Table 1.
HCC incidence and main risk factors among the NAFLD population.
| First Author (Country, Year) | Type of Study | Number and Type of Patients | Diagnostic Method for NAFLD/NASH | NAFLD Patients with Cirrhosis | Mean Follow Up | HCC Incidence | HCC-Independent Risk Factors (HR, 95% CI) among NAFLD (Multivariate Analysis) |
|---|---|---|---|---|---|---|---|
| Ascha (US, 2010) [37] | Retrospective cohort study | 195 NASH-cirrhosis 315 HCV-cirrhosis |
Histology or cryptogenic cirrhosis and MetS | 100% | 3.2 years | NASH: 2.6% | Any alcohol consumption (HR 3.6 (1.6–8.9)) Older age (HR 1.08 (1.02–1.1)) |
| Yatsuji (Japan, 2009) [39] | Prospective cohort study, observational | 68 NASH-cirrhosis 69 HCV-cirrhosis |
Histology | 100% | NR | NASH: 5-year occurrence rate = 11.3% | NA |
| Kanwal (US, 2018) [9] | Retrospective cohort study | 296,707 NAFLD 296,707 matched controls |
Elevated ALT and exclusion of other etiologies of liver disease | 0.4% at baseline | 9 years | NAFLD: 0.08 per 1000 person-years (PY) Subgroup analyses: -NAFLD + diabetes = 0.45 per 1000 PY -NAFLD + age > 65 = 0.41 per 1000 PY -NAFLD + age > 65 + Hispanic ethnicity = 0.93 per 1000 PY NASH-cirrhosis: 10.6 per 1000 PY (range 1.6–23, highest in older (>65 years) Hispanics). If cirrhosis + high FIB-4 = 13.55 per 1000 PY (11.93–15.33) |
Cirrhosis Age ≥ 65 years Hispanic ethnicity Diabetes Male sex Among cirrhosis, risk highest if: -Male sex -Hispanic ethnicity and age ≥ 65 years -Diabetes-FIB-4 score > 2.67 |
| Ioannou (US, 2019) [54] | Retrospective cohort study | 7068 NAFLD-cirrhosis 16,175 ALD-cirrhosis |
If comorbid with diabetes or BMI > 30 | 100% | 3.7 years | Annual incidence = 1.56% If FIB-4 > 3.65, annual incidence = 2.68% |
Older age (aHR ≈ 2.09 if age > 60) Male sex (Ahr = 1 versus 0.25 for female) Platelet count < 150 × 103 µL (aHR ≈ 2 to ≈3) Albumin < 3.7 g/dL (aHR ≈ 2 to ≈ 3) AST/ALT ratio > 8.8 (aHR ≈ 2 to ≈ 5) |
| Yang (US, 2020) [55] | Retrospective cohort study | 354 NASH-cirrhosis | Histology or history of steatosis or fatty liver at radiology | 100% | 47 months | 5-year cumulative incidence rate = 7.8% | Older age (per decade, HR = 1.8 (1.2–2.6)) Low albumin (HR 2.1 (1.5–2.9)) Diabetes (HR 4.2 (1.2–14.2)) |
| Yasui (Japan, 2011) [51] | Cross-sectional multicenter study | 87 NASH-related HCC cases | Histology | 51% | NR | NR | Advanced fibrosis (21%) and cirrhosis (51%), male sex (62%) and diabetes (59%), obesity (62%) and hypertension (55%) were highly prevalent in the population. Risk analysis was not performed. |
| Piscaglia (Italy, 2016) [13] | Multicenter observational prospective study | 145 NAFLD- related HCC cases 611 HCV-related HCC cases |
Histology or radiology | 53.8% | NR | NR | Causality not assessed but in comparison with the HCV cohort, NAFLD patients showed significantly higher prevalence of male gender, diabetes, hypertension and dyslipidemia |
| Kawamura (Japan, 2011) [56] | Retrospective cohort study | 6508 NAFLD patients | Ultrasound scan Only 16 patients had NAFLD-related HCC |
NR | 5.6 years | Overall incidence = 0.25% Annual incidence = 0.043% Cumulative HCC incidence: -4-year = 0.02% -8-year = 0.19% -12-year = 0.51% |
AST ≥ 40 IU/L (HR 8.20 (2.56–26.26)) Age > 60 (HR 4.27 (1.30–14.01)) Platelet count < 150 × 103/µL (HR 7.19 (2.26–23.26)) Diabetes (HR 3.21 (1.09–9.50)) APRI > 1.5 (i.e., significant fibrosis) (HR 25.03 (9.02–69.52)) |
| Lee (Taiwan, 2017) [11] | Population-based retrospective cohort study | 18,080 NAFLD patients | Not reported | NR | 6.3 years | Cumulative incidence at 1-year = 0.18%, increasing until up to 2.73% at 10 years | Age > 55 years (HR 7.78 (3.12–19.44)) ALT elevation (HR 6.80 (3.00–15.42)) 10-year cumulative incidence 4-fold higher in patients aged over 55 with ALT elevation |
| Tokushige (Japan, 2013) [57] | Prospective cohort study | 14,530 HCC cases: 84.1% viral aetiology; 7.2% alcoholic, 2% NAFLD; 5.1% cryptogenic |
Histology | 62% | NR | 5-years incidence = 11.3% | Older age (HR 1.103 (1.050–1.159) + Male gender (HR 4.680 (1.803–12.146)) Advanced liver fibrosis (HR 2.718 (1.745–4.233)) Higher GGT (HR 1.005 (1.001–1.009)) |
| Liu (UK, Switzerland, 2014) [58] | Prospective cohort study | 100 NAFLD-related HCC cases 275 NAFLD cases w/o HCC |
Histology or radiology | Among NAFLD-HCCs, 67% Among NAFLD w/o HCC, 26% |
NR | NR | Carriage of the PNPLA3 rs-738409 G > C polymorphism (2.26 (1.23–4.14)) Male gender (HR 11.11 (4.17–33.33)) Age (HR 1.24 (1.17–1.32)) Cirrhosis (HR 9.37 (3.82–23.00)) |
| Grimaudo (Italy, 2020) [59] | Prospective cohort study | 471 NAFLD cases | Histology or radiology | 11.5% | 64.6 months | Incidence rate in the non-cirrhotic vs. cirrhotic: - 1-year: 0.2% vs. 1.3% - 5 -years: 3.0% vs. 9.3% - 10-years: 4.2% vs. 13.5% |
Advanced fibrosis and cirrhosis (HR not reported) PNPLA3 G variant (HR 2.68 (1.01–7.26)). Among the subgroup of patients with F3–F4 fibrosis it was the only independent risk factor: HR 2.66 (1.02–7.13) |
NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steato-hepatitis; ALD, alcoholic liver disease; HCV, hepatitis C virus; HCC, hepatocellular carcinoma; NR, not reported; PY, person-years; GGT, gamma-glutamyl transpeptidase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; MetS, metabolic syndrome; HR, hazard ratio; aHR, adjusted hazard ratio; CI, confidence interval.