Achyut Saroj, Medical Science Liaison and Board-Certified Medical Affairs Specialist at Helio Genomics, shared on LinkedIn:
“How Is Risk-Based HCC Surveillance Changing the Game?
When evaluating an early cancer detection test, our first instinct is to ask, ‘How accurate is it?’ But in cancer surveillance, accuracy is only half the equation. The other half is baseline disease incidence.
In liver cancer surveillance, this mathematical reality creates a major clinical dilemma: even a test with solid diagnostic specs can result in more false alarms than true cancer detections if used in a low-risk cohort.
Positive Predictive Value (PPV), the probability that a patient with a positive diagnostic test result actually has cancer (e.g., MRI/CT confirmed HCC), is directly tied to disease prevalence.
Let’s consider a typical surveillance scenario in a low-prevalence population: Prevalence of Hepatocellular Carcinoma (HCC): 2%, Ultrasound Sensitivity: 50%, Ultrasound Specificity: 90%
Under these standard conditions, the resulting PPV is only 9%. That means 91% of all positive results are false positives. Because the non-diseased population is so much larger than the diseased subset, even a 90% specificity produces a volume of false positives that far outnumbers true positive cases.
How do we break out of this low-PPV trap? We cannot easily change a test’s physical limitations, but we can change the baseline incidence of the screened population.
Major hepatology organizations are shifting toward risk-based surveillance, restricting routine screening to cohorts that exceed specific annual incidence thresholds:
e.g., AASLD Threshold: Annual HCC incidence >1.0% per year.
To identify these high-risk subgroups, clinicians increasingly rely on validated risk-prediction scores:
PAGE-B and mPAGE-B
Practical scores using age, sex, platelets, and albumin to stratify risk in patients with chronic hepatitis B.
aMAP Score
An etiology-free model combining age, sex, albumin-bilirubin, and platelet counts to estimate risk across viral and non-viral etiologies.
Advanced Modalities
Exploring tools like non-contrast abbreviated MRI (aMRI) to improve sensitivity over standard ultrasound.
Emerging HelioLiver Liquid Biopsy Test
This AI -powered multianalyte test with abnormal results (methylation, AFP, AFPL3%, DCP, and patient age and gender) can identify high-risk cirrhotic patients with HCC lesions < 2cm (28.6% sensitivity vs 0% in ultrasound) in a prospective CLiMB trial published in the Journal of Hepatology
By focusing surveillance on higher-incidence subgroups, pre-test probability rises, PPV improves, and the overall benefit-to-harm ratio for patients is substantially enhanced.
Has your institution integrated risk-scoring models like PAGE-B, aMAP, and/or HelioLiver blood test into routine liver care workflows?”
Title: Surveillance of patients at risk of liver cancer: Expert opinions
Authors: Terry Cheuk-Fung Yip, Hamish Innes, Grace Lai-Hung Wong, Peter Jepsen
You can also read: Achyut Saroj: The Power of Methylation Score in Predicting Liver Cancer Progression