The efficacy of radiofrequency ablation versus cryoablation in the treatment of single hepatocellular carcinoma: A population‐based study

Abstract Background Radiofrequency ablation (RFA) is an effective treatment for single hepatocellular carcinoma (HCC), but it is difficult to use against tumors in some locations and often leads to incomplete ablation as a result of the heat‐sink effect. This study was conducted to evaluate the efficacy of cryoablation compared with that of RFA in the treatment of single HCC. Methods This retrospective study was conducted based on the Surveillance, Epidemiology, and End Results (SEER) database. From 2004 to 2015, patients aged 40 to 79 diagnosed with HCC were included in the study. A propensity score matching (PSM) model was used to reduce selection biases. Results Before PSM, the median overall survival (mOS) and median cancer‐specific survival (mCSS) in the RFA group were slightly longer than those in the cryoablation group (p > 0.05). In the subgroup analysis, the mOS and mCSS of patients with tumor sizes <3, 3–5, and >5 cm who received RFA treatment were longer than those of patients given cryoablation treatment, but there was no significant difference (p > 0.05). Similar results were presented in patients at American Joint Committee on Cancer (AJCC) stages I and II. After PSM, the mOS and mCSS were slightly better in the RFA group than the cryoablation group but without significant differences. Univariate and multivariate analysis showed that cryoablation treatment was not an unfavorable factor for OS and CSS before or after PSM (p > 0.05). In the multivariable competing risk model, non‐cancer‐specific death was taken as a competing factor and cryoablation was also not unfavorable for the survival of patients before and after PSM (p > 0.05). Conclusion Cryoablation is non‐inferior to RFA therapy for single HCC patients without lymph node invasion or distant metastasis.


| INTRODUCTION
Hepatocellular carcinoma (HCC) is the sixth most common cancer in the world and is the fourth leading cause of cancer-related deaths. 1 The incidence of HCC in the USA has increased in recent decades because of the increase in hepatitis C infections and non-alcoholic cirrhosis. 2 For early HCC, the European Association for the Study of the Liver recommends the use of transplantation, liver resection, or ablation as first-line treatments. 3 However, most patients with early HCC are not suitable candidates for transplantation because of a lack of available organs and Milan criteria restrictions. Liver resection can prolong the survival times of patients, but the associated complications are complex, and the 5-year tumor recurrence rate with this treatment can reach 70%. 4 Thus, minimally invasive techniques that can destroy tumors are more widely used. Radiofrequency ablation (RFA) is the most commonly used ablation technique and it can achieve a similar effect to liver resection in the treatment of HCC (with tumor diameters of less than 2 cm). 5,6 For tumors less than 3 cm in diameter, the 3-year overall survival rates of HCC patients range from 67% to 84%. 7 However, in the treatment of larger tumors, RFA often results in incomplete ablation because of the heat-sink effect (especially for tumors located near major blood vessels), resulting in increased tumor recurrence.
Recently, cryoablation has been more frequently used to treat solid tumors and has shown good efficacy. [8][9][10][11] It can destroy tumor cells by forming intracellular ice and extracellular crystals, which damage cell membranes and intracellular organelles, leading to tumor cell death. Several studies have shown that cryoablation can destroy tumor cells and induce the release of their intracellular contents into the extracellular space, which can trigger the immune response, killing residual tumor cells and reducing tumor recurrence. 12,13 During cryoablation, an ice ball forms around the tip of the probe within tumor tissue, and the ice can be precisely and intraprocedurally monitored via various imaging techniques such as ultrasound, magnetic resonance imaging, or computed tomography, facilitating more precise tumor treatment.
Nowadays, some scholars believe that patients with single HCC tumors that neither invade vessels nor metastasize to distant organs and have preserved liver function should be defined as early stage and be given radical treatment. 3,[14][15][16] Interest in applying cryoablation to the treatment of HCC is growing. However, it is unclear whether patients with single HCC tumors will benefit more from cryoablation or RFA. Therefore, we conducted a population-based study to compare the efficacy of cryoablation versus RFA in the treatment of single HCC tumors.

| Patients
The data used in this study were extracted from the Surveillance, Epidemiology, and End Results (SEER)-18 registries database, which includes approximately 28% of the population of the United States. The college ethics committee approved the retrospective study. The requirement to obtain informed consent was waived by the institutional review board, as it was a population-based study.
Patients with American Joint Committee on Cancer (AJCC) stages N1 or NX and M1 or MX were excluded. Patients with a survival month code of 0 were excluded because this indicated that the patient was lost to follow-up after diagnosis. Finally, 3614 patients were enrolled, of which 104 were treated with cryoablation and 3510 were treated with RFA ( Figure S1).

| Outcomes and statistical analysis
The endpoints of the study were overall survival (OS) and cancer special survival (CSS). OS was defined as the interval from the diagnosis of HCC to the death of the patient. CSS was defined as the interval from the diagnosis of HCC to death caused by HCC of the patient.
The data were extracted using SEER*Sat software (version 8.3.6). All continuous variables were transformed into categorical variables. Chi-squared test or Fisher's exact test were used to compare the differences in the baseline characteristics and the characteristics after PSM between the RFA group and cryoablation group. The OS and CSS were plotted on Kaplan-Meier curves and compared with the log-rank test. The Cox proportional hazard model was used to analyze the predictors for OS and CSS before and after PSM. Variables with p-values of less than 0.1 in univariate analysis were included in multivariate analysis. To compete the risk analysis, non-cancer-specific death was taken as the competing factor, and the Gray's test was used to identify statistical differences between the two groups. The Fine-Gray multivariable regression model was used to identify factors associated with the risk of survival of patients treated with RFA or cryoablation.
To reduce selection bias and the influence of potential confounding factors, PSM was used to generate better matching groups. The characteristics of age, gender, tumor stage, AJCC stage, size, marital status, and chemotherapy treatment were included in the PSM analysis. The optimal caliper was set as 0.1 and 498 patients were matched by the one to five nearest neighbor approach. Among these patients, 104 received cryoablation treatment and 394 received RFA treatment. p-values < 0.05 were considered statistically significant, except for in the univariate analysis and all statistical analyses were two-tailed. Analyses were conducted with SPSS (version 24.0) and Stata (version 14.0).

| Baseline characteristics
A total of 3614 patients were included in the study, of which 2769 patients were male, and 845 patients were female. Among them, 104 patients received cryoablation treatment and 3510 patients received RFA treatment. Before PSM, the patients in the cryoablation group were older than those in the RFA group (p = 0.003), the year of diagnosis in the cryoablation group was earlier than that in the RFA group (p = 0.002), and there were more white patients in the cryoablation group than the RFA group (p = 0.02) ( Table 1). After PSM, there were no significant differences in the characteristics between the two groups (All p > 0.05) ( Table 1).

| DISCUSSION
A hepatectomy is recommended for patients with single HCC without invasion of vessels, lymph nodes, or metastasis to distant organs. However, some patients cannot receive a resection because of their poor physical condition, highrisk status, or vulnerability to complications. 17,18 Minimally invasive treatments are used for these patients and can achieve satisfactory efficacies, especially RFA. 19 Substantial evidence shows that patients with single HCC receive survival benefits from RFA alone or in combination with other treatments. 20,21 However, the disadvantage of using RFA is that it is difficult to apply when the HCC is located under the diaphragm or is adjacent to other important organs. In these cases, cryoablation is recommended, but insufficient data limit the application of cryoablation in the treatment of HCC. In a high-quality study comparing the efficacies of cryoablation and RFA in the treatment of HCC patients with cirrhosis, tumor diameters of ≤4 cm and no more than two tumors, the rates of local tumor progression at 1, 2, and 3 years were higher in the cryoablation group than the RFA group (p = 0.041); however, there were no significant differences in the overall survival rates at 1, 3, or 5 years between T A B L E 1 Characteristics of patients between the two groups before and after PSM the two groups (p = 0.747). 22 Although, previous studies have shown that patients with unresectable liver cancer, local HCC, tumors larger than 2 cm and one or two tumors of no more than 4 cm who received cryoablation had similar survival benefits compared with those who received RFA. [22][23][24][25] The EASL guidelines classify patients with HCC more finely and recommends different treatments for different HCC patients. However, no studies have compared the efficacy of cryoablation with RFA for patients with single HCC not invading lymph nodes or metastasizing to distant organs and especially those patients with single HCC, early stage cancer, or with different tumor sizes. Thus, we conducted the current study to compare the efficacies of cryoablation and RFA in the treatment of a single HCC using the SEER database. Previous studies have shown HCC patients to have an OS range of 8 to 29 months when treated with cryoablation alone or in combination with other therapies. 24,26,27 The mOS in this study was longer than that in the previous study, mainly because the previous study recruited patients with larger tumor diameters and more tumors. Although, patients with single HCC were included in this study, the efficacy of cryoablation was found to be similar to that of RFA, which is consistent with previous studies that recruited patients with different stages of HCC and compared cryoablation with RFA efficacies. Prospective and retrospective studies have shown that patients with small HCC tumors received more survival benefits compared with those provided other treatments. Therefore, in this study, patients were divided into four subgroups according to tumor size and AJCC stage. The results of this study indicate that patients received RFA and cryoablation had similar mOS and mCSS as patients with tumors <3, 3-5, and >5 cm in diameter, as well as patients at AJCC stages I and II. RFA alone was not recommended as a first-line treatment for patients with single tumors larger than 5 cm, although, the results of cryoablation and RFA were similar in this study. Many studies have shown that patients with HCC tumors of >5 cm receive more survival benefits from RFA combined with other treatments. 21,28 Thus, for patients with single HCC tumors of >5 cm in diameter, the efficacy of cryoablation combined with other treatments compared with that of RFA combined with other treatments  Previous research has shown that tumor burden and liver function are independent predictors for patients with HCC who receive radical treatment (resection or ablation). 29,30 In this study, the patients' liver function was not included in the Cox proportional risk model, as the SEER database does not contain this information; however, patients' age, gender, year less advanced surgical techniques and/or treatment methods. Patients older than 60 years had a higher risk of comorbidities than younger patients. Thus, non-cancer-specific death was taken as a competing factor for comparing the efficacies between the RFA and cryoablation groups. However, cryoablation still did not show an increased mortality risk compared with RFA, indicating that patients with single HCCs receive similar survival benefits from cryoablation as from RFA. Although, the findings of the current study might provide evidence for clinicians when choosing the most suitable treatments for patients with single HCC who are not candidates for RFA, there are many unresolved problems in the cryoablation treatment of patients with HCC. Recently, studies have shown that HCC patients with single tumors up to 3 cm who receive RFA could experience similar survival benefits compared with those who receive liver resection, which led to guidelines recommending RFA as the first-line treatment for patients with HCC of no more than 3 cm. 3,19,20 However, whether similar survival benefits are bestowed on patients receiving cryoablation as patients receiving liver resection is still unknown and has not been the focus of any previous studies. Thus, future studies should aim to include these patients. This study had some limitations. First, this was a retrospective study, which inevitably produces selective bias, although, we used PSM to reduce this. Second, this was a population-based study, the central pathological review and the data on earlier tumor control endpoints, such as progression-free survival and time to progression, as well as information about complications, could not be obtained and evaluated, which may affect the accuracy of the results. However, previous studies have shown no statistical differences in the incidences of complications in patients that underwent cryoablation compared with those provided RFA. Nevertheless, a prospective randomized trial is needed to confirm the results of this study.

| CONCLUSION
This population-based study demonstrated that treatment with cryoablation provides similar survival benefits to patients with single HCC tumors treatment compared with RFA. Therefore, if patients with single HCCs are unsuitable for RFA or unwilling to receive RFA, cryoablation may be a selective choice.

ETHICAL APPROVAL AND CONSENT TO PARTICIPATE
The research do not need to be reviewed by the ethics committee because the data were from SEER database and the written informed consent was exempted. However, the data used in the research was permitted by the SEER database management department.