HCC Surveillance in Nonalcoholic Steatohepatitis
Patients with NASH cirrhosis should undergo biannual (every 6 months) surveillance with abdominal ultrasound plus alpha-fetoprotein (AFP), though ultrasound alone is acceptable if AFP is unavailable. 1, 2
Who Requires Surveillance
Definite Indications
- All patients with NASH cirrhosis require HCC surveillance regardless of disease activity or metabolic control 1, 2
- This recommendation applies to cirrhosis from any etiology, including NASH
The Non-Cirrhotic NASH Challenge
A critical caveat: 20-30% of NAFLD-related HCC occurs in patients without cirrhosis or advanced fibrosis 1. This creates a major surveillance gap, as current guidelines do not definitively recommend surveillance in non-cirrhotic NASH patients. However, the evidence base for surveillance in non-cirrhotic NASH is insufficient to make firm recommendations 1, 3.
For non-cirrhotic NASH patients, consider surveillance if they have:
- Advanced fibrosis (F3) detected by noninvasive testing
- Multiple metabolic risk factors (diabetes, older age >50, male gender, tobacco use) 3
Surveillance Protocol
Standard Approach
- Frequency: Every 6 months 2
- Modality: Abdominal ultrasound with or without AFP 2
- The combination of ultrasound plus AFP has higher sensitivity for early-stage HCC detection than ultrasound alone, though it still misses over one-third of early-stage HCC 4
Critical Limitations in NASH Patients
Obesity significantly impairs ultrasound effectiveness 1, 5. Patients with NASH have higher rates of obesity, making ultrasound technically difficult and reducing sensitivity for early HCC detection. This is a major reason why NASH patients are more likely to be diagnosed with HCC at advanced stages compared to other etiologies 6.
Emerging Alternatives for Obese Patients
When ultrasound quality is suboptimal due to obesity:
- Abbreviated MRI protocols show superior sensitivity and specificity compared to ultrasound in cohort studies 4
- GALAD score (gender, age, AFP, AFP-L3, DCP) demonstrates promising sensitivity for early-stage HCC detection 4
- These modalities are entering prospective trials but can be considered now in patients with poor ultrasound windows
Implementation Challenges
Surveillance Underuse
NASH patients have dramatically lower surveillance rates than other etiologies. In one study, 51.5% of NASH cirrhosis patients had no screening before HCC diagnosis, compared to 25.9% of HCV cirrhosis patients 7. This surveillance failure contributes to worse outcomes in NASH-related HCC.
Patients who undergo complete surveillance have significantly smaller tumors at diagnosis (P=0.006), though mortality differences have not been consistently demonstrated 7.
Common Pitfalls to Avoid
- Failure to recognize cirrhosis: Many NASH patients develop cirrhosis silently without obvious clinical decompensation
- Assuming non-cirrhotic NASH is safe: Remember that 20-30% of NASH-HCC occurs without cirrhosis 1
- Accepting poor-quality ultrasounds: In obese patients, document image quality and consider alternative modalities if visualization is inadequate
- Stopping surveillance after metabolic improvement: HCC risk persists even with weight loss or diabetes control
Risk Stratification Beyond Cirrhosis
For patients with advanced chronic liver disease, the PLEASE algorithm can stratify HCC risk 8:
- Platelet count <150×10⁹/L
- Liver stiffness ≥15 kPa
- Age ≥50 years
- Male sex
- Viral hepatitis status
- Presence of steatotic liver disease
High-risk patients (≥4 points) have a 2-year HCC incidence of 15.6%, compared to 1.7% in low-risk patients 8.
Quality Assurance
Surveillance is a process, not just ordering tests 2:
- Ensure ultrasound quality is documented
- Establish systems for tracking surveillance adherence
- Create protocols for appropriate follow-up of abnormal results
- Address patient-level barriers (obesity, access, awareness)
The evidence supporting mortality benefit from HCC surveillance remains limited, based primarily on observational studies rather than randomized trials 2, 9. However, surveillance does increase early-stage detection and access to curative treatment, which translates to 5-10 year median survival versus months without treatment 4.