Liver Cancer Screening Benefits Patients With Cirrhosis, Study Suggests

Screening for hepatocellular carcinoma (HCC) in liver disease patients with cirrhosis may indeed lead to a survival benefit, a retrospective analysis of data from two large U.S. health systems suggested.

Among more than 1,300 patients diagnosed with HCC, the median survival reached 37 months for those with screening-detected tumors compared with 19 months for those whose tumors were found incidentally or due to symptoms (unadjusted HR 0.65, 95% CI 0.56-0.75).

And this difference persisted in adjusted models that accounted for lead-time bias (HR 0.77 for a mean sojourn time of 6 months, 95% CI 0.67-0.89), Amit Singal, MD, of the University of Texas Southwestern Medical Center in Dallas, and colleagues detailed in JAMA Network Open.

Guidelines from the American Association for the Study of Liver Diseases recommend screening high-risk patients for HCC using ultrasound, with or without α-fetoprotein (AFP). And the recommendations extend to liver disease patients with cirrhosis.

But "the efficacy of HCC screening in patients with cirrhosis is controversial because of the lack of randomized data and the inherent biases of cohort studies, including lead-time and length-time biases," noted Singal and colleagues.

The controversy over whether to screen patients with cirrhosis was further fueled by a study in veterans that found no association between screening and reduced HCC-related mortality, though Singal and co-authors posited that this could have been due to downstream failures in care, including the underuse of curative therapies for early tumors.

In the current study, crude estimates showed 3- and 5-year survival rates of 46% and 32%, respectively, for the patients with screening-detected HCCs versus 25% and 8% for the group with non-screening cancers.

Analyses that corrected for both lead-time and length-time biases (using a mean sojourn time of 6 months) showed 3- and 5-year survival rates of 37% and 25% in the screening-detected group. "Lead-time bias occurs when screening leads to earlier cancer detection, so the time from diagnosis to death appears longer without an actual difference in mortality," the researchers explained, while "length-time bias results from screening being more likely to detect slow-growing indolent tumors that are less likely to be fatal, which confers an artificial perception of improved survival."

Overall, 42.3% of the cancers in the study were detected by screening. Tumor doubling time was similar for screening- and non-screening-detected HCCs (median 3.8 vs 5.6 months, respectively, P=0.40), numbers in line with a meta-analysis on the matter, and a similar proportion of patients in the two groups had indolent tumors (35.4% vs 38.1%; risk ratio 0.93, 95% CI 0.60-1.43).

"Overall, these data highlight the importance of promoting HCC screening implementation in practice," the authors wrote in conclusion, noting that the underuse of screening has been reported in several studies.

The retrospective cohort study from Singal and colleagues included 1,313 patients diagnosed with HCC at UT Southwestern Medical Center and Parkland Health, both in Dallas, from January 2008 to December 2022, including 556 whose tumors were detected via screening.

Patients had an average age of 62 years, and three-fourths were men. A little more than a third of patients were white (36%), 30% were Black, and 29% were Hispanic. Most (62.5%) had Child Pugh class A cirrhosis and the rest class B. (Patients with class C cirrhosis were excluded from the analysis, as screening is not indicated in this group due to the high competing risk of mortality.)

HCC in the screening group was detected by imaging alone in 44.6%, AFP alone in 10.6%, and by both methods in 44.7%. Most tumors in the non-screening group (75.3%) were found incidentally, with the rest detected following symptoms.

At the time of cancer diagnosis, a majority (56%) had Barcelona Clinic Liver Cancer stage 0/A disease, 20% had stage B disease, and 24% had stage C disease. Hepatitis C or B viruses were considered the cause in 60% and 7%, with alcohol responsible for 14% and metabolic dysfunction-associated steatotic liver disease for 12% of cases.

Patients with tumors detected by screening were more likely to have early-stage HCC (70.7% vs 45.7% in the non-screening group) and receive curative treatment (51.1% vs 33.5%). Exploratory analyses, however, showed similar responses to curative-intent treatment regardless of whether tumors were screen-detected, found incidentally, or found due to symptoms, with response rates exceeding 80% in each group. Response rates after embolic treatments were similar for patients with screen-detected or incidental HCCs (54.8% vs 59.9%), but lower among those whose tumors were detected due to symptoms (38.9%).

Study limitations included that some imaging may have been performed in other health systems, though outside clinical records were reviewed; potential misclassification bias; the risk for residual confounding; and that a higher proportion of patients with indolent tumors may have been used to estimate tumor doubling time.

Comment