Hepatocellular carcinoma surveillance with liver imaging is not associated with improved survival

Abstract

Objective: International guidelines recommend hepatocellular carcinoma (HCC) surveillance with ultrasound in high-risk patients with chronic liver diseases. However, there is low-strength evidence about the effects on mortality. The aim of our study was to assess the impact of surveillance on the clinical course and survival of HCC patients seen at a tertiary referral center in Germany.

Material and methods: We retrospectively evaluated the data of 401 HCC patients, who presented to our clinic between 1997 and 2015. Two groups were compared regarding patient and disease outcomes: one group included patients who received at least two ultrasound examinations for surveillance purposes prior to first diagnosis (n = 111). The other group consisted of patients with HCC at first presentation without foregoing HCC surveillance (n = 290).

Results: Median follow-up in the surveillance group was 76 months (range 4–310 months). Patients in the surveillance group had smaller median tumor sizes (3.5 cm vs. 4.5 cm; p < .001), fulfilled more often Milan criteria (64% vs. 42%; p < .001) and received more often liver transplantation (27% vs. 9%, p < .001) when compared with the non-surveillance group. However, HCC surveillance was not associated with an improved survival (14 months in the surveillance group vs. 12 months in the non-surveillance group, p = .375), hazard ratio regarding overall mortality for the surveillance group: 0.80 (95% CI: 0.62–1.04, p = .09).

Conclusions: HCC surveillance with ultrasound led to the detection of earlier disease stages but was not significantly associated with improved survival. Further prospective and long-term studies are needed to clarify benefits and harms of HCC surveillance programs on mortality.

Keywords:

• Liver cancer screening
• ultrasound surveillance
• liver cancer

Disclosure statement

There is no commercial affiliation or consultancy of any author that could be construed as a conflict of interest with respect to the submitted data. There is also no potential personal conflict of interest of any author.