Comparative efficacy and safety for second-line treatment with ramucirumab, regorafenib, and cabozantinib in patients with advanced hepatocellular carcinoma progressed on sorafenib treatment: A network meta-analysis

Abstract

Background: The present network meta-analysis was conducted to perform an indirect comparison among ramucirumab, regorafenib, and cabozantinib in patients with advanced hepatocellular carcinoma (HCC) progressed on sorafenib treatment.

Methods: A systematic review through Medline, Embase, and Cochrane library was developed, with eligible randomized clinical trials been included. Hazard ratios (HRs) including progression-free survival (PFS), overall survival (OS), odds ratios of disease control rate (DCR), objective response rate (ORR), and adverse events were compared indirectly with network meta-analysis using random model in software STATA version 13.0.

Results: A total of 4 randomized clinical trials including 2137 patients met the eligibility criteria and enrolled. Indirect comparisons showed that there was no statistical difference observed in the indirect comparison of PFS, OS, ORR, or DCR among agents of regorafenib, cabozantinib, and ramucirumab in advanced HCC patients with elevated α-fetoprotein (AFP) (400 ng/mL or higher). However, in patients with low-level AFP (lower than 400 ng/mL), regorafenib was the only agent associated with significant superiority in OS, compared with placebo (hazard ratio 0.67, 95% CI, 0.50-0.90).

Conclusions: The present network meta-analysis revealed that there might be no statistical difference observed in the indirect comparison of PFS, OS, ORR, or DCR among regorafenib, cabozantinib, or ramucirumab in advanced HCC patients with elevated AFP (400 ng/mL or higher). However, in patients with low-level AFP (lower than 400 ng/mL), regorafenib might be associated with significant superiority in OS, compared to placebo, which need further investigation in clinical practice.

Figure 1

Study selection procedure with PRISMA flow diagram. PRISMA = Preferred Reporting Items For Systematic Reviews and Meta-analyses.

Figure 2

Network of the comparisons.

Figure 3

Direct and indirect comparisons for PFS among Ram (ramucirumab), Reg (regorafenib), Cab (cabozantinib), and PLA (placebo). PFS = progression-free survival.

Figure 4

Direct and indirect comparisons for OS among Ram (ramucirumab), Reg (regorafenib), Cab (cabozantinib), and PLA (placebo). OS = overall survival.

Figure 5

Direct and indirect comparisons for ORR (A) and DCR (B) among Ram (ramucirumab), Reg (regorafenib), Cab (cabozantinib), and PLA (placebo). DCR = disease control rate; ORR = objective response rate.

Figure 6

Direct and indirect comparisons for PFS (A and C) and OS (B and D) by AFP level (A and B for low-level AFP; C and D for high level AFP), among Ram (ramucirumab), Reg (regorafenib), Cab (cabozantinib), and PLA (placebo). AFP = α-fetoprotein; OS = overall survival; PFS = progression-free survival.

Figure 7

Treatment comparison for SAEs (A), FAEs (B), and TEs (patients suffered dose reduction, (C); death during the therapy or within 30 days of treatment discontinuation, (D) (ORs, 95% CI). FAEs = fatal adverse events; OR = odds ratio; SAEs = severe adverse events; TEs = treatment events.

Figure 8

Treatment comparison for elevated ALT (A), hypertension (B), vomiting (C), and fatigue (D) at all grades (ORs, 95% CI). ORs = odds ratios.

Figure 9

Treatment comparison for elevated ALT (A), hypertension (B), vomiting (C), and fatigue (D) at high grade (ORs, 95% CI). OR = odds ratio.

Figure 10

Funnel plot for the detection of publication bias (Ram, ramucirumab; Reg, regorafenib; Cab, cabozantinib; PLA, placebo).