This study enrolled a large cohort of patients with early-stage HCC and Child–Pugh class A liver disease undergoing pRFA, who were stratified into four groups by age. Tumor numbers, BCLC stages, AFP level ≥ 20 ng/mL, and MELD score > 9 did not differ significantly between groups. The proportion of men was highest in Group 1 (p = 0.001), the HBsAg-positive proportion was highest in Group 1 (p < 0.001), and the anti-HCV-positive proportion was highest in Group 3 (p = 0.001). However, whether sex and chronic liver disease etiology were associated with OS among patients with early-stage HCC remains unclear.7 In contrast, tumor burden (i.e., tumor number and BCLC stage), AFP level, and liver function reserve (i.e., MELD score) are well-known factors associated with OS of patients with HCC [7]. We can assume that characteristics related to OS were similar between groups. Therefore, we did not perform propensity score matching between them. We found that five-year OS was lowest in those aged ≥ 80 years (p = 0.009). In addition, multivariate analysis showed that age ≥ 80 years was independently associated with inferior five-year OS. Therefore, we can conclude that age ≥ 80 years was associated with inferior five-year OS in patients with early-stage HCC and Child–Pugh class A liver disease undergoing pRFA. In contrast, age ≥ 80 years was not associated with five-year RFS.
Multiple studies have associated being elderly with inferior OS in patients with early-stage HCC undergoing pRFA. Takahashi et al. enrolled 461 patients with early-stage HCC and showed that OS was similar between elderly (i.e., aged ≥ 75 years) and non-elderly patients with HCC undergoing pRFA [17]. Similarly, Hiraoka et al. enrolled 206 patients with early-stage HCC and showed that OS was similar between elderly (i.e., aged ≥ 75 years) and non-elderly patients with HCC undergoing pRFA [18]. However, Nishikawa et al. enrolled 368 patients with early-stage HCC and showed that OS was inferior in elderly (i.e., aged ≥ 75 years) compared to non-elderly patients with HCC undergoing pRFA (p = 0.001) [19]. Nonetheless, Tateishi et al. showed no difference in three-year survival between patients aged > 68 years (76%) and < 68 years (79.2%) among 1000 patients treated with RFA [20]. Finally, Ochi et al. enrolled 1079 patients with early-stage HCC treated with RFA and found that those aged ≥ 80 years had significantly shorter OS than the other groups (p < 0.05) [8].
All these studies enrolled patients with Child–Pugh class A or B liver disease [8, 17–20]. Since decompensated cirrhosis may be a competing risk more strongly associated with death than older age, in this study, we only enrolled patients with Child–Pugh class A liver disease to clarify whether older age was associated with inferior OS in patients with early-stage HCC undergoing pRFA.
Similar to Ochi et al., [8] non-HCC-related deaths were most common among those aged ≥ 80 years in our study, although the difference was not statistically significant (p = 0.389).
A recent population-based study evaluated about 3.9 million patients in seven high-income Western countries with seven primary cancers (esophagus, stomach, colon, rectum, pancreas, lung, and ovary). It showed that the survival benefit was limited in elderly patients, whereas a significant improvement in the OS was noted in younger patients [21]. Although they did not assess HCC, their findings are consistent with ours regarding HCC.
Age ≥ 80 years was not associated with five-year RFS in our study, consistent with Ochi et al. [8] A MELD score of > 9 was associated with inferior five-year OS in our study. A recent review paper showed that the MELD score is commonly used in patients with HCC undergoing LR to predict postoperative liver failure and survival; it is also commonly used in patients with HCC undergoing LT to predict survival on the waiting list. The albumin-bilirubin (ALBI) score is used to predict survival in patients with HCC undergoing LR and non-surgical therapies [22]. However, the lack of albumin data in our HCC registry dataset precluded us from using the ALBI score to estimate patients’ liver functional reserve. In addition, while a MELD score of > 9 was originally used to predict postoperative liver decompensation in patients with HCC undergoing LR, [15] our findings suggest that it can also predict OS in patients with HCC undergoing pRFA.
Our study associated an AFP level of ≥ 20 ng/mL with inferior five-year OS and RFS. This cutoff was set according to the American Association for the Study of Liver Diseases guidelines, which recommend that among patients at risk of HCC and undergoing surveillance for HCC, those with an AFP level of ≥ 20 ng/mL should be examined using contrast-enhanced CT or MRI [23].
The strength of our study is that it only enrolled patients with Child–Pugh class A liver disease to clarify whether older age is associated with inferior OS in patients with early-stage HCC undergoing pRFA. However, it had some limitations. First, it was a retrospective single-center study. Second, the number of patients aged ≥ 80 years was limited. Future studies are needed to evaluate which treatment is most appropriate for patients with early-stage HCC and Child–Pugh class A liver disease. Stereotactic body radiotherapy (SBRT) may offer advantages in patients with early-stage HCC aged ≥ 80 years because it does not require general anesthesia. Future studies should also compare survival outcomes and treatment complications between pRFA and SBRT in patients with early-stage HCC aged ≥ 80 years.