Abstract
Background/Aims
Local ablation therapies (LATs), including radiofrequency (RFA) and microwave ablation (MWA), are established for small hepatocellular carcinoma (HCC) as standard treatment. External beam radiotherapy (EBRT), including stereotactic body radiotherapy (SBRT) and proton beam therapy (PBT), has emerged as a non-invasive alternative; however, direct comparative evidence remains limited. We conducted a systematic review and meta-analysis comparing the efficacy and safety of EBRT versus LAT in patients with HCC.
Methods
PubMed/MEDLINE, Embase, Cochrane Library, and KoreaMed were searched through June 2025 per PRISMA guidelines. Eligible studies included retrospective or prospective comparisons of EBRT (SBRT/PBT) with LAT (RFA/MWA) reporting recurrence or survival. Hazard ratios (HRs) and odds ratios (ORs) were pooled using random-effects models.
Results
Twelve studies (2,768 patients) were included. Compared with LAT, EBRT was associated with significantly lower local recurrence (OR 0.36, 95% confidence interval [CI] 0.25–0.50; P<0.001; I2=0%) and improved local progression-free survival (HR 0.41, 95% CI 0.33–0.50; P<0.001), regardless of tumor size. Progression-free survival had a trend favoring EBRT (HR 0.82, 95% CI 0.64–1.06; P=0.131), and overall survival did not differ significantly between modalities (HR 1.24, 95% CI 0.83–1.84; P=0.291). Severe toxicities (grade ≥3) were comparable between groups for acute (OR 0.78, 95% CI 0.27–2.24) and late events (OR 1.29, 95% CI 0.85–1.95).
Conclusions
EBRT was associated with better local control without significant differences in survival or toxicity compared with LAT. These findings support the safety and efficacy of EBRT for local treatment of patients with liver-confined HCC.
Keywords: Hepatocellular carcinoma, Stereotactic body radiotherapy, Proton beam therapy, Radiofrequency ablation, Microwave ablation
Graphical Abstract

INTRODUCTION
Hepatocellular carcinoma (HCC) is the most common primary liver malignancy and remains a leading cause of cancer-related death worldwide [1]. Although surgical resection and transplantation provide curative potential, many patients are ineligible due to advanced cirrhosis, comorbidities, or lack of donor availability. For early-stage, liver-confined tumors, percutaneous ablation—historically centered on radiofrequency ablation (RFA) and now encompassing microwave ablation (MWA)—is a well-established standard local ablative therapy (LAT). However, their efficacy is limited in larger lesions or tumors located in challenging subphrenic or perivascular sites [2-5].
In parallel, external beam radiotherapy (EBRT), including stereotactic body radiotherapy (SBRT) and proton beam therapy (PBT), has advanced considerably with modern imaging and delivery techniques. These approaches enable the delivery of ablative radiation doses with sharp falloff, achieving local control rates exceeding 80–90% even in tumors traditionally considered unsuitable for ablation. EBRT is noninvasive and broadly applicable across various tumor locations, making it particularly valuable for patients with high risk of post-ablation recurrence or when ablation is technically infeasible [6,7].
Reflecting these advances, international guidelines now incorporate EBRT as a locoregional treatment option for liver-confined HCC [6,8-12]. While both numerous retrospective studies [13-22] and recent phase 3 randomized controlled trials (RCTs) [23,24] evaluating PBT versus RFA or SBRT versus RFA have provided important data supporting its curative potential, prospective comparative evidence remains relatively limited. Furthermore, as MWA is increasingly utilized alongside RFA, a comprehensive synthesis of the accumulated evidence is warranted to define the relative efficacy and safety of EBRT, especially because previous meta-analyses [25-30] have yielded inconsistent findings. Therefore, we conducted systematic review and meta-analysis to evaluate the therapeutic outcomes and safety of EBRT compared with those of LATs in patients with HCC.
MATERIALS AND METHODS
Study protocol
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [31]. The study protocol was prospectively registered in PROSPERO (Registration No: CRD420251067260).
Search strategy
The literature search strategy was structured according to the Population, Intervention, Comparator, and Outcome framework (Supplementary Table 1). The population was patients with HCC; the intervention was EBRT, including SBRT or PBT; the comparator was LAT (RFA or MWA); and the outcomes were local control, survival endpoints, and treatment-related toxicities. A comprehensive search of PubMed/MEDLINE, Embase, the Cochrane Library, and KoreaMed was conducted to identify eligible studies published up to June 9, 2025. Search strings for each database are provided in Supplementary Table 2. Reference lists of included articles and recent meta-analyses were manually screened to identify additional studies. Only human studies published in English were included. Conference abstracts and letters were screened but excluded because full-text data were unavailable. No restrictions were placed on publication year. The search strategy was developed in consultation with a professional medical librarian at Soonchunhyang University College of Medicine, Bucheon.
Study selection
Studies were eligible if they were retrospective or prospective comparative trials that directly evaluated EBRT (SBRT or PBT) versus LAT (RFA or MWA) in patients with HCC. Eligible studies were required to report at least one oncologic endpoint, including local recurrence, local progression-free survival (LPFS), progression-free survival (PFS), overall survival (OS), or treatment-related toxicities. Studies were excluded if treatment was performed with neoadjuvant or adjuvant intent, if EBRT or LAT was planned as a bridge to liver transplantation, if patients had extrahepatic metastases, or if treatment was conducted solely with palliative intent. Studies employing combined modalities such as concurrent transarterial chemoembolization (TACE) with EBRT or ablation, overlapping cohorts from the same institution, or SBRT delivered in more than 10 fractions were excluded. Studies without available full-text articles were also excluded. All identified records were imported into a reference management system, and duplicate entries were removed. Titles and abstracts were independently screened by two investigators (SH Bae and SH Choi), and full texts of potentially eligible studies were reviewed in detail. Disagreements were resolved by consensus. PRISMA flow diagram of the study selection process is presented in Figure 1.
Data extraction and quality assessment
Data extraction was independently performed by two investigators using a standardized form. Extracted data included study characteristics (first author, year, and nation), patient and tumor characteristics (size and multiplicity), treatment details (SBRT or PBT, ablation modality, and use of propensity score matching analysis [PSM]), and clinical outcomes (local control, survival, and toxicity). The methodological quality of RCTs was assessed using the Risk of Bias 2 (RoB 2) tool [32], whereas that of non-randomized studies were evaluated using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) tool [33]. Discrepancies were resolved by consensus or consultation with a third investigator.
Statistical analysis
Meta-analyses were performed using the DerSimonian–Laird random-effects model to account for expected clinical and methodological heterogeneity among studies [34]. Hazard ratios (HRs) with 95% confidence intervals (CIs) were extracted for time-to-event outcomes (LPFS, PFS, OS) and odds ratios (ORs) for binary outcomes (local recurrence, toxicity). Results of individual studies and pooled effect estimates were displayed using forest plots, and study-level data were summarized in structured tables. Statistical heterogeneity was assessed using Cochran’s Q test [35] and quantified by the Higgins I2 statistic, with I2>50% considered indicative of substantial heterogeneity [36]. To explore potential study-level sources of heterogeneity in OS, univariate meta-regression was performed using tumor size, study design, Child–Pugh class A proportion, primary HCC (treatment-naïve) proportion, and tumor multiplicity as covariates. For comparisons between subgroups, a Q test based on analysis of variance and a random-effects model were used. Publication bias was explored through visual inspection of funnel plots and Egger’s regression test when feasible [37]; P>0.05 was considered to indicate no significant bias. A sensitivity analysis using the leave-one-out method was performed to explore the potential origins of heterogeneity for OS. Each study was sequentially excluded from the overall meta-analysis, and the pooled effect and heterogeneity were recalculated. Significant changes in the pooled effect indicated that the excluded study had a significant impact on the model’s conclusions. All statistical analyses were conducted using Rex Excel-based statistical analysis software (version 3.6.0; RexSoft, Seoul, Korea; https://rexsoft.org). A two-sided P<0.05 was considered statistically significant. Certainty of evidence for each outcome (LPFS, PFS, OS, acute toxicity, and late toxicity) was assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach, considering risk of bias, inconsistency, indirectness, imprecision, and other considerations.
RESULTS
Study selection and characteristics
The systematic search identified 1,311 records from PubMed/MEDLINE, Embase, Cochrane Library, and KoreaMed, along with 8 additional studies retrieved through manual screening of reference lists and other sources. After removing 61 duplicates, 1,258 studies remained for screening. Following title and abstract review, 1,220 records were excluded due to irrelevance or inadequate data. The remaining 38 articles were assessed for eligibility, and 26 were excluded after detailed review. Ultimately, 12 studies comprising 2,768 patients met all the inclusion criteria and were included in the final analysis [13-24] (Fig. 1). The included studies were published between 2015 and 2025 and originated from both Eastern and Western institutions. Two studies were RCTs, while the remaining were retrospective studies. Among these, four studies conducted PSM, and the data after PSM were extracted for analysis. SBRT was the most common EBRT modality, whereas two studies evaluated PBT. Eight studies compared SBRT with RFA, two compared SBRT with MWA, and two compared PBT with RFA. Eight studies were single-center analyses, and four were multicenter studies. Three studies included only primary HCC, three focused on recurrent HCC, and six included mixed or unspecified populations. Baseline patient characteristics, tumor size distribution, and treatment details are summarized in Table 1 and 2.
Figure 1.

PRISMA flow diagram of study selection. TACE, transarterial chemoembolization.
Table 1.
Characteristics of the included studies
| Author (year) | Country | Study design | Study period | Indication | No. of pts |
|---|---|---|---|---|---|
| Xi et al. (2025) | China | S/RCT | 2019–2022 | Recurrent HCC (single, ≤5 cm) | 166 |
| Ma et al. (2025) | China | M/PSM | 2015–2021 | Recurrent HCC (≤5 cm) | 84 |
| Yang et al. (2024) | China | S/R | 2017–2020 | Primary or recurrent HCC (single, ≤5 cm) | 288 |
| Seo et al. (2024) | Korea | S/PSM | 2016–2021 | Primary HCC | 83 |
| Maher et al. (2024) | Australia | S/R | 2016–2020 | Primary HCC (single, ≤3 cm) | 87 |
| Moon et al. (2023) | United States | M/R | 2012–2018 | HCC (unspecified) | 642 |
| Ueno et al. (2021) | Japan | S/PSM | 2014–2019 | Primary or recurrent HCC (single, ≤3 cm) | 93 |
| Kim et al. (2021) | Korea | S/RCT | 2013–2017 | Recurrent HCC (≤2 nodules, <3 cm) | 136 |
| Jeong et al. (2021) | Korea | S/R | 2013 | Primary or recurrent HCC (≤3 nodules, ≤3 cm) | 266 |
| Kim et al. (2020) | China, Japan, Hong Kong, Taiwan, and Korea | M/PSM | 2010–2016 | Primary or recurrent HCC (≤3 nodules, ≤6 cm) | 626 |
| Wahl et al. (2016) | United States | S/R | 2004–2012 | Inoperable HCC | 224 |
| Shiozawa et al. (2015) | Japan | M/R | 2011–2014 | Primary HCC (single) | 73 |
HCC, hepatocellular carcinoma; S, single center; M, multicenter; No. of pts, number of patients; PSM, retrospective study using propensity score matching analysis; R, retrospective study; RCT, randomized controlled trial.
Table 2.
Baseline patient and tumor characteristics in the included studies
| Author (year) | Treatment | RT dose (Gy) | mNo. of fx | No. of pts | Primary HCC (%) | Age (yr) | CP class A/B/C (%) | BCLC stage 0/A/B/C/D (%) | VI (%) | Size (cm) | Multiplicity (%) | mF/U (mo) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Xi et al. (2025) | SBRT | 36–54 | 3 | 83 | 0 | 56 (50–64)* | 100/0/0 | NR | 0 | 1.7 (1.3–2.3)* | 0 | 43 |
| RFA | 83 | 0 | 58 (51–64)* | 100/0/0 | NR | 0 | 1.6 (1.3–2.2)* | 0 | 43 | |||
| Ma et al. (2025) | SBRT | 37.5§ | 3§ | 42 | 0 | NR | 98/2/0 | NR | 0 | NR | 17 | NR |
| RFA | 42 | 0 | NR | 93/7/0 | NR | 0 | NR | 7 | NR | |||
| Yang et al. (2024) | SBRT | 36–54 | 3 | 122 | 51 | 56 (50–63)* | 98/2/0 | NR | 0 | 2.3±1.0† | 0 | 31 |
| RFA | 166 | 54 | 55 (48–63)* | 99/1/0 | NR | 0 | 2.1±0.8† | 0 | 40 | |||
| Seo et al. (2024) | PBT | 66§ | 10§ | 22 | 100 | 66±10.1† | 96/4/0 | 41/45/0/14/0 | 14 | 2.0 (1.7–2.8)* | 5 | 34 |
| RFA | 61 | 100 | 62±8.6† | 87/13/0 | 41/56/3/0/0 | 0 | 2.0 (1.7–2.3)* | 28 | 44 | |||
| Maher et al. (2024) | SBRT | 30–50 | 3 or 5 | 36 | 100 | 68±7.6† | 78/22/0 | 25/75/0/0/0 | 0 | 2.1 (1.4–3.0) | 0 | NR |
| MWA | 51 | 100 | 63±8.8† | 90/10/0 | 59/41/0/0/0 | 0 | 1.8 (1.3–2.5) | 0 | NR | |||
| Moon et al. (2023) | SBRT | NR | NR | 202 | 53 | 66±11.3† | 55/35/7 | 4/34/2/50/10 | 0 | 2.4 (1.8–3.4)* | 15 | 15 |
| RFA/MWA | 440 | 76 | 62±8.6† | 58/33/8 | 11/40/2/38/9 | 0 | 2.1 (1.6–2.6)* | 21 | 21 | |||
| Ueno et al. (2021) | SBRT | 40 | 5 | 31 | 10 | 75±9.1† | NR | 71/29/0/0/0 | 0 | 1.5±0.6† | 0 | 23 |
| RFA | 62 | 21 | 75±7.5† | NR | 74/26/0/0/0 | 0 | 1.5±0.6† | 0 | 27 | |||
| Kim et al. (2021) | PBT | 66 | 10 | 80 | 0 | 61 (40–82) | 96/4/0 | 6/54/35/5/0 | 0 | 1.2 (1.0–2.9) | 6 | 52 |
| RFA | 56 | 0 | 61 (44–77) | 98/2/0 | 7/56/32/5/0 | 0 | 1.2 (1.0–2.2) | 4 | 51 | |||
| Jeong et al. (2021) | SBRT | 30–60 | 3 or 4 | 87 | 5 | 63 (41–90) | 92/8/0 | 29/53/17/0/0 | 0 | 1.5 (0.8–2.8) | 11 | 50 |
| RFA | 179 | 48 | 60 (40–87) | 87/13/0 | 38/48/14/0/0 | 0 | 1.5 (0.5–3.0) | 16 | 50 | |||
| Kim et al. (2020) | SBRT | NR | NR | 313 | 16 | 66 (59–75)* | A: 88, B/C: 12 | 25/26/21/28/0 | 0 | 2.1 (1.5–3.1)* | 6 | 22 |
| RFA | 313 | 18 | 67 (59–74)* | A: 87, B/C: 13 | 26/25/23/26/0 | 0 | 2.2 (1.6–3.2)* | 7 | 30 | |||
| Wahl et al. (2016) | SBRT | 27–60 | 3 or 5 | 63 | NR | 62 (35–85) | 69/29/2 | NR | NR | 2.2 (0–10) | 22 | 13 |
| RFA | 161 | NR | 60 (31–81) | 50/42/8 | NR | NR | 1.8 (0.6–7.0) | 32 | 20 | |||
| Shiozawa et al. (2015) | SBRT | 60§ | 3–5 | 35 | 100 | 75‡ (55–89) | 80/20/0 | NR | NR | 2.9‡ (1.2–5) | 0 | 13 |
| RFA | 38 | 100 | 69‡ (42–86) | 82/18/0 | NR | NR | 1.8‡ (0.7–2.9) | 0 | 18 |
Means the interquartile range.
Means mean ± standard deviation.
Indicates mean value.
Indicates median value.
BCLC, Barcelona Clinic Liver Cancer; CP, Child–Pugh; HCC, hepatocellular carcinoma; mF/U, median follow-up; mNo. of fx, median number of fractions; MWA, microwave ablation; NR, not reported; PBT, proton beam therapy; RFA, radiofrequency ablation; RT, radiotherapy; SBRT, stereotactic body radiotherapy; VI, vascular invasion.
Quality assessment revealed that most non-randomized studies had a moderate to serious overall risk of bias according to the ROBINS-I tool, mainly due to confounding and selection of participants. Other domains—including classification of interventions, deviations from intended interventions, missing data, and outcome measurement—were generally rated as low risk. In contrast, the two RCTs demonstrated consistently low risk of bias across all RoB 2 domains. Detailed domain-level assessments for all studies are provided in Supplementary Figure 1.
Local recurrence and LPFS
EBRT was consistently associated with higher 1-, 3-, and 5-year LPFS rates compared with LAT (Table 3). Pooled analysis demonstrated that EBRT was associated with a significantly lower local recurrence rate compared with LAT (OR 0.36, 95% CI 0.25–0.50; P<0.001; I2=0%) (Fig. 2). EBRT was also associated with improved LPFS (HR 0.41, 95% CI 0.33–0.50; P<0.001) (Fig. 3A). Among studies reporting tumor size, subgroup analyses by tumor size confirmed that the advantage of EBRT was consistent for tumors ≤2 cm (HR 0.32, 95% CI 0.19–0.55) (Fig. 3B) and >2 cm (HR 0.34, 95% CI 0.20–0.59) (Fig. 3C).
Table 3.
Summary of treatment outcomes in the included studies comparing external beam RT and LAT
| Study (year) | LPFS (%) |
PFS (%) |
OS (%) |
|||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1-year |
3-year |
5-year |
1-year |
3-year |
5-year |
1-year |
3-year |
5-year |
||||||||||
| RT | LAT | RT | LAT | RT | LAT | RT | LAT | RT | LAT | RT | LAT | RT | LAT | RT | LAT | RT | LAT | |
| Xi et al. (2025) | 96 | 89 | 84 | 67 | 76 | 71 | 51 | 40 | 100 | 100 | 90 | 91 | ||||||
| Ma et al. (2025) | 81 | 86 | 45 | 29 | 98 | 98 | 86 | 64 | 79 | 41 | ||||||||
| Yang et al. (2024) | 76 | 75 | 59 | 48 | 99 | 100 | 90 | 94 | ||||||||||
| Seo et al. (2024) | 95 | 98 | 87 | 87 | 75 | 85 | 43 | 47 | 100 | 100 | 86 | 97 | ||||||
| Maher et al. (2024) | 97 | 82 | 67 | 55 | 94 | 97 | 69 | 80 | ||||||||||
| Moon et al. (2023) | 96 | 89 | 93 | 82 | 92 | 80 | 68 | 84 | 36 | 64 | 24 | 53 | ||||||
| Ueno et al. (2021) | 100 | 98 | 100 | 87 | 100 | 82 | 90 | 95 | 67 | 71 | 45 | 61 | ||||||
| Kim et al. (2021) | 97 | 86 | 88 | 78 | 81 | 72 | 54 | 54 | 21 | 23 | 11 | 11 | 96 | 99 | 79 | 87 | 65 | 70 |
| Jeong et al. (2021) | 99 | 97 | 95 | 94 | 96 | 98 | 72 | 84 | ||||||||||
| Kim et al. (2020) | 90 | 74 | 82 | 66 | 79 | 64 | 88 | 88 | 63 | 59 | 39 | 44 | ||||||
| Wahl et al. (2016) | 97 | 84 | 84 | 75 | 84 | 71 | 74 | 70 | ||||||||||
| Shiozawa et al. (2015) | 97 | 97 | 95 | 100 | ||||||||||||||
| Overall | 90–100 | 74–98 | 82–100 | 66–94 | 79–100 | 64–82 | 54–97 | 54–86 | 21–67 | 23–55 | 11 | 11 | 68–100 | 70–100 | 36–90 | 59–97 | 24–79 | 41–70 |
LAT, local ablation therapy; LPFS, local progression-free survival; OS, overall survival; PFS, progression-free survival; RT, radiotherapy.
Figure 2.

Forest plot comparing local recurrence rates between external beam radiotherapy and local ablation therapy. CI, confidence interval; OR, odds ratio.
Figure 3.

Forest plot comparing local progression-free survival between external beam radiotherapy and local ablation therapy in (A) all eligible studies, (B) tumors measuring ≤2 cm, and (C) tumors measuring > 2 cm. CI, confidence interval; HR, hazard ratio.
PFS
Three-year PFS rates ranged from 21% to 67% for EBRT versus 23% to 55% for ablation, while five-year rates were 11% in both arms (Table 3). Four studies were eligible for pooled analysis of PFS. EBRT was associated with a trend toward improved PFS compared with LAT (HR 0.82, 95% CI 0.64–1.06; P=0.131) (Fig. 4A). Low heterogeneity was observed (I²=26%).
Figure 4.

Forest plots comparing (A) progression-free survival and (B) overall survival between external beam radiotherapy and local ablation therapy. CI, confidence interval; HR, hazard ratio.
OS
Reported 1-, 3-, and 5-year OS rates varied widely across studies but did not differ significantly between the groups (Table 3). Nine studies including 2,388 patients were pooled for OS analysis. No significant difference was observed between EBRT and LAT (HR 1.24, 95% CI: 0.83–1.84; P=0.291; Fig. 4B). Considerable heterogeneity was observed across studies (I2=81%), prompting subgroup analyses to identify potential sources of variability. Univariate meta-regression identified study design as the dominant source of heterogeneity (P<0.001, R2=0.989), reducing the residual I2 to approximately 3%. Meanwhile, tumor size, Child–Pugh class A proportion, and tumor multiplicity were not significant. The proportion of primary HCC was also significantly associated with the OS estimate (P<0.001) (Supplementary Table 3). When stratified by study design, five RCT or PSM studies showed comparable OS between EBRT and ablation (HR 0.87, 95% CI 0.63–1.20; I2=36%), whereas four purely retrospective studies demonstrated inferior OS with EBRT (HR 2.07, 95% CI 1.66–2.59; I2=0%; Fig. 5). The difference between these two subgroups was statistically significant (P<0.001), and the residual heterogeneity after accounting for study design was minimal (I2=3%; P=0.400), suggesting that study design was the main contributor to between-study variability. Subgroup analysis by disease setting (primary vs. recurrent) showed no significant interaction (P=0.148; Supplementary Fig. 2).
Figure 5.

Forest plot of overall survival stratified by study design (RCT or retrospective studies using PSM vs. retrospective studies). CI, confidence interval; HR, hazard ratio; PSM, propensity score matching analysis; RCT, randomized controlled trials.
Treatment-related toxicity
Severe toxicities, defined as grade ≥3 events, were uncommon across both modalities (Supplementary Table 4). Acute toxicities within 3 months occurred at similar rates between EBRT and LAT (OR 0.78, 95% CI 0.27–2.24; P=0.649; I2=58%). Late toxicities, including hepatobiliary and gastrointestinal events, were also comparable (OR 1.29, 95% CI 0.85–1.95; P=0.233; I2=0%). EBRT showed a non-significant trend toward fewer acute events.
Sensitivity analysis and publication bias assessment
Sensitivity analysis showed that the sequential exclusion of individual studies did not significantly alter the pooled HRs of OS, confirming the robustness of the meta-analysis (Supplementary Fig. 3). Furthermore, Egger’s test did not detect statistically significant asymmetry (Supplementary Fig. 4). However, these findings should be interpreted cautiously because the number of included studies was limited, and the analyses were underpowered to reliably detect publication bias.
Certainty of evidence
Certainty of evidence was evaluated using the GRADE framework. The certainty was rated as high for LPFS, moderate for PFS, low for OS and acute toxicity, and very low for late toxicity. Detailed GRADE profiles are provided in Supplementary Table 5.
DISCUSSION
In this systematic review and meta-analysis, we compared the efficacy of EBRT with that of LATs for HCC and found that EBRT was associated with improved local tumor control without significant differences in survival and toxicity. Across 12 comparative studies involving 2,768 patients, EBRT was consistently associated with lower local recurrence and improved LPFS while maintaining overall safety comparable to RFA or MWA. These findings highlight an expanding role of modern radiotherapy as a potentially curative local treatment option in HCC, particularly when ablation may not be feasible or may be suboptimal.
A major strength of the present meta-analysis is that it is designed to reflect contemporary, clinically meaningful comparisons through stricter study selection than many prior syntheses. We focused on comparative studies with clearly defined local endpoints and curative treatment intent while excluding palliative regimens, combined modalities (e.g., TACE plus EBRT), and non-ablative fractionation patterns that do not represent current practice. We also attempted to minimize duplication and overlapping institutional cohorts to ensure data integrity. Consequently, the pooled estimates are more likely to represent the real-world performance of modern EBRT strategies rather than heterogeneous historical radiotherapy paradigms, which may partly explain the robust local effect estimates and minimal heterogeneity for local outcomes.
The observed association between EBRT and improved local control is biologically and technically plausible and consistent with accumulating prospective and high-quality observational evidence. Both SBRT and PBT can deliver ablative biologically effective doses with high conformity and steep dose gradients. By contrast, LATs (RFA or MWA) can be limited by perivascular constraints (including heatsink effects) and acoustic window limitations, potentially resulting in incomplete coverage or suboptimal margins [38-40]. EBRT is not constrained by tissue conductivity or vascular flow and can provide conformal target coverage across various tumor locations [6]. This distinction is clinically important in subphrenic, perivascular, or recurrent lesions, where LAT may perform less reliably.
Although these considerations indicate that EBRT may be particularly advantageous for larger tumors and lesions in technically challenging locations, the present meta-analysis could not formally establish such modality- and site-specific benefits. In an exploratory analysis by tumor size, the LPFS advantage of EBRT was observed in both tumors measuring ≤2 cm and >2 cm, for which durable local control with LAT can be difficult because of incomplete treatment or marginal failure. Several included studies that reported outcomes by tumor location suggested better local control with EBRT at sites such as perivascular, subphrenic, or dome locations [13-15,19], consistent with prior location-stratified analyses [39,40]. However, tumor-location definitions were heterogeneous across studies, and the number of studies per site was small, precluding a formal location-based subgroup analysis. Thus, the observations are only hypothesis generating and warrant confirmation in studies specifically designed to evaluate size- and location-specific outcomes.
Consistent with previous meta-analyses [25,29,30,41], the present study showed that EBRT was associated with significantly higher local control than RFA or MWA, yet this advantage did not translate into a clear survival benefit. This dissociation likely reflects two distinct contributions. First, the inferior OS associated with EBRT in the unadjusted retrospective studies is largely attributable to allocation bias rather than to a true survival disadvantage. Second, a limited effect of local control on OS is intrinsic to HCC rather than specific to EBRT. Long-term survival is strongly governed by multicentric intrahepatic recurrence and underlying liver dysfunction [42,43]; thus, focal therapies—although achieving local control comparable to that of surgical resection—cannot ultimately exceed the survival benefit achieved by curative surgical treatment, with liver transplantation uniquely addressing the entire cirrhotic liver. A similar dissociation has been observed with thermal ablation cases, in which improved local control has not consistently translated into longer survival. In this context, integrating local treatment with effective systemic or locoregional therapy may be needed to extend the benefit of local control to longer-term survival.
In our meta-analysis, OS did not differ significantly between EBRT and LAT, but heterogeneity was substantial. Much of this variability could be attributable to differences in study design: RCT and PSM studies [13,18,20,23,24] showed similar survival between the two treatments, whereas unadjusted retrospective series [14,16,17,19] tended to favor LAT. This was further supported by meta-regression, with study design accounting for much of the between-study heterogeneity in OS (R2=0.989), whereas tumor- and patient-level factors did not exert any influence. This likely reflects nonrandom treatment allocation that favored LAT. Particularly, EBRT was more often used for tumors in technically challenging locations and for patients with less favorable clinical characteristics, whereas LAT was generally applied to smaller, more accessible tumors with preserved liver function [44]. Given that such confounding by indication would tend to bias the unadjusted OS comparison against EBRT, the apparent OS difference was largely attenuated once these imbalances were accounted for in studies using PSM and RCTs. The proportion of treatment-naïve (primary) HCC was likewise associated with OS in the meta-regression. However, this reflects the same underlying difference: treatment-naïve patients were concentrated in the retrospective cohorts, whereas recurrent-disease patients were predominantly included in RCT or studies using PSM. The apparent difference by disease setting therefore might not represent an effect independent of trial design. The ongoing RCTs comparing EBRT with LAT (NCT05433701 and ACTRN12621001444875), which are the first RCTs to expand their scope to treatment-naïve HCC beyond recurrent cases, may provide definitive answers.
The GRADE certainty of evidence was high for LPFS, moderate for PFS, low for OS and acute toxicity, and very low for late toxicity. These ratings primarily reflect the predominance of non-randomized designs, heterogeneity and imprecision of the survival estimates, and variability in toxicity reporting across studies. Together with the consistency of local-control estimates, these findings support EBRT as a treatment associated with favorable local control and a comparable safety profile. Further randomized and prospective studies would help confirm these findings and improve their applicability to current practice.
Modern EBRT demonstrated a favorable safety profile in this meta-analysis, with severe hepatobiliary or gastrointestinal toxicities remaining uncommon and generally comparable to those observed after RFA or MWA. Historically, the risk of radiation-induced liver disease limited the use of EBRT in patients with cirrhosis. However, advances in intensity-modulated radiotherapy, respiratory motion management, and image guidance have enabled the safe delivery of high-dose, curative radiation while maximally sparing the surrounding normal liver and adjacent critical organs in stringently selected patients [6,7]. In addition, particle therapy offers further normal-tissue sparing through its finite range and absence of exit dose, reducing the integral dose to the uninvolved liver. PBT, the most widely studied particle modality in HCC, has shown favorable local control and safety [45,46]. However, current evidence has not demonstrated a clear efficacy difference from that of photon-based SBRT once an adequately high biologically effective dose is delivered [47,48]. Carbon-ion radiotherapy may offer additional benefit, with emerging data suggesting higher biological effectiveness while maintaining liver tolerance [49,50]. These advances likely contributed to the low toxicity rates observed across the included studies, supporting the feasibility of EBRT even in patients with borderline liver reserve.
From a clinical perspective, these findings have several implications for treatment selection. EBRT should be considered a curative option when LAT is contraindicated or technically challenging, including in tumors adjacent to major vessels, the diaphragm, or the biliary tree, where heatsink effects or limited acoustic windows compromise ablation efficacy. EBRT may also be advantageous in the recurrent setting, where post-ablation fibrosis and anatomical deformation impede accurate needle placement. Furthermore, the progressive adoption of PBT and the anticipated expansion of carbon-ion therapy may broaden the therapeutic applicability of radiation through improved sparing of uninvolved sections of the liver and enhanced tumoricidal potential. With increasing precision and biological adaptability, EBRT represents a noninvasive and reproducible modality that complements existing locoregional strategies and may offer consistent margin control across diverse anatomical scenarios.
Some limitations of this study warrant consideration. Most included studies were retrospective, and treatment allocation was influenced by tumor location, liver function, institutional expertise, and patient preference, which may have contributed to residual confounding despite subgroup and sensitivity analyses. Definitions of local recurrence and follow-up schedules were not fully uniform, potentially affecting endpoint ascertainment. Late toxicity was inconsistently reported, limiting precision for late-event estimates and contributing to lower certainty for that outcome. Although we restricted inclusion to ablative-intent EBRT whenever possible, heterogeneity in dose/fractionation, motion-management strategies, and LAT device technologies may introduce clinical variability. A direct comparison between SBRT and PBT was beyond the scope of this study and was not feasible as only two of the included studies evaluated PBT. In addition, the relatively small number of included studies limits the reliability and statistical power of formal publication bias assessments. Therefore, the results of funnel-plot inspection and Egger’s test should be interpreted cautiously. Furthermore, because this meta-analysis was based on study-level aggregate data, we could not perform individual-patient mediation or competing-risk analyses to formally quantify the relationship between local control and survival. These limitations highlight the need for further prospective studies with standardized endpoint definitions and toxicity reporting, including patient-reported outcomes and longitudinal liver function trajectories.
In conclusion, compared with RFA or MWA, EBRT was associated with improved local tumor control without compromising survival or safety in HCC. These findings highlight the role of EBRT as a curative local treatment option, particularly when ablation is infeasible or suboptimal. Further randomized and prospective registries are warranted to define optimal patient selection and guide the role of EBRT in HCC management.
Abbreviations
- CI
confidence interval
- EBRT
external beam radiotherapy
- GRADE
Grading of Recommendations Assessment
- HCC
hepatocellular carcinoma
- HR
hazard ratio
- LAT
local ablation therapy
- LPFS
local progression-free survival
- MWA
microwave ablation
- OR
odds ratio
- OS
overall survival
- PBT
proton beam therapy
- PFS
progression-free survival
- PICO
Population
- PRISMA
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
- PSM
propensity score matching analysis
- RCTs
randomized controlled trials
- RFA
radiofrequency ablation
- RoB 2
Risk of Bias 2
- ROBINS-I
Risk of Bias in Non-randomized Studies of Interventions
- SBRT
stereotactic body radiotherapy
- TACE
transarterial chemoembolization
Study Highlights
• EBRT was associated with a significantly lower risk of local recurrence and improved local progression-free survival compared with local ablation therapy in patients with HCC.
• Overall survival was comparable between the two treatment modalities, but considerable heterogeneity was observed across studies.
• The risk of severe toxicities (grade ≥3) was similar.
• EBRT may offer a safe and effective curative alternative for liver-confined HCC.
Footnotes
Data availability
The data extracted for this systematic review and the data used for all analyses are available from the corresponding authors upon reasonable request.
Authors’ contribution
Conception or design of the work: All authors; Data collection: Sun Hyun Bae, Seo Hee Choi, Tae Hyun Kim, Hee Chul Park; Data analysis and interpretation: All authors; Drafting the article: Sun Hyun Bae, Seo Hee Choi; Critical revision of the article: Sang Min Yoon, Jeong Il Yu, Hyun-Cheol Kang, Nalee Kim, Tae Hyun Kim, Hee Chul Park; Final approval of the version to be published: All authors.
Acknowledgements
This systematic review and meta-analysis was conducted as part of the evidence-review process for the development of the 2026 KLCA-NCC Korea Practice Guidelines for the Management of Hepatocellular Carcinoma. This work was supported in part by the National Cancer Center Grant (NCC-2610600-1).
Conflicts of Interest
The authors declare no conflicts of interest.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (http://www.e-cmh.org).
PICO framework defining the clinical question of this meta-analysis
Database search strategy
Univariate meta-regression analysis for overall survival
Summary of treatment-related toxicities in the included studies comparing external beam radiotherapy and local ablation therapy
GRADE certainty-of-evidence assessment for all clinical outcomes
Risk-of-bias assessment of included studies. (A) RoB 2 assessment for randomized controlled trials. (B) Summary distribution of ROBINS-I risk-of-bias ratings across non-randomized studies. (C) ROBINS-I traffic-light plot showing domain-level risk-of-bias judgments for each non-randomized study. RoB 2, Risk of Bias 2; ROBINS-I, Risk of Bias in Non-randomized Studies of Interventions.
Subgroup analyses of overall survival by disease setting (primary vs. recurrent HCC) between EBRT and LAT. CI, confidence interval; EBRT, external beam radiotherapy; HCC, hepatocellular carcinoma; HR, hazard ratios; LAT, local ablation therapy.
Leave-one-out sensitivity analysis for overall survival.
Funnel plots for the assessment of publication bias across outcomes. (A) Local recurrence. (B) Local progression- free survival. (C) Progression-free survival. (D) Overall survival. (E) Acute toxicity. (F) Late toxicity.
REFERENCES
- 1.Njei B, Rotman Y, Ditah I, Lim JK. Emerging trends in hepatocellular carcinoma incidence and mortality. Hepatology. 2015;61:191–199. doi: 10.1002/hep.27388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Nishikawa H, Osaki Y. Comparison of high-intensity focused ultrasound therapy and radiofrequency ablation for recurrent hepatocellular carcinoma. Hepatobiliary Surg Nutr. 2013;2:168–170. doi: 10.3978/j.issn.2304-3881.2013.03.01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Xu Q, Kobayashi S, Ye X, Meng X. Comparison of hepatic resection and radiofrequency ablation for small hepatocellular carcinoma: a meta-analysis of 16,103 patients. Sci Rep. 2014;4:7252. doi: 10.1038/srep07252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lee DH, Kim JW, Lee JM, Kim JM, Lee MW, Rhim H, et al. Laparoscopic liver resection versus percutaneous radiofrequency ablation for small single nodular hepatocellular carcinoma: comparison of treatment outcomes. Liver Cancer. 2021;10:25–37. doi: 10.1159/000510909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kang TW, Lim HK, Lee MW, Kim YS, Rhim H, Lee WJ, et al. Aggressive intrasegmental recurrence of hepatocellular carcinoma after radiofrequency ablation: risk factors and clinical significance. Radiology. 2015;276:274–285. doi: 10.1148/radiol.15141215. [DOI] [PubMed] [Google Scholar]
- 6.Apisarnthanarax S, Barry A, Cao M, Czito B, DeMatteo R, Drinane M, et al. External beam radiation therapy for primary liver cancers: an ASTRO clinical practice guideline. Pract Radiat Oncol. 2022;12:28–51. doi: 10.1016/j.prro.2021.09.004. [DOI] [PubMed] [Google Scholar]
- 7.Dudzinski SO, Newman NB, McIntyre J, Engineer R, Sanford NN, Wo JY, et al. Emerging evidence-based role for external-beam radiation therapy in hepatocellular carcinoma. Lancet Gastroenterol Hepatol. 2025;10:387–398. doi: 10.1016/S2468-1253(24)00267-X. [DOI] [PubMed] [Google Scholar]
- 8.European Association for the Study of the Liver EASL clinical practice guidelines on the management of hepatocellular carcinoma. J Hepatol. 2025;82:315–374. doi: 10.1016/j.jhep.2024.08.028. [DOI] [PubMed] [Google Scholar]
- 9. National Comprehensive Cancer Network (NCCN). NCCN clinical practice guidelines in oncology (NCCN Guidelines®): hepatocellular carcinoma. NCCN web site, < https://www.nccn.org/guidelines/guidelines-detail?category=1&id=1514 version 2.2026>. Accessed 9 Oct 2025.
- 10.Teng W, Wang HW, Lin SM, Diagnosis Group and Systemic Therapy Group of TLCA Management consensus guidelines for hepatocellular carcinoma: 2023 update on surveillance, diagnosis, systemic treatment, and posttreatment monitoring by the Taiwan Liver Cancer Association and the Gastroenterological Society of Taiwan. Liver Cancer. 2024;13:468–486. doi: 10.1159/000537686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Liu X, Xia F, Chen Y, Sun H, Yang Z, Chen B, et al. Chinese expert consensus on refined diagnosis, treatment, and management of advanced primary liver cancer (2023 edition) Liver Res. 2024;8:61–71. doi: 10.1016/j.livres.2024.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Korean Liver Cancer Association. National Cancer Center 2018 Korean Liver Cancer Association-National Cancer Center Korea practice guidelines for the management of hepatocellular carcinoma. Gut Liver. 2019;13:227–299. doi: 10.5009/gnl19024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ma ZH, Lin XL, Liu FH, Zhang JL, Yan ML, Song XC, et al. Radiofrequency ablation versus stereotactic body radiotherapy for recurrent hepatocellular carcinoma: a multicenter, propensity score matching analysis. BMC Cancer. 2025;25:424. doi: 10.1186/s12885-025-13800-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yang Z, Liu S, Hu L, Chen J, Wang J, Pan Y, et al. Stereotactic body radiotherapy is an alternative to radiofrequency ablation for single HCC ≤5.0 cm. JHEP Rep. 2024;6:101151. doi: 10.1016/j.jhepr.2024.101151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Seo SH, Yu JI, Park HC, Yoo GS, Choi MS, Rhim H, et al. Proton beam radiotherapy as a curative alternative to radiofrequency ablation for newly diagnosed hepatocellular carcinoma. Anticancer Res. 2024;44:2219–2230. doi: 10.21873/anticanres.17029. [DOI] [PubMed] [Google Scholar]
- 16.Maher AM, Shanker M, Liu HYH, Lee Y, Leggett D, Hodgkinson P, et al. Comparison of outcomes following surgical resection, percutaneous ablation or stereotactic body radiation therapy in early-stage, solitary and small (≤3 cm) treatmentnaïve hepatocellular carcinoma. Cancer Med. 2024;13:e6978. doi: 10.1002/cam4.6978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Moon AM, Kim HP, Singal AG, Owen D, Mendiratta-Lala M, Parikh ND, et al. Thermal ablation compared to stereotactic body radiation therapy for hepatocellular carcinoma: a multicenter retrospective comparative study. Hepatol Commun. 2023;7:e00184. doi: 10.1097/HC9.0000000000000184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ueno M, Takabatake H, Itasaka S, Kayahara T, Morimoto Y, Yamamoto H, et al. Stereotactic body radiation therapy versus radiofrequency ablation for single small hepatocellular carcinoma: a propensity-score matching analysis of their impact on liver function and clinical outcomes. J Gastrointest Oncol. 2021;12:2334–2344. doi: 10.21037/jgo-21-356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jeong Y, Lee KJ, Lee SJ, Shin YM, Kim MJ, Lim YS, et al. Radiofrequency ablation versus stereotactic body radiation therapy for small (≤ 3 cm) hepatocellular carcinoma: a retrospective comparison analysis. J Gastroenterol Hepatol. 2021;36:1962–1970. doi: 10.1111/jgh.15442. [DOI] [PubMed] [Google Scholar]
- 20.Kim N, Cheng J, Jung I, Liang JD, Shih YL, Huang WY, et al. Stereotactic body radiation therapy vs. radiofrequency ablation in Asian patients with hepatocellular carcinoma. J Hepatol. 2020;73:121–129. doi: 10.1016/j.jhep.2020.03.005. [DOI] [PubMed] [Google Scholar]
- 21.Wahl DR, Stenmark MH, Tao Y, Pollom EL, Caoili EM, Lawrence TS, et al. Outcomes after stereotactic body radiotherapy or radiofrequency ablation for hepatocellular carcinoma. J Clin Oncol. 2016;34:452–459. doi: 10.1200/JCO.2015.61.4925. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Shiozawa K, Watanabe M, Ikehara T, Matsukiyo Y, Kogame M, Kishimoto Y, et al. Comparison of percutaneous radiofrequency ablation and CyberKnife® for initial solitary hepatocellular carcinoma: a pilot study. World J Gastroenterol. 2015;21:13490–13499. doi: 10.3748/wjg.v21.i48.13490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kim TH, Koh YH, Kim BH, Kim MJ, Lee JH, Park B, et al. Proton beam radiotherapy vs. radiofrequency ablation for recurrent hepatocellular carcinoma: a randomized phase III trial. J Hepatol. 2021;74:603–612. doi: 10.1016/j.jhep.2020.09.026. [DOI] [PubMed] [Google Scholar]
- 24.Xi M, Yang Z, Hu L, Fu Y, Hu D, Zhou Z, et al. Radiofrequency ablation versus stereotactic body radiotherapy for recurrent small hepatocellular carcinoma: a randomized, open-label, controlled trial. J Clin Oncol. 2025;43:1073–1082. doi: 10.1200/JCO-24-01532. [DOI] [PubMed] [Google Scholar]
- 25.Pan YX, Fu YZ, Hu DD, Long Q, Wang JC, Xi M, et al. Stereotactic body radiotherapy vs. radiofrequency ablation in the treatment of hepatocellular carcinoma: a meta-analysis. Front Oncol. 2020;10:1639. doi: 10.3389/fonc.2020.01639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lee J, Shin IS, Yoon WS, Koom WS, Rim CH. Comparisons between radiofrequency ablation and stereotactic body radiotherapy for liver malignancies: meta-analyses and a systematic review. Radiother Oncol. 2020;145:63–70. doi: 10.1016/j.radonc.2019.12.004. [DOI] [PubMed] [Google Scholar]
- 27.Facciorusso A, Chierici A, Cincione I, Sacco R, Ramai D, Mohan BP, et al. Stereotactic body radiotherapy vs radiofrequency ablation for the treatment of hepatocellular carcinoma: a meta-analysis. Expert Rev Anticancer Ther. 2021;21:681–688. doi: 10.1080/14737140.2021.1891887. [DOI] [PubMed] [Google Scholar]
- 28.Zhang H, Chang N, Han T, Ma S, Qu G, Liu H, et al. Radiofrequency ablation versus stereotactic body radiotherapy for hepatocellular carcinoma: a meta-analysis. Future Oncol. 2021;17:4027–4040. doi: 10.2217/fon-2021-0263. [DOI] [PubMed] [Google Scholar]
- 29.Yang D, Lin K, Wang Y, Xie X, Xie X, Zhuang B. Stereotactic body radiation therapy versus radiofrequency ablation in hepatocellular carcinoma: an up-date meta-analysis. Abdom Radiol (NY) 2023;48:399–410. doi: 10.1007/s00261-022-03690-3. [DOI] [PubMed] [Google Scholar]
- 30.Rim CH, Lee JS, Kim SY, Seong J. Comparison of radiofrequency ablation and ablative external radiotherapy for the treatment of intrahepatic malignancies: a hybrid meta-analysis. JHEP Rep. 2022;5:100594. doi: 10.1016/j.jhepr.2022.100594. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. doi: 10.1136/bmj.l4898. [DOI] [PubMed] [Google Scholar]
- 33.Sterne JA, Hernán MA, Reeves BC, Savović J, Berkman ND, Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919. doi: 10.1136/bmj.i4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–188. doi: 10.1016/0197-2456(86)90046-2. [DOI] [PubMed] [Google Scholar]
- 35.Cochran WG. The comparison of percentages in matched samples. Biometrika. 1950;37:256–266. [PubMed] [Google Scholar]
- 36.Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–560. doi: 10.1136/bmj.327.7414.557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629–634. doi: 10.1136/bmj.315.7109.629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Han S, Sung PS, Park SY, Kim JW, Hong HP, Yoon JH, et al. Local ablation for hepatocellular carcinoma: 2024 expert consensus-based practical recommendations of the Korean Liver Cancer Association. Gut Liver. 2024;18:789–802. doi: 10.5009/gnl240350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Lee S, Kang TW, Cha DI, Song KD, Lee MW, Rhim H, et al. Radiofrequency ablation vs. surgery for perivascular hepatocellular carcinoma: propensity score analyses of long-term outcomes. J Hepatol. 2018;69:70–78. doi: 10.1016/j.jhep.2018.02.026. [DOI] [PubMed] [Google Scholar]
- 40.Song KD, Lim HK, Rhim H, Lee MW, Kang TW, Paik YH, et al. Hepatic resection vs percutaneous radiofrequency ablation of hepatocellular carcinoma abutting right diaphragm. World J Gastrointest Oncol. 2019;11:227–237. doi: 10.4251/wjgo.v11.i3.227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Wang L, Ke Q, Huang Q, Shao L, Chen J, Wu J. Stereotactic body radiotherapy versus radiofrequency ablation for hepatocellular carcinoma: a systematic review and meta-analysis. Int J Hyperthermia. 2020;37:1313–1321. doi: 10.1080/02656736.2020.1843719. [DOI] [PubMed] [Google Scholar]
- 42.Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet. 2018;391:1301–1314. doi: 10.1016/S0140-6736(18)30010-2. [DOI] [PubMed] [Google Scholar]
- 43.Tabrizian P, Jibara G, Shrager B, Schwartz M, Roayaie S. Recurrence of hepatocellular cancer after resection: patterns, treatments, and prognosis. Ann Surg. 2015;261:947–955. doi: 10.1097/SLA.0000000000000710. [DOI] [PubMed] [Google Scholar]
- 44.Khosla D, Singh G, Kapoor R. Understanding survival comparisons in nonrandomized treatment comparisons for patients with early-stage HCC. Hepatol Commun. 2024;8:e0281. doi: 10.1097/HC9.0000000000000281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Lee SU, Kim TH. Current evidence and the potential role of proton beam therapy for hepatocellular carcinoma. Clin Mol Hepatol. 2023;29:958–968. doi: 10.3350/cmh.2023.0274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Yoo GS, Yu JI, Park HC. Proton therapy for hepatocellular carcinoma: current knowledges and future perspectives. World J Gastroenterol. 2018;24:3090–3100. doi: 10.3748/wjg.v24.i28.3090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Hsieh RC, Lee CH, Huang HC, Wu SW, Chou CY, Hung SP, et al. Clinical and dosimetric results of proton or photon radiation therapy for large (>5 cm) hepatocellular carcinoma: a retrospective analysis. Int J Radiat Oncol Biol Phys. 2024;118:712–724. doi: 10.1016/j.ijrobp.2023.09.049. [DOI] [PubMed] [Google Scholar]
- 48.Uchinami Y, Katoh N, Abo D, Morita R, Taguchi H, Fujita Y, et al. Study of hepatic toxicity in small liver tumors after photon or proton therapy based on factors predicting the benefits of proton. Br J Radiol. 2023;96:20220720. doi: 10.1259/bjr.20220720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Abousaida B, Seneviratne D, Hoppe BS, Ko SJ, Asaithamby A, Cucinotta FA, et al. Carbon ion radiotherapy in the management of hepatocellular carcinoma. J Hepatocell Carcinoma. 2021;8:1169–1179. doi: 10.2147/JHC.S292516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Byun HK, Kim C, Seong J. Carbon ion radiotherapy in the treatment of hepatocellular carcinoma. Clin Mol Hepatol. 2023;29:945–957. doi: 10.3350/cmh.2023.0217. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
PICO framework defining the clinical question of this meta-analysis
Database search strategy
Univariate meta-regression analysis for overall survival
Summary of treatment-related toxicities in the included studies comparing external beam radiotherapy and local ablation therapy
GRADE certainty-of-evidence assessment for all clinical outcomes
Risk-of-bias assessment of included studies. (A) RoB 2 assessment for randomized controlled trials. (B) Summary distribution of ROBINS-I risk-of-bias ratings across non-randomized studies. (C) ROBINS-I traffic-light plot showing domain-level risk-of-bias judgments for each non-randomized study. RoB 2, Risk of Bias 2; ROBINS-I, Risk of Bias in Non-randomized Studies of Interventions.
Subgroup analyses of overall survival by disease setting (primary vs. recurrent HCC) between EBRT and LAT. CI, confidence interval; EBRT, external beam radiotherapy; HCC, hepatocellular carcinoma; HR, hazard ratios; LAT, local ablation therapy.
Leave-one-out sensitivity analysis for overall survival.
Funnel plots for the assessment of publication bias across outcomes. (A) Local recurrence. (B) Local progression- free survival. (C) Progression-free survival. (D) Overall survival. (E) Acute toxicity. (F) Late toxicity.
