What is the role of alpha-fetoprotein (AFP) in hepatocellular carcinoma (HCC)?

Role of AFP in Hepatocellular Carcinoma

AFP should be used in combination with ultrasound for HCC surveillance in at-risk patients, as the combined strategy increases sensitivity for early-stage HCC detection from 51.6% to 74.1%, despite a modest decrease in specificity from 87.9% to 83.9%. 1

Surveillance and Screening

The primary role of AFP is as a complementary surveillance tool rather than a standalone screening test. When used alone, AFP demonstrates poor sensitivity for early-stage HCC detection (49.1%) with specificity of 87.9% 1. However, the combination strategy of ultrasound plus AFP significantly improves early detection:

• Ultrasound alone: 51.6% sensitivity, 87.9% specificity
• Ultrasound + AFP: 74.1% sensitivity, 83.9% specificity 1

In prospective studies specifically, the improvement is even more pronounced—sensitivity increases from 40% with ultrasound alone to 60% with the combination 1. This represents a critical trade-off: accepting more false-positives to capture significantly more early-stage cancers, which aligns with patient preferences who favor sensitivity over specificity 1.

Etiology-Specific Considerations

AFP performance varies substantially by liver disease etiology, which is increasingly important as the epidemiology of HCC shifts:

• Active HCV: Highest AFP levels at diagnosis (median 32.7 ng/mL) 2
• Cured HCV (SVR): Lower AFP (8.3 ng/mL) with only 39.0% positivity in early-stage disease 3, 2
• MASLD: Lowest AFP (7.7 ng/mL) with only 34.5% positivity in T1a tumors 3
• ALD: Similarly low AFP (9.0 ng/mL) 2

For non-viral etiologies (MASLD, ALD, cured-HCV), consider lowering the AFP threshold to ≥10 ng/mL rather than the traditional ≥20 ng/mL cutoff. This increases sensitivity from 22-32% to 30-42% while maintaining specificity >94% 2. The incremental yield from adding DCP testing in AFP-negative cases is substantial: 21.1% in SVR-related and 30.5% in NBNC early-stage HCC 3.

Diagnostic Role

AFP alone is insufficient for HCC diagnosis—imaging findings of classic enhancement patterns on multiphasic CT or MRI are more definitive 4. Key diagnostic limitations:

• Only 18% of HCC patients have AFP >400 ng/mL
• 46% have normal AFP <20 ng/mL 4
• AFP can be elevated in pregnancy, chronic hepatitis, cholangiocarcinoma, colon metastases, lymphoma, and germ cell tumors 4

When AFP is rising without an identifiable liver mass on ultrasound, proceed to multiphasic contrast-enhanced CT or MRI 4. If imaging remains negative but suspicion persists, consider biopsy for suspicious lesions, though recognize the limitations of sampling error for small lesions 4.

Prognostic and Biological Significance

Beyond its diagnostic utility, AFP functions as an active oncogenic protein rather than merely a passive tumor marker:

• Regulates cancer stem cell properties through PI3K/Akt pathway activation 5
• Promotes immune evasion and suppresses anti-tumor immunity 6, 7
• Contributes to multidrug resistance in HCC treatment 8
• Correlates with angiogenesis and VEGF pathway activation 9

Persistently elevated AFP levels represent a risk factor for HCC progression and worse prognosis 8. This mechanistic understanding supports AFP's role as both a surveillance marker and potential therapeutic target.

Practical Algorithm

For HCC surveillance in cirrhosis:

1. Perform ultrasound + AFP every 6 months
2. Use AFP threshold ≥10 ng/mL for non-viral etiologies (MASLD, ALD, cured-HCV)
3. Consider adding DCP testing in high-risk populations, particularly non-viral etiologies
4. If AFP rising or ultrasound shows nodule → multiphasic CT/MRI
5. If imaging inconclusive → multidisciplinary discussion ± biopsy

Critical pitfall: Do not rely on AFP alone for diagnosis or to exclude HCC—up to 46% of HCC cases have normal AFP levels 4. The combination strategy misses over one-third of early-stage HCCs even with optimal use 1.