Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association
October 15, 2026 · via Europe PMC
Optimal Incidence Threshold for Cost-Effective Hepatocellular Carcinoma Surveillance in Alcohol-Associated Cirrhosis.
Hajjar A, Julien J, Nemutlu G, Sahinkoc HM, Peters MLB, Kanwal F, Chhatwal J.
AI summary — for doctors, generated from the abstract only
A modeling study found HCC surveillance in alcohol-associated cirrhosis to be cost-effective at incidence above about 0.45 per 100 person-years.
What was studied
A Markov microsimulation of biannual ultrasound plus AFP surveillance versus none in US patients, separately for active drinkers and abstinent patients.
Key findings
- Surveillance becomes cost-effective at an annual HCC incidence of about 0.45 per 100 person-years at $100,000 per QALY
- ICER was $18,952 per QALY in active drinkers and $56,000 per QALY in abstinent patients
- Per 100,000 patients, surveillance generated 2,083 additional life-years and 992 QALYs in active drinkers, and 3,357 life-years and 1,700 QALYs in abstinent patients
Clinical relevance
It supports lowering the threshold in guidelines, so more patients could be offered surveillance.
Who should care
Limitations
It is a simulation dependent on model inputs and assumptions in the US setting.
Surveillance is likely cost-effective well below the traditional 1 per 100 person-years threshold.
This is an AI-generated summary of the published abstract, not a clinical recommendation. It may not reflect the full text or subsequent updates — verify against the original source before acting on it.
Abstract
Background
Current guidelines recommend routine hepatocellular carcinoma (HCC) surveillance in patients with cirrhosis when annual HCC incidence exceeds 1 per 100 person-years, a threshold derived from older studies and not specific to alcohol-associated cirrhosis. However, contemporary data suggest that HCC incidence in this population is often below 1%, leaving the value of routine surveillance uncertain.
Methods
We developed a Markov-based state-transition microsimulation model to simulate the natural history of HCC and subsequent treatment pathways in patients with alcohol-associated cirrhosis in the United States setting. The model compared biannual surveillance with ultrasound and alpha-fetoprotein (AFP) versus no surveillance separately for two distinct cohorts: patients who continued to actively consume alcohol and patients who had achieved sustained abstinence. Inputs included the natural history of alcohol-associated cirrhosis, HCC incidence, test performance, surveillance adherence, and contemporary treatment. Sensitivity analyses examined the robustness of model assumptions and findings.
Results
Biannual surveillance becomes cost-effective when annual HCC incidence reached approximately 0.45 per 100 person-years using willingness-to-pay threshold of $100,000 per quality-adjusted life year (QALY). Among active drinkers, surveillance yielded an incremental cost-effectiveness ratio (ICER) of $18,952 per QALY; among abstinent patients, the ICER was $56,000 per QALY. Per 100,000 patients, routine surveillance generated an estimated 2,083 additional life-years and 992 QALYs in active drinkers and 3,357 life-years and 1,700 QALYs in abstinent patients.
Conclusion
Routine biannual HCC surveillance with ultrasound and AFP is cost-effective for patients with alcohol-associated cirrhosis in the United States when HCC incidence exceeds 0.45 per 100 person-years, substantially below the historically cited 1 per 100 person-years threshold. Updating clinical guidelines to reflect these contemporary thresholds could improve early HCC detection in this high-risk population.
DOI
10.1016/j.cgh.2026.09.035
PMID
42880653
Research metadata (titles, abstracts, authors, publication dates) is sourced from Europe PMC, which aggregates content from PubMed/MEDLINE, PubMed Central, and other providers under the U.S. National Library of Medicine (NLM/NCBI). NLM does not claim copyright on abstracts; the original journal publishers or authors may hold rights to the full articles. AI-generated summaries are derived only from the abstract, never the full text, and are always labeled as AI-generated.
Want more like this?
Browse more from the feed. A free account unlocks the full feed, saving and communities.
Back to the research feed