PSMA PET/CT–Avid Liver Lesions in Prostate Cancer: How to Recognize Hepatic Hemangiomas and Avoid False-Positive Metastatic Staging
A Case-Based Review of Imaging Findings, Differential Diagnosis, Clinical Consequences, and AI-Assisted Interpretation
A focal liver lesion with increased prostate-specific membrane antigen (PSMA) uptake in a patient with prostate cancer raises concern for metastatic disease. However, PSMA uptake is not synonymous with malignancy. Hepatic hemangiomas, benign vascular lesions, can demonstrate increased uptake and mimic liver metastases on PSMA positron emission tomography/computed tomography (PET/CT). The key diagnostic principle is to correlate tracer uptake with CT morphology, contrast-enhancement patterns, clinical risk, and, when necessary, liver magnetic resonance imaging (MRI). In this case, two PSMA-avid liver lesions in a man in his 60s undergoing prostate cancer staging were ultimately diagnosed as benign hemangiomas in an external pathology report. Recognizing this pitfall can prevent incorrect upstaging and unnecessary investigations or treatment.
Why This Case Matters
PSMA PET/CT has become an important imaging tool for prostate cancer staging and recurrence assessment. Its ability to identify sites of disease can influence treatment planning, particularly in patients with high-risk disease. The prospective proPSMA trial demonstrated the value of PSMA PET/CT for more accurate staging than conventional imaging in men with high-risk prostate cancer [1].
Greater detection capability, however, creates a diagnostic challenge: not every focus of increased tracer uptake represents cancer.
This distinction is especially important in the liver. Confirmed hepatic metastases may substantially alter the clinical assessment of prostate cancer, whereas a benign liver lesion should not be used to classify a patient as having metastatic disease.
The central lesson is straightforward:
A PSMA-avid lesion is an imaging finding, not a histologic diagnosis.
Accurate interpretation requires integration of functional imaging, anatomic morphology, contrast-enhancement behavior, prior examinations, and the patient's clinical context.
Clinical Scenario: Two PSMA-Avid Liver Lesions
A man in his 60s underwent PSMA PET/CT for prostate cancer staging. Increased tracer uptake was identified at the prostatic apex. Two additional focal areas of increased PSMA uptake were observed in the liver.
Contrast-enhanced CT demonstrated two hyperenhancing hepatic lesions. An external pathology report subsequently classified both lesions as benign hemangiomas.
The available case information does not specify the patient's prostate-specific antigen (PSA) concentration, Gleason grade group, lesion dimensions, standardized uptake values (SUVs), detailed MRI findings, or subsequent treatment and follow-up. These parameters should not be inferred when interpreting the case.
The diagnostic challenge was to determine whether the hepatic uptake represented prostate cancer metastases or a benign vascular process.
In a patient with a known or suspected malignancy, metastatic disease must remain in the differential diagnosis. Nevertheless, the presence of cancer should not override imaging characteristics that suggest a benign alternative.
What Is PSMA, and Why Is Uptake Not Cancer-Specific?
Prostate-specific membrane antigen is a cell-surface protein that is frequently overexpressed in prostate cancer. Radiotracers targeting PSMA allow PET imaging to identify areas of tracer accumulation associated with prostate cancer.
Despite its name, PSMA is not exclusively expressed in prostate cancer. PSMA-related uptake can occur in nonprostatic tissues and in a range of benign and malignant conditions. Uptake patterns must therefore be interpreted in light of the tracer used, the expected biodistribution, and the morphologic characteristics of the corresponding lesion [2–4].
One important mechanism involves PSMA expression associated with tumor neovasculature. This biology helps explain why PSMA-targeted imaging can demonstrate uptake outside conventional prostate cancer sites.
A hepatic hemangioma is a benign vascular lesion composed of blood-filled vascular spaces. Increased PSMA uptake in a hepatic hemangioma has been described in the literature, making it a recognized interpretive pitfall [5].
The exact biological mechanism responsible for uptake in an individual hemangioma should not be assumed from PET appearance alone. The practical point is that benign vascular lesions can be PSMA-avid and can imitate metastatic disease.
Imaging Findings: Integrating PET and Contrast-Enhanced CT
1. PSMA PET: Identify the Abnormality
PSMA PET identifies focal tracer uptake in the liver. In this case, two hepatic foci were identified alongside increased uptake at the prostatic apex.
These findings appropriately raise the question of metastatic disease, but they do not establish it.
The radiologist should first determine whether the hepatic foci correspond to discrete anatomic lesions on the CT component. The location, number, distribution, and morphology of the lesions should be assessed, together with the presence or absence of other suspicious sites of disease.
Tracer intensity alone cannot reliably distinguish a hemangioma from a metastasis in every case.
2. CT Morphology: Identify the Corresponding Lesion
The CT component provides anatomic information that complements the functional data from PET.
In this case, contrast-enhanced CT demonstrated two hyperenhancing liver lesions corresponding to the sites of increased PSMA uptake. The external pathology report identified both as benign hemangiomas.
Hyperenhancement is an important observation, but it is not independently diagnostic of a hemangioma. The enhancement pattern across imaging phases, lesion morphology, and clinical context must be considered together.
Figure 1. Axial CT of the liver.
Axial CT demonstrates a hepatic lesion requiring correlation with the corresponding PSMA PET finding and contrast-enhancement characteristics.
ALT text: Axial CT image showing a focal liver lesion evaluated during PSMA PET/CT staging for prostate cancer.
3. Contrast Enhancement: Look for the Hemangioma Pattern
The classic CT appearance of a hepatic hemangioma on dynamic contrast-enhanced imaging includes:
Peripheral, discontinuous nodular enhancement during the arterial phase.
Progressive centripetal enhancement on subsequent phases.
Persistent enhancement of the enhancing portions of the lesion on delayed images.
These features reflect the vascular architecture of a typical hemangioma. Not every hemangioma demonstrates the complete classic pattern. Small lesions, atypical vascular lesions, and lesions with sclerotic changes may require additional characterization.
The case description identifies the lesions as hyperenhancing but does not provide sufficient phase-by-phase detail to reconstruct their complete enhancement pattern. The final diagnosis should therefore be understood in the context of the reported imaging assessment and external pathology report, rather than an independently inferred enhancement sequence.
Figure 2. Coronal PSMA PET.
Coronal PSMA PET demonstrates focal tracer uptake in the liver, prompting evaluation for both malignant and benign causes.
ALT text: Coronal PSMA PET image showing focal hepatic tracer uptake during prostate cancer staging.
Figure 3. Axial PSMA PET/CT.
Axial PET/CT demonstrates the relationship between hepatic tracer uptake and the corresponding anatomic structures.
ALT text: Axial PSMA PET/CT image demonstrating a PSMA-avid liver focus requiring CT correlation.
Figure 4. Contrast-enhanced CT of the liver.
Contrast-enhanced CT demonstrates two hyperenhancing hepatic lesions that were subsequently diagnosed as benign hemangiomas in an external pathology report.
ALT text: Contrast-enhanced CT showing two hyperenhancing hepatic lesions in a patient undergoing prostate cancer staging.
Differential Diagnosis of PSMA-Avid Liver Lesions
A systematic differential diagnosis is essential whenever focal hepatic uptake is identified on PSMA PET/CT.
Hepatic metastases from prostate cancer
Prostate cancer metastases must be considered, particularly when the overall clinical and imaging findings support disseminated disease. However, hepatic metastases should not be diagnosed solely because a lesion demonstrates increased PSMA uptake.
The distribution of disease, prostate cancer risk profile, morphology of the liver lesions, and findings in other organs should be evaluated together.
Importantly, the absence of PSMA uptake does not exclude a metastasis. Uptake can vary with tumor biology, tracer characteristics, and disease phenotype [3, 7, 10].
Hepatic hemangioma
A hemangioma is a key benign mimic in this setting. A characteristic vascular enhancement pattern on CT or MRI can strongly support the diagnosis. PSMA uptake does not invalidate otherwise convincing benign imaging characteristics.
When a lesion is confidently characterized as a hemangioma, it should not be classified as a prostate cancer metastasis merely because it is PSMA-avid.
Focal nodular hyperplasia
Focal nodular hyperplasia (FNH) is another benign hepatic lesion that may demonstrate characteristic vascular enhancement. MRI can be particularly useful when CT findings are not definitive.
FNH should be considered according to the patient's clinical context and the lesion's morphologic and enhancement characteristics. PSMA uptake alone is insufficient to establish this diagnosis.
Hepatocellular carcinoma
Hepatocellular carcinoma should be considered when the patient has relevant liver disease risk factors or imaging features suggestive of a primary hepatic malignancy. Appropriate clinical information and multiphasic CT or MRI findings are important. Laboratory markers may contribute to the assessment when clinically indicated, but no unreported laboratory values should be assumed.
Metastases from another primary malignancy
The differential diagnosis should remain broad when the clinical history, lesion morphology, or distribution of disease suggests an alternative primary malignancy.
The purpose of a structured differential is not to list every possible diagnosis indiscriminately. It is to identify the most relevant alternatives and use imaging features to determine which require further evaluation.
A Practical Diagnostic Algorithm for Radiologists
When PSMA-avid liver lesions are identified, a structured interpretation can reduce the risk of false-positive metastatic staging.
Step 1: Review the clinical context.
Assess the available PSA level, prostate biopsy findings, Gleason grade group, prior imaging, treatment history, and evidence of biochemical recurrence, when applicable.
Step 2: Localize the hepatic uptake.
Determine the number and distribution of lesions and whether each focus corresponds to a discrete abnormality on CT.
Step 3: Characterize CT morphology.
Evaluate lesion margins, attenuation, vascular appearance, and the relationship between the PET focus and the anatomic finding.
Step 4: Review contrast-enhancement behavior.
Look for the characteristic peripheral nodular enhancement and progressive centripetal fill-in of a typical hemangioma. Avoid treating nonspecific hyperenhancement as definitive proof of a benign or malignant diagnosis.
Step 5: Compare with prior imaging.
Previously documented stability and established benign imaging characteristics may help resolve uncertainty. Stability should be interpreted in the context of lesion type, duration, and clinical risk.
Step 6: Obtain liver MRI when needed.
If CT findings are indeterminate or atypical, MRI can provide additional tissue characterization.
Step 7: Integrate the complete disease pattern.
Review other potential sites of metastasis, the clinical probability of metastatic prostate cancer, and the totality of imaging findings. If uncertainty persists and would materially affect management, multidisciplinary review and appropriately selected additional evaluation may be warranted.
The final report should communicate both the degree of diagnostic confidence and the reason for that confidence.
When Is Liver MRI Useful?
MRI is particularly valuable when contrast-enhanced CT does not establish a confident diagnosis.
A typical hepatic hemangioma may demonstrate:
Low or intermediate signal on T1-weighted imaging.
Markedly high signal on T2-weighted imaging.
Characteristic enhancement on dynamic contrast-enhanced sequences.
Diffusion and apparent diffusion coefficient (ADC) characteristics that can help in the differential diagnosis when interpreted with the complete examination.
No single MRI feature should be interpreted in isolation. Diffusion-weighted imaging, for example, is supportive rather than independently definitive.
A practical imaging sequence is:
PSMA PET identifies the focus → CT characterizes the anatomy and enhancement → MRI resolves remaining uncertainty.
Additional imaging should be selected according to the actual diagnostic question rather than ordered automatically for every PSMA-avid liver lesion.
Diagnostic Risk: How a Benign Lesion Can Change Cancer Staging
A false-positive diagnosis of hepatic metastasis can have consequences beyond the imaging report.
Potential consequences include:
Incorrect assignment of metastatic disease.
Changes in the proposed treatment strategy.
Additional imaging or invasive diagnostic procedures.
Unnecessary patient anxiety.
Increased healthcare utilization and cost.
These are potential consequences of diagnostic misclassification, not documented outcomes in this individual case.
The opposite error also matters. A suspicious hepatic lesion should not be dismissed as benign merely because hemangiomas can show PSMA uptake. The objective is neither to overcall nor to undercall disease, but to reach the most defensible diagnosis using all available evidence.
This is where radiologic pattern recognition and disciplined clinical reasoning become essential. The presence of a known malignancy increases the importance of evaluating a lesion carefully; it does not eliminate the need to consider benign alternatives.
Could AI Help Prevent This Diagnostic Pitfall?
Artificial intelligence could support the evaluation of PSMA-avid hepatic lesions, but its role should be defined carefully.
Potential applications include:
PET/CT lesion matching. Image-registration and lesion-localization tools could help identify the CT abnormality corresponding to a focal PET signal.
Quantitative image analysis. Software could measure lesion dimensions, characterize attenuation or enhancement patterns, and compare findings across examinations when suitable imaging data are available.
Longitudinal comparison. AI-assisted tools could flag a lesion documented on previous examinations, allowing the radiologist to assess interval change and previously established benign characteristics.
Structured reporting support. Decision-support systems could prompt consideration of CT morphology, dynamic enhancement, relevant clinical risk factors, and the need for MRI when a lesion remains indeterminate.
These are potential applications, not validated performance claims for the present case. No AI system was reported as having been used, and no sensitivity, specificity, diagnostic accuracy, or time-saving result can be attributed to this examination.
Why AI Cannot Replace Radiologic Judgment
A system that identifies increased tracer uptake may correctly detect a focal abnormality while still failing to determine whether it is a metastasis or a hemangioma.
Potential limitations include false-positive alerts, missed lesions, image-registration errors, differences in scanner protocols, variation among PSMA tracers, and reduced generalizability across patient populations.
An AI model trained predominantly on common metastatic patterns may also perform poorly on uncommon benign mimics if its training data do not adequately represent them.
The most useful role for AI is therefore to support lesion detection, image correlation, quantitative assessment, and workflow prioritization—not to convert tracer uptake directly into a definitive cancer diagnosis.
Final interpretation must remain under qualified clinical oversight, with uncertainty communicated when the evidence is insufficient.
Clinical Workflow and Healthcare-System Implications
The diagnostic pathway illustrates why effective imaging workflows must combine functional imaging, anatomic correlation, and appropriate escalation to problem-solving examinations.
A poorly integrated workflow may encourage readers to focus on the most conspicuous PET abnormality while overlooking the diagnostic information contained in the CT component. In high-volume imaging environments, a structured review process can help reduce this risk.
Useful workflow measures include:
Reviewing PET and CT findings together rather than interpreting tracer uptake in isolation.
Maintaining access to prior imaging when available.
Using structured reporting language for indeterminate lesions.
Escalating unresolved cases to liver MRI or multidisciplinary review when appropriate.
Monitoring discrepancies between imaging impressions and subsequent diagnostic conclusions.
AI may reinforce these processes, but introducing software does not automatically improve diagnostic accuracy. Clinical validation, local workflow testing, monitoring for false positives and false negatives, and clear responsibility for final reporting remain necessary.
From a healthcare economics perspective, avoiding an unsupported metastatic diagnosis may prevent downstream testing and management changes. However, the economic benefit of any specific workflow or AI intervention must be demonstrated rather than assumed.
Treatment and Prognosis: Why Correct Classification Matters
Most hepatic hemangiomas are benign and do not require treatment when the diagnosis is secure and the patient is asymptomatic. Intervention may be considered in selected cases involving significant symptoms, complications, substantial diagnostic uncertainty, or other specific clinical concerns.
PSMA uptake alone is not an indication for hemangioma surgery or other lesion-directed treatment.
By contrast, confirmed hepatic metastases from prostate cancer may have important implications for disease classification and treatment planning. The distinction between a benign vascular lesion and metastatic disease is therefore clinically consequential.
In this case, both hepatic lesions were reported as benign hemangiomas by an external pathology service. The available information does not establish the patient's subsequent treatment, long-term follow-up, or overall oncologic outcome. Those outcomes should not be inferred from the benign classification of the liver lesions alone.
Common Interpretation Pitfalls
Pitfall 1: Equating increased PSMA uptake with metastasis.
Tracer uptake must be interpreted alongside anatomy, enhancement, and clinical context.
Pitfall 2: Treating hyperenhancement as diagnostic by itself.
A complete assessment of the enhancement pattern is more informative than the presence of enhancement alone.
Pitfall 3: Ignoring the CT component of PET/CT.
The anatomic information may provide the clue that prevents a false-positive metastatic diagnosis.
Pitfall 4: Assuming every hemangioma is typical.
Atypical or incompletely characterized lesions may require MRI or other appropriate evaluation.
Pitfall 5: Assuming a negative PSMA study excludes metastasis.
Some prostate cancer metastases may demonstrate limited or absent uptake.
Pitfall 6: Ordering invasive tests before reviewing available imaging evidence.
Prior studies and characteristic imaging findings may resolve uncertainty and help avoid unnecessary procedures.
Frequently Asked Questions
Can a hepatic hemangioma show increased PSMA uptake?
Yes. Increased PSMA uptake in hepatic hemangiomas has been reported. This is an important benign mimic of prostate cancer metastasis and demonstrates why tracer uptake alone cannot establish malignancy.
What CT finding most strongly supports a typical hepatic hemangioma?
Peripheral, discontinuous nodular enhancement followed by progressive centripetal fill-in is a characteristic pattern. Persistent enhancement of the enhancing portions on delayed images can provide additional support.
When should liver MRI be considered?
MRI is useful when CT does not demonstrate a confidently benign pattern, when findings are atypical, or when further characterization would affect clinical management.
Does a PSMA-avid liver lesion automatically change prostate cancer staging?
No. The lesion must be characterized and interpreted in the context of the complete examination and clinical information. A benign hemangioma should not be classified as a metastasis solely because it shows PSMA uptake.
Can AI reliably distinguish hemangiomas from liver metastases on PSMA PET/CT?
AI may support lesion detection, PET/CT correlation, quantitative assessment, and longitudinal comparison. However, reliable differentiation requires appropriate clinical validation, representative training data, and human oversight. No AI performance data are available for this case.
Key Takeaways
PSMA uptake is not specific enough to diagnose prostate cancer metastasis in isolation.
Hepatic hemangiomas can be PSMA-avid and mimic metastatic lesions.
CT morphology and dynamic contrast-enhancement patterns are central to lesion characterization.
Liver MRI can help resolve indeterminate findings.
Incorrect metastatic staging can lead to unnecessary investigations and changes in management.
AI may support image correlation and workflow consistency, but it cannot replace comprehensive radiologic interpretation.
In this case, two PSMA-avid hepatic lesions were ultimately diagnosed as benign hemangiomas in an external pathology report.
The key lesson is that functional imaging identifies abnormal tracer distribution, while integrated imaging interpretation determines its clinical meaning. A careful review of PSMA PET, CT morphology, enhancement characteristics, and clinical context can help distinguish a benign hepatic hemangioma from metastatic prostate cancer and support more appropriate patient management.
References
M. S. Hofman et al., “Prostate-specific membrane antigen PET-CT in patients with high-risk prostate cancer before curative-intent surgery or radiotherapy (proPSMA): a prospective, randomised, multicentre study,” The Lancet, vol. 395, pp. 1208–1216, 2020, doi: 10.1016/S0140-6736(20)30314-7.
M. S. Hofman, R. J. Hicks, T. Maurer, and M. Eiber, “Prostate-specific membrane antigen PET: Clinical utility in prostate cancer, normal patterns, pearls, and pitfalls,” RadioGraphics, vol. 38, no. 1, pp. 200–217, 2018, doi: 10.1148/rg.2018170108.
S. Sheikhbahaei et al., “Prostate-Specific Membrane Antigen PET Imaging of Prostate Cancer: An Update on Important Pitfalls,” Seminars in Nuclear Medicine, vol. 49, no. 4, pp. 255–270, 2019, doi: 10.1053/j.semnuclmed.2019.02.006.
K. Z. Keidar et al., “68Ga-PSMA PET/CT in prostate cancer patients—patterns of disease, benign findings and pitfalls,” Cancer Imaging, vol. 18, article 39, 2018, doi: 10.1186/s40644-018-0175-3.
H. Bhardwaj, M. Stephens, M. Bhatt, and P. A. Thomas, “Prostate-Specific Membrane Antigen PET/CT Findings for Hepatic Hemangioma,” Clinical Nuclear Medicine, vol. 41, no. 12, pp. 968–969, 2016, doi: 10.1097/RLU.0000000000001384.
W. P. Fendler et al., “PSMA PET/CT: joint EANM procedure guideline/SNMMI procedure standard for prostate cancer imaging 2.0,” European Journal of Nuclear Medicine and Molecular Imaging, vol. 50, pp. 1466–1486, 2023, doi: 10.1007/s00259-022-06089-w.
R. Mei et al., “PET/CT Variants and Pitfalls in Prostate Cancer: What You Might See on PET and Should Never Forget,” Seminars in Nuclear Medicine, vol. 51, no. 6, pp. 621–632, 2021, doi: 10.1053/j.semnuclmed.2021.06.016.
R. G. Gore et al., “Management of Incidental Liver Lesions on CT: A White Paper of the ACR Incidental Findings Committee,” Journal of the American College of Radiology, vol. 14, no. 11, pp. 1429–1437, 2017, doi: 10.1016/j.jacr.2017.07.018.
R. W. Foley et al., “Fluorine-18 labelled prostate-specific membrane antigen (PSMA)-1007 positron-emission tomography-computed tomography: normal patterns, pearls, and pitfalls,” Clinical Radiology, 2020, doi: 10.1016/j.crad.2020.06.031.
S. Mattoni et al., “PSMA PET for the Evaluation of Liver Metastases in Castration-Resistant Prostate Cancer Patients: A Multicenter Retrospective Study,” Cancers, vol. 14, no. 22, article 5680, 2022.
Continue Learning
Medical Disclaimer: This article is intended for medical education and professional discussion. It does not replace individualized medical evaluation, diagnostic judgment, or treatment recommendations. Imaging findings should be interpreted by qualified clinicians in the context of the patient's complete clinical history and relevant diagnostic studies.
Comments
Post a Comment