray 2024, PMID 38572751), and HCC usually arises on a diseased liver in a population shifting from viral hepatitis toward metabolic (MASLD) and alcohol-related disease (Huang 2021, PMID 33349658). Every decision runs through two organs at once: tumor stage and liver function are combined in the BCLC system to allocate resection, ablation, transplantation, locoregional, or systemic therapy (Reig 2022, PMID 34801630). Surveillance of at-risk patients with semi-annual ultrasound ± AFP is guideline-standard but rests on one positive RCT plus observational data, and misses over a third of early tumors — a live controversy (Zhang 2004, PMID 15042359; Kansagara 2014, PMID 24934699; Arvind 2025, PMID 39943795). First-line systemic therapy is now immunotherapy-combination based (atezolizumab–bevacizumab, durvalumab–tremelimumab, nivolumab–ipilimumab), yet none of HCC's recurrent driver mutations is druggable (Finn 2020, PMID 32402160; Schulze 2015, PMID 25822088). The biggest unknown: whether surveillance and treatment paradigms built on viral-era cirrhosis transfer to the growing MASLD population, a substantial fraction of whom develop HCC without cirrhosis (Huang 2021, PMID 33349658).

Definition and diagnostic anchors

  • HCC is the malignant transformation of hepatocytes; it accounts for approximately 90% of primary liver cancers (EASL 2025, PMID 39690085).
  • Diagnosis is unusual among solid tumors: in cirrhotic livers a lesion showing arterial-phase hyperenhancement with venous/delayed washout on multiphase CT/MRI can be diagnosed non-invasively, without biopsy; nodules <1 cm are followed with short-interval imaging instead (Della Corte 2012, PMID 22846857; AASLD guidance: Singal 2023, PMID 37199193).
  • The 2025 EASL guidelines standardize imaging procedures and diagnostic criteria and emphasize multidisciplinary, multiparametric decision-making (EASL 2025, PMID 39690085).

Headline epidemiology

Fact Figure Source
Global liver cancer rank (deaths) 3rd leading cause of cancer death Bray 2024, PMID 38572751
HCC share of primary liver cancer ~90% EASL 2025, PMID 39690085
Fastest-growing HCC cause (USA/France/UK) NAFLD/MASLD Huang 2021, PMID 33349658
Annual HCC incidence, NASH cirrhosis 0.5–2.6%/yr Huang 2021, PMID 33349658
Annual HCC incidence, non-cirrhotic NAFLD ~0.1–1.3 per 1,000 patient-yr Huang 2021, PMID 33349658
HR for HCC, non-cirrhotic NAFLD vs matched controls (Korea) 2.69 (95% CI 1.33–5.44) Lee 2025, PMID 40345323

The etiologic transition. The classical drivers — chronic HBV (vaccine-preventable; Shepard 2006, PMID 16754644) and HCV (now curable) — are receding where vaccination and antivirals reach, while metabolic and alcohol-related liver disease grow: alcohol-associated cirrhosis and its HCC burden are projected to rise globally (Huang 2023, PMID 36258033), and NAFLD/MASLD-related HCC is projected to increase with the obesity epidemic (Huang 2021, PMID 33349658). This matters clinically because MASLD-HCC more often arises without cirrhosis, outside the surveilled population (Huang 2021, PMID 33349658; Lee 2025, PMID 40345323).

Surveillance: the standing debate

  • For: the single large RCT (Shanghai, 18,816 HBV-infected adults) found biannual AFP + ultrasound reduced HCC mortality 37% (rate ratio 0.63, 95% CI 0.41–0.98) despite 58% adherence (Zhang 2004, PMID 15042359). A meta-analysis of 59 cohort studies (145,396 patients) associates surveillance with better early detection (RR 1.86), curative treatment receipt (RR 1.83), and survival (HR 0.67) after lead-time adjustment (Singal 2022, PMID 35139400).
  • Against/caution: a systematic review graded the mortality evidence "very low strength," noting the RCT's methodologic flaws (Kansagara 2014, PMID 24934699); modeling suggests ~13 deaths averted but 150 false-positive experiences per 1,000 surveilled over 5 years (number needed to harm ≈7) (Taylor 2017, PMID 28605060).
  • Practice reality: ultrasound sensitivity for early HCC is poor in obesity and non-viral etiologies; ultrasound+AFP still misses >1/3 of early-stage tumors; surveillance is underused, especially in racial-minority and low-income groups (Arvind 2025, PMID 39943795).

Mechanism and molecular biology (sketch)

Exome sequencing of 243 tumors defined the landscape: TERT promoter mutation is the earliest and most common event; mutation groups center on CTNNB1 (alcohol-associated), TP53 (HBV-associated), and AXIN1; etiology leaves risk-factor-specific mutational signatures (alcohol+tobacco, aflatoxin B1); only ~28% of tumors carried an alteration targetable by an FDA-approved drug — and none of the three dominant drivers (TERT, CTNNB1, TP53) is druggable (Schulze 2015, PMID 25822088). TERT promoter mutation and chromosome 8p loss also characterize NAFLD-related HCC (Ki Kim 2016, PMID 27511114). Consequence: unlike lung adenocarcinoma, HCC systemic therapy is not genomically stratified; treatment selection uses stage and liver function, not mutations.

Treatment landscape

Allocation follows BCLC 2022, which maps tumor burden + liver function + performance status to treatment intent (Reig 2022, PMID 34801630), synthesized with AASLD 2023 guidance (Singal 2023, PMID 37199193) and EASL 2025 guidelines (PMID 39690085).

Modality Anchor evidence Key numbers
Transplantation (early HCC + cirrhosis) Milan criteria: single ≤5 cm or ≤3 nodules ≤3 cm (Mazzaferro 1996, PMID 8594428) 4-yr survival 85% within criteria
Expanded transplant selection Metroticket 2.0: AFP + number+size model (Mazzaferro 2018, PMID 28989060) c-statistic 0.78; outperformed Milan/UCSF for 5-yr HCC-specific survival
First systemic agent SHARP: sorafenib vs placebo (Llovet 2008, PMID 18650514) OS 10.7 vs 7.9 mo, HR 0.69
TKI alternative REFLECT: lenvatinib non-inferior to sorafenib (Kudo 2018, PMID 29433850) OS 13.6 vs 12.3 mo, HR 0.92
Second line after sorafenib RESORCE: regorafenib (Bruix 2017, PMID 27932229) OS 10.6 vs 7.8 mo, HR 0.63
First-line standard since 2020 IMbrave150: atezolizumab + bevacizumab vs sorafenib (Finn 2020, PMID 32402160) superior OS and PFS; practice-changing (Kulik 2020, PMID 33376711)
Dual checkpoint option HIMALAYA: STRIDE tremelimumab + durvalumab (Abou-Alfa 2022, PMID 38319892) first dual-IO OS benefit vs sorafenib (de Castria 2022, PMID 36399155)
Newest IO doublet CheckMate 9DW: nivolumab + ipilimumab vs lenvatinib/sorafenib (Yau 2025, PMID 40349714) OS 23.7 vs 20.6 months; HR 0.79 (95% CI 0.65–0.96)

Frontier glimpse

  • Personalized, risk-stratified surveillance and new tools (abbreviated MRI, biomarker panels such as GALAD) are named priorities of EASL 2025 (PMID 39690085) but validation is incomplete (Arvind 2025, PMID 39943795).
  • IO combinations are moving earlier: EASL 2025 addresses combination immunotherapy "at various stages," transitions between locoregional and systemic therapy, and transplant integration (PMID 39690085).
  • Selection biomarkers for IO vs TKI remain undefined; sequencing after IO-doublet failure is unstudied territory (Kulik 2020, PMID 33376711).

The HCC care cascade

HCC outcomes depend on transitions between domains rather than any single test or drug:

Transition Core decision Main failure mode Deep page
Chronic liver disease → risk recognition Is incidence high enough for surveillance? Cirrhosis not recognized; non-cirrhotic MASLD gap Cirrhosis and risk stratification
Risk → repeated surveillance Which six-month test is adequate? Underuse or poor ultrasound visualization Surveillance and early detection
Abnormal test → diagnosis Does multiphase imaging establish HCC? Lost follow-up or misapplied non-invasive criteria Diagnosis and imaging
Diagnosis → allocation What do tumor extent, liver reserve, and performance permit? Tumor-only staging Staging and treatment allocation
Early disease → local cure Resection, ablation, or transplant? Ignoring portal hypertension or recurrence biology Surgical resection and ablation
Intermediate disease → local/systemic choice Is selective embolization likely to help? Repeated ineffective TACE and liver injury Locoregional therapy
Advanced disease → systemic sequence Which IO combination or TKI fits risk? Cross-trial ranking and Child-Pugh extrapolation Systemic therapy
Response → reassessment Continue, migrate, downstage, or stop? Size-only response and delayed decompensation recognition Diagnosis and imaging

Two outcomes must be tracked

Every intervention has at least two outcome axes:

  1. Cancer control: response, recurrence, progression, vascular invasion, metastasis, and HCC-specific death.
  2. Liver preservation: bilirubin, albumin, ascites, encephalopathy, portal-hypertension events, renal dysfunction, transplant eligibility, and non-HCC death.

An intervention can improve local control while shortening the time to decompensation. Conversely, etiologic therapy can improve liver survival while residual HCC risk persists. Trials and clinical summaries should therefore avoid collapsing progression and liver failure into an unexplained “treatment discontinuation” endpoint.

How the evidence changes by population

Population Stronger evidence Important boundary
Chronic HBV Vaccination, surveillance RCT, antiviral-era risk models Regional calibration and non-cirrhotic HCC
HCV cirrhosis after cure Residual-risk cohorts and meta-analysis When surveillance can stop
Alcohol-associated cirrhosis Incidence and burden estimates Few HCC-specific prevention trials
MASLD/MASH Rapidly expanding epidemiology Non-cirrhotic surveillance and ultrasound failure
Child-Pugh A advanced HCC Multiple phase 3 systemic trials Does not represent decompensated practice
Child-Pugh B/C Real-world/selected evidence Competing liver death and toxicity
Transplant recipients/candidates Selection and recurrence models Checkpoint rejection risk

See epidemiology and etiology, MASH/MASLD and HCC, and red flags and safety concerns for the populations in which standard pathways are least secure.

Reading the evidence safely

  • Surveillance survival associations are not equivalent to randomized mortality effects.
  • Imaging specificity in an at-risk liver does not transport to a low-risk incidental lesion.
  • Transplant cohorts are selected for favorable biology and cannot estimate untreated natural history.
  • PFS improvement after combined TACE/systemic therapy does not guarantee overall-survival improvement.
  • A biomarker associated with outcome is not predictive until it modifies relative treatment benefit.
  • Cross-trial median survival cannot rank regimens with different populations and subsequent therapy.
  • Patient-reported burden and caregiver capacity are outcomes, not footnotes.

Ten numbers that define the current field

Domain Quantified anchor Why it matters
Etiology HBV 52% and HCV 21% of global HCC in 2022, approximately 345,434 and 134,418 cases Viral prevention remains central despite the metabolic transition (Cao 2026, PMID 42135055)
Surveillance interval Six versus twelve months increased early detection, RR 1.17 (95% CI 1.08–1.26) Six months is evidence-based but not supported equally in every etiology/risk stratum (Yang 2023, PMID 36921104)
Diagnostic modality LR-5 sensitivity 61% with MRI versus 48% with CT; specificity 93% versus 96% Modality selection trades sensitivity, specificity, access, and transplant consequences (Kim 2022, PMID 35849177)
Small-tumor local cure SURF randomized 302 patients with ≤3 tumors, each ≤3 cm, and found no material OS advantage for surgery over RFA Resection and ablation are context-dependent alternatives, not a universal hierarchy (Kawaguchi 2025, PMID 40554738)
Downstaging 55.16% successfully downstaged and 31.52% transplanted by intention to treat across 25 studies Post-transplant outcomes alone omit failure before transplant (Tan 2023, PMID 35181565)
TARE versus TACE TRACE randomized 72 patients and favored TARE for time to progression Technique/dosimetry can matter more than broad modality labels (Dhondt 2022, PMID 35258371)
Durable systemic benefit STRIDE five-year OS 19.6% versus 9.4% with sorafenib; HR 0.76 (95% CI 0.65–0.89) Immunotherapy creates a survival tail, but most patients still do not reach five years (Rimassa 2025, PMID 40222621)
Early-detection biomarker HES V2.0 exceeded GALAD true-positive rate by 7.2% at fixed 10% false-positive rate Biomarker comparisons must fix the harm budget, not report AUC alone (El-Serag 2025, PMID 38899967)
Transplant after ICI Rejection 26.4% among 91 pre-transplant ICI-exposed recipients Conversion response can trade against graft loss (Rezaee-Zavareh 2025, PMID 38996924)
Patient priorities Ten months of maintained daily function was valued at least as highly as ten months of OS in a 200-patient discrete-choice study Efficacy tables are incomplete without function and toxicity (Li 2023, PMID 36900262)

The major unresolved tensions

  1. Population benefit versus individualized risk. Cirrhosis-based surveillance is implementable, but static eligibility misses changing risk after viral cure and leaves non-cirrhotic MASLD unresolved (Cao 2026, PMID 42135055).
  2. Sensitivity versus downstream harm. MRI and blood panels detect more disease than ultrasound or AFP alone in selected cohorts, but false-positive imaging, biopsy, anxiety, contrast, and capacity must be counted (Kim 2022, PMID 35849177; El-Serag 2025, PMID 38899967).
  3. Local control versus liver preservation. A technically complete procedure can still worsen portal hypertension or reserve enough to shorten survival; treatment allocation must integrate both tumor and liver endpoints (Dhondt 2022, PMID 35258371).
  4. Expansion versus fairness in transplantation. Dynamic response and AFP justify moving beyond a rigid morphology rule for selected patients, while intention-to-treat dropout and competing candidates limit unchecked expansion (Tan 2023, PMID 35181565).
  5. Trial efficacy versus real-world eligibility. Pivotal systemic trials predominantly enrolled Child-Pugh A, ECOG 0–1 patients; decompensated, transplant, bleeding, and autoimmune populations remain evidence-poor (Rimassa 2025, PMID 40222621).
  6. Response versus cure. Neoadjuvant/adjuvant and locoregional-systemic combinations can improve response or PFS without proving that recurrence or death is prevented.
  7. Tumor biology versus host biology. Genomic and immune classes matter, but etiology, fibrosis, microbiome, portal circulation, and organ reserve shape the same treatment response.
  8. Survival versus lived benefit. Patients may rationally reject a regimen with greater median efficacy if its bleeding, hand-foot, hypertension, or logistical burden threatens the outcome they value most (Li 2023, PMID 36900262).

Open questions

  • Does surveillance reduce mortality in non-HBV cirrhosis? The only RCT was HBV-based (Zhang 2004, PMID 15042359) and the observational estimate carries high heterogeneity (Singal 2022, PMID 35139400) against a "very low strength" grading (Kansagara 2014, PMID 24934699).
  • Which MASLD patients without cirrhosis warrant surveillance, given incidence ~0.1–1.3/1,000 patient-yr overall but HR 2.69 vs controls and higher risk in older males with diabetes (Huang 2021, PMID 33349658; Lee 2025, PMID 40345323)?
  • Can TERT, CTNNB1, or TP53 be drugged, or immune-classes exploited, to give HCC a genomically stratified therapy like lung adenocarcinoma's (Schulze 2015, PMID 25822088)?
  • How should first-line therapy be chosen among three IO-combination regimens when the 2026-08-30 PubMed search found no randomized direct comparison (Finn 2020, PMID 32402160; Abou-Alfa 2022, PMID 38319892; Yau 2025, PMID 40349714)?
  • Do transplant selection models calibrated in viral-era cohorts (Metroticket 2.0, Mazzaferro 2018, PMID 28989060) hold in MASLD-era candidates with different competing mortality?

References

  1. Bray F, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74:229-263. PMID 38572751
  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. Huang DQ, El-Serag HB, Loomba R. Global epidemiology of NAFLD-related HCC: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol. 2021;18:223-238. PMID 33349658
  4. Huang DQ, et al. Global epidemiology of alcohol-associated cirrhosis and HCC: trends, projections and risk factors. Nat Rev Gastroenterol Hepatol. 2023;20:37-49. PMID 36258033
  5. Lee CH, et al. Hepatocellular carcinoma risk in ICD-coded non-cirrhotic nonalcoholic fatty liver disease refined by fatty liver index: a nationwide South Korean cohort study. Clin Res Hepatol Gastroenterol. 2025;49:102612. PMID 40345323
  6. Shepard CW, et al. Hepatitis B virus infection: epidemiology and vaccination. Epidemiol Rev. 2006;28:112-125. PMID 16754644
  7. Zhang BH, Yang BH, Tang ZY. Randomized controlled trial of screening for hepatocellular carcinoma. J Cancer Res Clin Oncol. 2004;130:417-422. PMID 15042359
  8. Kansagara D, et al. Screening for hepatocellular carcinoma in chronic liver disease: a systematic review. Ann Intern Med. 2014;161:261-269. PMID 24934699
  9. Singal AG, et al. HCC surveillance improves early detection, curative treatment receipt, and survival in patients with cirrhosis: a meta-analysis. J Hepatol. 2022;77:128-139. PMID 35139400
  10. Taylor EJ, et al. Modeling the benefits and harms of surveillance for hepatocellular carcinoma: information to support informed choices. Hepatology. 2017;66:1546-1555. PMID 28605060
  11. Arvind A, Redmon K, Singal AG. Persisting challenges in the early detection of hepatocellular carcinoma. Expert Rev Anticancer Ther. 2025:1-12. PMID 39943795
  12. Della Corte C, Colombo M. Surveillance for hepatocellular carcinoma. Semin Oncol. 2012;39:384-398. PMID 22846857
  13. Reig M, et al. BCLC strategy for prognosis prediction and treatment recommendation: the 2022 update. J Hepatol. 2022;76:681-693. PMID 34801630
  14. Singal AG, et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78:1922-1965. PMID 37199193
  15. Schulze K, et al. Exome sequencing of hepatocellular carcinomas identifies new mutational signatures and potential therapeutic targets. Nat Genet. 2015;47:505-511. PMID 25822088
  16. Ki Kim S, et al. TERT promoter mutations and chromosome 8p loss are characteristic of nonalcoholic fatty liver disease-related hepatocellular carcinoma. Int J Cancer. 2016;139:2512-2518. PMID 27511114
  17. Mazzaferro V, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996;334:693-699. PMID 8594428
  18. Mazzaferro V, et al. Metroticket 2.0 model for analysis of competing risks of death after liver transplantation for hepatocellular carcinoma. Gastroenterology. 2018;154:128-139. PMID 28989060
  19. Llovet JM, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359:378-390. PMID 18650514
  20. Kudo M, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet. 2018;391:1163-1173. PMID 29433850
  21. Bruix J, et al. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;389:56-66. PMID 27932229
  22. Finn RS, et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med. 2020;382:1894-1905. PMID 32402160
  23. Abou-Alfa GK, et al. Tremelimumab plus durvalumab in unresectable hepatocellular carcinoma. NEJM Evid. 2022;1:EVIDoa2100070. PMID 38319892
  24. Yau T, et al. Nivolumab plus ipilimumab versus lenvatinib or sorafenib as first-line treatment for unresectable hepatocellular carcinoma (CheckMate 9DW): an open-label, randomised, phase 3 trial. Lancet. 2025;405:1851-1864. PMID 40349714
  25. Kulik L, et al. Potential impact of IMbrave150 results in the evolving treatment landscape of advanced hepatocellular carcinoma: a multidisciplinary expert opinion. J Hepatocell Carcinoma. 2020;7:423-433. PMID 33376711
  26. de Castria TB, et al. Tremelimumab and durvalumab in the treatment of unresectable, advanced hepatocellular carcinoma. Future Oncol. 2022;18:3769-3782. PMID 36399155
  27. Cao M, et al. HCC attributable to HBV, HCV, and other risk factors globally: updated meta-analysis. Gut. 2026. PMID 42135055
  28. Yang J, et al. Comparative effectiveness of HCC screening intervals and modalities. Chin Med J (Engl). 2023;136:1322-1330. PMID 36921104
  29. Kim YY, et al. Diagnostic performance of CT versus MRI LI-RADS category 5: a meta-analysis. Eur Radiol. 2022;32:6723-6729. PMID 35849177
  30. Kawaguchi Y, et al. Surgery versus ablation for HCC: SURF randomized and cohort trials. J Clin Oncol. 2025;43:2628-2638. PMID 40554738
  31. Tan DJH, et al. UNOS down-staging criteria for liver transplantation: meta-analysis of 25 studies. Clin Gastroenterol Hepatol. 2023;21:1475-1484. PMID 35181565
  32. Dhondt E, et al. Y-90 radioembolization versus DEB-TACE: TRACE randomized trial. Radiology. 2022;303:699-710. PMID 35258371
  33. Rimassa L, et al. Five-year OS update from HIMALAYA. J Hepatol. 2025;83:899-908. PMID 40222621
  34. El-Serag HB, et al. HES V2.0 outperforms GALAD for detection of HCC. Hepatology. 2025;81:465-475. PMID 38899967
  35. Rezaee-Zavareh MS, et al. Pre-transplant ICI use and post-transplant outcomes: IPD meta-analysis. J Hepatol. 2025;82:107-119. PMID 38996924
  36. Li D, et al. Patient preferences for unresectable HCC treatments: a discrete-choice experiment. Cancers. 2023;15. PMID 36900262