Liver Resection Combined with Radiotherapy versus Liver Resection Alone in the Treatment of Hepatocellular Carcinoma: A Population-based Study

Radiotherapy has been used in the treatment of hepatocellular carcinoma (HCC) more widely. However, little research focus on comparing the ecacy of patients with liver resection combined with radiotherapy with that received liver resection alone. The study was conducted to evaluate whether the ecacy of liver resection combined with radiotherapy in the treatment of patients with HCC is better than liver resection alone. The study utilized the data from the Surveillance, Epidemiology, and End Results 18 registry (SEER-18). Patients diagnosed with HCC between 2004 and 2015 who received liver resection or the combination of liver resection and radiotherapy were included in the analysis. The propensity score matching model (PSM) was used to reduce selection bias and potential confounding factors. risk


Introduction
Hepatocellular carcinoma (HCC) is one of the most malignant cancers and the second most lethal cancer worldwide [1]. In the United States, the number of HCC cases and deaths has increased dramatically in recent decades [2,3]. The European Association for the Study of the Liver (EASL) guidelines recommend radical treatments, such as liver resection, liver transplantation, and ablation, as the most effective treatment for patients with early HCC [1]. For patients with a single tumor no larger than 3 cm, the e cacy of radiofrequency ablation (RFA) is similar to that of liver resection. However, but for patients with tumors larger than 3 cm, liver resection provides a better outcome than RFA because of the limited range of ablation and the heat sink effect of RFA [4,5]. Liver resection might in ict more damage to the patients than RFA, limiting its use in patients with a single tumor smaller than 3 cm. However, liver resection might be used more widely in the treatment of HCC. Recent studies documented that patients with intermediate-stage HCC had better overall survival after with liver resection than after transarterial chemoembolization (TACE), which used to be recommended as the rst-line treatment for intermediate HCC patients [6][7][8]. Although liver resection and RFA can both prolong the overall survival of patients with HCC, the 5-years tumor recurrence rate in patients subjected to these therapies remains high, leading to the death of patients in the short term [9,10]. The combination of liver resection and other treatments might represent a good choice for HCC patients.
tumors as patients treated with RFA, but provided an advantage in case of tumors larger than 2 cm [11]. Another study compared the e cacy of stereotactic ablative radiotherapy with liver resection in HCC patients with small tumors and found that both modalities resulted in similar progression-free survival (PFS) and overall survival (OS) [12]. Therefore, in the future, radiotherapy might be used more widely in HCC treatment.
Radiotherapy includes internal radiation therapy and external radiation therapy. Both approaches are used in HCC treatment and ensure adequate e cacy [13,14]. Typically, radiotherapy is used as an adjuvant treatment for HCC and combined with other treatment protocols. A previous study indicated that radiotherapy combined with TACE prolongs the OS of HCC more than TACE alone [15]. A randomized controlled trial found that the e cacy of radiotherapy combined with liver resection in small HCC tumors that invaded portal vein, de ned as advanced HCC, was better than that of liver resection alone [16]. However, whether patients with different stage HCC can get survival bene t from the combined treatment is unknown. Therefore, we compared the e cacy of radiotherapy combined with liver resection and liver resection alone using a populationbased cancer registry.

Methods And Materials
This retrospective study was conducted using the Surveillance, Epidemiology, and End Results (SEER 18) database, which covers approximately 28% of the United States population and includes cancer incidence, demographics, the rst course of treatment, and mortality from the time of diagnosis.
The study included patients diagnosed with HCC (International Classi cation of Disease for Oncology, Third Edition (ICD-O-3), histology code 8170/3, 8171/3, 8172/3, 8173/3, 8174/3 and 8175/3, site code C220) between 2004 and 2015. The patients that were younger than 85 years or older than 45 years at the time of diagnosis were excluded. The patients whose exact information (summary stage, AJCC 6 th edition stage, whether they received surgical treatment, tumor size, length of survival) was unknown or unclear were excluded. Together, 5187 patients were included in the analysis; among them, 139 patients were treated by a combination of liver resection and radiotherapy, and 5048 patients received liver resection alone ( Figure 1).
The study was approved by the SEER program managers and the Institutional Review Board.

Variables and outcomes
The characteristics of patients included gender, ethnicity, marital status, age at HCC diagnosis, year of HCC diagnosis, tumor stage according to the AJCC 6 th edition, tumor size, number of tumors, administration of chemotherapy. The endpoints of the analysis were overall survival (OS) and cancer-speci c survival (CSS). OS was de ned as from the time of HCC diagnosis to the death of patients. CSS was de ned as from the time of HCC diagnosis to death caused by cancer.

Statistical analysis
The data used in this study was extracted by the SEER*Stat software (version 8.3.6). The characteristics of patients treated with liver resection combined with radiotherapy and treated with liver resection only were compared by Chi-square test or Fisher's exact test before and after propensity score matching (PSM). Kaplan-Meier curves for OS and CSS were plotted and compared using the log-rank test. A univariate and multivariate logistic regression model was used to analyze the predictors for OS and CSS. The predictors with the P-value of less than 0.1 in univariate analysis were included in multivariate analysis. PSM was conducted to reduced confounding factors and selection biases. PSM involved gender, ethnicity, marital status, age at diagnosis, year of diagnosis, summary stage, AJCC stage, tumor size, number of tumors, and administration of chemotherapy. The optimal caliper was set as 0.01, and 568 patients were generated using the 1:4 nearest neighbor approach. The P-value of less than 0.05 was considered statistically signi cant, and all statistical analyses were two-tailed. SPSS 25.0 (IBM, Chicago, IL, USA) and GraphPad Prism 8.0 (GraphPad Software, San Diego, CA) were used to perform statistical calculations.

Patients
A total of 5187 patients were included in this study before PSM; 3736 were male, and 1451 were female. Among them, 5048 patients were treated with liver resection alone (liver resection alone group), and 139 patients were treated with liver resection combined with radiotherapy (combined treatment group). There was a statistically signi cant difference in age at diagnosis, tumor stage, AJCC stage, tumor size, ethnicity, and administration of chemotherapy between the two groups before PSM (P<0.05). No statistically signi cant differences of characteristics between the two groups were present after PSM (P>0.05) ( Table 1). After PSM, higher AJCC stage and black race were independent predictors of shorter OS. Earlier years of HCC diagnosis, more advanced AJCC stage, and black race were independent predictors of shorter CSS (P<0.05)

Discussion
Published guidelines do not recommend radiotherapy as the standard treatment for any stage HCC, which limited its usage for this type of tumor 1 . Several decades ago, external radiotherapy was not recommended to treat patients with HCC because the radiation could not be precisely positioned according to tumor location, size, and shape, which might lead to the damage of normal liver tissue and surrounding vital organs, and liver failure. However, modern imaging techniques, including the emergent internal radiotherapy and stereotactic body radiotherapy (SBRT), as well as the progress in understanding the tolerance of liver parenchyma to radiation, led to the wide use of radiotherapy in the treatment of HCC as an adjuvant therapy [17,18].
Several recent studies documented that HCC patients receiving radiotherapy combined with other treatments had longer survival than those treated with radiotherapy alone [19][20][21]. A meta-analysis that compared the e cacy of TACE combined with radiotherapy with TACE alone treatment in patients with portal vein tumor thrombus (PVTT) demonstrated that the combined treatment yielded a higher objective response rate (ORR) and better survival bene ts than the TACE alone treatment [22]. Wei and coworkers conducted a randomized clinical trial to compare the e cacy of hepatectomy combined with neoadjuvant radiotherapy with that of hepatectomy alone in patients with resectable HCC and portal vein tumor thrombus. The trial demonstrated that the 6, 12, 18, and 24 months OS and disease-free survival was signi cantly higher in the combined treatment group (P < 0.001) [16]. The present investigation compared the e cacy of the combination of liver resection and radiotherapy as the treatment for HCC with liver resection alone. Before PSM, the mOS and mCSS in the liver resection alone group were longer than in the combined treatment group (P < 0.05). However, the results documented that patients subjected to the combined treatment did not have better OS and CSS than patients treated with liver resection alone, after PSM.
Liver resection is recommended as the rst-line treatment for early HCC by the EASL guideline, and it can increase the 1-, 3-, and 5-years survival 1 . However, liver resection can result in signi cant damage. Patients with HCC often present poor liver function and physical condition due to cirrhosis, which is one of the main reasons why liver resection is not widely used in HCC treatment. Although liver damage induced by radiotherapy has been markedly reduced, radiation-induced hepatotoxicity might trigger more extensive liver injury when combined with liver resection, leading to the death of patients due to liver failure [23].
A recent study by Lewandowski and collaborators concluded that radiotherapy could extend the survival of patients with early HCC. The results indicated that 90% of patients showed tumor response according to the EASL criteria, of which 59% showed a complete response with a median time to progression of 2.4 years and mOS of 6.7 years [24]. Another analysis utilizing PSM compared the e cacy of radiotherapy and RFA in HCC patients with no more than three tumors, 3 cm or smaller in size. The results showed that the 3-year local recurrence rate in the radiotherapy group was lower than in the RFA group (P < 0.01), and the 3-year survival rate in the radiotherapy group was similar as in the with RFA group (P = 0.86). These ndings indicate that in comparison with RFA, radiotherapy ensures superior local control and comparable OS [25]. However, in the current study, mOS and mCSS of patients with AJCC stage I and II and with single tumor not larger than 5 cm, was comparable between the liver resection alone group and the combined treatment group. These results support the conclusion that radiotherapy may not provide additional survival bene ts in HCC patients who received liver resection.
Although the mOS and mCSS in the liver resection alone group were longer than in the combined treatment group, in multivariate logistic regression analysis, the combination of liver resection and radiotherapy was an independent factor negatively affecting OS (P = 0.033), but not CSS (P = 0.216) before PSM. After PSM, the combined treatment did not predict the outcomes of patients.
Some limitations of the current study should be acknowledged. First, this research was designed as a retrospective analysis, and the selection bias was inevitable. However, the bias was reduced by employing the PSM model. Second, this study was population-based, liver function and physical condition of the patients were not consistently from the SEER data, and these factors might affect the patient outcomes. However, in contrast to previously published research, the present study included a large number of subjects, which strengthens the reliability of the results. Future randomized clinical trials are needed to con rm the results reported here.
Chuansheng Zheng and Fan Yang designed the work. Lei Chen acquired the data from the SEER database. Lei Chen and Xiaopeng Guo, and Shi Chen analyzed the data. Lei Chen, Shi Chen, Yanqiao Ren, and Tao Sun wrote the manuscript. Chuansheng Zheng, Fan Yang, and Xiaopeng Guo reviewed the manuscript. All authors approved the version to be published.

Figure 1
Flowchart of patients selection