Effects of TACE combined with sorafenib on survival of patients with hepatocellular carcinoma based on ALBI grades and tumor load layers And Establishment of prognosis prediction model

doi:10.21203/rs.3.rs-1971351/v1

Download PDF

Research Article

Effects of TACE combined with sorafenib on survival of patients with hepatocellular carcinoma based on ALBI grades and tumor load layers And Establishment of prognosis prediction model

https://doi.org/10.21203/rs.3.rs-1971351/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Hepatocellular Carcinoma (HCC) is the fourth most common malignant tumor in China and the second leading cause of cancer-related death. Transcatheter arterial chemoembolization (TACE) and targeted therapy are widely used in the treatment of HCC. To make a better treatment strategy from many treatment options, this study analyzed the survival of patients with liver cancer based on Albumin-Bilirubin (ALBI) grades and tumor load layer, and established a nomogram model.

Methods

The clinical data of 354 patients with unresectable HCC and recurrent HCC after radical resection who received TACE combined with or without sorafenib(SORA) in our hospital from 2017 to 2020 were investigated retrospectively.After 1:2 propensity score matching (PSM), eighty-one cases in TACE + SORA group and 162 cases in TACE group were included.

Results

The survival benefit of TACE combined with sorafenib is better than that of TACE alone in patients with advanced liver cancer and recurrent primary liver cancer after radical surgery. Compared with TACE alone, patients with good liver function reserve (ALBI grade 1_5 points) can benefit significantly from TACE combined with SORA (median PFS: 11.8 months vs 7.0 months, P=0.000; median OS: 27.6 months vs 23.0 months, P=0.019). For patients with poor liver function (such as ALBI grade 1_ grade B), combined therapy can not significantly prolong the survival of patients with liver cancer(median PFS: P=0.370; median OS: P=0.131). ALBI grades is superior to Child-Pugh grades in predicting the prognosis of HCC. The prognosis of patients with large tumor burden ((layer 2 and layer 3) is better than that of patients with TACE alone(layer 2: 24.5 months vs 18.5 months, layer 3: 15.0 months vs 12.2 months; P<0.05). The nomogram prediction model can predict the prognosis of patients with unresectable liver cancer and patients with recurrence of liver cancer(C-index and AUC: 0.846 and 0.703, respectively)(C index range in predicting OS in 1, 2 and 3 years: 0.707-0.928), and it is also a good risk stratification model. Nomograph model based on ALBI grades is better than nomograph model based on Child-Pugh grades (0.844 vs 0.839).

Conclusions

Compared with TACE alone, the combination of TACE and sorafenib could improve the prognosis of patients, especially the patients with better liver function (ALBI grade 1_5points) and the patients with higher tumor load,such as layer 2 and 3. The nomogram model established in this study performed well in predicting prognosis, and the model could be used for risk stratification of patients.

Hepatocellular Carcinoma

TACE

Sorafenib

ALBI grade

Tumor load layer

Prognosis

Nomogram

According to the latest global cancer burden datas in 2020 released by the International Agency for Research on Cancer (IARC) of the World Health Organization [1], the incidence rate of hepatocellular carcinoma was fifth in China, second in mortality. Almost 70% of patients of HCC were in the middle or later stages (Barcelona Clinic Liver Cancer stage [BCLC] B or C)，the opportunity of radical treatments such as surgical operation, liver transplantation and ablation has been lost in the definite diagnosis. Even in patients undergoing radical surgery, the risk of recurrence is still high, because of the highly malignant degree of liver cancer itself [2]. Thus, the non-surgical treatment plays a very important role in the treatment of liver cancer.

TACE is one of the most widely used non-surgical methods in the treatment of primary liver cancer. TACE has been included in the treatment of primary liver cancer in many clinical practice guidelines[3,4,5,6,7]. Sorafenib is a tyrosine kinase inhibitor (TKI), which can act on vascular endothelial growth factor receptor (VEGFR), playing an important role in antitumor activity through anti-angiogenesis and anti-proliferation effects. Sorafenib is the most widely used systemic antiangiogenic drug in the treatment of liver cancer, which can significantly prolong the survival of patients (compared with placebo, median overall survival(OS): 10.7 vs 7.9 months, P < 0.001；6.5 vs 4.2months, P < 0.001[8,9]). Other molecular targeted drugs that are gradually rising[10,11,12].

Multiple guidelines and consensus[7,13,14]recommended that it is urgent to improve the curative effect by combined and multidisciplinary treatment in recent years. The results of a randomized, open, multicenter trials of tactics (Gut, 2020) [15] showed that, for HCC patients without vascular invasion or extrahepatic spread who could not get a profit from surgical resection or ablation, TACE combined with sorafenib was more effective than TACE alone in prolonging progression free survival (PFS)(median PFS: 25.2months vs 13.5months；P=0.006). The results of OS research are presented at the American Society of Clinical Oncology (ASCO) meeting in 2021[16],TACE combined with sorafenib did not significantly improve OS compared with TACE alone, but the median OS of TACE combined with sorafenib was longer (36.2months vs 30.8months, P=0.40,ΔOS=5.4months).

Different from other solid tumors, the prognosis and treatment of HCC were closely related to liver function as well as tumor characteristics[17]. According to the time-dependent ROC analysis(Hiraoka et al.[18]) and artificial neural network (ANN) model(Wang et al.[19]), ALBI grade has better discrimination ability in predicting survival, especially in the long period of time. The ‘Six-and-twelve’model had good performance and discrimination ability[20], based on which, patients were divided into three layers. The median survival time was significantly different which were 49.1 (95％confidence interval[CI] 43.7-59.4) months, 32.0 (95％CI 29.9-37.5) months and 15.8 (95％CI 14.1-17.7) months respectively.

In this study, we conducted the propensity analysis about the influence of liver function and tumor load on the prognosis of patients who could not undergo radical treatment and patients with recurrent liver cancer after radical hepatectomy. Furthermore, we could help to formulate individualized treatment strategies and select the dominant groups who benefit significantly from TACE combined with sorafenib, through the accurate grasp of patients' disease status (such as liver function, tumor load, etc.) .The study also established a linear prediction model, and then had a more accurate grasp of the prognosis of patients.

Study populations

Study populations We retrospectively analyzed the clinical data of 354 patients diagnosed with primary liver cancer in the fourth hospital of Hebei Medical University from January 2017 to March 2020. This study has been reviewed by our ethics committee.

Study groups Patients were divided into two groups according to different treatment methods, one group was TACE + SORA group (they were treated with TACE and sorafenib during the study), the other group was TACE group (they were treated with TACE alone, and served as control group).

Inclusion criterias 1) Adult liver cancer patients aged 18 and over; 2) patients with liver function of Child-Pugh A or B liver function; 3) Patients with general good performancen(the score of Eastern Cooperative Oncology Group[ECOG] is 0 or 1); 4) patients with advanced liver cancer who were not suitable for surgery or ablation and patients with recurrent advanced liver cancer after radical hepatectomy; 5) all patients received at least two cycles of TACE; 6) patients were definitely diagnosed as primary liver cancer, through liver biopsy, postoperative pathology and typical imaging findings.

Exclusion criterias 1) Patients with incomplete basic information; 2) Patients who interrupted treatment due to inability to tolerate adverse drug reactions; 3) and other organs of patients with malignant tumor; 4) Patients with other serious diseases, such as severe heart and kidney dysfunction, severe infection, bleeding etc, and 5) Patients lost to follow-up for any reason. According to the inclusion and exclusion criterias, 354 patients were included in the study, as shown in Fig. 1.

All methods were performed in accordance with the relevant guidelines and regulations.

Data

We collected all available clinical information for the patients, including the age, gender, basic liver disease or etiology, routine blood test, coagulation function, blood biochemical indexes and tumor conditions of the patients. Calculate the ALBI score and tumor burden score based on the following formulas:

ALBI score=(log10Bilirubin (µmol/L)×0.66) - (Albumin (g/L)×0.085)[18]; ( ALBI grade 1, 2 and 3 were≤–2.60; <–2.60 to ≤–1.39;>–1.39,respectively);

“Tumor burden score”=Maximum diameter of tumor + Tumor numbers; (Tumor load layers : layer 1was 6 points or less; layer 2 was from 6 points (excluding 6 points) to12 points; layer 3was more than 12 points)[20].

Survival indexes and Follow-up

The OS and PFS of patients were recorded. Also, other survival indexes were PFS rate at 3,6,12 months and OS rate at 12,18,24,36 months.

The follow-up methods included telephone follow-up, reexamination of imaging data and re-admission records. The follow-up time was up to November 15,2020. According to the modified response evaluation criteria in solid tumors (mRECIST), the imaging response of patients with disease treatment was evaluated through computed tomography (CT) or magnetic resonance imaging(MRI) every two or three months.

Administration of sorafenib

According to the medication instructions, we recommend the dosage of sorafenib to the patients. The initial dose was 400mg twice per day. In the course of taking medicine, after the occurrence of sorafenib related adverse reactions, we took some steps, such as the dosage reduction(400mg once per day or 400mg once every two days), discontinue administration or discontinue treatment temporarily. To give symptomatic treatment in time when necessary.

Statistical methods

The P-P chart was used to test the normality of measurement data,which was expressed by mean ± standard deviation. T-test of independent samples was carried out for the measurement data that obey normal distribution. Nonparametric Mann-Whitney U test of two independent samples was performed for data that did not conform to normal distribution. The counting datas were expressed by frequency and percentage, and were compared between groups by Pearson chi square test or Fisher exact probability method. The baseline datas of patients, such as age, gender, liver function status and tumor factors were included in the logistic regression model. The corresponding propensity scores were calculated, and then we had a PSM. Using 1:2 nearest neighbor matching method, the matching capacity difference (caliper value) was 0.1(A number between 0 and 1 was used as the maximum tendency difference allowed when matching). The matched variables were age, WBC, PLT, laboratory indexes of liver function, ALBI grade, Child-Pugh grade, tumor load layer, intrahepatic metastasis, extrahepatic metastasis and tumor thrombus. A total of 243 patients were included (TACE + SORA group was 81 patients, TACE group was 162 patients). Kaplan Meier survival curves of OS and PFS(expressed as median [unit: month]) were established for patients with propensity score matching(81:162), and Log-Rank test was performed. Univariate Cox regression analysis was used to screen the prognostic factors of HCC. The factors with P value < 0.05 were included in multivariate Cox regression analysis to further clarify the independent risk factors related to prognosis. The SPSS 22.0 (IBM, USA) and R (R for windows 2.15.3) software were used, and the psmatching 3.04 plug-in was installed. Statistical analysis was performed using the above methods.

Based on the Cox proportional hazards regression model, the study used RStudio (RStudio for Windows 1.4.1103) software to construct the nomogram model related to the prognosis and establish the linear prognosis prediction model. The C-index of the prediction model was calculated and the consistency was tested.The ROC curves of nomogram model was established and AUC is calculated. The calibration curves of nomogram model were established to further verify the prediction efficiency of the prediction model. X-tile (version 3.6.1) software was used to determine the cut-off value of the prediction model, and Kaplan Meier survival curve was established.

When P value was less than or equal to 0.05, there was statistical difference.

Baseline characteristics of TACE + SORA and TACE groups

A total of 354 patients with advanced liver cancer and recurrent primary liver cancer after radical surgery were included in the study.There were 101 cases in TACE + SORA group and 253 cases in TACE group. The median follow-up time was 17.0(4.0-41.0)months.The mean follow-up time was 18.4±6.9 months. There were statistical differences in age (P=0.003), the level of α-L-Fucosidase(AFU) (P=0.016), extrahepatic metastasis whether or not(P＜0.1), hemangioma thrombus whether or not(P=0.001) BCLC staging (P=0.016) and ECOG score (P=0.002) between TACE + SORA group and TACE group, as shown in Table 1. After PSM according to the 1:2 (81cases were in TACE + SORA group, 162 cases were in TACE group), the baseline datas were well balanced. After PSM, P values were > 0.05, as shown in Table 1.

Table 1 Comparison of baseline characteristics of patients between TACE+SORA group and TACE group (pre-PSM & post-PSM)
Parameters		pre-PSM			post-PSM
Parameters		TACE+SORA(n=101)	TACE (n=253)	P	TACE+SORA (n=81)	TACE (n=162)	P
Age	（years）	55.7±11.1	59.6±11.2	0.003	57.43±10.74	57.80±10.60	0.799
Sex	Male	87(86.1%)	213(84.2%)	0.645	68(84.0%)	136(84.0%)	1.000
Sex	Female	14(13.9%)	40(15.8%)	0.645	13(16.0%)	26(16.0%)	1.000
Etiologies	HBV	82(81.2%)	210(83.0%)	0.476	65(80.2%)	135(83.3%)	0.720
	HCV	3(3.0%)	14(5.5%)		3(3.7%)	7(4.3%)
	Others	16(15.8%)	29(11.5%)		13(16.0%)	20(12.3%)
ALT（U/L)		50.45±45.97	49.94±49.45	0.929	51.18±50.32	53.53±51.92	0.737
AFU（U/L)		34.53±15.55	30.24±13.63	0.016	32.77±14.44	31.07±14.22	0.385
Albmiun（g/L)		39.90±5.63	39.91±5.09	0.998	39.67±5.81	39.28±5.80	0.621
AFP levels	≤400ng/mL	66(65.3%)	143(56.5%)	0.127	50(61.7%)	105(64.8%)	0.637
AFP levels	>400ng/mL	35(34.7%)	110(43.5%)	0.127	31(38.3%)	57(35.2%)	0.637
ALBI grades	1	57(56.4%)	137(54.2%)	0.472	45(55.6%)	88(54.3%)	0.983
	2	42(41.6%)	114(45.1%)		35(43.2)	72(44.4%)
	3	2(2.0%)	2(0.8%)		1(1.2%)	2(1.2%)
Child-Pugh grades	5points	64(63.4%)	166(65.6%)	0.727	51(63.0%)	103(63.6%)	0.960
	6points	22(21.8%)	56(22.1%)		18(22.2%)	37(22.8%)
	B	17(16.8%)	31(12.3%)		12(14.8%)	22(13.6%)
Intrahepatic metastasis	No	35(34.7%)	89(35.2%)	0.926	30(37.0%)	74(45.7%)	0.199
Intrahepatic metastasis	Yes	66(65.3%)	164(64.8%)	0.926	51(63.0%)	88(54.3%)	0.199
Extrahepatic metastasis	No	83(82.2%)	187(73.9%)	0.099	66(81.5%)	120(82.1%)	0.906
Extrahepatic metastasis	Yes	18(17.8%)	66(26.1%)	0.099	15(18.5%)	29(17.9%)	0.906
Tumor thrombus	No	45(44.6%)	162(64.0%)	0.001	42(51.9%)	85(51.2%)	0.928
Tumor thrombus	Yes	56(55.4%)	91(36.0%)	0.001	39(48.1%)	79(48.8%)	0.928
Tumor burdens	≤6	19(18.8%)	50(19.8%)	0.608	15(18.5%)	34(21.0%)	0.890
	6＜and≤12	52(51.5%)	116(45.8%)		42(51.9%)	81(0.0%)
	＞12	30(29.7%)	87(34.4%)		24(29.8%)	47(29.0%)
BCLC staging	B	36(35.6%)	126(49.8%)	0.016	31(38.3%)	71(43.8%)	0.408
BCLC staging	C	65(64.4%)	127(50.2%)	0.016	50(61.7%)	91(56.2%)	0.408
ECOG score	0	88(87.1%)	181(71.5%)	0.002	66(81.5%)	122(75.3%)	0.278
ECOG score	1	13(12.9%)	72(28.5.0%)	0.002	15(18.5%)	40(24.7%)	0.278

Distribution of liver cancer patients with different child Pugh grades in different ALBI grades

The distribution of liver cancer patients with different child Pugh grades in different ALBI grades was shown in Table 2. In 243 patients, patients of ALBI grade 1,2,3 contained 133(54.7%) cases、107(44.0%) cases and 3(1.2%)cases. In patients with ALBI grade 1, patients with liver function of Child-Pugh 5points, 6 points and grade B were 111(83.5%) cases, 18(13.5%) and 4(3.0%) cases. In patients with ALBI grade 1, patients with liver function of Child-Pugh 5points, 6 points and grade B were 43(40.2%), 37(34.6%) and 27(25.2%) cases. All of the 3 patients with ALBI grade 3 were Child-Pugh grade B.

In 154 patients with Child-Pugh grade A_5 points, the distribution in ALBI grade 1 was more than that in ALBI grade 2(72.1% vs 27.9%). In 55 patients with Child-Pugh grade A_6 points, patients with ALBI grade 1 were less than those with ALBI grade 2 (32.7% vs 67.3%). In 31patients with Child-Pugh grade B, patients with ALBI grade 2were more than those with ALBI grade 1(12.9% vs 87.1%). There was no significant difference in the distribution of liver cancer patients with different liver function status between the two groups (P=0.851). The number of cases in individual groups was less than 5 (TACE+SORA group: ALBI grade 1_6 points, ALBI grade 1_grade B,ALBI grade 3_grade B; TACE group: ALBI grade 1_grade B, ALBI grade 3_grade B).

Table 2 Distribution of liver cancer patients with different child Pugh grades in different ALBI grades
Liver function status		TACE+SORA	TACE	Total	P
ALBI G1	5points	39	72	111	0.851
	6points	4	14	18
	Grade B	2	2	4
ALBI G2	5points	12	31	43
	6points	14	23	37
	Grade B	9	18	27
ALBI G3	Grade B	1	2	3
Total		81	162	243

Survival comparison between TACE + SORA group and TACE group and the influence of liver function and tumor load on survival

1. Survival comparison between TACE + SORA group and TACE group

As shown in Figure 2, for PFS and OS, the survival benefit of TACE + SORA group was significantly better than that of TACE group (median PFS 9.0 months vs 6.0 months, median OS 24.0 months vs 17.0 months; P=0.000).

2. Effects of ALBI grades on prognosis of patients receiving TACE + SORA or TACE

The prognosis of 240 patients with liver cancer of grade 1 and 2 was analyzed(Figure 3). In HCC patients with liver function of ALBI grade 1, there were significant differences in median PFS(11.2 months vs 7.0 months, P=0.000) and OS (25.5 months vs 21.0 months, P=0.006) between TACE + SORA group and TACE group. In HCC patients with liver function of ALBI grade 2, there were significant differences in median PFS(6.4 months vs 5.0months, P=0.026) and OS (18.3months vs 13.4 months，P=0.006) between TACE + SORA group and TACE group.

3. Effect of liver function on survival of patients with hepatocellular carcinoma receiving TACE + SORA or TACE

Except for the groups with less than 5 patients with different liver function status in the study population (Table 3), the survival of 232 patients (TACE+SORA group: ALBI grade 1_5 points,ALBI grade 2_5 points, ALBI grade 2_6 points, ALBI grade 2_grade B; TACE group: ALBI grade 1_5 points, ALBI grade 1_6 points, ALBI grade 2_5 points,ALBI grade 2_6 points, ALBI grade 2_grade B) with liver cancer was analyzed (Figure 4).

Patients with liver function of ALBI grade 1_5 points would had a better prognosis of TACE + SORA group than TACE group(median PFS: 11.8 months vs 7.0 months, P=0.000; median OS: 27.6 months vs 23.0 months, P=0.019). In patients with liver function of ALBI grade 2_5 points, the survival benefit of patients receiving TACE + SORA was also higher than that of patients receiving TACE only (median PFS：6.8 months vs 5.0 months, P=0.046; median OS: 22.5 months vs 15.0 months, P=0.005).

In patients with liver function of ALBI grade 2_6 points, there was no significant difference in short-term prognosis between TACE + SORA and TACE (median PFS: 6.3 months vs 5.9 months, P=0.494); however there were significant differences in long-term prognosis(median OS: 16.8 months vs 13.5 months，P=0.001). In patients with liver function of ALBI grade 2_ grade B, there was no significant difference in the prognosis of TACE + SORA group or TACE group (median PFS：6.0 months vs 4.4 months, P=0.370; median OS:14.8 months vs 11.0 months，P=0.131). Also, the median PFS and OS of patients with liver function of ALBI grade 1_6 points in TACE group were 6.5months and 15.8months.

Figure 4 showed that, compared with TACE alone, patients with good liver function (especially ALBI grade 1_5 points) could benefit from TACE + SORA. Compared with TACE alone, combined treatment strategy does not benefit patients with poor liver function (such as ALBI grade 2_grade B) significantly.

4. Effect of tumor burden layers on the prognosis of patients receiving TACE + SORA or TACE

The study analyzed the prognosis of patients with different tumor loads receiving different treatment methods(Figure 5). In patients with HCC of layer 1, there was difference in median PFS (11.9 months vs 7.8months,P=0.000) between TACE + SORA group and TACE group. But there was no significant difference in long-term survival (27.7 months vs 23.9 months, P=0.402). In patients with HCC of layer 2, TACE combined with SORA can significantly improve the prognosis of patients(median PFS: 9.8 months vs 6.0 months,median OS: 24.5 months vs 18.5 months, P＜0.05). In patients with HCC of layer 3, combined treatment could also significantly improve the prognosis of patients (median PFS: 6.3 months vs 4.7 months, P=0.118; median OS: 15.0 months vs 12.2 months, P=0.001).

The results showed that the median OS of TACE + SORA or TACE was not significantly different in patients with lower tumor load (layer 1) (P=0.402); patients with high tumor load in TACE + SORA group had a long-term benefit than TACE group (P value＜0.05).

5. Survival rates of patients with different liver functions and tumor loads

Table 3 showed the proportions of PFS survival in HCC patients with different liver function and tumor load treated with TACE + SORA or TACE. The 3, 6, 12 months survival rate of patients treated with combined therapy was higher than that of TACE alone in the whole population (98.8% vs 86.3%, 81.5% vs 46.8%, 23.5% vs 7.0%). Based on liver function status and tumor load layers, the PFS rate of TACE + SORA group was higher than that of TACE group.

Table 4 showed the proportions of OS survival in HCC patients with different liver function and tumor load treated with TACE + SORA or TACE. The 12, 24, 36 months survival rate of patients treated with combined therapy was higher than that of TACE alone in the whole population (92.6% vs 79.9%, 49.9% vs 20.9%, 8.9% vs 5.6%). Similarly, based on liver function status and tumor load layers, the OS rate of TACE + SORA group was higher than that of TACE group.

Table 3 The proportion of patients with liver cancer with different liver function and tumor load receiving TACE +SORA or TACE (PFS)
		TACE+SORA			TACE
		3-months PFS（%）	6-months PFS（%）	12-months PFS（%）	3-months PFS（%）	6-months PFS（%）	12-months PFS（%）
Overall patients		98.8	81.5	23.5	86.3	46.8	7.0
ALBI grades	1	100.0	95.6	37.8	93.1	55.0	8.2
	2	100.0	65.7	5.7	83.1	36.6	2.8
Child-Pugh grades	5	100.0	90.2	35.3	88.2	48.4	4.0
	6	94.4	77.8	5.6	86.5	48.6	8.1
	B	91.7	50.0	0.0	77.3	36.4	9.1
ALBI grade1_5points 5分		100.0	94.9	41.0	93.0	53.3	5.8
ALBI grade1_6points		—	—	—	92.9	57.1	7.1
ALBI grade1_grade B		—	—	—	—	—	—
ALBI grade2_5points 5分		100.0	91.7	16.7	83.5	36.8	3.3
ALBI grade2_6points		100.0	71.4	0.0	91.3	43.5	8.7
ALBI grade2_grade B		100.0	55.6	0.0	72.2	27.8	0.0
Tumor burden layer	1	100.0	100.0	46.7	94.1	67.6	6.1
	2	100.0	85.7	31.0	86.5	45.7	6.2
	3	95.8	62.5	12.5	80.0	35.6	4.4

Table 4

The proportion of patients with liver cancer with different liver function and tumor load receiving TACE +SORA or TACE (OS).

Proportional hazard regression analysis of prognosis of patients with hepatocellular carcinoma and establishment of prognosis model

1. Univariate and multivariate Cox regression analysis of survival in patients with hepatocellular carcinoma

Univariate Cox regression analysis showed that treatment regimen, albumin,ALBI grade, Child-Pugh grade, tumor load layer, intrahepatic metastasis, extrahepatic metastasis and tumor thrombus were risk factors for prognosis (P＜0.05; Table 5).

The factors with P<0.05 in univariate analysis were included in multivariate Cox regression analysis. The multivariate Cox regression analysis showed that treatment regimen (P=0.002), albumin level (P=0.021),ALBI grade (P=0.005), Child-Pugh grade (P=0.004), tumor load (P=0.027), intrahepatic metastasis (P=0.041) and extrahepatic metastasis (P=0.019) were independent risk factors for the prognosis of HCC patients.

Table 5 Multivariate Cox regression analysis of survival time of HCC
	Univariate analysis				Multivariate analysis
	Sig.	HR	HR(95%CI)		Sig.	HR	HR(95%CI)
			Lower	Upper			Lower	Upper
Cure（T+S vs T）	0.000	0.527	0.377	0.737	0.002	0.577	0.404	0.822
Sex（Male vs Femal）	0.817	1.050	0.694	1.588
ALT（U/L）	0.231	1.002	0.999	1.004
Alb（g/L)	0.000	0.841	0.815	0.869	0.021	0.929	0.872	0.989
AFP（≤400 vs>400ng/ml）	0.127	0.788	0.580	1.070
ALBI grades(1/2/3)	0.001	0.113	0.033	0.387	0.005	0.065	0.009	0.442
Child-Pugh grades(5/6/B)	0.000	2.753	2.271	3.338	0.004	0.006	0.000	0.188
Tumor burden layers(1/2/3)	0.008	3.427	2.594	4.526	0.027	0.520	0.292	0.928
In-metastasis (No vs Yes)	0.002	0.529	0.388	0.720	0.041	0.696	0.491	0.985
Ex-metastasis (No vs Yes)	0.000	6.125	4.136	9.070	0.019	0.471	0.251	0.884
Tumor thrombus(No vs Yes)	0.034	2.427	1.781	3.307	0.313	0.825	0.568	1.198

2. Establishment of nomogram prediction model for prognosis and risk stratification based on the model

2.1 Establishment of nomogram prediction model

In order to further predict the prognosis of patients in TACE + SORA group and TACE group according to liver function and tumor load, we tried to establish linear prediction model and risk stratification model. Through univariate and multivariate Cox regression analysis, seven independent prognostic risk factors were identified, including treatment regimen, albumin level, ALBI grade, Child-Pugh grade, tumor load layer, intrahepatic metastasis and extrahepatic metastasis. Based on Cox regression analysis, seven variables were included in nomogram function at the same time, and nomogram diagram was constructed (Figure 6). The prognosis of patients was predicted after scoring by constructing nomograms. We could more intuitively reflect the predictive effect of different variables on the prognosis of patients in Cox regression analysis.The higher the total score was, the better the patients' expected survival was.

According to the nomogram, the clinical indicators of patients were scored, and the total score was 180. The specific scores were as follows (unit: points) : TACE+SORA 25 points, TACE 0 points, the score of patients with and without intrahepatic metastasis were 0 and 13 respectively, the score of patients with and without extrahepatic metastasiswere 0 and 16 respectively, patients with ALBI grade 1,2 and grade 3were 0, 4 and 8 points, patients with Child Pugh 5 points,6 points and grade B were 11.3points,0 and 0.5point, tumor load layer1, 2, 3were 32.5 points, 13.5 points and 0 point, The albumin score was as follows: (Alb-20)*(100/35).

2.2 Validation of nomogram prediction model

2.2.1 C-index and AUC of nomogram prediction model

Further, the prediction performance of the prediction model was evaluated and verified. The C-index and AUC of the nomogram model based on ALBI grade and Child-Pugh grade were 0.846 (95%CI: 0.762-0.913) and 0.703(0.631-0.776), respectively. The C-index and AUC of the nomogram model based on ALBI grade or Child-Pugh grade separately were 0.844(0.741-0.915) and 0.839(0.768-0.898). It also performed well in predicting 1-year OS, 2-year OS and 3-year OS,The AUC values of ROC curve were as follows: 0.928(0.891-0.965) , 0.792(0.735-0.849) and 0.707(0.635-0.779)(Figure 7).

The results of Figure 7 showed that the nomogram model combined with ALBI grade and Child-Pugh grade had a good predictive ability for the prognosis of patients with unresectable or recurrent liver cancer after TACE combined with sorafenib or TACE alone, which was better than the nomogram model only included in ALBI grade or Child-Pugh grade; At the same time, the model based on ALBI grade was better than that based on Child-Pugh grade.

ROC curves at different time points show that, the performance of ALBI grade in predicting prognosis was better than Child-Pugh grade(AUC of 1 year: .787 vs 0.774; AUC of 2 years: 0.683 vs 0.668; AUC of 3 years : 0.631 vs 0.582; Figure 9). The ability to predict the survival of patients within 2 years was similar; with the extension of time, ALBI was more effective in predicting the survival (ΔAUC of 1, 2,3-years OS were 0.013,0.015 and 0.049, Figure 8).

2.2.2 Construction of calibration curve of nomogram prediction model

The calibration curves of linear prediction model were further constructed by using RStudio software (Figure 9). According to the calibration curves, the nomogram model could well fit the predicted values when predicting OS of 12 months, 18 months, 24 months and 36 months. The actual predicted value fluctuated around the predicted value in a certain range. This showed the good prediction ability of the model again. The calibration curves showed that within the study population, the prognostic nomogram model could accurately predict OS at 12, 18, 24 and 36 months.

2.3 Nomogram prediction model could be used for risk stratification of prognosis of liver cancer patients

According to the nomogram, seven clinical variables were scored and summed. The total score of the linear prediction model was 180 points, and the score range of 243 patients was 34.1-174.7 points. Using X-tile software to determine the optimal cut-off value (62.3 points and 104.2points, Figure 10B),divide patients into three risk groups. According to the score, the risk stratification of patients was < 62 (high risk), ≥ 62 and < 104 (medium risk), ≥ 104 and < 180 (low risk). The corresponding survival curves of different risk groups were shown in Figure 10B. When the patient's score was ≥ 104 points and < 180 points, the survival curve was like pink curve in Figure 10; When the patient's score was <62 points, the survival curve like blue curve in Figure 10.

The survival curve based on risk stratification (Figure 10) shows that, the higher the total score was, the longer the expected survival was (low-risk group); the higher the total score was, the shorter the expected survival was (high-risk group) .

The survival analysis of 354 patients with advanced liver cancer who could not be treated by radical resection and those with recurrent liver cancer after radical resection showed that compared with TACE alone, the combination of TACE and sorafenib could improve the prognosis of patients, especially the patients with better liver function (ALBI grade 1_5points) and the patients with higher tumor load,such as layer 2 and 3. The nomogram model established in this study performed well in predicting prognosis, and the model could be used for risk stratification of patients.

In the overall population, the survival benefit of TACE combined with sorafenib was better than that of TACE alone.

In this study, patients receiving TACE + SORA or TACE benefited significantly from PFS and OS (median PFS: 9.0 months vs 6.0 months, median OS: 24.0 months vs 17.0 months, P＜0.05; Figure 2). This study included patients with vascular invasion and extrahepatic spread, also a certain number of patients with postoperative recurrence of liver cancer, which might be the reasons for the differences in the results. Some studies had confirmed that hemangioma thrombus was an independent risk factor for the prognosis of patients with liver cancer[19]. Both TACTICS trials and this study confirmed that the combination therapy was superior to TACE alone, but more patients with higher malignancy were included in this study[15,16]. An important reason for the positive results of the TACTICS study might be that more than half of the patients with initial combination therapy were included in the study.A large, multicenter, prospective, noninterventional study (OPTIMIS study[21])showed that, early combination with sorafenib can prolong the median OS. A national cohort study based on a large sample (TACE combined with sorafenib group,n=426; sorafenib as a control group, n=168) showed that in BCLC stage C and liver function (Child-Pugh A) patients showed consistency, this trial fully confirmed the survival benefit of combination therapy for patients with advanced liver function reserve (median OS were 381days and 204 days,respectively; P=0.021[22].

In theory, the combination of TACE and anti angiogenesis drugs is effective and feasible. The theory of potential synergistic effect of targeted drugs combined with TACE is still developing, especially at the molecular level[23,24,25,26,27].

ALBI grade was superior to Child-Pugh grade in predicting the prognosis of HCC. Compared with TACE alone, patients with good liver function reserve of ALBI grade 1_5 points could benefit significantly from TACE + SORA.

ALBI grade was a more novel index of liver function than Child-Pugh grade in recent studies. The prognosis of ALBI grade was at least not worse than Child Pugh grade[18,28.29.30,31,32]. Sorafenib could be used in patients with unresectable liver cancer and liver function of ALBI grade 1 or partial patients with liver function of ALBI grade 2. Patients with liver function of ALBI grade≥2 or Child-Pugh grade B were not suggesting using sorafenib[33,34]. In the study of Zhong et al, patients with liver cancer of ALBI grade 3 were excluded because of the safety of treatment[28]. A retrospective study of HCC patients with BCLC stage B who were newly treated showed that[35], the median OS of ALBI grade 1, 2 and 3 were 41.7, 18.3 and 14.3 months (P＜0.05). Professor Kudo mentioned two ideas that were, “TACE unsuitability” and “The liver function reserve decreased to Child Pugh grade B” [36]. In many clinical studies about tyrosine kinase inhibitors, patients with poor liver function reserve of Child-Pugh grade B or C were excluded. Partial patients with liver function of Child-Pugh grade B or patients with liver function of Child-Pugh grade C were not recommended to use sorafenib.

Most patients with ALBI grade 1 were also classified as Child-Pugh A(Table 2). A similar distribution was found in a multicenter retrospective study[19].Of the 504 study cohort[28], 256 patients with ALBI grade 1,in which there was 248(49.2%) patients of Child-Pugh grade A, 8(1.6%) patients of Child-Pugh B. In 248 cases of ALBI grade 2, there was 188(37.3%) of Child-Pugh grade A, 60(11.9%) of Child-Pugh grade B. The combination of ALBI grade and Child-Pugh grade reduced the heterogeneity of liver cancer patients with different liver function status. In the study of Zhong et al[28], ALBI grade 1_5 points, ALBI grade 2_5 points, the median OS of patients with liver function of ALBI grade 2_6 points and ALBI grade 2_ grade B were 24.5、11.0、11.8、8.0 months. Among patients of ALBI grade 1, the prognosis of patients with different Child Pugh grades were different (P=0.038). In patients with ALBI grade 2, there was no significant difference in the prognosis of patients with different Child Pugh grades(P=0.230). In the two studies, the proportion of patients with Child-Pugh 5 points in patients with liver function of ALBI grade 1 was the largest, which might be the factor that the median OS of patients with Child Pugh 5 score in this study is close to that of other studies(27.6 months vs 24.5 months). It could be considered that patients with good liver function (especially the liver function was ALBI grade 1_ 5 points) receiving TACE + SORA treatment could obtain better survival than TACE alone (median OS: 27.6 months vs 23.0 months, P=0.019).

In this study, for patients with relatively poor liver function(ALBI grade 2_grade B), the long-term benefits of combined therapy are limited (median OS: 14.8 months vs 11.0 months, P=0.131), which was consistent with the view that targeted drugs were not recommended for patients with poor liver function. Patients with liver function of ALBI grade 2 or above [33] or Child Pugh C [3] are not recommended to use sorafenib. With the increase of times of TACE, the liver function reserve had a negative relative change [37].Some Japanese experts put forward the theoretical supports of interventional combined with targeted therapy: targeted drugs having high response rate; liver function reserve can be reserved; normalizing blood vessels and improve the delivery and distribution of chemotherapeutic drugs; preventing recurrence and metastasis by inhibiting hypoxia inducible factor after TACE; targeted drugs being suggested to used early and in the whole process [38,39]. Therefore, in order to benefit more patients with limited TACE efficacy, early combined with sorafenib was an effective treatment strategy, which could not only maintain liver function or reduce liver function damage to a certain extent, but also significantly prolonged the survival of patients.

In addition, this study also found that in the patients with Child-Pugh score of 5, whether PFS or OS, patients with ALBI grade 1 were better than those with ALBI grade 2(P value＜0.05). In patients of HCC with liver function of Child-Pugh score of 6 and grade B，there was a trend of prolonged survival in patients with ALBI grade 1, although P value＞0.05. ALBI grade included albumin and bilirubin, which reflected the synthesis and excretion function of liver. In addition to the above two objective factors, Child-Pugh grade also included coagulation function (Prothrombin time[PT], International normalized ratio[INR]), ascites and hepatic encephalopathy, directly or indirectly reflecting the anabolism and detoxification function of liver. The latter two indicators were subjective factors. In clinical practice, some patients with liver cancer might not be able to use Child-Pugh grade for liver function evaluation because of the lack of coagulation function indicators (PT or INR). ALBI grade reflected liver function more accurately, while Child-Pugh grade contained more clinical factors. And the latter algorithm was more complex. Because there were no subjective factors of ascites and hepatic encephalopathy in the patients with Child-Pugh score of 5, it was more strict for the patients with Child-Pugh score of 5, and the ALBI grade could better distinguish these patients. In addition, in terms of 12-month PFS rate，patients with ALBI grade 1_5 points were significantly higher than those with ALBI grade 2_5 points (41.0% vs 16.7%,Table 3); there were also significant differences in OS rate (24-months OS rate: 72.5% vs 38.9%; Table 4). Combined with the research results of Zhong et al., it could be considered that compared with Child-Pugh grade, ALBI grade had better effect in guiding medication and predicting prognosis of patients.

The prognosis of patients with high tumor burden (layer 2, 3) who received TACE + SORA was better than that of patients who received TACE alone.

Another important indicator for the prognosis of solid tumor patients is tumor load. Tumor size and number are two important factors to evaluate tumor load. Based on "6 and 12 scores"[20], tumor load layer was established and analyzed in this study. At present, there are no other reports about the hierarchical model. First the study analyzed the impact of tumor load layer on the prognosis of TACE group and TACE + SORA group. Whether in TACE + SORA group or TACE group, the survival of patients with different tumor load stratification was different (median PFS and OS: P＜0.05). The results showed that the smaller the tumor load was, the better the survival benefit was.

In the diagnosis and treatment of liver cancer, experts and scholars had proposed a variety of tumor load concepts (“Milan Criteria”, “Up-to-seven Criteria”, “Up-to-eleven Criteria”)and liver cancer staging concepts (“Clinical staging of hepatocellular carcinoma in China”, “BCLC Staging”), which were related to tumor size and tumor number [40]). These concepts play an important role in the diagnosis and treatment of liver cancer. At present, there was no other related research on predicting the prognosis of liver cancer by dividing 6 points and 12 points. The median OS of tumor burden from low to high were 49.1 months, 32.0 months and 15.8 months, respectively. There were significant differences in the survival curves of the three levels, whether in the trial cohort or in the validation cohort. Compared with this study, patients with vascular invasion or extrahepatic spread and the score of ECOG > 0 were excluded. Similarly, fewer prognostic factors were included in this study, resulting in a significant difference in the median OS between the two studies. In Cox risk regression model, tumor load is an independent prognostic factor, which could predict the survival of patients with liver cancer. This study once again confirmed the predictive effect of the hierarchical model on prognosis. Because there were very few similar studies, the feasibility of the hierarchical model still needed to be further confirmed.

The results showed that patients with low tumor burden did not benefit significantly from TACE combined with sorafenib, which might be related to TACE reducing liver function reserve and increasing the risk of tumor recurrence. However, the combination of TACE combined with sorafenib could significantly prolong the survival time of patients with large tumor load, which further confirmed that the curative effect of TACE + SORA was better than TACE. Therefore, on the basis of TACE treatment, sorafenib could be combined as early as possible to prolong the survival time and improve the survival benefit for patients with advanced liver cancer or recurrent liver cancer with large tumor load and certain liver function reserve.This study confirmed the good performance of the linear model in guiding the treatment decision of patients with high tumor load. However, the guiding role of treatment for patients with low tumor load needed to be further explored.

This study was mainly about the survival analysis of TACE and TACE + SORA, which was consistent with most studies. The curative effect of combination therapy was better than that of TACE alone. Hierarchical analysis showed that TACE + SORA was superior to TACE alone in the treatment of high tumor burden patients with better liver function. Therefore, patients with relatively good liver function and high tumor load or patients with recurrence of liver cancer after TACE + SORA had the obvious advantage of TACE + SORA.

The nomogram prediction model had a good performance in predicting the prognosis of patients with unresectable liver cancer and patients with postoperative recurrence of liver cancer, and it was also a good risk stratification model.

As mentioned above, many studies have confirmed that ALBI grade, Child-Pugh grade, tumor load and other factors had an impact on the prognosis of liver cancer patients. The nomograph model established in this study included the above-mentioned factors. A nomogram model based on ALBI grade and Child-Pugh grade showed a good predictive effect on the prognosis of patients who were treated with TACE combined with sorafenib or TACE alone (C-index was 0.846), and was better than the nomogram model included in ALBI grade and Child-Pugh grade (C-index were 0.844 and 0.839). A multicenter study [41] of TACE combined with sorafenib established nomogram models based on ALBI grade and Child-Pugh grade respectively, which included tumor number, tumor size (＜5, 5-10,＞10), extrahepatic metastasis and other factors. The C-index were 0.823 and 0.802. At the same time, multivariate Cox analysis of this multicenter study showed that extrahepatic metastasis was associated with the prognosis of HCC patients [41].As was shown in figure 9, ALBI grade was better than Child-Pugh grade in predicting the survival of patients with liver cancer (ΔAUC of 1,2,3year OS were 0.013, 0.015 and 0.049 sequentially). In the study of predicting the prognosis of patients with advanced liver cancer treated with TACE, it was mentioned that the predictive ability of the two liver function grades was similar on Prognosis within 1 year. With the extension of observation time, the performance of ALBI grade was stronger; The C-index of ALBI grade and Child-Pugh grade were 0.63(95%CI: 0.53-0.60) and 0.57(95%CI: 0.57-0.68) [42].

The nomogram model of this study includes both ALBI grade and Child-Pugh grade, which helped us to have a more accurate grasp of the baseline situation of patients. Also, the model was based on the updated tumor load stratification (≤6, 6＜and ≤12,＞12). In addition, the nomogram model also included albumin level, which was closely related to ALBI grade and Child-Pugh grade; and intrahepatic and extrahepatic metastasis closely related to tumor load.

The nomogram model provided a good research idea for clinical scholars, which could predict the survival rate and risk stratification of patients at the same time. The model could help clinicians to obtain the clinical data and predict the prognosis intuitively. The linear prediction model had not been verified by external data, so further research was needed to confirm. Although the prediction model of this study had a good performance, it lacked external data validation, and a small number of patients with ALBI grade 3 were included in the model, which might reduce the ability to evaluate the prognosis of these patients.

There were more and more researches on tumor microenvironment and molecular mechanism [43,44,45,46]. People have a better understanding of TACE and sorafenib resistance mechanism [47,48].In the era of immunotherapy, based on the improvement of tumor microenvironment[TME] by Immune checkpoint inhibitors, immunotherapy combined with anti angiogenic therapy and TACE is expected to bring better benefits to patients with liver cancer in the future.

Advantages：

Though a number of basic clinical data of patients were compared, the bias caused by large sample size difference was reduced after propensity score matching.
Patients were screened strictly according to the inclusion and exclusion criteria to avoid the ‘all comers’ error[49]. It was also one of the factors to obtain positive results in this study.
The study adopted the latest ALBI grade. Compared with the traditional Child-Pugh grade with subjective factors, the former is calculated by a formula, which is more objective and accurate.The ALBI grade has the feasibility in clinical practice.
This study adopted a novel tumor load layer, which had strong clinical operability, and confirmed the effectiveness of this stratification in predicting the prognosis of patients with liver cancer.
The linear prediction model of prognosis based on a number of clinical data showed its correlation with prognosis intuitively and vividly. The performance of the model was preliminarily verified.

Disadvantages：

This is a retrospective study, which could not avoid the bias of the nature of the study itself. In this study, multiple clinical variables of the two groups were statistically tested as much as possible, and the propensity score was matched according to 1:2.
Although the tumor load layer adopted in this study was effective in this study, it lacked more clinical cases to verify. Therefore, whether this tumor load layer was suitable for more patients with liver cancer, especially for patients with low tumor load, needs to be confirmed by further research.
In this study, no external validation group was set up to verify the nomogram prediction model. Thus more studies are needed to confirm the effectiveness of the model.

HCC, Hepatocellular Carcinoma; TACE, Transcatheter arterial chemoembolization; ALBI, Albumin-Bilirubin; SORA, Sorafenib; PSM,propensity score matching; AUC,Area Under Curve; IARC,International Agency for Research on Cancer; BCLC,Barcelona Clinic Liver Cancer; TKI,Tyrosine kinase inhibitor; VEGFR,Vascular endothelial growth factor receptor; OS, Overall survival;PFS, Progression free survival, ASCO,American Society of Clinical Oncology; ANN,Artificial neural network; CI,Confidence interval; ECOG,Eastern Cooperative Oncology Group; mRECIST,modified response evaluation criteria in solid tumors; CT,Computed Tomography; MRI,Magnetic Resonance Imaging;AFU, α-L-fucosidase;HBV, Hepatitis B virus; HCV, Hepatitis C virus; ALT, Alanine aminotransferase; AFP,Alpha fetoprotein; ROC,Receiver operating characteristic; PT, Prothrombin time; INR, International normalized ratio; TME, Tumor microenvironment.

Acknowledgements

Not applicable.

Authors' contributions

First of all, as the first author, Qianying Qie conducted the design of the project, data collection and analysis, and writing work, and participated in the entire process of the manuscript. Second, as the corresponding author,Professor Fei Yin conducted guidance during the creation of the article. Last, authors, Jianfei Shi, Man Zhao, Xiaolei Yin, Xin Han, Xiaonan Yin,Ning Li and Lili Mi contributed to data analysis or revising the article.
Funding

Not applicable.

Availability of data and materials

Data is available from the corresponding author upon reasonable request.

Ethics approval and consent to participate

This study was in accordance with the Helsinki Declaration. And all study protocols were approved by the ethics committee of The fourth Hospital of Hebei Medical University. The need for informed consent was waived by the ethics committee, because of the retrospective nature of the study.

Consent for publication
Not applicable.

Competing interests

The authors declare that they have no competing interests.

World Cancer Resport: Cancer research for cancer prevention. 2020.
Poon RT, Fan ST, Lo CM, Ng IO, Liu CL, Lam CM, Wong J. Improving survival results after resection of hepatocellular carcinoma:a prospective study of 377 patients over 10 years.Ann Surg. 2001;234:63-70.
ealth administration of the people's Republic of China. Diagnosis and treatment of primary liver cancer (Version2019). Chinese J Dige Surg.2020;19(1):1-20.
Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet 2018;391:1301–1314.
European association for the study of the liver. EASL clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 2018;69:182–236.
Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, Zhu Andrew X, Murad MH, Marrero JA. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology. 2018;67(1):358–380.
Omata M, Cheng AL, Kokudo N, Kudo M, Lee JM, Jia JD, Tateishi R, HanKH, Chawla YK, Shiina S, Jafri W, Payawal DA, Ohki T, Ogasawara S, Chen PJ, Lesmana CR A, Lesmana LA, Gani RA, Obi S, Dokmeci AK, Sarin SK. Asia-Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int 2017;11:317–370.
Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc JF,Oliveira de AC, Santoro A, Raoul JL, Forner A, Schwartz M, Porta C, Zeuzem S, Bolondi L, Greten TF, Galle PR, Seitz JF, Borbath I, Häussinger D, Giannaris T, Shan MH , Moscovici M, Voliotis D,Bruix Jordi, for the SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359:378-390.
Cheng AL, Kang YK, Chen ZD, Tsao CJ, Qin SK, Kim JS , Luo RC, Feng JF, Ye SL, Yang TS, Xu JM, Sun Y, Liang HJ, Liu JW, Wang JJ, Tak WY , Pan HM, Burock K, Zou J, Voliotis D, Guan ZZ. Efficacy and safety of sorafenib in patients in the Asia-Pacific region with advanced hepatocellular carcinoma: a phase III randomised, double-blind, placebo-controlled trial. Lancet Oncol. 2009; 10: 25–34.
Kudo M, Finn RS, Qin SK, Han KH, Ikeda KJ, Piscaglia F, Baron A, Park JW, Han GH, Jassem J, Blanc JF , Vogel A, Komov D, Evans TRJ, Lopez C, Dutcus C, Guo M, Saito K, Kraljevic S, Tamai T, Ren M, Cheng AL. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet. 2018; 391(10126):1163–1173.
Bruix J, Qin SK, Merle P, Granito A, Huang YH, Bodoky G, Pracht M, Yokosuka O, Rosmorduc O, Breder V, Gerolami R, Masi G, Ross PJ, Song TQ, Bronowicki JP, Isabelle OH, Kudo M, Cheng AL, Llovet JM, Finn RS, LeBerre MA, Baumhauer A, Meinhardt G, Han GH, on behalf of the RESORCE Investigator. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;389(10064):56–66.
D'Alessio A, Prete MG, Cammarota A, Prete MG, Cammarota A, Personeni N, Rimassa L.The Role of Cabozantinib as a Therapeutic Option for Hepatocellular Carcinoma: Current Landscape and Future Challenges. J Hepatocell Carcinoma. 2021-03-29;8:177-191.
Wang WH, Ceng ZC. Chinese radiotherapy guidelines for hepatocellular carcinoma (2020 edition). J Int Oncol, January 2021; 48(1):1-10.
Chinese Society of Liver Cancer.中国肝癌多学科综合治疗专家共识(2020). Chinese Journal of the Frontiers of Medicine Science. 2020,12(12):28-36.
Kudo M, Ueshima K, Ikeda M, Torimura T, Tanabe N, Aikata, Izumi N,Yamasaki T,Nojiri S, HinoK, Tsumura H, Kuzuya T, Isoda N, Yasui K, Yoshimura K, Okusaka T, Furuse J, Kokudo N, Okita K, Arai Y. Randomized, open label, multicenter, phase II trial comparing transarterial chemoembolization (TACE) plus sorafenib with TACE alone in patients with hepatocellular carcinoma (HCC): TACTICS trial. J Clin Oncol. 2018;36(4_suppl):206.
TACTICS: Final overall survivial (OS) data from a randomized, open label, multicenter, phase Ⅱ trial of transcatheter arterial chemoembolization (TACE) therapy in combination with sorafenib as compared with alone in patients (pts) with hepatocelluar carcinoma (HCC). Data released at the 2021 ASCO GI.
Schutte K, Bornschein J, Malfertheiner P. Hepatocellular carcinoma–epidemiological trends and risk factors. Digest Dis.2009;27(2):80–92.
Hiraoka A, Kumada T, Michitaka K, Kudo M. Newly Proposed ALBI Grade and ALBI-T Score as Tools for Assessment of Hepatic Function and Prognosis in Hepatocellular Carcinoma Patients. Liver Cancer. 2019;8(5) :312-325.
Wang ZX, Fan QL, Wang MM, Wang EX, Li HC, Liu L.Comparison between Child-Pugh Score and albumin-bilirubin grade in patients treated with the combination therapy of transarterial chemoembolization and sorafenib for hepatocellular carcinoma. Ann Transl Med. 2020; 8(8):537.
Wang QH, Xia DD, Bai W, Wang EX, Sun JH, China HCC-TACE Study Group.Development of a prognostic score for recommended TACE candidates with hepatocellular carcinoma: A multicentre observational study. J Hepatol. 2019 May ;70(5) :893-903.
Raoul JL, Decaens T, Burak K, Koskinas J，Villadsen GE，Heurgue-Berlot A，Bayh I，Cheng AL，Kudo M，Lee HC. Practice patterns and deterioration of liver function after transarterial chemoembolization (TACE) in hepatocellular carcinoma (HCC): Final analysis of OPTIMIS in Europe and Canada. Ann Oncol.2018;29 (Suppl 8):viii240.
Liebl, MC ; Hofmann, TG. Identification of responders to immune checkpoint therapy: which biomarkers have the highest value? J Eur Acad Dermatol Venereol.2019 ;33(Suppl 8):52-56.
Kudo M. A New Treatment Option for Intermediate-Stage Hepatocellular Carcinoma with High Tumor Burden: Initial Lenvatinib Therapy with Subsequent Selective TACE. Liver Cancer.2019 Oct ;8(5) :299-311.
Jiang HC, Meng QH, Tan HT, Pan SH, Sun B, Xu RA, Sun XY. Antiangiogenic therapy enhances the efficacy of transcatheter arterial embolization for hepatocellular carcinomas. Int J Cancer. 2007; 121(2):416–24.
Zheng L, Guo CY, Chen CS, Xiao JC, Hu HT, Cheng HT, Zong DW, Jiang L, Li HL. Sorafenib improves lipiodol deposition in transarterial chemoembolization of Chinese patients with hepatocellular carcinoma: A long-term, retrospective study. Oncotarget. 2017;8: 97613–97622.
Richly H, Schultheis B, Adamietz IA, Kupsch P, Grubert M, Hilger RA, Ludwig M, Brendel E, Christensen O, Strumberg D. Combination of sorafenib and doxorubicin in patients with advanced hepatocellular carcinoma: results from a phase I extension trial. Eur J Cancer. 2009; 45: 579-587.
Kim BK, Shim JH, Kim SU, Park JY, Kim DY,; Ahn SH, Kim KM, Lim YS, Han KH, Lee HC. Risk prediction for patients with hepatocellular carcinoma undergoing chemoembolization: development of a prediction model. Liver Int. 2016;36:92–99.
Zhong BY,Yan ZP, Sun JH, Zhang L, Hou ZH, Yang MJ, Zhou GH, Wang WS, Li Z, Huang P, Zhang S, Zhu XL, Ni CF. Prognostic Performance of Albumin– Bilirubin Grade With Artifificial Intelligence for Hepatocellular Carcinoma Treated With Transarterial Chemoembolization Combined With Sorafenib.Front Oncol.2020;10:525461.
Kao WY, Su CW, Chiou YY, Chiu NC, Liu CA, Fang KC, Huo TI, Huang YH, Chang CC, Hou MC, Lin HC, Wu JC. Hepatocellular Carcinoma: Nomograms Based on the Albumin-Bilirubin Grade to Assess the Outcomes of Radiofrequency Ablation. Radiology.2017;285(2):670–680.
Edeline J, Blanc JF, Johnson P, Boris CG, Ross P, MaYT, King J, A. Hubner R, Sumpter K, Darby S, Evans J, Iwuji C,Swinson D, Collins P, Patel K, Muazzam I, H. Palmer D, Meyer T. A multicentre comparison between Child Pugh and Albumin-Bilirubin scores in patients treated with sorafenib for Hepatocellular Carcinoma. Liver Int Off J Int Assoc Study Liver. 2016;36(12):1821–8.
Hiraoka A, Kumada T, Kudo M, Hirooka M, Tsuji K, Itobayashi E, Kariyama K, Ishikawa T, Tajiri K, Ochi H, Tada T, Toyoda H , Nouso K, Joko K, Kawasaki H, Hiasa Y, Michitaka K, and on behalf of the Real-Life Practice Experts for HCC (RELPEC) Study Group and HCC 48 Group (hepatocellular carcinoma experts from 48 clinics).Albumin-Bilirubin (ALBI) Grade as Part of the Evidence-Based Clinical Practice Guideline for HCC of the Japan Society of Hepatology: A Comparison with the Liver Damage and Child-Pugh Classifications. Liver Cancer. 2017; 6(3):204–215.
Hansmann J, Evers MJ, Bui JT, Lokken RP, J. Lipnik A, C. Gaba R, E. Ray C. Albumin-Bilirubin and Platelet-Albumin-Bilirubin Grades Accurately Predict Overall Survival in High-Risk Patients Undergoing Conventional Transarterial Chemoembolization for Hepatocellular Carcinoma. J Vasc Interventional Radiol JVIR. 2017; 28(9):1224–1231.e2.
gasawara S, Chiba T, Ooka Y, Suzuki E, Kanogawa N, Saito T, Motoyama T, Tawada A, Kanai F, Yokosuka O. Liver function assessment according to the Albumin-Bilirubin (ALBI) grade in sorafenib-treated patients with advanced hepatocellular carcinoma. Invest New Drugs. 2015;33(6):1257–62.
King J, Palmer DH, Johnson P, Ross P, Hubner RA, Sumpter K, Darby S, Braconi C, Iwuji C, Swinson D, Collins P, Patel K, Nobes J, Muazzam I, Blesing C, Kirkwood A, Nash S, Meyer T. Sorafenib for the Treatment of Advanced Hepatocellular Cancer - a UK Audit. Clin Oncol-UK. 2017;29(4):256–62.
Lee IC, Hung YW, Liu CA, Lee RC, Su CW, Huo TI, Li CP, Chao Y, Lin HC, Hou MC, Huang YH. A new ALBI-based model to predict survival after transarterial chemoembolization for BCLC stage B hepatocellular carcinoma.Liver Int.2019 Sep ;39(9) :1704-1712.
Kudo M. Extremely High Objective Response Rate of Lenvatinib: Its Clinical Relevance and Changing the Treatment Paradigm in Hepatocellular Carcinoma. Liver Cancer. 2018 Sep;7(3):215–224.
Hiraoka A, Kumada T, Kudo M, Hirooka M, Koizumi Y, Hiasa Y, Tajiri K, Toyoda H, Tada T, Ochi H, Joko K, Shimada N, Deguchi A, Ishikawa T, Imai M, Tsuji K. Hepatic Function during Repeated TACE Procedures and Prognosis after Introducing Sorafenib in Patients with Unresectable Hepatocellular Carcinoma: multicenter Analysis. Dig Dis. 2017;35(6):602–610.
Kudo M, Trevisani F, Abou-Alfa GK, Rimassa L. Hepatocellular carcinoma: therapeutic guidelines and medical treatment. Liver Cancer. 2017;6:16–26.
Kudo M. A New Treatment Option for Intermediate-Stage Hepatocellular Carcinoma with High Tumor Burden: Initial Lenvatinib Therapy with Subsequent Selective TACE. Liver Cancer.2019;8(5) :299-311.
Mazzaferro V, Llovet JM, Miceli R, Bhoori S; Schiavo M, Mariani L, Camerini T, Roayaie S, Schwartz M E, Grazi GL, Adam R, Neuhaus P, Salizzoni M, Bruix J, Forner A, Carlis LD, Cillo U, K Burroughs A, Troisi R, Rossi M, Gerunda GE, Lerut J, Belghiti J, Boin I, Gugenheim J, Rochling F, Hoek BV, Majno Pietro. Predicting survival after liver transplantation in patients with hepatocellular carcinoma beyond the Milan criteria: a retrospective, exploratory analysis. Lancet Oncol. 2009;10:35–43.
hong BY,Teng GJ. Multicentric validation and prognostic prediction of transarterial chemoemolization monotherapy or combined with sorafenib for hepatocellular carcinoma. Southeast University, 2020.
Wang ZX, Wang EX, Xia DD. Value of Child-Pugh score versus albumin-bilirubin grade in predicting the prognosis of unresectable hepatocellular carcinoma treated by transarterial chemombolization. J Clin Hepatol. 2020;36 (1). Jan.2020.
eist RS, Duda DG, Sahani DV, Ancukiewicz M, Fidias P, Sequist LV, Temel JS, Shaw AT, Pennell NA, Neal JW, Gandhi L, Lynch TJ, Engelman JA, Jain RK. Improved tumor vascularization after anti-VEGF therapy with carboplatin and nab-paclitaxel associates with survival in lung cancer. Proc Natl Acad Sci USA 2015;112:1547–52.
Lee WS, Yang H, Chon HJ , Kim C. Combination of anti-angiogenic therapy and immune checkpoint blockade normalizes vascular-immune crosstalk to potentiate cancer immunity. Experimental & molecular medicine. 2020;52:1475-1485.
Yang J, Yan J, Liu BR. Targeting VEGF/VEGFR to Modulate Antitumor Immunity. Front Immunol 2018;9:978.
Barker HE, Paget JT, Khan AA, Harrington KJ. The tumour microenvironment after radiotherapy: mechanisms of resistance and recurrence. Nat Rev Cancer. 2015(7):409-25.
Ye S, Zhao XY, Hu XG, Yang L, Li T, Xu QR, Yang HM, Huang DS, Yang L. TP53 and RET may serve as biomarkers of prognostic evaluation and targeted therapy in hepatocellular carcinoma. Oncology reports. 2017(37): 2215-2226.
Pan JP , Lu CL, Jun W, Wu YF, Shi XL, Ding YT.The up-regulation of P62 levels is associated with resistance of sorafenib in hepatocarcinoma cells. Int J Clin Exp Pathol. 2019;12(7):2622-2630.
Park JW, Kim YJ, Kim DY, Bae SH, Paik SW, Lee YJ, Kim HY, Lee HC, Han SY, Cheong JY, Kwon OS, Yeon JE, Kim BH, Hwang J. Sorafenib with or without concurrent transarterial chemoembolization in patients with advanced hepatocellular carcinoma: The phase III STAH trial. J Hepatol.2019 Apr ;70(4) :684-691.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Effects of TACE combined with sorafenib on survival of patients with hepatocellular carcinoma based on ALBI grades and tumor load layers And Establishment of prognosis prediction model

Status:

Version 1

Abstract

Figures

Background

Methods

Study populations

Administration of sorafenib

Statistical methods

Results

Discussions

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1