Efficacy and Dose-Response Relationship of Stereotactic Body Radiotherapy for Abdominal Lymph Node Metastases from Hepatocellular Carcinoma

Abstract

Background

The objective of this study was to investigate the treatment efficacy of stereotactic body radiotherapy (SBRT) and evaluate the influence of radiation dose on local control and survival in patients with abdominal lymph node metastases (LNM) from hepatocellular carcinoma (HCC).

Patients and methods

Between 2010 and 2020, data of 148 patients with HCC with abdominal LNM, including 114 who underwent SBRT and 34 who received conventional fractionation radiation therapy (CFRT), were collected. A total radiation dose of 28-60 Gy was delivered in 3-30 fractions, with a median biologic effective dose (BED) of 60 Gy (range, 39-105 Gy). Freedom from local progression (FFLP) and overall survival (OS) rates were analyzed.

Results

With a median follow-up of 13.6 months (range, 0.4-96.0 months), the 2-year FFLP and OS rates of the entire cohort were 70.6% and 49.7%, respectively. Median OS of the SBRT group was longer than the CFRT group (29.7 vs. 9.9 months, P = .007). A dose-response relationship was observed between local control and BED in either the entire cohort or the SBRT subgroup. Patients who received SBRT with a BED ≥60 Gy had significantly higher 2-year FFLP and OS rates than those who received a BED <60 Gy (80.1% vs. 63.4%, P = .004; 68.3% vs. 33.0%, P < .001). On multivariate analysis, BED was an independent prognostic factor for both FFLP and OS.

Conclusions

SBRT achieved satisfactory local control and survival with feasible toxicities in patients with HCC with abdominal LNM. Moreover, the findings of this large series suggest a dose-response relationship between local control and BED.

This article reports on treatment efficacy of stereotactic body radiotherapy and evaluates the influence of radiation dose on local control and survival in patients with abdominal lymph node metastases from hepatocellular carcinoma.

Implications for Practice.

For patients with abdominal lymph node metastases (LNM) from hepatocellular carcinoma (HCC), stereotactic body radiotherapy (SBRT) could achieve satisfactory local control and survival with non-significant treatment-related toxicities. Moreover, there is a dose-response relationship between biologic effective dose (BED) and local control, suggesting SBRT with BED ≥60 Gy is a promising therapeutic strategy for these patients.

Introduction

Hepatocellular carcinoma (HCC) is the sixth most common type of cancer and the third leading cause of cancer-related death worldwide.¹ The incidence of abdominal lymph node metastases (LNM) in patients with HCC is 4% to 11%, while that reported in autopsy studies is 25%-42%.² According to a Japanese nationwide survey, patients with node-positive HCC have similar prognosis to those with locally advanced HCC; that is, they have a worse survival and a higher recurrence risk compared with those without LNM.^3,4

The optimal treatment for abdominal LNM from HCC has not yet been established. Regional lymphadenectomy had no impact on the survival of patients who underwent liver resection or liver transplantation, and only selective lymphadenectomy of single LNM improved the patients’ prognosis.^2,4,5 Although transcatheter arterial chemoembolization and percutaneous ablation have been used to treat LNM in the specific region of the abdomen, severe complications may occur, and their efficacy remains uncertain.^6,7 Targeted therapy is recommended as the standard treatment for patients with HCC with extrahepatic metastases, such as sorafenib. However, sorafenib has not shown survival benefits in subgroups of patients with extrahepatic metastases, with an objective response rate (ORR) of <5%.^8,9

Several studies have been conducted to examine the effectiveness of traditional radiotherapy in patients with HCC with abdominal LNM. Specifically, the response rates were reported to be 56.7%-86.7%; the 1- and 2-year overall survival (OS) rates were 41.0% and 19.9%, respectively, with a median survival time of 5.8-19.0 months; this result suggests that external beam radiotherapy is an effective palliative treatment and may prolong the survival of patients with HCC with LNM.^10-12 In majority of patients with HCC, the abdominal LNs are adjacent to the gastrointestinal tract; however, previous studies mostly used 2-dimensional or 3-dimensional conformal radiotherapy (3DCRT), which provided limited protection to the surrounding normal organs and increased the incidence of gastrointestinal complications. The incidence rates of radiotherapy-induced gastroduodenal ulcer and gastrointestinal bleeding were as high as 22%-30.7% and 20%-22%, respectively.^10-13

Stereotactic body radiation therapy (SBRT) is an effective treatment option for patients with HCC, providing high local control and minimal treatment-related toxicity.^14,15 In addition, a dose-response relationship was observed between local control and biologically effective dose (BED) of SBRT in patients with HCC with intrahepatic lesions.¹⁶ However, the evidence supporting the efficacy and safety of SBRT for abdominal LNM from HCC is limited. Therefore, this retrospective study aimed to investigate the treatment efficacy of SBRT and evaluate the influence of radiation dose on local control and survival in patients with HCC with abdominal LNM.

Patients and Methods

Patient Selection

The data of consecutive patients with HCC with abdominal LNM who underwent radiotherapy in our institution between March 2010 and August 2020 were obtained from the prospectively maintained database and retrospectively analyzed. (1) Patients aged >18 years who were primarily diagnosed with HCC based on the biopsy results or the noninvasive criteria of the European Association for the Study of Liver guidelines,¹⁷ (2) whose abdominal LNM was diagnosed by contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI), (3) with a Child-Pugh classification of A or B, and (4) with an Eastern Cooperative Oncology Group (ECOG) performance status score of ≤2 were included in the study. Combination treatment with targeted and immunotherapy was defined as the use of targeted therapy or immunotherapy within 2 months before and/or after radiotherapy. This study was approved by the Ethics Committee, and the requirement for obtaining informed consent was waived due to the retrospective nature of the study.

Radiation Treatment

During CT simulation, the patients were placed in a supine position with arms raised above the head and immobilized with a vacuum bag. Contrast-enhanced 4-dimensional CT (4DCT) scans were acquired during quiet breathing with a 2.5-3.0-mm slice thickness on a 16-slice positron emission tomography/CT scanner (General Electric Medical Systems, Waukesha, WI).

Gross tumor volume (GTV) was defined as positive abdominal LNMs visualized on CT/MRI images. For patients with uncontrolled intrahepatic tumor and/or tumor thrombosis, the intrahepatic lesion and tumor thrombosis were also included as GTV. Internal target volume (ITV) was defined as the combined GTVs on all phases of the 4DCT scan. Planning target volume (PTV) was constructed by adding the ITV and the expansions of 0.5-0.7 cm to account for the interfractional motion variability and daily setup errors. SBRT was defined as a technique that utilizes a radiation dose of ≥5 Gy/fraction, and the total number of fractions is not more than 8. Conventional fractionation radiation therapy (CFRT) was defined as a technique that utilizes a radiation dose of <5 Gy/fraction. SBRT was delivered using volumetric modulated arc therapy technique (VMAT), with one or 2 arcs using 6-MV beams. CFRT was delivered via intensity-modulated radiotherapy (IMRT) or 3DCRT. To compare the radiation doses among patients treated with different fractionation schemes, the α/β ratio of 10 was used to calculate the BED. The dose-fractionation schedules are provided in Supplementary Table S1. The organs at risk (OARs) included normal liver, kidneys, stomach, duodenum, small intestine, colon, and spinal cord. Details regarding the dose constraints for the OARs are shown in Supplementary Table S2.

Follow-up

Patients were evaluated weekly for toxicities during the period of radiotherapy and every 3 months after radiotherapy for the first 2 years and at least every 6 months thereafter. Freedom from local progression (FFLP) was defined as the time from the initiation of radiotherapy to the date of in-field recurrence, which was evaluated using contrast-enhanced CT or MRI based on the consensus between radiation oncologist and radiologist. The treatment response of LNM was assessed using the Response Evaluation Criteria in Solid Tumors version 1.1.¹⁸ ORR was defined as complete response rate plus partial response rate. Treatment-related acute toxicities were defined as toxicities occurring within 3 months after radiotherapy, while late treatment-related toxicities were defined as toxicities occurring >3 months after radiotherapy. The toxicities were evaluated according to the Common Terminology Criteria for Adverse Events version 5.0.

Statistical Analysis

The Kaplan-Meier method was used to analyze the FFLP and OS rates, the Log-rank test was performed to compare the differences between the study groups, and a Cox proportional hazard model was used to perform the univariate and multivariate analyses. Variables with a P-value of <.05 in the univariate analysis were included in the multivariate analysis. Logistic regression analysis was used to analyze the dose-response relationship between local control and BED. The chi-square test or Fisher’s exact test was used to compare the tumor response rates and toxicities in different groups. All statistical analyses were performed using the SPSS software package (version 26.0; SPSS Inc, Chicago, IL). All tests were 2 sided, and a P-value of <.05 was considered significant.

Results

Patient and Treatment Characteristics

A total of 148 patients with HCC with LNM receiving radiotherapy were included in this study. The baseline patient and tumor characteristics are listed in Table 1. Approximately 81.8% of the patients were male, and the median age at LNM diagnosis was 52 years (range, 18-77 years). Majority of the patients (97.3%) had a compensated liver function with Child-Pugh class A. With regard to the number of abdominal LNM, 111 patients (75.0%) had multiple LNM, while only 37 (25.0%) patients had solitary LNM. In terms of systemic treatment, 34 patients (23.0%) received targeted therapy, 6 (4.1%) received immunotherapy, and 16 (10.8%) received both treatments.

Table 1.

Patient and tumor characteristics (n = 148).

Characteristic	No. of patients	%
Sex
Male	121	81.8
Female	27	18.2
Age (years)
<52	70	47.3
≥52	78	52.7
ECOG performance status
0	114	77
1-2	34	23
Child-Pugh classification
A	144	97.3
B	4	2.7
History of hepatitis
HBV/HCV	116	78.4
No	32	21.6
Serum AFP level before radiotherapy (ng/mL)
<400	95	64.2
≥400	44	29.7
PIVKA-II level before RT (mAU/mL)
<40	37	25.0
≥40	54	36.5
Previous local therapy for intrahepatic tumors
Hepatectomy ± TACE/RFA	92	62.2
TACE	36	24.3
RFA ± TACE	14	9.5
Systemic therapy combined with radiotherapy
Targeted therapy	34	23.0
Immunotherapy	6	4.1
Targeted therapy + immunotherapy	16	10.8
None	92	62.2
Intrahepatic tumors
Uncontrolled	70	47.3
Controlled	78	52.7
Tumor thrombosis
Present	25	16.9
Absent	123	83.1
Distant metastasis
No	122	82.4
Yes	26	17.6
Other lesions except abdominal LNM
Absent	62	41.9
Present	86	58.1
LN size (cm)
<4	110	74.3
≥4	38	25.7
Abdominal pain related to LNM
Yes	14	9.5
No	134	90.5
Radiotherapy technique
3DCRT	15	10.1
VMRT/IMRT	133	89.9
Radiation modality
CFRT	34	23.0
SBRT	114	77.0
Dose/fx (Gy)
<8	112	75.7
≥8	36	24.3
BED (Gy)
<60	67	45.3
≥60	81	54.7
Treatment era
2010-2015	46	31.1
2015-2020	102	68.9

Abbreviations: ECOG, Eastern Cooperative Oncology Group; HBV, hepatitis B virus; HCV, hepatitis C virus; AFP, alpha-fetoprotein; RT, radiotherapy; TACE, transcatheter arterial chemoembolization; RFA, radiofrequency ablation; LNM, lymph node metastasis; 3DCRT, 3-dimensional conformal radiation therapy; VMAT, volume modulated radiotherapy; IMRT, intensity-modulated radiation therapy; CFRT, conventional fractionation radiation therapy; SBRT, stereotactic body radiotherapy; BED, biological effective dose.

In total, 114 patients (77.0%) received SBRT, while 34 patients (23.0%) received CFRT. The prescribed dose ranged from 28 to 60 Gy delivered in 3-30 fractions, with a median BED of 60 Gy (range, 39-105 Gy). Based on the cutoff point of median BED, 67 patients (45.3%) received <60 Gy and 81 (54.7%) received ≥60 Gy, respectively.

Tumor Response and Patterns of Failure

In terms of the best treatment response of target lesions after radiotherapy, 46 (31.1%) patients achieved complete response, 74 (50.0%) achieved partial response, and the ORR was 81.1%. Figure 1 shows one patient with abdominal LNM who achieved complete response after SBRT. The ORRs in the BED <60 Gy and ≥60 Gy groups were 73.1% and 87.7%, respectively (P = .025; Fig. 2A). In addition, the ORR in the SBRT group was significantly higher than that in the CFRT group (85.1% vs. 67.6%, P = .023; Supplementary Fig. S1A).

Figure 1.

Representative magnetic resonance images of hepatocellular carcinoma (HCC) with lymph node metastases (LNM) before and after stereotactic body radiotherapy (SBRT). (A) Before radiotherapy. (B) Dose distribution curves of SBRT. (C) One month after SBRT. (D) Three months after SBRT; patients with LNM achieved complete response.

Figure 2.

Effect of radiation dose on local control and survival in the entire cohort of HCC patients with LNM. (A) Objective response rates between different biologically effective dose (BED) groups. (B) Dose-response curve for local control and BED. Kaplan-Meier estimates of freedom from local progression (C) and overall survival (D) between different BED groups.

Local failure in abdominal LNM, intrahepatic progression outside the PTV, and distant metastases were observed in 34 (23.0%), 78 patients (52.7%), and 70 patients (47.3%), respectively. The local control rates were 70.2% and 81.5% in the BED <60 Gy and BED ≥60 Gy groups, respectively (P = .106). Furthermore, the dose-response relationship between local control and BED was analyzed: the local failure probability decreased as the dose increased (odds ratio: 0.95 per 1 Gy BED, P = .034; Fig. 2B). A positive dose-response relationship was also observed between BED and local control in subgroups according to LNM size (<4 cm vs. ≥4 cm), LNM number (solitary vs. multiple), and treatment era (before 2016 vs. after 2016; Supplementary Fig. S2). Additionally, the local control rate in the SBRT group was higher than that in the CFRT group but without statistical difference (78.9% vs. 67.6%, P = .174; Supplementary Fig. S1B).

Survival Outcomes

The median follow-up period of the entire group was 13.6 months (range, 0.4-96.0 months). During the study period, 74 patients (50.0%) died. Median OS of the entire cohort was 22.0 months (95% confidence interval [CI], 9.8-34.1 months), with 1- and 2-year FFLP rates of 79.9% and 70.6%, and 1- and 2-year OS rates of 65.0% and 49.7%, respectively. The BED ≥60 Gy group showed a superior 2-year FFLP rate compared with the BED <60 Gy group (76.4% vs. 63.0%, respectively, P = .01, Fig. 2C). Meanwhile, the BED ≥60 Gy group had a significantly improved OS compared with the BED <60 Gy group (2-year OS: 63.5% vs. 32.7%; median OS: 50.4 vs. 14.0 months; P < .001; Fig. 2D). In terms of radiation modality, patients who received SBRT had a significantly longer median OS time than those treated with CFRT (29.6 vs. 9.9 months, P = .007; Supplementary Fig. S1D).

Prognostic Factors for the Entire Cohort

Results of the univariate and multivariate analyses of the prognostic factors for FFLP in the entire cohort are listed in Table 2. The multivariate analysis revealed that the BED was the only independent prognostic factor for local control (hazard ratio [HR] = 0.42, P = .012). In addition, multivariate analysis indicated that BED ≥60 Gy remained a significantly favorable predictor of OS (HR = 0.41, P = .008; Table 3). Meanwhile, patients treated after 2016 had a better OS (HR = 0.55, P = .017). However, uncontrolled intrahepatic tumors (HR = 2.76, P < .001), the presence of tumor thrombus (HR = 2.02, P = .019), the size of LNM ≥4 cm (HR = 1.19, P = .014), and grades 3-4 lymphopenia (HR = 1.91, P = .013) were adverse prognostic factors for OS.

Table 2.

Univariate and multivariate analysis of prognostic factors for FFLP in the entire cohort (n = 148).

Variable	Univariable analysis		Multivariable analysis
	HR (95% CI)	P-value	HR (95% CI)	P-value
Sex		.200
Male	1
Female	1.62 (0.78-3.37)
Age (years)		.239
<52	1
≥52	1.50 (0.76-2.96)
ECOG performance status		.123
0	1
1-2	1.79 (0.86-3.74)
Child–Pugh scores		.177
A	1
B	2.69 (0.64-11.29)
History of hepatitis		.702
No	1
HBV/HCV	1.18 (0.51-2.69)
AFP evaluation before RT (ng/mL)		.500
<400	1
≥400	0.76 (0.34-1.69)
PIVKA-II level before RT (mAU/mL)		.651
<40	1
≥40	1.21 (0.51-2.81)
Intrahepatic tumors		.631
Controlled	1
Uncontrolled	1.18 (0.60-2.34)
Tumor thrombus		.822
Negative	1
Positive	0.89 (0.31-2.53)
Distant metastasis		.253
No	1
Yes	1.63 (0.71-3.74)
Other lesions except abdominal LNM		.979
Absent	1
Present	1.01(0.51-1.98)
LNM size (cm)		.688
<4	1
≥4	1.18 (0.53-2.60)
Abdominal pain related to LNM		.725
No	1
Yes	1.24 (0.38-4.05)
BED (Gy)		.012		0.012
<60	1		1
≥60	0.42 (0.22-0.83)		0.42 (0.22-0.83)
Radiation modality		.145
CFRT	1
SBRT	0.59 (0.29-1.20)
Combined with systemic therapy		.328
No	1
Yes	1.40 (0.71-2.76)
Treatment era		.721
2010-2015	1
2016-2020	1.14 (0.55-2.39)
Treatment-related lymphopenia		.886
Grades 0-2	1
Grades 3-4	0.95 (0.44-2.02)

Abbreviations: ECOG, Eastern Cooperative Oncology Group; HBV, hepatitis B virus; HCV, hepatitis C virus; RT, radiotherapy; LNM, lymph node metastasis; BED, biological effective dose; CFRT, conventional fractionation radiation therapy; SBRT, stereotactic body radiotherapy.

Table 3.

Univariate and multivariate analysis of prognostic factors for overall survival in the entire cohort (n = 148).

Variable	Univariable analysis		Multivariable analysis
	HR (95% CI)	P-value	HR (95% CI)	P-value
Sex		.151
Male	1
Female	0.63 (0.34-1.18)
Age (years)		.397
<52	1
≥52	1.22 (0.77-1.94)
ECOG performance status		.171
0	1
1-2	1.45 (0.85-2.48)
Child–Pugh scores		.528
A	1
B	1.58 (0.38-6.47)
History of hepatitis		.773
No	1
HBV/HCV	1.09 (0.62-1.90)
AFP evaluation before RT (ng/mL)		.505
<400	1
≥400	1.19 (0.72-1.96)
PIVKA-II level before RT (mAU/mL)		.181
<40	1
≥40	1.63 (0.80-3.33)
Intrahepatic tumors		<.001
Controlled	1
Uncontrolled	3.17 (1.95-5.14)
Tumor thrombus		.001
Negative	1
Positive	2.63 (1.50-4.58)
Distant metastasis		.031
No	1
Yes	1.89 (1.06-3.56)
Other lesions except abdominal LNM		<.001		<.001
Absent	1		1
Present	3.82 (2.24-6.50)		4.15 (2.38-7.23)
LNM size (cm)		.001		.047
<4	1		1
≥4	2.21 (1.37-3.58)		1.68 (1.01-2.79)
Abdominal pain related to LNM		.572
No	1
Yes	1.27 (0.55-2.94)
BED (Gy)		<.001		.019
<60	1		1
≥60	0.45 (0.28–0.72)		0.50 (0.27–0.80)
Radiation modality		.008
CFRT	1
SBRT	0.52 (0.32-0.84)
Combined with systemic therapy		.609
No	1
Yes	1.14 (0.69-1.87)
Treatment era		.008		.008
2010-2015	1		1
2016-2020	0.53 (0.33-0.85)		0.51 (0.31-0.84)
Treatment-related lymphopenia		<0.001		0.007
Grades 0-2	1		1
Grades 3-4	2.66 (1.67-4.23)		2.02 (1.21-3.37)

Abbreviations: ECOG, Eastern Cooperative Oncology Group; HBV, hepatitis B virus; HCV, hepatitis C virus; RT, radiotherapy; LNM, lymph node metastasis; BED, biological effective dose; CFRT, conventional fractionation radiation therapy; SBRT, stereotactic body radiotherapy.

Prognostic Factors for the SBRT Group

Regarding the SBRT group (n = 114), the ORRs in the BED <60 Gy and ≥60 Gy groups were 78.0% and 90.6%, respectively (P = .060; Fig. 3A). Moreover, a dose-response relationship was also observed between local control and BED in the SBRT group, (odds ratio, 0.94 per 1 Gy, P = .046; Fig. 3B). Patients who received SBRT with a BED ≥60 Gy had significantly more favorable 2-year FFLP and OS rates than those who received a BED <60 Gy (80.1% vs. 63.4%, P = .004; 68.3% vs. 33.0%, P < .001; Fig. 3C and 3D). The univariate and multivariate analysis of prognostic factors for FFLP and OS are shown in Supplementary Tables S3 and S4. BED was a significant prognostic factor for both FFLP and OS (HR = 0.34, P = .011; HR = 0.52, P = .039). Intrahepatic tumors, multiple LNMs, and grades 3-4 lymphopenia were independent prognostic factors for OS as well.

Figure 3.

Effect of radiation dose on local control and survival in the SBRT group of HCC patients with LNM. (A) Objective response rates between different biologically effective dose (BED) groups. (B) Dose-response curve for local control and BED. Kaplan-Meier estimates of freedom from local progression (C) and overall survival (D) between different BED groups.

Toxicity

The toxicities that developed in the entire cohort during and after radiotherapy are listed in Supplementary Table S5. Overall, acute toxicities commonly occurred but were relatively mild; 3 patients (2.0%) had grade 3 leukopenia, 3 (2.0%) had grade 3 hyperbilirubinemia, 6 (4.1%), 5 (3.4%), and 46 (31.1%) patients with grades 3-4 thrombocytopenia, transaminase and lymphopenia, respectively. Compared with the CFRT group, the incidence of grade ≥3 lymphopenia was significantly lower in the SBRT group (55.9% vs. 23.7%, P < .001).

Late grade 4 toxicities rarely occurred; among the study patients, 2 (1.4%) had anemia, 1 (0.7%) had an elevated transaminase level, and 1 (0.7%) had hyperbilirubinemia. A total of 11 patients (7.5%) had grades 1-2 gastroduodenal ulcer, while only 1 patient had grade 3 ulcer. In addition, 4 patients had grades 1-2 gastrointestinal bleeding. None of the patients had grade 4 or 5 gastroduodenal ulcer or bleeding. The incidence of late toxicities was not significantly different between the 2 BED groups.

Discussion

SBRT is a promising treatment modality and has been increasingly used in patients with HCC owing to its proven efficacy and safety. However, the current evidence supporting the feasibility of SBRT for abdominal LNM from HCC is solely based on a limited number of small sample studies.^19-21 On the basis of a relatively large cohort of patients, this study identified that SBRT could provide satisfactory local control and survival with tolerable treatment-related toxicities in patients with HCC with abdominal LNM. More importantly, a dose-response relationship was observed between local control and BED either in the entire cohort or in the SBRT group.

CFRT is a commonly used dose-fractionation modality for abdominal LNM from HCC in the past years. Nevertheless, its benefit is limited, with a median survival time of 9-10 months.^10-13 The median OS of patients who received CFRT (9.9 month) in our study was also within this range. Owing to the higher BED to be delivered precisely to the target lesion, SBRT has the potential to increase disease control and prolong survival compared with CFRT. A recent small sample study found that SBRT delivered in 45-49 Gy/6-9 fractions for patients with HCC with LNM achieved a 2-year FFLP of 90% and a 2-year OS of 28.6%, showing encouraging outcomes.²¹ In the current study, the SBRT group showed not only significantly higher ORR but also better OS than the CFRT group, suggesting that the SBRT might be the better modality of choice for patients with HCC with abdominal LNM. Therefore, further randomized studies directly comparing these 2 modalities are warranted.

Previous studies have demonstrated the dose-response relationship in patients with HCC with intrahepatic lesions. A pooled analysis of an Asian liver radiation therapy group study reported that patients who received a BED of ≥100 Gy had significantly better FFLP and OS than those who received a BED <100 Gy.¹⁶ Similarly, Huang et al found that BED was notably associated with the prognosis in patients with HCC after SBRT.²² However, the optimal radiation dose for LNM from HCC has not yet been established. Owing to the proximity to the gastrointestinal tract in most cases with abdominal LNM, it is difficult to use a BED (≥100 Gy) as high as that used for intrahepatic lesions. Therefore, a BED range of 40-65 Gy is commonly used in the literature, especially in studies using 2-dimensional or 3DCRT technique. In a retrospective study including 65 patients with HCC with abdominal LNM, the ORR significantly increased as the BED increased from <45 to 45-54 to ≥55 Gy (38.9% vs. 65.7% vs. 83.3%, respectively, P = .037).²⁰ Similarly, Kim et al reported that a higher BED (≥60 Gy) was associated with improved OS in patients with HCC with LNM (P = .042).²³ The findings in this work showed that a BED of ≥60 Gy had significant benefits in terms of FFLP, and a dose-response relationship was also observed between local control and BED across LNM size, LNM number, and treatment era. Meanwhile, the BED of ≥60 Gy group showed excellent OS compared with the BED of <60 Gy group not only in the whole cohort but also in the SBRT subgroup. These results suggested that using SBRT and increasing the radiation dose might be helpful for achieving local control; and a superior FFLP is necessarily to have a positive effect on the survival outcomes in patients with HCC with LNM. Thus, prospective investigations that include homogeneous patients with LNM could provide more concrete evidence regarding the effect of BED on patients’ survival, and to determine the optimal dose range.

Several reports have investigated the prognostic factors in patients with HCC with LNM after radiotherapy.^10-13 Kim et al reported that patients with LNM-related symptoms had worse survival than those without related symptoms,²³ but it was not observed in our results, which may be due to the limited number of patients with related symptoms in the current study. Several studies have consistently demonstrated that intrahepatic tumor status and the occurrence of tumor thrombosis were significant prognostic factors, which is also confirmed by our findings.^10-13 In addition, the LNM-related factors such as LNM size was found to be predictors for survival outcomes in our study, which is consistent with the findings reported by Chen et al.²⁴

Previous studies have demonstrated that although higher radiation dose could be more effective in achieving tumor control, the incidence of treatment-related complications might also increase in patients with HCC.^12,25 For abdominal tumors, gastrointestinal toxicity was the most noticeable radiation-related adverse event. Zeng et al reported that more than 40% of patients treated with a BED of >67.2 Gy experienced fatal gastrointestinal bleeding and suggested the delivery of a conventionally fractionated BED of <67.2 Gy.¹² Similarly, Kim et al found that majority of patients with a radiotherapy-related gastrointestinal ulcer received a BED of >64.8 Gy.²³ Therefore, they recommended a BED of 60-64.8 Gy as an appropriate dose for abdominal LNM. By contrast, in the present study, only 1 patient (0.7%) experienced severe gastrointestinal ulcer or bleeding. Moreover, the incidence of late toxicities was not significantly different between the 2 BED groups. Similarly, Matoba et al evaluated 15 patients with LNM treated with SBRT delivered using IMRT and reported that none of the patients developed grade ≥3 toxicity.²¹ The discrepancy in the toxicity rates might be related to the different radiation techniques used among these studies. Considering its efficacy and safety, SBRT with a higher BED (≥60 Gy) delivered by IMRT or VMAT should be recommended for patients with HCC with abdominal LNM.

Lymphopenia was the most common acute toxicity reported in the current study, and severe lymphopenia is a significant adverse prognostic factor for OS. Previous studies reported that radiation could increase the FasL expression levels in bone marrow stem cells and serum levels of pro-inflammatory markers, thus inducing lymphocyte deficiency.²⁶ Treatment-related severe lymphopenia was associated with tumor progression and worse prognosis,²⁷ which was also confirmed by our results. Cho et al reported that SBRT reduced the risk of lymphopenia compared with CFRT in lung cancer.²⁸ Similar to the findings of the previous study, this study also reported a lower incidence of lymphopenia in the SBRT group. The reduction in fractions, which could decrease the risk of lymphopenia and improve immune surveillance in patients who underwent SBRT, may have contributed to the improved survival outcomes.

Previous studies have reported that the combination of radiotherapy with targeted therapy and/or immunotherapy is safe and may reduce the risk of metastatic recurrence and improve local and systemic immunity.^29,30 Theoretically, the combination of SBRT and systemic therapy could provide clinical benefits for patients with HCC with abdominal LNM. However, in our study, the combination of SBRT with systemic therapy is not a significant prognostic factor for FFLP or OS.

The major potential reason might be the limited number of patients who received systemic therapy in this cohort due to the lack of coverage by health insurance in China before 2020. Moreover, the baseline characteristics are not balanced between patients who received systemic therapy and those who did not, which may affect the therapeutic efficacy. Therefore, prospective studies are warranted to investigate the synergistic efficacy of systemic treatment combined with SBRT in patients with HCC with LNM.

This study has some limitations. It was retrospective in nature and conducted in a single institution. Hence, selective bias cannot be avoided. Moreover, the results might be influenced by the long treatment span and the variety of dose-fractionation schedules. In addition, the dose-response relationship between BED and local control should be validated by external data and a prospective cohort study.

Conclusions

SBRT achieved satisfactory local control and survival with non-significant treatment-related toxicities in patients with abdominal LNM from HCC. Moreover, results of this large series suggest a dose-response relationship between local control and BED, and a BED of ≥60 Gy significantly affected the local control rate as well as survival.

Funding

This work was supported by the grant from “5010 program” of Sun Yat-Sen University (2019013).

Conflict of Interest

The authors indicated no financial relationships.

Author Contributions

Conception/design: Y.Z., M.X. Provision of study material or patients: Q.L., S.L., Y.Y., M.L. Collection and/or assembly of data: Y.W., Q.L., J.Z., M.X. Data analysis and interpretation: All authors. Manuscript writing: All authors. Final approval of manuscript: All authors.

Data Availability

Research data are stored in an institutional repository (www.researchdata.org.cn) and will be shared upon request to the corresponding author.