Transarterial chemoembolization combined with lenvatinib vs transarterial chemoembolization combined with sorafenib for unresectable hepatocellular carcinoma: A systematic review and meta-analysis

Abstract

BACKGROUND

Lenvatinib and sorafenib are tyrosine kinase inhibitors that are effective in the treatment of unresectable hepatocellular carcinoma (uHCC). The efficacy of which of them is better suited to combine transarterial chemoembolization (TACE) for the treatment of uHCC is ripe.

AIM

To compare the effectiveness of TACE combined with lenvatinib (TACE-lenvatinib) and TACE combined with sorafenib (TACE-sorafenib) in the treatment of uHCC, this study was carried out.

METHODS

Publicly available studies comparing the efficacy of TACE-lenvatinib and TACE-sorafenib in the treatment of uHCC were collected from PubMed, Embase and Cochrane Library, with a cut-off date of December 2024. Stata SE 15 software was used for statistical analysis.

RESULTS

A total of six studies involving 547 patients were included, 248 in the TACE-lenvatinib group and 299 in the TACE-sorafenib group. Meta-analysis results showed that TACE-lenvatinib was more effective than TACE-sorafenib in complete response [relative risk (RR) = 1.81, 95% confidence interval (CI): 1.11-2.96, P = 0.02], partial response (RR = 1.38, 95%CI: 1.12-1.70, P = 0.002), objective response rate (RR = 1.47, 95%CI: 1.24-1.74, P < 0.0001) and disease control rate (RR = 1.22, 95%CI: 1.00-1.49, P = 0.05). TACE-lenvatinib was significantly lower than TACE-sorafenib in progressive disease rate (RR = 0.54, 95%CI: 0.39-0.74, P = 0.002). No significant difference was found in stable disease rate (RR = 0.89, 95%CI: 0.60-1.33, P = 0.58) between the two groups. TACE-lenvatinib was significantly more effective than TACE-sorafenib in overall survival (hazard ratio = 2.00, 95%CI: 1.59-2.50, P < 0.05) and progression free survival (hazard ratio = 2.04, 95%CI: 1.49-2.86, P < 0.05). As regards adverse events, TACE-lenvatinib was better in reducing the incidence of hypertension than TACE-sorafenib, while no significant difference was found in overall adverse events, abdominal pain, fever, fatigue, nausea and vomiting, decreased appetite, liver dysfunction, hand-foot skin reaction, diarrhea, thrombocytopenia, and rash between the two groups.

CONCLUSION

In patients with uHCC, TACE-lenvatinib induced a better tumor response rate and survival outcome than TACE-sorafenib, while TACE-lenvatinib resulted in a higher incidence of hypertension than TACE-sorafenib. However, these conclusions are derived from currently available medical evidence, and further confirmation by more rigorously designed randomized controlled studies is still needed.

Key Words: Hepatocellular carcinoma; Lenvatinib; Sorafenib; Transcatheter arterial chemoembolization; Meta-analysis

Core Tip: In this study, we compared the efficacy of transarterial chemoembolization (TACE)-lenvatinib and TACE-sorafenib against unresectable hepatocellular carcinoma. The TACE-lenvatinib group had higher tumor response rate than the TACE-sorafenib group. We also found that both overall survival and progression-free-survival were higher in the TACE-lenvatinib group than in the TACE-sorafenib group. The TACE-lenvatinib group had a higher incidence of hypertension. Therefore, the combination of TACE and lenvatinib resulted in better tumor response rates and survival outcomes in unresectable hepatocellular carcinoma patients compared to TACE-sorafenib, although the former was associated with a higher incidence of hypertension.

INTRODUCTION

Liver cancer is the most common and third deadliest cancer worldwide, and hepatocellular carcinoma (HCC) is the predominant subtype that accounts for 75%-85% of all liver cancer-related deaths[1]. Surgical resection is still considered a curative approach for HCC, although two-thirds of the patients are diagnosed in the middle to advanced stages of the disease, which precludes the possibility of surgery[2-4]. Transarterial chemoembolization (TACE) has been established as the standard local therapy for unresectable HCC (uHCC), whereas targeted drugs is the mainstay of treating advanced HCC, such as sorafenib and lenvatinib[5,6]. Sorafenib was the first drug approved for the first-line treatment of uHCC, and has been effective when used in combination with TACE[7,8]. While TACE contributes to local tumor control through ischemia and necrosis, sorafenib inhibits TACE-induced upregulation of vascular endothelial growth factor (VEGF), thereby inhibiting tumor re-vascularization and progression.

Lenvatinib is a novel tyrosine kinase inhibitor (TKI) that targets VEGF receptors (VEGFRs) 1-3, fibroblast growth factor receptors (FGFRs) 1-4, and platelet-derived growth factor receptor alpha[8,9] The REFLECT trial[8] demonstrated a significant improvement in the objective response rate (ORR) and progression-free survival (PFS) of liver cancer patients treated with lenvatinib compared to that of the sorafenib arm. Currently, lenvatinib and sorafenib are recommended as first-line therapeutic agents for patients with uHCC. However, monotherapy regimens based on TKIs have shown limited efficacy with response rates below 20%[7,10,11]. A growing number of studies have suggested that the combination of TACE with sorafenib or lenvatinib may increase the tumor response rate and provide a survival benefit for patients with uHCC[12,13]. The aim of this study was to compare the efficacy and safety of TACE combined with lenvatinib (TACE-lenvatinib) with that of TACE combined with sorafenib (TACE-sorafenib) in patients with uHCC.

MATERIALS AND METHODS

The systematic review was reported in accordance with PRISMA guidelines[14]. This study was exempt from the requirement of institutional review-board approval or informed consent from patients on account of the public availability of the data.

Selection of studies

The Embase, PubMed, and Cochrane Library databases were searched for relevant articles published till December 2024. The search strategy is described in Supplementary Table 1. The original authors were contacted for extra information if necessary. In case of multiple studies from the same authors or medical centers with duplicate samples, only the most recent, highest quality studies were selected. The inclusion criteria were as follows: (1) The target population of the study is patients diagnosed with histologically or clinically confirmed uHCC, including patients with TACE-refractory or non-refractory HCC; (2) Comparative analysis of the efficacy of TACE-lenvatinib and TACE-sorafenib; (3) No restriction on study sample size; (4) No restriction on follow-up time; (5) No restriction on language type; (6) Human subjects; and (7) Studies assessing treatment response based on Response Evaluation Criteria for Solid Tumors and adverse events (AEs) according to Common Terminology Criteria for AEs v5.0. The exclusion criteria were as follows: (1) Incomplete information, inability to extract valid data and/or contact authors, duplicate publications, unpublished studies; (2) Single-arm studies reporting only the efficacy of TACE-lenvatinib or TACE-sorafenib; and (3) Reviews, case reports, and animal experiments. A risk assessment was conducted for all randomized control trials (RCTs) according to the “risk assessment tool” recommended by the Cochrane Collaboration Network. The quality of the cohort studies was assessed with the Newcastle-Ottawa Scale as detailed in Supplementary Table 2.

Statistical analysis

Stata SE 15 software was used for statistical analysis. The relative risk (RR) was calculated to compare effect sizes for dichotomous variables by the Mantel-Haenszel method, and hazard ratios (HRs) were used to compare effect sizes for survival data by the inverse variance method. Heterogeneity among studies was qualitatively evaluated using a χ²-based Q test with I² statistics; I² < 30%, 30% ≤ I² ≤ 50%, I² > 50% were defined as low, moderate, and high heterogeneity respectively. A random model which based on the Wald-type framework was used in this study. Sensitivity analysis was performed by removing one study at a time to assess whether the results were markedly affected by a single study. Publication bias was quantified through Begg’s test and Egger’s test, with P < 0.05 as the threshold for significance. Funnel plots were used for the qualitative assessment of publication bias.

RESULTS

Search results and study selection

A total of 63 articles were retrieved, of which 38 remained after excluding the duplicates. We then excluded reviews, case reports and other types of literature, and incorporated the remaining 6 articles[15-20] in the meta-analysis. The detailed steps of literature search are outlined in Figure 1. The 6 studies[15-20] included a total of 547 patients, of which 248 (45.34%) received TACE-sorafenib and 299 (54.66%) received TACE-sorafenib. The characteristics of these studies are summarized in Tables 1 and 2.

Figure 1 PRISMA flow diagram. TACE: Transarterial chemoembolization.

Table 1 Basic characteristics and quality assessment of included studies.

Ref.	Country	Period	Medical center	Study type	TACE + LEN vs TACE + SOR, n	Gender (F/M), TACE + LEN	Gender (F/M), TACE + SOR	Age, TACE + LEN	Age, TACE + SOR	Quality
Ding et al[15], 2021	China	2018-2020	Beijing Ditan Hospital	R	32 vs 32	7/25	5/27	57 ± 11	56 ± 11	5
Xue et al[16], 2021	China	2017-2020	First Affiliated Hospital of Sun Yat-Sen University	RCT	50 vs 100	4/46	3/97	54 (49-61)	54 (49-63)	6
Xu et al[17], 2023	China	2019-2021	First Affiliated Hospital of Anhui Medical University	R	43 vs 41	18/25	17/24	55 ± 12.26	52.5 ± 11.47	6
Yang et al[18], 2021	China	2017-2020	West China Medical School	R	38 vs 38	4/34	4/34	55.18 ± 10.94	54.39 ± 12.17	NA
Zhang et al[19], 2022	China	2018-2021	The First Affiliated Hospital with Nanjing Medical University	R	53 vs 59	9/44	7/52	57.7 ± 11.8	58.8 ± 11.1	7
Zhao et al[20], 2022	China	2018-2020	Chinese People’s Liberation Army General Hospital, Beijing	R	32 vs 29	1/31	4/25	57.38 ± 9.44	55.90 ± 8.18	7

R: Retrospective cohort studies; RCT: Randomized controlled trial; M: Male; F: Female; TACE: Transarterial chemoembolization; LEN: Lenvatinib; SOR: Sorafenib; NA: Not available.

Table 2 Other basic characteristics and quality assessment of included studies.

Ref.	Prior HBV infection, TACE + LEN1	Prior HBV infection, TACE + SOR1	ECOG-PS (0-1/2), TACE + LEN	ECOG-PS (0-1/2), TACE + SOR	Child-Pugh (A/B), TACE + LEN	Child-Pugh (A/B), TACE + SOR	Invading blood vessels, TACE + LEN1	Invading blood vessels, TACE + SOR1	Distant metastasis, TACE + LEN1	Distant metastasis, TACE + SOR1	AFP (< 400/≥ 400, ng/mL), TACE + LEN	AFP (< 400/≥ 400, ng/mL), TACE + SOR
Ding et al[15], 2021	30/2	29/3	24/7	22/10	22/10	28/4	32/0	32/0	19/13	23/9	16/16	14/18
Xue et al[16], 2021	46/4	97/3	37/13	84/16	41/9	84/16	36/14	81/19	27/23	45/55	NA	NA
Xu et al[17], 2023	38/5	36/5	20/23	23/18	32/11	28/13	17/26	15/26	11/32	12/29	23/20	26/15
Yang et al[18], 2021	36/2	34/4	27/11	27/11	37/1	37/1					14/24	15/23
Zhang et al[19], 2022	50/3	55/4	43/10	45/14	52/1	57/2	36/17	41/18	36/17	39/20	20/33	25/34
Zhao et al[20], 2022	22/1	21/11	12/11	18/14	19/4	27/5	10/13	18/14	1/22	22/10	NA	NA

¹No/Yes.

TACE: Transarterial chemoembolization; LEN: Lenvatinib; SOR: sorafenib; ECOG: Eastern Cooperative Oncology Group; AFP: Alpha fetoprotein; NA: Not available.

Results of the meta-analysis

The efficacy and safety of TACE-lenvatinib and TACE-sorafenib were compared in terms of tumor response rates, long-term outcomes, and AEs.

Tumor response rate

Six studies[15-20] reported complete response (CR), partial response (PR), stable disease, progressive disease (PD), ORR, and disease control rate (DCR). The CR (RR = 1.81, 95%CI: 1.11-2.96, P = 0.02), PR (RR = 1.38, 95%CI: 1.12-1.70, P = 0.002), ORR (RR = 1.47, 95%CI: 1.24-1.74, P < 0.0001) and DCR (RR = 1.22, 95%CI: 1.00-1.49, P = 0.05) were significantly higher in the TACE-lenvatinib group compared to the TACE-sorafenib group. On the other hand, the TACE-lenvatinib group had lower PD rate than the TACE-sorafenib group (RR = 0.54, 95%CI: 0.39-0.74, P = 0.0002). The rate of stable disease was similar in both groups (RR = 0.89, 95%CI: 0.55-1.44, P > 0.05). The results are shown in Table 3 and Figure 2A.

Figure 2 Forest map. A: Forest map of complete response, partial response, stable disease, and progressive disease; B: Forest map of objective response rate and disease control rate; C: Forest map of overall survival and disease-free survival. RR: Relative risk; CI: Confidence interval; CR: Complete response; PR: Partial response; SD: Stable disease; PD: Progressive disease; ORR: Objective response rate; DCR: Disease control rate; HR: Hazard ratio; OS: Overall survival; PFS: Progression-free survival.

Table 3 Tumor response rate and long-term outcome.

Measured outcomes	Studies, n	Heterogeneity test, I2 (%)	Heterogeneity test, P value	Model	RR/HR	95%CI	P value
Complete response	6	0	0.95	Random	1,81	1.11-2.96	0.02
Partial response	6	0	0.78	Random	1.38	1.12-1.70	0.002
Stable disease	6	50	0.08	Random	0.89	0.60-1.33	0.58
Progressive disease	6	0	0.67	Random	0.54	0.39-0.74	0.0002
Objective response rate	6	0	0.80	Random	1.47	1.24-1.74	< 0.0001
Disease control rate	6	82	< 0.0001	Random	1.22	1.00-1.49	0.05
Overall survival	4	39.8	0.173	Random	0.28	0.19-0.42	< 0.05
Disease free survival	4	0	0.482	Random	0.54	0.42-0.69	< 0.05

RR/HR: Relative risk /hazard ratio; CI: Confidence interval.

Long-term outcomes

Four studies[16,18-20] reported overall survival (OS) and disease-free survival. The TACE-lenvatinib group showed significantly higher OS (HR = 2.00, 95%CI: 1.59-2.50, P < 0.05) and PFS (HR = 2.04, 95%CI: 1.49-2.86, P < 0.05) than the TACE-sorafenib group. The results are shown in Table 3 and Figure 2B and C.

AEs

The included studies[15-20] reported all AEs, in addition to abdominal pain, fever, fatigue, nausea, vomiting, decreased appetite, liver dysfunction, hypertension, hand- foot skin reaction, diarrhea, thrombocytopenia, and rashes. The incidence of all AEs (RR = 0.97, 95%CI: 0.79-1.20, P = 0.80), abdominal pain (RR = 0.95, 95%CI: 0.65-1.41, P = 0.81), fever (RR = 0.95, 95%CI: 0.72-1.25, P = 0.69), fatigue (RR = 1.28, 95%CI: 0.79-2.05, P = 0.32), nausea and vomiting (RR = 1.25, 95%CI: 0.85-1.83, P = 0.25), decreased appetite (RR = 1.18, 95%CI: 0.81-1.71, P = 0.40), liver dysfunction (RR = 1.04, 95%CI: 0.62-1.75, P = 0.87), hand-foot skin reaction (RR = 0.74, 95%CI: 0.18-3.04, P = 0.68), diarrhea (RR = 1.38, 95%CI: 0.55-3.44, P = 0.49), thrombocytopenia (RR = 1.18, 95%CI: 0.71-1.98, P = 0.52), and rashes (RR = 0.88, 95%CI: 0.19-4.16, P = 0.88) were similar in both treatment groups. However, the TACE-lenvatinib group had a significantly higher incidence of hypertension (RR = 2.53, 95%CI: 1.19-5.39, P = 0.02) compared to the TACE-sorafenib group. The results are shown in Table 4 and Supplementary Table 3.

Table 4 Adverse events.

Adverse events	Studies, n	Heterogeneity test, I2(%)	Heterogeneity test, P value	Model	RR	95%CI	P value
All adverse events	3	0	0.53	Random	0.97	0.79-1.20	0.80
Abdominal pain	4	0	0.68	Random	0.95	0.65-1.41	0.81
Fever	3	0	0.66	Random	0.95	0.72-1.25	0.69
Fatigue	4	0	0.68	Random	1.28	0.79-2.05	0.32
Nausea and vomiting	5	0	0.70	Random	1.25	0.85-1.83	0.25
Decreased appetite	5	0	0.76	Random	1.18	0.81-1.71	0.40
Liver dysfunction	4	0	0.46	Random	1.04	0.62-1.75	0.87
Hypertension	5	0	0.44	Random	2.53	1.19-5.39	0.02
Hand-foot skin reaction	5	75	0.003	Random	0.74	0.18-3.04	0.68
Diarrhea	6	39	0.16	Random	1.38	0.55-3.44	0.49
Thrombocytopenia	3	0	0.71	Random	1.18	0.71-1.98	0.52
Rash	6	65	0.04	Random	0.88	0.19-4.16	0.88

RR: Relative risk; CI: Confidence interval.

Sensitivity analysis and publication bias

According to the sensitivity analyses, the results of the meta-analysis were stable. We did not detect publication bias with Begg’s test and Egger’s test which are shown in Supplementary Table 4. The funnel plots with pseudo 95% confidence limits are shown in Supplementary Figure 1.

DISCUSSION

A growing body of research suggests that TACE in combination with sorafenib may increase tumor response rates and provide a survival benefit in patients with uHCC[12,13]. However, TACE can upregulate VEGF, which is a key factor driving neoangiogenesis and tumor progression[21]. While sorafenib inhibits VEGF[22], sorafenib also activates the FGF family of proteins which may weaken VEGF inhibition[23]. In contrast, lenvatinib exerts inhibitory effects on the FGF/FGFR as well as the VEGF/VEGFR pathways. In addition, lenvatinib may also inhibit other signaling pathways with pro-angiogenic effects, such as platelet-derived growth factor, which is not affected by sorafenib[24,25]. It stands to reason that the differences in the mechanisms of both TKIs may influence tumor control in the TACE-lenvatinib and TACE-sorafenib groups. Given its broad-spectrum inhibitory action against VEGFRs 1-3, platelet-derived growth factor receptor alpha, FGFRs 1-4, and RET, lenvatinib in combination with TACE could theoretically result in stronger inhibition of tumor neovascularization, and therefore a greater reduction in tumor growth.

In this study, we compared the efficacy of TACE-lenvatinib and TACE-sorafenib against uHCC. The TACE-lenvatinib group had higher CR, PR, ORR, and DCR, and significantly lower PD rates than the TACE-sorafenib group. Although TACE-lenvatinib group exhibited superior tumor response as per the Response Evaluation Criteria for Solid Tumors criteria, the tumor response often does not reflect the direct clinical benefits of anti-neoplastic agents, such as improved survival, improved quality of life for patients, or reduction of tumor-related symptoms. Therefore, we also analyzed the long-term survival outcomes in both groups, and found that both OS and PFS were higher in the TACE-lenvatinib group than in the TACE-sorafenib group. The combination of TACE and sorafenib is considered clinically safe[7,8]. Consistent with this, there were no difference in the incidence of AEs between the two treatment groups in our study, although the TACE-lenvatinib group had a higher incidence of hypertension. Furthermore, the most common side effects of lenvatinib are hand-foot skin reactions, diarrhea and hypertension, whereas fever, pain and transient transaminase elevation primarily result from TACE. These treatment-related AEs are predominantly grade 1 or 2, which can be resolved or eliminated with appropriate and timely management.

A previous meta-analysis[26] showed that although TACE-lenvatinib was significantly more effective than TACE-sorafenib in patients with uHCC, the former was associated with serious AEs. The authors concluded that TACE-sorafenib should be used as an alternative to TACE-lenvatinib in case of serious AEs. However, two studies in that meta-analysis[27,28] included patients with uHCC refractory to TACE. In fact, some uHCC patients are unresponsive to TACE due to considerable tumor heterogeneity, and repeated ineffective TACE can even aggravate liver injury and worsen prognosis. The Global Liver Cancer Society proposed the concept of “TACE failure refractoriness” as a criterion for timely termination of TACE in order to avoid repeated ineffective treatments[29-31]. In the event that TACE is ineffective, the response of patients depends on the efficacy of the TKI, making the combination treatment redundant. According to the principle of intention to treat, a patient with uHCC cannot know before TACE whether the treatment will be effective. Therefore, inclusion of patients who are unresponsive to TACE[27,28] likely introduced bias in the meta-analysis. Therefore, we excluded these studies in our meta-analysis, and our conclusions were close to real-world unresectable intermediate and advanced HCC who are about to receive their initial treatment, including those refractory and non-refractory to TACE.

While retrospective studies can reflect real-world scenarios, their inherent bias cannot be avoided. Meta-analyses of well-designed non-randomized comparative studies are likely to be as accurate as RCTs[32]. We performed comparative meta-analysis, which is better accepted in the medical literature as a complementary tool to qualitative reviews[33]. We believe that our findings are significant and provide a comprehensive picture of the real-world clinical efficacy of combining TACE and lenvatinib in the treatment of uHCC. Taken together, TACE-based combination therapy warrants further investigation in future RCTs.

The limitations of the study were as followed: First, the included studies were retrospective, which may have led to selection bias. Second, most of the studies on uHCC patients did not perform further subgroup analysis, which limited the further subdivision of the conclusions. Finally, most of the studies were conducted on limited number of cases from single medical centers, which reduced the interpretive power for the conclusions.

CONCLUSION

The combination of TACE and lenvatinib resulted in better tumor response rates and survival outcomes in uHCC patients compared to TACE-sorafenib, although the former was associated with a higher incidence of hypertension. However, further validation with more rigorously designed RCTs and longer follow-up are still needed in the future.