Noncontrast-enhanced MRI superior to ultrasound but less cost-effective for HCC screening

Noncontrast-enhanced magnetic resonance imaging (NCEMRI) demonstrates superiority over ultrasonography (US) for the surveillance of hepatocellular carcinoma (HCC) in at-risk populations, although this strategy is more expensive than US, according to a Singapore study.

A total of 482,000 patients (mean age 40 years) underwent simulation and were followed up for 40 years. The average total costs were SGD 1,193 with no surveillance, SGD 8,099 with US surveillance, and SGD 9,720 with NCEMRI surveillance. [Singapore Med J 2024;65:23-29]

In addition, the quality-adjusted life years (QALYs) were 7.460, 11.195, and 11.366 with no surveillance, US surveillance, and NCEMRI surveillance, respectively.

“Future local cost-effectiveness analyses should include stratifying surveillance methods with a variety of imaging techniques (US, NCEMRI, CEMRI) based on patients’ risk profiles,” the researchers said. “This would enhance our understanding of the cost-effectiveness and impact on the overall outcome of patients using various imaging tools for HCC surveillance.”

HCC surveillance

A similar analysis comparing CEMRI with US for HCC surveillance was conducted in a study by Kim and colleagues. They found that CEMRI was more cost-effective than US. [Hepatology 2019;69:1599-1513]

“Our study showed a lower overall gain in QALY of 0.18 using NCEMRI instead of US, compared to 0.22 incremental QALY using CEMRI compared to US shown by Kim and colleagues,” the researchers said.

In another study, Andersson and colleagues performed cost-effective analysis comparing different imaging surveillance strategies (US vs MRI vs computed tomography [CT]) and surveillance intervals (annual vs semi-annual).

The research team observed an incremental cost-effectiveness ratio (ICER) exceeding USD 100,000 for MRI surveillance. They concluded that MRI was least cost-effective relative to US or CT. [Clin Gastroenterol Hepatol 2008;6:1418–1424]

“This could be because in the study by Andersson and colleagues, full multisequence CEMRI was performed, which increases the cost of the scan compared to CT or US,” the researchers said. “The findings are similar to that of our study, which showed that NCEMRI is not a cost-effective surveillance modality, even without intravenous contrast.”

Transition probability

In other cost-effective analyses, cohorts with liver cirrhosis were simulated. In regions where other risk factors (eg, chronic hepatitis C or alcoholic cirrhosis) might contribute to HCC, investigators applied higher transition probabilities, which ranged from 1.5 percent to 5 percent. [Clin Gastroenterol Hepatol 2008;6:1418-1424; AJR Am J Roentgenol 2019;213:17-25; Hepatology 2019;69:1599-1513]

“The higher transition probabilities applied in these studies will tend to lead to increased cost-effectiveness, as more cases of early HCC are picked up when the disease is still curable,” the researchers said.

In the current study, Markov modelling and microsimulation were used to perform the cost-effectiveness analysis of no surveillance, US surveillance, and NCEMRI surveillance. The researchers simulated at-risk patients and followed them for 40 years to estimate the patients’ disease status, direct medical costs, and effectiveness. They also calculated the QALYs and ICER.