Abstract
International liver societies recommend hepatocellular carcinoma (HCC) surveillance for those at high-risk of developing HCC. While previous studies have shown the benefits of surveillance for middle-aged patients, but its necessity for elderly patients is unclear. This study aimed to assess the benefits of HCC surveillance in the elderly by comparing diagnosis mode of HCC. Consecutive, elderly patients aged 75 years or older who were newly diagnosed with HCC were screened at our institution between January 2009 and December 2021. Patients were grouped into those who were diagnosed with HCC during surveillance (n = 235, surveillance group) and those who were diagnosed with HCC due to symptoms (n = 184, symptomatic group). The study outcome was overall survival. It was compared in the overall cohort and a propensity score (PS)-matched cohort. Early-stage diagnosis was more frequent in the surveillance group than in the symptomatic group (mUICC stage I/II: 72.3% vs. 39.1%, p < 0.001). The overall survival rate was better in the surveillance group than in the symptomatic group (median 4.4 vs. 2.1 years, log-rank p < 0.001). In multivariable-adjusted models, the hazard ratio (HR) of mortality of the surveillance group compared to the symptomatic group was 0.64 (95% confidence interval (CI): 0.47-0.87). However, further adjustment for the tumor stage markedly attenuated this association, which was no longer statistically significant (adjusted HR = 0.75; 95% CI: 0.54-1.02). In the PS-matched cohort analysis, outcomes were similar when the PS matching variables included the tumor stage. In contrast, when PS matching variables did not include the tumor stage, outcomes were better for the surveillance group. The surveillance group of elderly patients showed better survival than the symptomatic group, which was largely explained by earlier tumor stage at diagnosis. This suggests that the overall outcome of elderly HCC patients could be improved by increasing surveillance-detected cases compared to symptom-driven cases.