What is the accuracy of alpha-fetoprotein (AFP) for diagnosing hepatocellular carcinoma (HCC)?

Accuracy of Alpha-Fetoprotein for Hepatocellular Carcinoma Diagnosis

Alpha-fetoprotein (AFP) alone has limited accuracy for hepatocellular carcinoma diagnosis, with a sensitivity of only 60% at the commonly used cutoff of 20 ng/mL, making it inadequate as a standalone screening or diagnostic test. 1

Sensitivity and Specificity at Different Cutoff Values

• At a cutoff of 20 ng/mL, AFP has a sensitivity of approximately 60% and specificity of 80-94% for detecting HCC at any stage, meaning it would miss 40% of HCC cases 1
• For early-stage tumors, the sensitivity drops further to only 32-49% 1
• When the AFP cutoff is raised to 200 ng/mL, sensitivity decreases to 22.4% while specificity improves to 66-97% 1
• At a cutoff of 400 ng/mL, sensitivity drops even further to 17.1% 1, 2

Positive Predictive Value Considerations

• The positive predictive value (PPV) of AFP is heavily dependent on the prevalence of HCC in the population being tested 1
• At a cutoff of 20 ng/mL and HCC prevalence of 50%, the PPV is 84.6% 1
• However, in most liver clinics where HCC prevalence is around 5%, the PPV at 20 ng/mL drops significantly to only 41.5% 1
• Even at a higher cutoff of 400 ng/mL, the PPV is only about 60% when HCC prevalence is 5% 1

Clinical Utility in Specific Contexts

• AFP has high diagnostic value in cirrhotic patients with a liver mass when levels exceed 200 ng/mL, with a positive predictive value approaching 99% 1, 3
• AFP accuracy varies significantly based on the etiology of liver disease:
  • In HCV-negative patients, a lower cutoff of 11 ng/mL provides better diagnostic accuracy 4
  • In HCV-positive patients, a higher cutoff of 59 ng/mL is more accurate 4
• AFP shows significantly higher accuracy for detecting HCC in HIV-positive patients with cirrhosis compared to HIV-negative patients (c-statistic 0.81 vs 0.59) 4

Role in Current Guidelines

• AFP is no longer recommended as a standalone screening test by major Western guidelines (EASL-EORTC and AASLD) due to its limited sensitivity 1
• However, some Asian guidelines (APASL, JSH) still include AFP in their surveillance algorithms 1
• The Korean Liver Cancer Study Group (KLCSG) guidelines suggest that in high-risk patients with AFP ≥200 ng/mL, typical HCC characteristics on a single dynamic imaging study are sufficient for diagnosis 1
• Current best practice is to use ultrasound combined with AFP for surveillance, which increases sensitivity for early-stage HCC to 63% compared to 45% with ultrasound alone 1, 3

Improving AFP Diagnostic Accuracy

• Combining AFP with other biomarkers can improve diagnostic performance:
  • Des-gamma-carboxy prothrombin (DCP/PIVKA-II) 1
  • Lens culinaris agglutinin-reactive AFP (AFP-L3) 1, 3
• Using longitudinal AFP measurements rather than single threshold measurements may improve accuracy by detecting concerning trends 1
• The GALAD model (incorporating gender, age, AFP-L3, AFP, and DCP) has shown sensitivity >60% for early HCC detection 1

Common Pitfalls

• AFP may be elevated in non-HCC conditions, including:
  • Chronic liver disease with significant inflammation 1, 5
  • Acute hepatitis with elevated transaminases 1
  • Non-HCC malignancies such as cholangiocarcinoma 1, 5
• Relying solely on AFP for screening will miss a significant proportion of early HCCs 1, 6
• AFP performance can vary significantly based on patient characteristics, requiring individualized interpretation 4, 5

In conclusion, while AFP has limitations as a standalone test, it retains value when significantly elevated (>200 ng/mL) in patients with liver masses and when used in combination with appropriate imaging studies. For optimal diagnostic accuracy, AFP should be interpreted in the context of the patient's specific risk factors and combined with high-quality imaging.