HCC surveillance may be cost-effective at lower incidence in alcohol-associated cirrhosis

Share

Routine hepatocellular carcinoma (HCC) surveillance may remain cost-effective in patients with alcohol-associated cirrhosis at substantially lower incidence rates than previously assumed, including among patients who continue to drink alcohol, according to a modeling study published in Clinical Gastroenterology and Hepatology.

Researchers found that twice-yearly surveillance with ultrasound plus alpha-fetoprotein (AFP) became cost-effective when annual HCC incidence reached approximately 0.45 per 100 person-years, using a willingness-to-pay threshold of $100,000 per quality-adjusted life-year (QALY).

Jagpreet Chhatwal, PhD

“The key takeaway is that routine HCC surveillance is cost-effective in patients with alcohol-associated cirrhosis at a substantially lower annual HCC incidence than previously assumed,” senior author Jagpreet Chhatwal, PhD, director of the Center for Health Technology Assessment at Mass General Brigham and associate professor of radiology at Harvard Medical School, told GI & Hepatology News.

He added that surveillance provided clinical benefit and good economic value in both patients who continued to drink and those who had achieved abstinence.

Binu V. John, MD, MPH, associate professor of medicine at the University of Miami Miller School of Medicine and chief of gastroenterology and hepatology at the Miami VA Healthcare System, was not involved in the study. He said the findings support offering HCC surveillance even to patients who continue to drink.

“Ongoing alcohol use should not be a reason to withhold surveillance,” he said.

Model compared surveillance with no surveillance

Ali Hajjar, PhD

Ali Hajjar, PhD, and colleagues developed a Markov-based microsimulation model to compare twice-yearly ultrasound plus AFP with no surveillance in U.S. patients with compensated alcohol-associated cirrhosis. Simulated patients began surveillance at age 50 and were followed until death.

The researchers modeled two groups separately: patients who continued actively drinking and patients who had achieved sustained abstinence. The model incorporated HCC incidence, tumor progression, surveillance adherence, test performance, treatment options, liver-related and non-liver-related mortality, quality of life, and costs. It assumed only abstinent patients younger than 75 were eligible for liver transplantation. The model’s base-case HCC incidence was 8.29 per 1,000 person-years (about 0.83 per 100).

Among active drinkers, surveillance produced an incremental cost-effectiveness ratio (ICER) of $18,952 per QALY. Among patients who had stopped drinking, the ICER was $56,000 per QALY.

“The ICER is the extra cost of surveillance divided by the extra health it buys,” Dr. John explained. “A QALY is one year of life in full health. So surveillance spends about $19,000 for every healthy life-year it adds.”

That figure falls below the $100,000-per-QALY threshold the researchers applied, which they noted reflects updated recommendations.

At the population level, surveillance added an estimated 2,083 life-years and 992 QALYs per 100,000 active drinkers. Among abstinent patients, the corresponding gains were 3,357 life-years and 1,700 QALYs per 100,000 patients.

Surveillance also shifted HCC detection toward earlier-stage disease in both groups. The ICER stayed below $100,000 per QALY in all one-way sensitivity analyses, and probabilistic analyses showed an 86% likelihood of cost-effectiveness at that threshold among active drinkers and 82% among abstinent patients.

Incidence threshold

The 0.45 threshold compares with the roughly 1 per 100 person-years historically cited to guide surveillance, according to the study.

“Our findings suggest that the incidence thresholds historically used to establish the cost-effectiveness of HCC surveillance deserve reconsideration, particularly for patients with alcohol-associated cirrhosis,” Dr. Chhatwal said.

He noted that the commonly cited 1% annual incidence threshold was based largely on older studies, whereas the current model incorporated contemporary surveillance performance, treatment options, and healthcare costs.

For patients with established cirrhosis, Dr. John said, the implication is relatively straightforward.

“The HCC risk of patients with compensated alcohol-associated cirrhosis is well above this new suggested threshold,” he said. “For patients with cirrhosis, the case for surveillance is clear.”

The larger unanswered question may involve patients closer to the surveillance threshold, he said, such as those with advanced fibrosis but without cirrhosis or patients whose HCC risk declines after sustained abstinence.

“For them, individualized risk tools such as liver stiffness measurement may help better guide the surveillance threshold, and we need more data in this population to determine additional patients who may benefit,” he added.

Per-patient gains

Dr. John cautioned that favorable cost-effectiveness does not mean the absolute benefit for each patient is large, as surveillance added approximately 0.01 QALYs per active drinker and 0.017 QALYs among abstinent patients.

“The gain per patient is small,” he said. “Surveillance added only a few days of quality-adjusted life per patient, because competing mortality is high in this population.”

The incremental lifetime cost of surveillance was $188 per patient among active drinkers and $952 among abstinent patients.

“The ratio looks favorable because the extra cost is also small,” he noted. Earlier detection, he added, may partly offset surveillance costs through less expensive and more effective treatment of smaller tumors.

Abstinent patients gained more health but had a higher ICER of about $56,000 per QALY.

“This is not because surveillance works less well in them,” Dr. John explained. “They live longer, so they accumulate more years of surveillance and cirrhosis care, which costs more. The higher cost reflects longer survival.”

Dr. Chhatwal said surveillance remained beneficial in both groups.

“Both groups can benefit from HCC surveillance, although the magnitude of benefit differs,” he said. “Alcohol abstinence remains a critical goal in managing alcohol-associated cirrhosis, but surveillance should not be withheld solely because a patient continues to drink. Likewise, achieving abstinence does not eliminate the need for ongoing HCC surveillance.”

Adherence and future research

The study was based on a simulation model reflecting U.S. costs and transplant practices, which the authors said may not apply to other healthcare systems. It also evaluated only ultrasound plus AFP and did not assess newer blood-based biomarkers or alternative imaging strategies.

The researchers also analyzed active drinkers and abstinent patients as separate groups and did not model changes in alcohol consumption over time.

“In clinical practice, patients may achieve abstinence, relapse, or change their alcohol consumption, potentially affecting both HCC risk and treatment eligibility,” Dr. Chhatwal said. He added that the model also did not explicitly account for metabolic comorbidities such as diabetes.

Future studies, he said, should validate the findings using real-world data, assess how changes in alcohol consumption affect the benefits of surveillance, and evaluate more individualized strategies that incorporate additional patient-level risk factors.

Fewer than one-quarter of patients in the adherence data used by the model received consistent surveillance, defined as at least one ultrasound per year. Among those with alcohol-associated cirrhosis, the figure was 4.3%, although surveillance remained cost-effective when the model applied those rates.

“The main challenge is not whether surveillance is worth doing,” Dr. John said. “It is delivering it reliably to this group of patients who are often hard to retain in care.”

The study authors reported no conflicts of interest.