Long-term prognosis after curative resection is unsatisfactory because of the high incidence of recurrence. Due to strict follow-up plans and the development of imaging tools, an increasing number of patients with recurrent HCC are detected in the early stage. As previous studies have reported, the mean size of recurrent HCC is 2–4 cm in diameter, and 32–43% of recurrent HCC cases contain 2–3 nodules not larger than 3 cm[10–12]. In this study, the pattern of recurrence was in accordance with previous reports that 62.7% of patients were detected in the early stage (according to the Barcelona Clinical Liver Cancer Staging system), and multiple tumours were detected in 62.6%. Therefore, the appropriate management of recurrent HCC in the early stage is central to improving the overall long-term efficacy of HCC. For this research purpose, the inclusion criteria mentioned previously were established at the beginning state of this study. Salvage liver transplantation might be the ideal treatment, but it plays a limited role in areas where there is the shortage of organs. Repeat hepatectomy has been performed as the most effective therapy, with a 5-year survival rate ranging from 37–70%. Unfortunately, the repeat hepatectomy rate is unsatisfactory for various reasons. One of the reasons is that many patients prefer treatments with minimal damage. In our population, the repeat hepatectomy rate was 14.8%, which is similar to the rates observed in other studies which have repeat hepatectomy rates of approximately 10–30%. Therefore, non-surgical treatment is critical for prolonging survival time after recurrence.
Although various therapeutic modalities have been used to treat recurrent HCC, the effects of different treatment methods have not been sufficiently compared. There is no standard strategy used to select the modality for multiple recurrent tumours. Due to the fact that the patients enrolled in this retrospective study had the criteria of “2–3 tumours, each < 3 cm in size at recurrence”, TACE and RFA were the most frequent treatments performed. It is generally accepted that RFA is a reliable, effective and safe therapy for intra-hepatic recurrent HCC, especially for the small single tumour, but a substantial number of patients undergo TACE as the first-line treatment for several reasons, including objective medical parameters, subjective concerns, or insurance coverage[14–19]. The prognosis of TACE therapy after recurrence compared to that of repeat resection or RFA is very poor, with a reported 5-year survival rate of 0–27%. According to our results, TACE therapy demonstrated a significantly worse prognosis than RFA. TACE is still considered a useful modality based on the fact that the 3-year survival rate of patients with recurrence but without treatment is 8%. Only in patients with unfavourable tumour conditions or poor liver function, TACE may be the first-line treatment of choice.
The use of RFA for multiple tumours is limited because of the presence of undiscovered minute lesions and the increased probability of insufficient ablation. Due to the development of contrast-enhanced intraoperative ultrasound, minute tumours that are undetected during preoperative examinations have been confirmed in 9–23% of patients with HCC[21, 22]. On the other hand, the rates of local recurrence attributed to insufficient ablation at an ablated site after RFA vary from 3–26%[22, 23]. In theory, RFA combined with TACE has advantages as a supplemental treatment for uncompleted necrotic areas; however, to date, there have been few reports on combined treatment. Although no evidence was reported in the published work, our study suggested that combined treatment is superior to TACE and RFA, and RFA was better than TACE. To minimize the possibility of selection bias, patients were divided into subgroups on the basis of tumour size, tumour number and recurrence time. Subgroup analyses showed a survival benefit of CT over TACE and RFA, except for patients with 3 tumours. Considering the small number of patients, the results of patients with 3 tumours do not provide strong evidence for the survival benefit of CT.
Independent prognostic factors for post-recurrence survival include tumour status, liver reserve, and recurrence status[24–26]. Recurrence time was identified as a prognostic factor in the multivariate analysis with a high hazard ratio (HR = 2.85). Early recurrence was defined as intrahepatic, regional, or systemic recurrence within 1 year after curative resection. Previous studies have demonstrated that differences in pathogenesis lead to better outcome for late recurrence (> one year after resection) than for early recurrence[27–29]. The results suggest that aggressive treatment should be adopted as frequently as possible for patients with late recurrence.
There are some limitations in this study: (1) it is a single-institutional and non-randomized study; (2) treatments for recurrence were more likely to be self-selected by patients; and (3) some bias could have been present due to the preferences of doctors. Nevertheless, a prospective, randomized trial will be needed to confirm these results.
In summary, although there was inevitable selection bias in terms of the patients and treatments, our study suggested several important points for multiple recurrent HCCs in the early stage: first, long-term survival was achievable when an aggressive regimen was used to improve survival in patients with HCC recurrence, especially in patients with late recurrence; second, RFA combined with TACE offered the best chance of survival for patients who did not undergo repeat hepatectomy; and finally, compared with TACE, RFA still demonstrated a survival benefit for managing patients who had HCC recurrence with a total of 2–3 tumours.