AFP-Negative Hepatocellular Carcinoma: CT Imaging Diagnosis, Differential Diagnosis, AI Applications, and Clinical Management
Why Normal Alpha-Fetoprotein Does Not Exclude Liver Cancer: A Comprehensive Radiology and Precision Imaging Guide
Clinical Hook
"Can liver cancer exist even when AFP is completely normal?"
A 48-year-old man underwent a routine abdominal ultrasound during a health screening. He had no abdominal pain, no jaundice, and no constitutional symptoms. Surprisingly, ultrasonography revealed a 2 cm hepatic nodule within a cirrhotic liver. Subsequent contrast-enhanced CT demonstrated arterial phase hyperenhancement (APHE) followed by portal venous washout, highly suggestive of hepatocellular carcinoma (HCC). However, his serum alpha-fetoprotein (AFP) level remained within the normal range (<20 ng/mL). This clinical scenario illustrates one of the most important diagnostic pitfalls in hepatology: normal AFP does not exclude liver cancer.
This case underscores why modern liver cancer diagnosis relies heavily on imaging rather than tumor markers alone.
Learning Objectives
After completing this article, readers will be able to:
- Understand the biological characteristics of AFP-negative hepatocellular carcinoma.
- Recognize the hallmark CT imaging features of HCC.
- Explain why arterial phase hyperenhancement and portal venous washout are highly specific imaging biomarkers.
- Differentiate AFP-negative HCC from other hypervascular hepatic lesions.
- Understand current treatment options for early-stage disease.
- Explore how artificial intelligence is transforming liver imaging and precision oncology.
Anatomy Review
Figure 1. Normal Liver Anatomy Relevant to HCC Imaging
Anatomy Review
The liver receives approximately 70% of its blood supply from the portal vein and 30% from the hepatic artery. This dual blood supply is fundamental to understanding contrast-enhanced liver imaging. During hepatocarcinogenesis, progressive neoangiogenesis shifts tumor perfusion toward predominantly arterial blood flow, producing the characteristic imaging appearance of hepatocellular carcinoma. This physiologic alteration explains why HCC enhances intensely during the arterial phase and subsequently demonstrates washout during the portal venous phase. The uploaded case emphasizes these imaging principles in explaining the observed enhancement pattern.
Case Presentation
Patient Profile
| Parameter | Finding |
|---|---|
| Age | 48 years |
| Sex | Male |
| Presentation | Incidentally detected liver nodule during routine screening |
| Symptoms | None |
| Background | Cirrhotic liver |
| AFP | Normal (<20 ng/mL) |
| Initial Examination | Liver ultrasonography |
Clinical History
The patient underwent routine health screening without liver-related complaints. Screening ultrasonography identified a solitary approximately 2 cm solid hepatic nodule within a cirrhotic liver, prompting multiphasic contrast-enhanced CT evaluation. The source document notes that AFP remained within the normal range despite imaging findings suspicious for HCC.
Clinical Question
Can hepatocellular carcinoma be confidently diagnosed despite a normal AFP level?
CT Findings
Figure 2. Multiphasic Contrast-Enhanced CT
Arterial Phase
Radiologic Interpretation:
The hepatic lesion demonstrates marked arterial phase hyperenhancement (APHE) compared with the surrounding liver parenchyma.
Key Imaging Features
- Hypervascular lesion
- Approximately 2 cm diameter
- Cirrhotic liver background
- Strong arterial enhancement
The uploaded case identifies arterial hyperenhancement as one of the defining imaging characteristics supporting the diagnosis of HCC.
Figure 3. Multiphasic Contrast-Enhanced CT
Portal Venous Phase
Radiologic Interpretation:
During the portal venous phase, contrast material rapidly clears from the lesion, producing washout and making the tumor appear hypoattenuating relative to the surrounding liver.
Diagnostic Features
- Portal venous washout
- Relative hypoattenuation
- Classic vascular behavior of HCC
The source document highlights the combination of arterial enhancement plus portal venous washout as the representative imaging pattern for hepatocellular carcinoma.
MRI Findings
Although this clinical case primarily emphasizes multiphasic CT findings, liver MRI may provide additional characterization in similar patients. The uploaded document repeatedly stresses the importance of CT or MRI when AFP is normal but imaging suspicion persists, without presenting a specific MRI dataset for this patient.
Potential MRI evaluation in comparable cases may include:
- T1-weighted imaging
- T2-weighted imaging
- Diffusion-weighted imaging (DWI)
- Hepatobiliary phase imaging (when applicable)
These MRI considerations are general imaging practice and are not reported as findings from this specific case.
Final Diagnosis
AFP-Negative Hepatocellular Carcinoma
Diagnostic Basis
- Normal serum AFP (<20 ng/mL)
- Cirrhotic liver
- Incidentally detected hepatic nodule
- Arterial phase hyperenhancement
- Portal venous washout
- Imaging pattern highly consistent with hepatocellular carcinoma
The uploaded case defines AFP-negative HCC as HCC occurring with AFP levels below 20 ng/mL and emphasizes that a normal AFP value does not exclude malignancy.
Imaging Pearls
One of the greatest misconceptions in hepatology is that a normal alpha-fetoprotein (AFP) level excludes hepatocellular carcinoma (HCC). This case clearly demonstrates the opposite. The uploaded source emphasizes that up to approximately 20% of HCCs may present with AFP levels below 20 ng/mL, making imaging indispensable for diagnosis.
Imaging Pearl 1
Normal AFP never rules out HCC.
AFP should be regarded as an adjunctive biomarker rather than a definitive screening test. Imaging findings remain the cornerstone of diagnosis when clinical suspicion exists.
Imaging Pearl 2
Arterial Phase Hyperenhancement (APHE) is the earliest imaging hallmark of hypervascular HCC.
In this case, the lesion enhanced more intensely than the surrounding liver during the arterial phase because tumor angiogenesis shifts blood supply toward the hepatic artery. The source document explains this mechanism in relation to the vascular remodeling of HCC.
Imaging Pearl 3
Portal Venous Washout strongly favors HCC.
Washout reflects rapid clearance of contrast material from the tumor relative to the surrounding liver. The uploaded case identifies the combination of arterial enhancement and washout as the representative imaging pattern of hepatocellular carcinoma.
Imaging Pearl 4
Always interpret liver nodules within the clinical background.
A 2 cm hypervascular lesion has a very different diagnostic implication in a cirrhotic liver than in a normal liver.
Imaging Pearl 5
Ultrasound is frequently the first imaging modality that detects AFP-negative HCC.
The patient's lesion was discovered incidentally during routine health screening ultrasonography before definitive CT evaluation.
Imaging Pearl 6
Tumor size alone is insufficient.
Small lesions may already demonstrate classic vascular characteristics diagnostic of HCC.
Imaging Pearl 7
Multiphasic CT is essential.
Single-phase CT examinations cannot adequately evaluate arterial enhancement or washout.
Imaging Pearl 8
Washout reflects vascular physiology rather than simple contrast loss.
Irregular tumor microvasculature causes contrast to leave the lesion more rapidly than surrounding hepatic parenchyma.
Imaging Pearl 9
Evaluate vascular invasion carefully.
The uploaded source reminds readers that radiologists assess vascular invasion, lymph nodes, metastases, tumor multiplicity, and surgical feasibility when interpreting HCC.
Imaging Pearl 10
CT interpretation should follow a systematic approach.
The uploaded case recommends evaluating:
- Cirrhosis
- Lesion size
- Arterial enhancement
- Washout
- Capsule
- Vascular invasion
as the principal imaging workflow.
Pathophysiology
Figure 4. Suggested illustration
From Chronic Liver Injury to AFP-Negative Hepatocellular Carcinoma
Pathophysiologic Overview
The uploaded document explains that hepatocellular carcinoma develops through repeated cycles of chronic liver injury and regeneration. Persistent inflammation and fibrosis promote progressive genomic instability, allowing dysplastic hepatocytes to evolve into malignant clones. AFP-negative tumors produce little or no alpha-fetoprotein, making imaging especially important for diagnosis.
Unlike conventional AFP-positive tumors, AFP-negative HCC may remain clinically silent because laboratory abnormalities are absent despite malignant transformation.
Epidemiology
Table 1. Clinical Characteristics of AFP-Negative HCC
| Characteristic | Observation |
|---|---|
| Typical presentation | Often asymptomatic |
| Detection | Health screening ultrasound |
| AFP level | <20 ng/mL |
| HBV positivity | May be lower |
| AST/ALT ratio | May be lower |
| Diagnostic importance | Imaging predominates |
These epidemiologic observations are summarized directly from the uploaded document.
Clinical Manifestations
The uploaded source emphasizes that AFP-negative HCC is frequently asymptomatic during the early stage.
When symptoms eventually occur, they may include:
- Right upper quadrant pain
- Chronic fatigue
- Weight loss
- Poor appetite
- Hepatomegaly
- Jaundice
- Ascites
Because early disease often lacks symptoms, routine surveillance imaging remains critically important in patients with cirrhosis.
Differential Diagnosis
Table 2. Hypervascular Liver Lesions
| Disease | Arterial Enhancement | Washout | Distinguishing Feature |
|---|---|---|---|
| AFP-negative HCC | Strong | Present | Cirrhosis + APHE + washout |
| High-grade dysplastic nodule | Mild | Usually absent | Premalignant lesion |
| Hemangioma | Peripheral nodular | Absent | Progressive fill-in |
| FNH | Homogeneous | Rare | Central scar |
| Hepatic adenoma | Variable | Variable | Young women |
| Intrahepatic cholangiocarcinoma | Progressive | Uncommon | Fibrosis, bile duct dilation |
This table reflects the comparative imaging characteristics described in the uploaded case review.
High-Grade Dysplastic Nodule
Usually demonstrates little arterial enhancement and lacks definite washout.
Hemangioma
Shows peripheral nodular enhancement with progressive centripetal fill-in rather than washout.
Focal Nodular Hyperplasia
Typically demonstrates homogeneous enhancement with a central scar and little or no washout.
Hepatic Adenoma
Often associated with oral contraceptive use and may contain fat or hemorrhage.
Intrahepatic Cholangiocarcinoma
Usually demonstrates delayed progressive enhancement with associated fibrosis and bile duct dilatation instead of the classic APHE/washout combination.
Clinical Management
The uploaded document summarizes currently accepted therapeutic approaches while noting that treatment recommendations are supplemented from contemporary clinical practice rather than derived from the case itself.
Surgical Resection
Preferred for:
- Solitary lesion
- Preserved liver function
- No vascular invasion
The uploaded source notes that a solitary 2 cm lesion such as this case may be an appropriate surgical candidate.
Liver Transplantation
Recommended for patients with advanced cirrhosis meeting accepted transplant criteria.
Radiofrequency Ablation (RFA)
Particularly effective for tumors measuring approximately 2–3 cm.
Microwave Ablation
Increasingly used because of improved thermal efficiency.
Transarterial Chemoembolization (TACE)
Appropriate for patients who are not candidates for surgical therapy.
Radiation Therapy
Stereotactic Body Radiation Therapy (SBRT) may be considered in selected patients.
Systemic Therapy
The uploaded document lists examples of targeted therapies, including:
- Sorafenib
- Lenvatinib
It also notes that immunotherapy has become an important component of modern HCC management.
Prognostic Considerations
The uploaded source notes several favorable features commonly associated with AFP-negative HCC:
- Earlier detection
- Smaller tumors
- Less vascular invasion
However, it also cautions that diagnosis may be delayed because clinicians can be falsely reassured by normal AFP levels.
Artificial Intelligence Perspective
Figure 5. Enterprise AI Workflow for AFP-Negative Hepatocellular Carcinoma
Why Artificial Intelligence Matters
The uploaded case demonstrates an important diagnostic challenge: serum AFP remained normal despite imaging findings strongly suggestive of hepatocellular carcinoma. Consequently, imaging—not laboratory biomarkers—became the decisive diagnostic modality.
This clinical scenario illustrates where AI can provide significant value by assisting radiologists in detecting subtle lesions and standardizing interpretation. The following AI discussion expands beyond the uploaded case and represents general applications of AI in medical imaging, not findings reported in the source document.
AI-Assisted Liver Lesion Detection
Modern deep learning systems can automatically:
- Detect small hepatic nodules
- Segment liver tumors
- Estimate tumor volume
- Measure lesion growth over time
- Compare serial examinations
These capabilities may reduce perceptual errors, particularly for small lesions discovered during surveillance.
Automated Multiphasic CT Analysis
AI systems can evaluate enhancement characteristics across multiple phases, including:
- Arterial Phase Hyperenhancement (APHE)
- Portal Venous Washout
- Capsule appearance
- Threshold growth
- Ancillary imaging features
Such automated analysis can support standardized reporting alongside radiologist interpretation.
AI-Based LI-RADS Support
Rather than replacing radiologists, AI may assist by identifying imaging features that contribute to standardized liver lesion categorization.
Radiomics
Radiomics converts medical images into quantitative biomarkers.
Potential quantitative features include:
- Texture heterogeneity
- Entropy
- Shape complexity
- Edge sharpness
- Enhancement kinetics
- Perfusion characteristics
Radiomics may eventually help distinguish:
- Early HCC
- High-grade dysplastic nodules
- Intrahepatic cholangiocarcinoma
- Benign hypervascular lesions
These applications are an area of active research and are not described in the uploaded case document.
Foundation Models in Medical Imaging
Large multimodal foundation models may eventually integrate:
- CT
- MRI
- Ultrasound
- Laboratory results
- Pathology
- Clinical notes
into unified diagnostic support systems.
Potential applications include:
- Automated report drafting
- Imaging summarization
- Tumor staging assistance
- Longitudinal disease monitoring
These are emerging technologies beyond the scope of the uploaded source.
Clinical Decision Support Systems (CDSS)
This model emphasizes that AI functions as a decision-support tool rather than an autonomous decision-maker.
AI Strengths
Potential advantages include:
- Improved detection of small lesions
- Consistent measurement of tumor size
- Standardized reporting
- Automated follow-up comparison
- Reduced interobserver variability
- Workflow efficiency
AI Limitations
Despite rapid progress, AI has important limitations:
- Imaging artifacts
- Scanner variability
- Limited generalizability
- False-positive detections
- Dependence on training datasets
- Need for expert clinical oversight
Normal AFP does not simplify interpretation, and AI cannot independently establish a diagnosis without appropriate clinical and imaging correlation.
Future of Precision Imaging
Radiogenomics
Radiogenomics seeks to correlate imaging phenotypes with underlying molecular characteristics.
Potential future applications include:
- Predicting tumor biology
- Estimating recurrence risk
- Guiding targeted therapies
- Selecting immunotherapy candidates
Digital Twin
Digital Twin technology aims to create a computational representation of an individual patient's liver by integrating:
- CT imaging
- MRI
- Laboratory data
- Histopathology
- Genomic information
- Longitudinal follow-up
Potential clinical applications include:
- Treatment simulation
- Surgical planning
- Outcome prediction
- Personalized surveillance
Federated Learning
Medical imaging datasets are often distributed across institutions.
Federated learning allows AI models to be trained collaboratively while patient data remain within each institution, supporting privacy-preserving model development.
Synthetic Data
Synthetic imaging datasets may help address data scarcity by generating realistic training examples for AI development.
Potential benefits include:
- Improved model robustness
- Rare disease augmentation
- Enhanced educational resources
Careful validation remains essential before clinical deployment.
Precision Oncology
Future liver cancer management may integrate:
- Imaging biomarkers
- Molecular biomarkers
- Clinical characteristics
- AI-assisted risk prediction
to support increasingly individualized therapeutic strategies.
The Future Radiology Workflow
Key Takeaways
The uploaded case illustrates that AFP-negative hepatocellular carcinoma may be diagnosed primarily through characteristic imaging findings, particularly arterial phase hyperenhancement and portal venous washout.
The AI and precision imaging concepts presented in this section are forward-looking extensions intended to place the case within the broader context of enterprise medical imaging and should be interpreted as general educational material rather than source-derived findings.
Clinical Pearls
Pearl 1
A normal AFP level does not exclude hepatocellular carcinoma. The uploaded case specifically emphasizes that AFP-negative HCC exists and that some patients with HCC have AFP values within the normal range.
Pearl 2
Routine abdominal ultrasound can detect clinically silent liver cancer before symptoms develop, as illustrated by this asymptomatic 48-year-old patient whose lesion was discovered during health screening.
Pearl 3
In cirrhotic patients, every newly detected hepatic nodule deserves careful evaluation because HCC risk is substantially increased. The uploaded source repeatedly highlights this clinical principle.
Pearl 4
Arterial Phase Hyperenhancement (APHE) is one of the most important CT imaging findings supporting HCC diagnosis.
Pearl 5
Portal venous washout is another hallmark imaging feature that, when combined with APHE, strongly favors hepatocellular carcinoma.
Pearl 6
Multiphasic CT is considerably more informative than AFP alone in patients with suspicious hepatic lesions. The uploaded case explicitly emphasizes the importance of CT or MRI when AFP is normal.
Pearl 7
Interpret imaging findings in the context of the patient's liver background. A hypervascular lesion in a cirrhotic liver carries a different implication than one in a non-cirrhotic liver.
Pearl 8
Radiologists systematically evaluate:
- Cirrhosis
- Lesion size
- APHE
- Washout
- Capsule
- Vascular invasion
when interpreting possible HCC.
Pearl 9
Differential diagnosis remains essential because not every hypervascular hepatic lesion represents HCC. The uploaded document discusses dysplastic nodules, hemangioma, focal nodular hyperplasia, hepatic adenoma, and intrahepatic cholangiocarcinoma.
Pearl 10
Small tumors may still be curable when detected early.
Pearl 11
Treatment selection depends on:
- Tumor burden
- Liver function
- Vascular invasion
- Surgical candidacy
The uploaded source outlines these considerations.
Pearl 12
Early detection through surveillance imaging can expand curative treatment options.
Pearl 13
AI may support lesion detection and workflow efficiency but should complement—not replace—expert radiologic interpretation.
Pearl 14
Multidisciplinary collaboration remains central to optimal HCC management.
Pearl 15
Imaging continues to be the cornerstone of AFP-negative HCC diagnosis.
Quiz
Question 1
Which imaging combination most strongly supports hepatocellular carcinoma in this case?
A. Peripheral nodular enhancement with delayed fill-in
B. Progressive delayed enhancement
C. Arterial phase hyperenhancement with portal venous washout
D. Central scar with homogeneous enhancement
Answer: C
Question 2
According to the uploaded case, AFP-negative HCC refers to:
A. AFP >400 ng/mL
B. AFP >200 ng/mL
C. AFP >100 ng/mL
D. AFP <20 ng/mL
Answer: D
Question 3
Which lesion commonly demonstrates progressive centripetal fill-in?
A. HCC
B. FNH
C. Hemangioma
D. Cholangiocarcinoma
Answer: C
Question 4
Which modality first detected the lesion?
A. MRI
B. PET
C. CT
D. Ultrasound
Answer: D
Question 5
The patient described in this case was:
A. 32-year-old woman
B. 55-year-old man
C. 48-year-old man
D. 70-year-old woman
Answer: C
Frequently Asked Questions (FAQ)
Can liver cancer occur with a normal AFP level?
Yes. The uploaded source explains that AFP-negative HCC exists and stresses that a normal AFP level does not exclude hepatocellular carcinoma.
Why was CT so important in this case?
Because multiphasic CT demonstrated arterial phase hyperenhancement and portal venous washout despite normal AFP values.
Why was the patient asymptomatic?
According to the uploaded document, AFP-negative HCC is frequently asymptomatic during the early stage and may be detected incidentally during health screening.
What imaging findings are most characteristic?
The uploaded case identifies:
- APHE
-
Portal venous washout, as the representative imaging pattern.
Conclusion
This case highlights a crucial lesson in modern liver imaging: normal AFP values should never be used to rule out hepatocellular carcinoma. In this asymptomatic 48-year-old man, screening ultrasound detected a small hepatic nodule, and multiphasic CT revealed the characteristic combination of arterial phase hyperenhancement and portal venous washout, leading to the diagnosis of AFP-negative HCC.
For clinicians and radiologists, the case reinforces the importance of integrating surveillance imaging, structured CT interpretation, and clinical context when evaluating patients with cirrhosis or newly detected hepatic nodules.
Reference
[1] J. M. Llovet, R. K. Kelley, A. Villanueva, et al., "Hepatocellular carcinoma," Nature Reviews Disease Primers, vol. 7, no. 1, Art. no. 6, Jan. 2021.
[2] J. A. Marrero, L. M. Kulik, C. B. Sirlin, et al., "AASLD Practice Guidance on Prevention, Diagnosis, and Treatment of Hepatocellular Carcinoma," Hepatology, vol. 78, no. 6, pp. 1922–1965, 2023.
[3] European Association for the Study of the Liver (EASL), "EASL Clinical Practice Guidelines on the Management of Hepatocellular Carcinoma," Journal of Hepatology, vol. 81, 2024.
[4] M. Ronot and V. Vilgrain, "Imaging of Hepatocellular Carcinoma: Diagnostic and Therapeutic Implications," Radiology, vol. 296, no. 3, pp. 493–507, 2020.
[5] American College of Radiology, LI-RADS® v2024 CT/MRI Manual. Reston, VA, USA: American College of Radiology, 2024.
[6] A. Forner, M. Reig, and J. Bruix, "Hepatocellular carcinoma," The Lancet, vol. 391, no. 10127, pp. 1301–1314, Apr. 2018.
[7] F. Nomura, K. Ohnishi, and Y. Tanabe, "Clinical features and prognosis of hepatocellular carcinoma with reference to serum alpha-fetoprotein levels," Cancer, vol. 64, no. 8, pp. 1700–1707, 1989.
[8] R. J. Gillies, P. E. Kinahan, and H. Hricak, "Radiomics: Images Are More than Pictures, They Are Data," Radiology, vol. 278, no. 2, pp. 563–577, Feb. 2016.
[9] G. Litjens, T. Kooi, B. E. Bejnordi, et al., "A Survey on Deep Learning in Medical Image Analysis," Medical Image Analysis, vol. 42, pp. 60–88, Dec. 2017.
[10] E. J. Topol, "High-performance medicine: the convergence of human and artificial intelligence," Nature Medicine, vol. 25, no. 1, pp. 44–56, Jan. 2019.
Comments
Post a Comment