The impacts of physical factors on huge hepatocellular carcinoma treated by transarterial chemoembolization combined with radiotherapy

Abstract

Aims: To assess the influence of various physical factors on the outcome of transarterial chemoembolization combined with γ-ray hypofractionated radiation therapy (TACE-γHRT) for unresectable huge (≥10 cm) hepatocellular carcinoma (UH-HCC) patients.

Materials & methods: A total of 162 UH-HCC patients with different tumor locations treated with TACE-γHRT and a retrospective analysis was conducted to evaluate the impacts of selected physical parameters on clinical outcomes.

Results: The selected physical factors influenced the clinical outcomes significantly. No adverse events exceeding grade 3 were observed in the enrolled patients.

Conclusion: Higher P₇₀ and marginal dose, smaller tumor size and tumor location of neither skin nor gastrointestinal tracts involved were independent predictors for better overall survival and progression free survival.

Plain language summary

Article highlights.

• The option for the treatment of UH-HCC is very limited in clinical practice.

• TACE-γHRT is an efficient and feasible modality for the management of UH-HCC.

• The modality of TACE-γHRTa may be a better choice than the TACE-γHRTs for patients with UH-HCC.

• UH-HCC should be further stratified according to the tumor location for tailored treatment.

• Tumor location, tumor size, marginal dose, and P₇₀ played significant role in UH-HCC patients treated with TACE-γHRT.

• No severe adverse events were observed in any of the patients.

• NSNG patients can achieve better results by comparison with other location patients.

• There is a dearth of more data on effective treatments tailored to the subset of UH-HCC patients.

1. Introduction

The incidence of hepatocellular carcinoma (HCC) has been steadily increasing on a global scale [1–3]. It is projected that the incidence of HCC will exceed one million cases by 2025 [4]. Huge HCC (H-HCC), defined as having a diameter of 10 cm or more, accounts for approximately 20% of all HCC cases [5]. Various treatment modalities have been recommended for managing small to medium-sized HCC, including radiofrequency ablation (RFA), transarterial chemoembolization (TACE) [6,7] and hepatic resection [8]. However, H-HCC patients face limited options due to factors such as poor hepatic function, advanced disease stage, or other contraindications. Although TACE has been explored as a potential management approach for H-HCCs [9,10], postoperative adjuvant TACE has shown benefits for patients at high risk of tumor recurrence [11]. Nevertheless, the efficacy of TACE as a standalone treatment for H-HCC is generally suboptimal [12]. To date, surgical resection remains the primary curative treatment for H-HCC [13,14]. However, a notable challenge associated with resection is the high recurrence rate in HCC cases [10,15], with H-HCC patients experiencing a reported 5-year recurrence rate of 60–80% following surgery [16]. This substantial recurrence rate significantly compromises long-term patient survival [15]. Moreover, resection is a viable option only for a highly selected group of H-HCC patients, with fewer than 30% considered suitable candidates for surgical tumor removal [17]. One of the paramount considerations in managing HCC is the preservation of liver function. Initial radiotherapy for HCC did not adequately protect normal liver tissue [18], and its effectiveness was consequently limited. Thus, radiotherapy was historically considered an unsuitable option for HCC patients. However, with recent advancements in computer technology and treatment techniques, radiotherapy now offers improved protection of normal liver tissues, leading to enhanced efficacy in treating HCC [19]. Several studies have highlighted the promising therapeutic outcomes of stereotactic body radiotherapy (SBRT) for HCC [20–22]. In fact, SBRT has even been regarded as the primary treatment option for unresectable HCC [23]. Nevertheless, there is a paucity of reports on the management of unresectable huge HCC (UH-HCC). Our previous work [24] is one of the limited works to report the treatment of UH-HCC. Consequently, there is a dearth of more data on effective treatments tailored to the subset of UH-HCC patients.

In this study, we retrospectively analyzed the impacts of various physical factors, including P_x (the percentage of tumor volume encompassed by x% isodose curve within the entire tumor), marginal dose, fractional dose, tumor size, tumor location and treatment posture (supine or prone), on the clinical outcomes of UH-HCC patients undergoing TACE combined with γ-ray hypfraction radiotherapy (TACE-γHRT).

2. Materials & methods

2.1. Patient cohort

This study was conducted with the approval of the Ethics Committee of our hospital(900th hospital of PLA), and prior written consent was obtained from all patients before undergoing TACE-γHRT. A total of 1123 patients with HCC received treatment at our department between May 2009 and May 2022. The eligibility criteria were described in our previous work. Briefly, 1) Diagnosis of UH-HCC; 2) No previous history of liver radiotherapy; 3) No extrahepatic metastasis; 4) Classified as Child-Pugh (CP) Class A or B; and 5) Completion of TACE followed by γHRT as part of their treatment. Patients were treated with TACE-γHRT solely in either the supine or prone position (TACE-γHRTs) before December 2013, whereas those were treated in both supine and prone positions alternately (TACE-γHRTa) between January 2014 and May 2022. Ultimately, a total of 59 patients were treated with TACE-γHRTs, and 103 patients received TACE-γHRTa, making them eligible for inclusion in this study. According to tumor location, patients with the minimum distances between the edge of tumors and skin ≤3 cm were classified as skin involved patients (n = 62, 23 patients treated with TACE-γHRTs and 39 patients treated with TACE-γHRTa); Patients with the minimum distances between the edge of tumors and gastrointestinal tracts (GITs) ≤0.5 cm were classified as GITs involved patients (n = 53, 22 patients treated with TACE-γHRTs and 31 patients treated with TACE-γHRTa); The remainder of the patients were classified as neither skin nor GITs involved (NSNG, n = 47, 14 patients treated with TACE-γHRTs and 33 patients treated with TACE-γHRTa). Some tumors (n = 21) involved both skin and GITs. Therefore, patients were ultimately stratified into skin involved without GITs involved (SIWG, n = 51), GITs involved without skin involved (GIWS, n = 43), both skin and GITs involved (BSAG, n = 21) and NSNG (n = 47).

2.2. Treatment

Patients underwent one course of TACE. Selective digital subtraction angiography of superior mesenteric artery, celiac artery and hepatic artery was performed to identify tumor arterial supply. TACE procedures involved the infusion of iodizel and cisplatin, followed by the use of gelatin sponge cubes for embolization. Computed tomography (CT) was utilized to confirm the coverage areas. Great care was taken in the selection of tumor-feeding vessels to optimize embolization and preserve liver function.

Patients underwent γHRT treatment within 2 to 4 weeks after TACE. The γHRT was administered using a γ-ray body radiotherapy system. Patients treated with TACE-γHRTs received treatment in either the supine or prone position alone, whereas patients treated with TACE-γHRTa received treatment in both supine and prone positions alternately. All treatment plans meeting the following criteria were considered eligible: 1) Well-tolerance of normal tissues; 2) Planning target volume (PTV) enveloped by 50% or 55% isodose lines; 3) Presence of ≥70% isodose curves within the gross tumor volume (GTV); And 4) prescription dose normalized at the 50% or 55% isodose curve. Treatment plans for patients undergoing TACE-γHRTa had to meet additional criteria, including: a) The 60% isodose curve encompassing at least 90% of the PTV; b) The 70% isodose curve encompassing at least 60% of the PTV; and c) One plan for prostrate position and the other for prone position. The prescription dose was determined based on the function of the remaining liver tissue and the predicted toxicities of other normal tissues. Each patient received one fraction of radiation therapy per day, with a one-day interval for rest after every six consecutive fractions, resulting in a treatment course lasting 12 to 14 days.

2.3. Statistical analysis

The response was assessed by considering the initial tumor volume (pre-TACE) and the modified Response Evaluation Criteria in Solid Tumors (mRECIST) was used in the evaluation of the response. Overall survival (OS) was calculated starting from the first day of TACE-γHRT treatment to the last follow-up or death. Progression-free survival (PFS) was calculated starting from the first day of TACE-γHRT treatment to the progression or death. Kaplan-Meier curves were employed for OS and PFS assessment, while the log-rank test was utilized to compare these survival metrics between patients subjected to the two treatment modalities. Multivariate analysis of the relationship between OS (or PFS) and various parameters was performed using the forward stepwise Cox regression model. And statistical significance was determined at a two - tailed p value of <0.05. In order to minimize the effects of selection biases, propensity score matching (PSM) analysis was performed. The association among covariates was evaluated through Spearman correlation or Cramer's V coefficients. Statistical analysis was conducted using SPSS version 22 (SPSS Inc., Chicago, IL) software.

3. Results

The detailed characteristics of all enrolled patients are summarized in Table 1. HCC diagnosis was confirmed through histological or cytological evidence (n = 135), radiological evidence combined with elevated alpha-fetoprotein (AFP) levels (>400 ng/ml; n = 11), or a minimum of two types of radiological evidence (n = 16). The marginal dose, mean dose and fractional dose were as follows: 38.4 ± 3 Gy, 51.2 ± 2.6 Gy and 3 ± 0.2 Gy for the SIWG patients, 36.8 ± 2.8 Gy, 48.6 ± 2.4 Gy and 2.8 ± 0.2 Gy for the GIWS patients, 36.8 ± 2.8 Gy, 48.6 ± 2.2 Gy and 2.8 ± 0.2 Gy for the BSAG patients, 41.6 ± 3.2 Gy, 55.3 ± 3.1 Gy and 3.4 ± 0.4 Gy for the NSNG patients, respectively.

Table 1.

Demographic and clinical characteristics of patients enrolled in this study (n = 162).

Characteristics	Tumor location (no. of patient, %)
	SIWG	GIWS	BSAG	NSNG	p-value
Treatment modality					.003
TACE-γHRTs	19 (11.7)	18 (11.1)	8 (4.9)	14 (8.6)
TACE-γHRTa	32 (19.8)	25 (15.4)	13 (8)	33 (20.4)
Gender					.621
male	26 (16)	23 (14.2)	14 (8.6)	24 (14.8)
female	25 (15.4)	20 (12.3)	7 (4.3)	23 (14.2)
ECOG PS					.038
0–1	27 (16.7)	22 (13.6)	12 (7.4)	24 (14.9)
2	24 (14.8)	21 (13.0)	9 (5.6)	23 (14.2)
Tumor size (cm)					.008
median	14.8	14.6	14.8	15.3
range	10.3–18.3	10.8–18.0	10.4–18.2	10.2–18.5
BCLC stage				.	.043
A-B	3 (1.8)	5 (3.1)	2 (1.2)	3 (1.8)
C	48 (29.6)	38 (23.5)	19 (11.7)	44 (27.2)
Child-Pugh class					.054
A	27 (16.7)	23 (14.2)	12 (7.4)	29 (17.9)
B	24 (14.8)	20 (12.3)	9 (5.6)	18 (11.1)
AFP (ng/ml)					.035
>400	41 (25.3)	32 (19.8)	19 (11.7)	39 (24.1)
≤400	10 (6.2)	11 (6.8)	2 (1.2)	8 (4.9)
C/H confirmation					.732
Yes	41 (25.3)	34 (21)	21 (13.0)	39 (25.3)
No	10 (6.2)	9 (5.6)	0	8 (2.2)
Dose					.016
marginal	38.4 ± 3 Gy	36.8 ± 2.8 Gy	36.8 ± 2.8 Gy	41.6 ± 3.2 Gy
mean	51.2 ± 2.6 Gy	48.6 ± 2.4 Gy	48.6 ± 2.2 Gy	55.3 ± 3.1 Gy
P70					.006
median	51.3	50.9	50.6	51.6
range	34.6–72.4	31.8–71.9.	31.9–71.6	34.7–73.3

BCLC: Barcelona clinic liver cancer; ECOG PS: Eastern Cooperative Oncology Group performance status.

Routine blood and liver function assessments were conducted on a weekly basis throughout the treatment course. To evaluate tumor size within the irradiated fields, all patients underwent abdominal MRI examinations and liver function assessments monthly for the first 3 months following completion of γHRT and subsequently at intervals of 3–6 months.

Within the enrolled patients, the objective response (OR) rate was 87.7%. Among these, 23 (14.2%) patients achieved complete response (CR), while 119 (73.5%) patients achieved partial response (PR) within 6 months. Additionally, 13 patients (8%) exhibited stable disease (SD) and 7 patients (4.3%) experienced progressive disease (PD) during the same period. NSNG cohort achieved 100% of OR rate which was constituted of 29.8% (14/47) of CR rate and 70.2% (33/47) of PR rate. The CR, PR, SD and PD rates for SIWG patients were 13.7% (7/51), 74.5% (38/51), 9.8% (5/51) and 2% (1/51), respectively. And the GIWS patients achieved 4.7% (2/43), 74.4% (32/43), 11.6 (5/43) and 9.3% (4/43) of CR, PR, SD and PD rates, respectively. The CR, PR, SD and PD rates for BSAG cohort of patients were 0% (0/21), 76.2% (16/21), 14.3% (3/21) and 9.5% (2/21), respectively. The differences in tumor responses between patients with different tumor locations are illustrated in Figure 1. These findings indicate that the NSNG patients, in comparison to other cohorts of patients, can achieve more favorable effectiveness, particularly in terms of achieving OR.

Figure 1.

The response of patients with different tumor locations.

According to the Kaplan-Meier curve analysis, the median OS for the entire patients was 18.2 months. Among NSNG patients, the median OS reached 36.3 months, while the median OS for SIWG, GIWS and BSAG patients were 22.3, 14.9, 8.9 months respectively. The median PFS for all enrolled patients was 15.3 months, with SIWG patients experiencing a median PFS of 18.8 months, GIWS patients showing a median PFS of 8.7 months, NSNG patients exhibiting a median PFS of 31.8 months, and BSAG patients experiencing a median PFS of 6.6 months. For OS rates at 1, 3 and 5 years, they were as follows: 57.8%, 21.3% and 9.2% for the entire patient cohort; 85.1%, 53.2% and 26.5% for NSNG patients; 69.2%, 21.1% and 5.8% for SIWG patients; 52.3%, 2.3% and 0% for GIWS patients and 9.5%, 0% and 0% for BSAG patients, respectively. Regarding PFS rates at the same intervals, they stood at 52.8%, 17.7% and 6.8% for all enrolled patients; 81.4%, 42.6% and 19.2% for NSNG patients; 51.4%, 16.3% and 4.6% for SIWG patients;. 46.5%, 2.1% and 0% for GIWS patients and 4.8%, 0%, 0% for BSAG patients, respectively. Figures 2A, F depict the OS and PFS for the entire patients with different tumor locations treated with TACE-γHRT.

Figure 2.

OS of the enrolled patients (A); comparison of OS for SIWG (B), GIWS (C), NSNG (D) and BSAG (E) patients treated with TACE-γHRTs and TACE-γHRTa; PFS of the enrolled patients (F).

In SIWG patients, those treated with TACE-γHRTa achieved much longer median OS and PFS than those treated with TACE-γHRTs (The median OS of 28.8 months versus 12.3 months, P < 0.001; The median PFS of 24.6 months versus 8.7 months, P < 0.001, respectively). The GIWS, NSNG and the BSAG patients undergoing TACE-γHRTa had, to some extent, longer OS and PFS also than those treated with TACE-γHRTs. Figures 2B–D and Figure 2E showed the differences of OS for the SIWG, GIWS, NSNG and the BSAG patients treated with different treatment modalities. The median PFS were 8.9 versus 8.1 months (p = 0.037) for GIWS patients treated with TACE-γHRTa and TACE-γHRTs. And 32.4 versus 24.8 months (P < .001), 6.8 versus 6.3 months (p = 0.052) were the PFS achieved for NSNG, BSAG patients subjected to TACE-γHRTa and TACE-γHRTs.

The Spearman correlation coefficient of 0.947 demonstrates a positive correlation between marginal and fractional dose. Additionally, a positive correlation with a coefficient of 0.714 was observed between P₆₀ and P₇₀. The positive correlations between these variables are presented in Tables 2 & 3. Because of the positive correlation, the multivariable analysis selected marginal dose and P₇₀ over fractional dose and P₆₀ for further evaluation.

Table 2.

Correlations between marginal and fractional dose.

			Fractional dose	Marginal dose
Spearman's rho	Fractional dese	Correlations Coefficient	1.000	.947a
		Sig. (2-tailed)		.000
		N	162	162
	Marginal dose	Correlation Coefficient	.947a
		Sig. (2-tailed)	.000
		N	162	162

^aCorrelation is significant at the 0.01 level (2-tailed).

Table 3.

Correlations between P₇₀ and P₆₀.

			P60	P70
Spearman's rho	P60	Correlations Coefficient	1.000	.714a
		Sig. (2-tailed)		.000
		N	162	162
	P70	Correlation Coefficient	.714a
		Sig. (2-tailed)	.000
		N	162	162

^aCorrelation is significant at the 0.01 level (2-tailed).

Multivariate Cox regression analysis after PSM identified several physical factors associated with OS in patients with UH-HCC when employing the Cox regression hazard model (p = 0.05). These factors include tumor location (p = 0.017), treatment modality (p = 0.005), P₇₀ (p = 0.002), tumor size (p = 0.012) and the marginal dose (p = 0.009). The outcomes of the Cox regression analysis regarding OS after PSM are presented in Figure 3A. The results indicate TACE-γHRTa, in comparison to TACE-γHRTs, may offer superior OS for UH-HCC patients. Additionally, a higher marginal dose, increased P₇₀, and smaller tumor size were identified as independent predictors for improved OS. Moreover, The multivariate Cox regression analysis after PSM indicated that a higher marginal dose (p = 0.007) and P₇₀ (p = 0.032), tumor location of NSNG (p = 0.021), the treatment modality of TACE-γHRTa (p = 0.008), and smaller tumor size (p = 0.018) were independent predictors for enhanced PFS. The results of the Cox regression analysis for PFS after PSM are detailed in Figure 3B. These results suggest that NSNG patients can achieve better OS and PFS than other tumor location patients in the treatment of UH-HCC by TACE-γHRT.

Figure 3.

The forest plot of Cox regression analysis regarding OS after PSM (A); the forest plot of the Cox regression analysis for PFS after PSM (B).

All enrolled patients completed their TACE-γHRT treatment successfully. A reduction in leukocyte count was observed in every patient during γHRT treatment. Most patients (n = 118; 72.8%) experienced the thrombocytopenia following the completion of γHRT, and a reduction of leukocyte count occurred in some patients (n = 44; 27.2%) during γHRT. However, both leukocyte and platelet counts returned to normal levels after appropriate management. Other mild complications, such as grade 1 dermatitis, nausea, diarrhea, and so on, also occurred in some patients. But these symptoms spontaneously resolved within a few days or weeks after treatment completion. Among SIWG patients (n = 51), radiation-induced dermatitis was noted in 41 (80.4%) patients between 1 to 3 months after concluding γHRT. Of these cases, 15 (29.4%) were classified as grade 1, and 24 (47.1%) as grade 2. Unfortunately, 2 (3.9%) patients experienced grade 3 of radiation-induced dermatitis, which posed considerable challenges. The grade 3 radiation-induced dermatitis was also observed in BSAG patients (n = 1, 4.8%). However, the grade 3 radiation-induced dermatitis did not observed in NSNG or GIWS patients. Moreover, the incidence of radiation-induced dermatitis for NSNG and GIWS patients were both lower by comparison. The grade 1, 2, and 3 of radiation-induced dermatitis incidences for NSNG and GIWS patients were 8.5% (n = 4), 4.3% (n = 2), and 0%, 18.6% (n = 8), 11.6% (n = 5), and 0%, respectively. Those were 29.4% (n = 15), 47.1% (n = 24), 3.9% (n = 2) for SIWG patients and 14.3% (n = 3), 42.9% (n = 9), 4.8% (n = 1) for BSAG patients, respectively. The incidences of radiation-induced dermatitis of grade 2 and 3 for SIWG patients treated with TACE-γHRTa declined from 84.2% to 25% (P < .001) and 10.5% to 0% (P < .001) by comparison with those treated with TACE-γHRTs. The receiver operating characteristic (ROC) curve analysis for SIWG patiens indicated that a larger tumor size could predict more severe dermatitis in patients treated with TACE-γHRTs (Figure 4A). However, tumor size did not predict dermatitis severity in patients treated with TACE-γHRTa (Figure 4B). These findings suggest that TACE-γHRTa may offer a more favorable toxicity profile than TACE-γHRTs in the treatment of SIWG UH-HCC patients. In terms of radiation-induced dermatitis incidence, there were no significant differences between patients treated with TACE-γHRTa and TACE-γHRTs for other patients. The grade 1, 2, and 3 of radiation-induced dermatitis incidences were 7.1%, 7.1%, and 0% for NSNG patients treated with TACE-γHRTa and 9.1%, 3%, and 0% for those treated withTACE-γHRTs, respectively. Those incidences were 20%, 12%, and 0% for GIWS patients treated with TACE-γHRTs and 16.7%, 11.1% and 0% for patients treated with TACE-γHRTa, respectively. Those incidences were 12.5%, 37.5%, and 12.5% for BSAG patients treated with TACE-γHRTs and 15.4%, 46.2%, and 0% for patients treated with TACE-γHRTa, respectively. By comparison with other patients, the BSAG patients experienced not only higher incidence of radiation-induced dermatitis. The BSAG patients also experienced higher incidence of gastroenteritis which was composed of 23.8% (n = 5) of mild gastroenteritis, 38.1% (n = 8) of moderate gastroenteritis, and 4.8% (n = 1) of severe gastroenteritis. Additionally, the GIWS patients suffered higher incidence of gastroenteritis also. Moreover, the higher incidence of gastroenteritis for BSAG and GIWS patients could not be reduced by changed the treatment modality of TACE-γHRTs into TACE-γHRTa. The 25%, 37.5%, 12.5% of mild, moderate, and severe gastroenteritis incidences were the BSAG patients suffered when patients were treated with TACE-γHRTs. By comparison, those incidences were 23%, 38.4%, 0% when the BSAG patients were treated with TACE-γHRTa. Those incidences were 22.2%, 44.4%, 5.6% for GIWS patients treated with TACE-γHRTs and 24%, 36%, 4% for those treated with TACE-γHRTa, respectively. No other adverse events (AEs) of grade 3 or higher were observed among the enrolled patients. The main AEs occurred in the cohorts of patients are shown in Figure 5.

Figure 4.

The ROC curve made between dermatitis and tumor size for SIWG patients treated with TACE-γHRTs (A) and treated with TACE-γHRTa (B).

Figure 5.

The comparison of main AEs occurred in the cohorts of patients.

4. Discussion

Our findings strongly suggest that TACE-γHRTa presents a promising treatment modality for patients with UH-HCC. Several key physical factors, including treatment modality, P₇₀, marginal dose, tumor size, and tumor location, exert significant impacts on clinical outcomes. Specifically, TACE-γHRTa, higher P₇₀, marginal dose, smaller tumor size, and tumor location of NSNG, were identified as independent predictors for improved OS. Moreover, these factors were also identified as independent predictors for superior PFS.

The outcomes of this study support the efficacy of TACE-γHRT in the treatment of UH-HCC patients. Firstly, the OS rates observed in UH-HCC patients treated with TACE-γHRT in our study align closely with those reported in previous studies [25,26]. Notably, one study by Shiro M et al. [25] reported a 16% dropout rate among patients, although their study included a much smaller cohort (25 patients) compared with our study (162 patients). Another study by Lo CH et al. [26] did not specify the UH-HCC. Moreover, the 1-, 3-year OS rates of 57.8%, 21.3% for all enrolled patients in our study mathed with the 1-, 2-year rates (46.5%, 23.9%) reported by Kim YJ et al. [27]. Secondly, the OR rate (87.7%) for all enrolled patients in our study closely matches the OR rates (88.4%, 90.1%, and 91%) reported in previous studies [28–30], although their studies did not specifically focus on huge HCC. Considering the the factors of UH-HCC in our study, we deemed TACE-γHRT to be an efficient treatment modaligy for UH-HCC.

The outcomes of this study also indicate the TACE-γHRT is a safe treatment modality for UH-HCC patients. Although previous studies have demonstrated low radiation therapy (RT) toxicity [31,32], some studies have reported grade 3 AEs [33,34]. In our study, grade 3 AE (severe gastroenteritis) occurred in GIWS and BSAG cohorts. And radiation-induced grade 3 dermatitis was observed in SIWG and BSAG patients treated with TACE-γHRTs. By comparison with the grade 3 AEs reported by Aditya J et al. [22] or the grade 4 toxicities in the study of Nalee K et al. [35], we deemed the toxicities in our study to be within acceptable limits. Notably, no AEs of grade 3 or higher occurred in NSNG cohort. The grade 3 dermatitis was not observed also when patients were treated with TACE-γHRTa. The rates of grade 2 radiation-induced dermatitis were also substantially lower in TACE-γHRTa patients compared with TACE-γHRTs patients. Therefore, TACE-γHRTa is a safe treatment modality for UH-HCC, especially for NSNG or SIWG patients.

The outcomes of our study demonstrated that UH-HCC patients treated with TACE-γHRTa achieved longer median OS and higher objective response (OR) rates than those treated with TACE-γHRTs. Especially, SIWG cohort treated with TACE-γHRTa could achieve much longer median OS than those treated with TACE-γHRTs (28.8 months versus 12.3 months). These improvements can primarily be attributed to modifications made in the treatment protocols. Specifically, the prescription dose normalization at the 50% or 55% isodose curve was consistent for both TACE-γHRTs and TACE-γHRTa patients. However, TACE-γHRTa treatment plans were required to meet the criteria of the 70% isodose curve encompassing at least 60% of the PTV, with a limitation on the 70% isodose curve within the GTVs. In contrast, no such requirements were imposed on the treatment plans for TACE-γHRTs patients. These differences resulted in higher doses within the GTVs of TACE-γHRTa patients. Additionally, the marginal doses for TACE-γHRTa patients were higher than those for TACE-γHRTs patients, contributing to the observed longer OS. This rationale aligns with the findings of a study by Teraoka Y et al. [36], which highlighted the positive correlation between higher dose levels and increased OS. The enhanced OR rate observed in our study can also be attributed to the higher doses within the GTVs and the increased marginal dose surrounding the PTVs. It is worth noting that NSNG cohort in our study have achieved a 100% of OR rate. We also observed that TACE-γHRTa exhibited more favorable toxicity profiles than TACE-γHRTs in the treatment of UH-HCC.

The results in our study indicated the tumor location played an important role in the UH-HCC patients treated with TACE-γHRT. Firstly, the tumor location of NSNG patients can achieve longer OS, PFS, higher OR, and more favorable toxicities by comparison with other cohorts of patients. It is worth noting that NSNG cohort in our study have achieved a 100% of OR rate. Furthermore, both the 1-, 3-, 5-year OS rates (85.1%, 53.2%, 26.5%) and PFS rates (81.4%, 42.6%, 19.2%) for NSNG patients in our study are comparable with the corresponding rates of huge HCC treated by liver resection in the report of Ehab E et al. [37] (85.1%, 53.2%, 26.5% versus 76.6%, 39.5%, 39.5%; 81.4%, 42.6%, 19.2% versus 72.2%, 37.5%, 0%, respectively). The 1-, 3-year OS rates (85.1%, 53.2%) for NSNG patients in this study also matched the results reported in previous work [38,39], in which the 1-, 3-year OS rates of 88.8%, 69% and 80.2%, 55.4%for large and huge HCC treated by curative resection were reported, respectively. Interestingly, our study revealed a bit better 5-year OS rate (26.5%) for NSNG patients when compared with the 2-year OS rate (24%) of UH-HCC patients treated with proton beam therapy [40]. Additionally, no grade 3 or higher AE was observed in NSNG patients. The incidences of grade 1, 2 AE for NSNG cohort were much lower than other cohorts of patients cohort. Secondly, the treatment modality of TACE-γHRTa can reduce the incidences of radiation-induced grade 3 dermatitis for SIWG and BSAG patients. The results of ROC in our study illustrates the bigger tumor size predicts more severe dermatitis for SIWG UH-HCC treated with TACE-γHRTs. But the tumor size cannot predict severe dermatitis when SIWG UH-HCC patients are treated with TACE-γHRTa. Although cannot the bigger tumor size predict other AEs for other UH-HCC patients either, it predict worse OS and PFS. The impact of tumor size on OS/PFS aligns with the conclusion reported by Kim YJ et al. [27]. However, the grade 3 AE (severe gastroenteritis) observed in GIWS and BSAG cohorts cannot be reduced by changing the treatment modality of TACE-γHRTs into TACE-γHRTa. Otherwise, the differences of OS between SIWG patients treated with TACE-γHRTa and TACE-γHRTs were more distinct when compared with other cohorts of patients. These results suggest UH-HCC should be further stratified according to the tumor location for tailored treatment.

5. Conclusion

This study provides compelling evidence supporting TACE-γHRT as an effective and safe modality for treating UH-HCC. UH-HCC patients subjected to TACE-γHRTa exhibited superior responses and outcomes, including a higher CR rate, prolonged OS, and more favorable toxicity profiles when compared with those treated with TACE-γHRTs. Additionally, the results emphasize the significant role played by tumor location in the context of TACE-γHRT for UH-HCC. NSNG patients can achieve better results by comparison with other cohorts of patients. It is beneficial for UH-HCC patients to be stratified according to the tumor location.

The selection bias is a limitation of the present study because of its retrospective nature. The prospective study, as a future direction, would avoid the limitations of retrospective study significantly.

Author contributions

JJ Shen: drafted the work and revised it, approval of the submitted version; NB Zhong: design of the work, approval of the submitted version; ZH Chen: data acquisition, approval of the submitted version; JH Lin: data analysis and interpretation, approval of the submitted version; DY Ma: revised the work, approval of the submitted version.

Financial disclosure

The authors have no financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Ethical conduct of research

Ethics approval and consent to participate: Ethics approval and participation consent were obtained from 900^th Hospital of PLA (approval no.:2022060).

Data availability statement

The data and materials have not been archived in a publicly accessible repository due to ethical restrictions. Data without identifiers can be made available upon reasonable request and with appropriate ethics approval.

Patient consent for publication

Prior written consent was obtained from each patient for publication.