Transarterial chemoembolization versus radioembolization as a treatment option for single large (>5 cm) hepatocellular carcinoma: a retrospective study in South Korea

This study was approved by the Institutional Review Board (IRB) of Severance hospital (approval number: 4-2023-0442). All methods were performed in accordance with the relevant guidelines and regulations of the IRB. All participants provided written informed consent for participation in the study.

From January 2011 to May 2017, patients who received TACE or TARE as a first-line treatment for single large hepatocellular carcinoma (>5 cm) at Severance hospital will be reviewed retrospectively. We excluded subjects who had any of the following: (1) Child-Pugh score over seven; (2) patients with infiltrative tumor; (3) if TACE or TARE was used as an adjunctive therapy other than the primary treatment, or as a follow-up treatment following TACE or TARE during the same hospital stay. Patient groups are classified according to the treatment method for the patients collected within the above-mentioned guided range.

We compare the baseline characteristics of each patient group with the current history, laboratory test results, liver function using Child-Pugh score. Tumor related factor was also compared including size, presence of portal vein invasion or satellite nodule, and tumor markers. Tumor markers included alpha-fetoprotein (AFP) and protein induced by vitamin K absence or antagonist-II (PIVKA-II).

As part of the treatment effect analysis in this study, the proportion of patients who underwent resection or LT because curative surgery was possible after the procedure was analyzed for each group.

In addition, we analyzed the treatment response through image work-up 3 months after the procedure to analyze the treatment effect. Treatment response was analyzed using modified Response Evaluation Criteria in Solid Tumors (mRECIST).[12] Complete response (CR), partial response (PR), and stable disease (SD), excluding progressive disease (PD), were defined as disease control and the ratio was calculated separately. If the patient underwent curative resection before image work-up at 3 months or had a follow-up loss, it was excluded from the treatment response analysis.

Survival analysis was performed as the last method to analyze the treatment effect. The survival rate of each patient group is analyzed using Kaplan-Meier curve.[13] Progression free survival and time to secondary therapy will be also calculated by calculating the duration until progression and the second treatment. If the patient has a follow-up loss at Severance, use health insurance data to check for survival.

Nursing records are used to compare the frequency of fever, nausea, vomiting, and abdominal pain during hospitalization to compare adverse effects between treatment groups. Fever was defined as the case where the tympanic membrane temperature was above 37.8 degrees, this is because the nurses notify the doctors from this point on in the institution. Abdominal pain, nausea and vomiting were determined based on patients with Grade 2 or higher of the Common terminology criteria for adverse events (CTCAE) Version 5.0.[14] Not only side effects but also length of hospitalization was calculated and compared for each group.

Additionally, to compare the therapeutic effect as well as the treatment stability, we checked the liver function test day after procedure, three weeks after the procedure and three months after the procedure. At three months, liver function tests as well as Child-Pugh scores were separately calculated. And the difference between the two groups was compared.

All statistical analyses were performed using SPSS version 27 for Windows (SPSS Inc., Chicago, Illinois, USA). Differences in categorical variables between the groups were analyzed using the chi-square test or Fisher’s exact test. Continuous variables were compared using Student’s t-test.[15] Survival statistics were evaluated using Kaplan-Meier method.[16] Statistical significance was set at two-sided p < 0.05.

Although there have been studies comparing TARE with other treatment modalities or comparing TARE with TACE in various type of HCC, this study is the first to compare TACE and head-to-head in single large HCC. We could not demonstrate significant difference in overall survival or progression free survival between treatment modalities. However, although there are limitations due to the small number of subjects, we confirmed that the TARE group showed a superior disease control rate compared to TACE at three months after treatment. And we also proved that TARE was superior in terms of side effects and length of hospitalization despite having the same baseline characteristics.

We hypothesized that TARE would be superior to TACE in terms of efficacy and safety in single large HCC. This is because in previous studies, TARE showed comparable outcomes to resection for single large HCC, and TARE is known to be more effective and safer than TACE in several types of HCC.[5, 9, 17-19] In our results, TARE showed a better treatment response at three months and was also superior in terms of side effects. However, this has not been demonstrated in OS and PFS, which can be considered for the following reasons. As will be mentioned later, it is possible that this is due to the small number of subjects, which is the biggest drawback of this study.[20] However, it is unreasonable to simply apply that reason because there were significant results in the treatment response and side effects at 3 months. One possibility is that there was no benefit to OS as a result of TARE-induced long-term toxicity and complications.[21, 22] And since the response by mRECIST is based on the radiologic image, it is only an indirect reflection of the tumor response, and therefore the behavior of the actual tumor may be different.[12, 23] However, these are only hypotheses and we are not sure as there is no head-to-head comparison with TACE.

Although excellence in survival was not demonstrated, our study is meaningful in the following aspects. A high disease control rate over a specific period of time reduces the need for other treatments. In addition, there are few side effects after the procedure, and the average hospitalization length is shortened by about two days. Although medical costs vary from country to country, these results suggest that TARE has an advantage in terms of medical costs.[24, 25]

Although there are positive results, this study has many limitations. First of all, since this study was conducted at one institution, it is highly likely that there would be various biases such as the type of patient group or the level of procedure.[26] Second, the number of subjects was very small, so some results were different between the two groups, but statistical significance was not obtained with a slight difference.[20] And for some results, the possibility that the p-value was exaggerated because the result of one group was measured as 0 cannot be excluded.[20, 26] Third, this study was a retrospective design, and although there was little difference in baseline characteristics between the two groups, the level of evidence cannot be considered high.[27] Lastly, the subjects of this study were patients from about 12 years ago at the longest and about six years ago at the shortest time, and patients whose treatment time has elapsed considerably, and may show different treatment tendencies from the present.

Therefore, because of these limitations, it is thought that the following additional efforts are needed to reconsider the reliability of this study. A large number of subjects must be collected to increase the reliability of research results and to obtain more results[20]. And Multicenter studies are needed to reduce bias. In addition, a randomized controlled study is necessary to increase the level of evidence[27], but it is difficult in reality. This is because there are studies that, although not limited to single large HCC, suggest that TARE is superior to TACE, and the BCLC guideline revised in 2022 recommends TARE treatment for single large HCC less than 8 cm.[1] Therefore, if various results are collected and meta-analysis is performed, the level of evidence can be increased.[27]