Published online Nov 15, 2026. doi: 10.4251/wjgo.123385
Revised: June 29, 2026
Accepted: July 16, 2026
Published online: November 15, 2026
Processing time: 172 Days and 18.8 Hours
Sorafenib and lenvatinib remain the standard first-line therapies for unresectable hepatocellular carcinoma (HCC); however, disease progression is nearly univer
To compare the clinical efficacy of regorafenib plus PD1 inhibitor vs monotherapy as a second-line treatment for patients with unresectable HCC.
This study analyzed 140 advanced HCC patients who experienced disease progression after first-line therapy between February 2019 and August 2024. Patients were divided into two cohorts, the regorafenib plus PD1 inhibitor group [regorafenib plus PD1 inhibitor (RP) group, n = 101] and the regorafenib monotherapy group [regorafenib monotherapy (R) group, n = 39]. Clinical outcomes were evaluated using objective response rate, disease control rate (DCR), progression-free survival, and overall survival (OS). Subgroup analyses and mul
The RP group achieved a higher DCR, with better control of both intrahepatic and extrahepatic lesions than the R group. Median PFS (5.4 months vs 2.7 months, P = 0.003) and OS (54.8 months vs 25.3 months, P = 0.041) were longer in the RP group. Patients with prior ICI exposure achieved longer OS with RP therapy (P = 0.006), whereas no significant OS difference was observed in ICI-naive patients. First-line treatment type (lenvatinib vs sorafenib) did not affect second-line outcomes. Multivariable analysis identified second-line PD-1 inhibitor therapy as an independent predictor of PFS, whereas no independent predictors of OS were identified.
For patients with unresectable HCC who fail first-line treatment, the combination of regorafenib and PD-1 inhibitors confers better survival benefits and disease control than regorafenib monotherapy.
Core Tip: This multicenter retrospective study provides real-world evidence supporting regorafenib plus programmed cell death-1 (PD-1) inhibitors as second-line therapy for unresectable hepatocellular carcinoma after first-line treatment failure. Combination therapy improved disease control and prolonged progression-free and overall survival. Notably, survival benefit was mainly observed in patients with prior immune checkpoint inhibitor exposure, whereas first-line targeted therapy (lenvatinib or sorafenib) did not influence second-line outcomes. Second-line PD-1 inhibitor therapy independently pre
- Citation: Ma F, Zheng LY, Song DJ, Zhao ZW, Cao F, Hao WY, Qiu JQ, Lu D, Zheng JP. Regorafenib combined with programmed death inhibitor after initial treatment failure in unresectable hepatocellular carcinoma. World J Gastrointest Oncol 2026; 18(11): 123385
- URL: https://www.wjgnet.com/1948-5204/full/v18/i11/123385.htm
- DOI: https://dx.doi.org/10.4251/wjgo.123385
Hepatocellular carcinoma (HCC) is one of the most prevalent malignancies globally, ranking sixth in terms of global incidence and third in cancer-related mortality[1]. Similarly, HCC is also one of the most prevalent cancers in China[2]. Due to the insidious nature of early clinical symptoms of HCC, most patients are diagnosed at an advanced stage and thus lose the opportunity for radical surgical resection[3]. For unresectable HCC, the main therapeutic strategies include local therapy, systemic anti-tumor therapy, conversion therapy, and supportive care[4]. In recent years, significant advances have been made in systemic therapy. First-line tyrosine kinase inhibitors (TKIs), such as sorafenib and lenvatinib, have been validated to confer significant survival benefits in multiple clinical trials[5]. Moreover, the phase III IMbrave150 trial demonstrated that atezolizumab combined with bevacizumab improves overall survival (OS) in patients with unresectable HCC[6]. However, some patients experience disease progression or intolerable toxicity, requiring second-line or subsequent treatment[7].
Regorafenib was the first agent approved by the United States Food and Drug Administration for second-line treatment of patients with advanced HCC based on the results of RESORCE trial[8,9]. Previous evidence has revealed that regorafenib can inhibit vascular endothelial growth factor receptor (VEGFR), modulate the JAK1/2-STAT1 and MAPK pathways, and downregulate programmed cell death ligand-1 (PD-L1) expression in HCC[10]. Additionally, it can normalize tumor vasculature and promote CD8+ T cell infiltration, thereby potentiating the antitumor efficacy of programmed cell death-1 (PD-1) antibodies[11]. These effects suggest that the combination of regorafenib and PD-1 inhibitors may further improve the treatment efficacy. A real-world study has demonstrated that regorafenib combined with PD-1 inhibitors as a second-line therapy could prolongs the median progression-free survival (PFS) compared with the regorafenib monotherapy (11.50 months vs 5.13 months) in patients with unresectable HCC[12]. In addition, this study further explored the prognostic differences in the second-line therapy among patients who received lenvatinib or sorafenib as first-line therapy, and assessed whether different first-line targeted treatments influenced the efficacy of second-line regorafenib combined with PD-1 inhibitors or monotherapy. Herein, based on multicenter retrospective data, this study evaluated the clinical efficacy of the combined therapeutic regimen for unresectable HCC and identified relevant prognostic factors. Subgroup analyses were further performed to assess the prognostic impacts of first-line immune checkpoint inhibitor (ICI) exposure and different first-line TKI regimens (lenvatinib vs sorafenib).
This multicenter retrospective study included 181 patients with HCC who received second-line regorafenib-based therapy after developing disease progression or experiencing intolerable toxicity to prior first-line treatment (e.g., sorafenib or lenvatinib). Patients were treated at four institutions: Zhejiang Cancer Hospital, The Fifth Affiliated Hospital of Wenzhou Medical University, The First Affiliated Hospital of University of Science and Technology of China, and The People’s Hospital of Pingyang, between February 2019 and August 2024. This retrospective multicenter study was approved by medical ethics committee of Zhejiang Cancer Hospital [approval No. IRB-2026-762 (IIT)] and other participating centers. It adhered to the Helsinki Declaration’s ethical guidelines, exempting patients from informed consent.
The inclusion criteria were as follows: (1) Age ≥ 18 years; (2) Clinically or pathologically confirmed diagnosis of HCC; (3) China Liver Cancer (CNLC) stage II or III; and (4) Receipt of regorafenib monotherapy or regorafenib combined with a PD1 inhibitor after first-line treatment failure. Patients were excluded from the study if they met any of the following exclusion criteria: (1) Absence of evaluable efficacy data; (2) Missing essential baseline clinical information; or (3) Presence of additional malignancies other than primary HCC. Figure 1 outlines the patient inclusion and exclusion process.
Baseline data were collected within one week before the initiation of second-line treatment. Recorded parameters included (1) Age; (2) Sex; (3) History of comorbidities (hypertension, diabetes mellitus); (4) Hepatitis B/C virus infection status; (5) Eastern Cooperative Oncology Group performance status; (6) Child-Pugh class; (7) CNLC stage; (8) Tumor size; (9) Portal vein tumor thrombus; (10) First-line treatment regimen; (11) Prior exposure to ICIs during first-line therapy; and (12) Initial daily dose of regorafenib. Laboratory examinations included: (1) Complete blood count; (2) Liver function tests; (3) Coagulation function tests; and (4) Serum tumor biomarkers.
In the second-line therapy, patients received oral regorafenib at a daily dose of 40-160 mg (adjusted based on tolerability) for 21 consecutive days followed by a 7-day treatment-free interval, constituting a 28-day cycle. Patients in the com
The primary endpoint was PFS, and the secondary endpoints were OS, objective response rate (ORR), and disease control rate (DCR). PFS was measured from initiation of second-line treatment to disease progression or death from any cause, whichever occurred first. OS was measured from initiation of second-line treatment to death from any cause. ORR and DCR were defined as the proportions of patients achieving complete response (CR) or partial response (PR), and CR, PR, or stable disease (SD), respectively.
Treatment efficacy was assessed starting from the date of the first regorafenib administration. Tumor response was evaluated every 4-8 weeks using computed tomography or magnetic resonance imaging according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 until disease progression, treatment modification, treatment discontinuation, or death. All analyses were conducted based on data available at the final followup date (May 2025).
Categorical variables were presented as n (%) and were compared using the χ2 test or Fisher’s exact test. Continuous variables were expressed as median (interquartile range) or mean ± SD, according to data distribution, and were compared using the independentsamples t-test or Mann-Whitney U test.
OS and PFS were estimated with the Kaplan-Meier method, with betweengroup differences assessed using the Logrank test. Subgroup analyses were performed to explore prognostic variations stratified by different firstline re
In this study, we retrospectively identified 181 patients with advanced HCC who were treated with either regorafenib plus PD1 inhibitors or regorafenib monotherapy as secondline therapy at four participating institutions (Zhejiang Cancer Hospital, The Fifth Affiliated Hospital of Wenzhou Medical University, The First Affiliated Hospital of the University of Science and Technology of China, and The People’s Hospital of Pingyang) between February 2019 and August 2024. Of these, 140 patients met the inclusion criteria and were included in the final analysis: 101 in the regorafenib plus PD1 inhibitor (RP) group and 39 in the regorafenib monotherapy (R) group. No significant differences were observed in baseline characteristics between the two groups including demographic features, comorbidities, laboratory parameters and tumor characteristics (Table 1).
| Characteristics | Overall (n = 140) | R group (n = 39) | RP group (n = 101) | P value |
| Age | 0.811 | |||
| < 65 | 99 (70.7) | 27 (69.2) | 72 (71.3) | |
| ≥ 65 | 41 (29.3) | 12 (30.8) | 29 (28.7) | |
| Gender | 0.652 | |||
| Female | 22 (15.7) | 7 (17.9) | 15 (14.9) | |
| Male | 118 (84.3) | 32 (82.1) | 86 (85.1) | |
| Smoking | 0.839 | |||
| No | 95 (67.9) | 28 (71.8) | 67 (66.3) | |
| Yes | 39 (27.9) | 10 (25.6) | 29 (28.7) | |
| Unknown | 6 (4.3) | 1 (2.6) | 5 (5.0) | |
| Drinking | 0.203 | |||
| No | 103 (73.6) | 26 (66.7) | 77 (76.2) | |
| Yes | 30 (21.4) | 12 (30.8) | 18 (17.8) | |
| Unknown | 7 (5.0) | 1 (2.6) | 6 (5.9) | |
| Hepatitis B infection | 0.908 | |||
| No | 31 (22.1) | 8 (20.5) | 23 (22.8) | |
| Yes | 107 (76.4) | 31 (79.5) | 76 (75.2) | |
| Unknown | 2 (1.4) | 0 (0.0) | 2 (2.0) | |
| Hepatitis C infection | 0.106 | |||
| No | 134 (95.7) | 38 (97.4) | 96 (95.0) | |
| Yes | 1 (0.7) | 1 (2.6) | 0 (0.0) | |
| Unknown | 5 (3.6) | 0 (0.0) | 5 (5.0) | |
| Diabetes | 1.000 | |||
| No | 126 (90.0) | 36 (92.3) | 90 (89.1) | |
| Yes | 13 (9.3) | 3 (7.7) | 10 (9.9) | |
| Unknown | 1 (0.7) | 0 (0.0) | 1 (1.0) | |
| Hypertension | 0.285 | |||
| No | 99 (70.7) | 25 (64.1) | 74 (73.3) | |
| Yes | 41 (29.3) | 14 (35.9) | 27 (26.7) | |
| First-line ICI therapy | 0.224 | |||
| No | 53 (37.9) | 14 (35.9) | 39 (38.6) | |
| Yes | 72 (51.4) | 18 (46.2) | 54 (53.5) | |
| Unknown | 15 (10.7) | 7 (17.9) | 8 (7.9) | |
| Tumor size | 0.323 | |||
| ≤ 5 cm | 46 (32.9) | 12 (30.8) | 34 (33.7) | |
| > 5 cm | 25 (17.9) | 10 (25.6) | 15 (14.9) | |
| Unknown | 69 (49.3) | 17 (43.6) | 52 (51.5) | |
| CNLC | 0.715 | |||
| IIa + IIb | 50 (35.7) | 13 (33.3) | 37 (36.6) | |
| IIIa + IIIb | 90 (64.3) | 26 (66.7) | 64 (63.4) | |
| Child-Pugh class | 0.714 | |||
| A | 74 (52.9) | 22 (56.4) | 52 (51.5) | |
| B | 21 (15.0) | 5 (12.8) | 16 (15.8) | |
| C | 2 (1.4) | 0 (0.0) | 2 (2.0) | |
| Unknown | 43 (30.7) | 12 (30.8) | 31 (30.7) | |
| ECOG PS | ||||
| 0 | 22 (15.7) | 5 (12.8) | 17 (16.8) | 0.908 |
| 1 | 26 (18.6) | 9 (23.1) | 17 (16.8) | |
| 2 | 9 (6.4) | 2 (5.1) | 7 (6.9) | |
| 3 | 1 (0.7) | 0 (0.0) | 1 (1.0) | |
| Unknown | 82 (58.6) | 23 (59.0) | 59 (58.4) | |
| Portal vein invasion tumor | ||||
| No | 66 (47.1) | 19 (48.7) | 47 (46.5) | 0.702 |
| Yes | 30 (21.4) | 9 (23.1) | 21 (20.8) | |
| Unknown | 44 (31.4) | 11 (28.2) | 33 (32.7) | |
| Sites of metastasis | ||||
| Lung | 28 (45.2) | 9 (40.9) | 19 (47.5) | |
| Bone | 8 (12.9) | 3 (13.6) | 5 (12.5) | |
| Lymph node | 16 (25.8) | 6 (27.3) | 10 (25.0) | |
| Others | 10 (16.1) | 4 (18.2) | 6 (15.0) | |
| Laboratory indicators, median [IQR] | ||||
| Platelet (× 109/L) | 151.00 [111.00, 217.25] | 152.00 [110.50, 178.00] | 151.00 [111.00, 222.00] | 0.577 |
| Hemoglobin (g/L) | 132.00 [107.00, 144.25] | 127.00 [114.50, 145.50] | 134.00 [107.00, 143.00] | 0.961 |
| ALT (IU/L) | 35.00 [23.00, 65.00] | 37.00 [23.00, 54.00] | 34.00 [23.00, 73.00] | 0.705 |
| AST (IU/L) | 41.50 [29.00, 82.75] | 46.00 [31.00, 104.00] | 41.00 [27.00, 74.00] | 0.265 |
| ALB (g/L) | 39.70 [36.12, 43.90] | 39.70 [36.45, 43.90] | 39.70 [36.20, 43.90] | 0.917 |
| TBIL (μmol/L) | 17.20 [11.15, 23.30] | 16.30 [13.20, 26.30] | 17.30 [10.65, 21.90] | 0.430 |
| DBIL (μmol/L) | 7.95 [5.18, 11.73] | 7.70 [5.00, 12.00] | 8.00 [5.20, 11.35] | 0.631 |
| PT (s) | 12.55 [11.80, 13.55] | 12.70 [11.90, 13.85] | 12.50 [11.70, 13.30] | 0.503 |
| INR | 1.03 [0.98, 1.11] | 1.04 [0.90, 1.10] | 1.02 [0.99, 1.13] | 0.465 |
| AFP (ng/mL) | 124.20 [7.60, 2000.00] | 101.10 [7.85, 1978.90] | 124.20 [7.20, 2403.61] | 0.970 |
| Initial daily dose of regorafenib | ||||
| 40 mg | 6 (4.3) | 1 (2.6) | 5 (5.0) | 0.057 |
| 80 mg | 65 (46.4) | 13 (33.3) | 52 (51.5) | |
| 120 mg | 12 (8.6) | 6 (15.4) | 6 (5.9) | |
| 160 mg | 18 (12.9) | 5 (12.8) | 13 (12.9) | |
| Unknown | 39 (27.9) | 14 (35.9) | 25 (24.8) |
Treatment efficacy was evaluated according to RECIST 1.1 criteria for overall, intrahepatic, and extrahepatic lesions. No patient achieved a CR in the overall cohort. Among all patients, PR was observed in 2.85% of patients, SD in 57.86%, and PD in 39.29% in the overall assessment, resulting in an ORR of 2.85% and a DCR of 60.71% (Supplementary Table 1). For intrahepatic lesions, PR and SD rates were 2.85% and 73.57%, respectively, yielding an ORR of 2.85% and a DCR of 76.43%. For extrahepatic lesions, the corresponding rates were 2.14% for PR and 25.00% for SD, with an ORR of 2.14% and a DCR of 27.14%.
Patients were further stratified by treatment regimen (RP group vs R group) for lesion-based response analysis (Table 2). PR was observed only in the RP group, with rates of 3.96% for intrahepatic lesions and 5.77% for extrahepatic lesions, whereas no PR was recorded in the R group. For intrahepatic lesions, the RP group achieved a higher DCR than the R group (83.17% vs 58.97%, P = 0.007), accompanied by a lower PD rate (16.83% vs 38.46%, P = 0.007). A similar pattern was observed for extrahepatic lesions, with a higher DCR (59.62% vs 31.82%, P = 0.004) and a lower PD rate (40.38% vs 68.18%, P = 0.042) in the RP group. The recorded treatment-related adverse events are summarized in Supple
| Best Response | Overall response | Intrahepatic response | Extrahepatic response | ||||||
| RP group | R group | P value | RP group | R group | P value | RP group | R group | P value | |
| CR | 0 (0.00) | 0 (0.00) | 1.000 | 0 (0.00) | 0 (0.00) | 1.000 | 0 (0.00) | 0 (0.00) | 1.000 |
| PR | 4 (3.96) | 0 (0.00) | 0.576 | 4 (3.96) | 0 (0.00) | 0.575 | 3 (5.77) | 0 (0.00) | 0.550 |
| SD | 66 (65.35) | 15 (38.46) | 0.005 | 80 (79.21) | 23 (58.97) | 0.031 | 28 (53.85) | 7 (31.82) | 0.126 |
| PD | 31 (30.69) | 24 (61.54) | 0.001 | 17 (16.83) | 15 (38.46) | 0.007 | 21 (40.38) | 15 (68.18) | 0.042 |
| ORR | 4 (3.96) | 0 (0.00) | 0.576 | 4 (3.96) | 0 (0.00) | 0.575 | 3 (5.77) | 0 (0.00) | 0.550 |
| DCR | 70 (69.31) | 15 (38.46) | 0.001 | 84 (83.17) | 23 (58.97) | 0.007 | 31 (59.62) | 7 (31.82) | 0.0042 |
For the entire cohort, the median OS was 54.6 months [95%CI: 25.3-not available (NA)], and the median PFS was 4.3 months (95%CI: 3.5-5.43) (Figure 2). In the RP group, median OS was 54.8 months (95%CI: 27.6-NA), compared with 25.3 months (95%CI: 15.0-NA) in the R group, representing a significant survival benefit for the combination therapy (HR = 0.506, 95%CI: 0.260-0.984, P = 0.041) (Figure 3A). Similarly, median PFS was 5.4 months (95%CI: 4.33-7.33) in the RP group vs 2.7 months (95%CI: 2.47-3.77) in the R group, with a longer PFS observed in the RP group (HR = 0.562, 95%CI: 0.380-0.831, P = 0.003) (Figure 3B).
Stratified analysis based on prior ICI exposure status revealed that OS exhibited no significant difference between the RP and R groups in patients without first-line ICI exposure (Figure 4A). In contrast, among patients with firstline ICI exposure, the OS was longer in the RP group than in the R group (P = 0.006) (Figure 4B). When stratified by the type of firstline targeted therapy, no significant difference in OS was observed between the two groups for patients receiving either lenvatinib or sorafenib (Figure 5). The findings on PFS were consistent with those on OS, demonstrating alignment between the two endpoints (Supplementary Figure 1). Among patients who received first-line treatment for HCC, lenvatinib did not improve PFS or OS compared with sorafenib (Supplementary Figure 2). Among patients with documented PD-1 inhibitor information, PFS and OS were comparable between those receiving camrelizumab and tislelizumab (Supplementary Figure 3).
For OS, univariate Cox regression analysis identified diabetes (P = 0.008), alanine aminotransferase (P = 0.052), alpha-fetoprotein (AFP) (P = 0.020), prior first-line ICI therapy (P = 0.083), Child-Pugh class (P = 0.024), and second-line PD-1 inhibitor therapy (P = 0.044) as candidate variables for multivariable analysis (Table 3). However, none remained independently associated with OS in the multivariable analysis (Table 3).
| Characteristics | Univariate analysis | Multivariate analysis | ||
| HR (95%CI) | P value | HR (95%CI) | P value | |
| Gender (female vs male) | 1.551 (0.604-3.984) | 0.362 | ||
| Age (< 65 vs ≥ 65 years) | 0.937 (0.428-2.054) | 0.871 | ||
| Hepatitis B infection (no vs yes) | 0.910 (0.426-1.944) | 0.807 | ||
| Hepatitis C infection (no vs yes) | 0.000 (0.000-inf) | 0.997 | ||
| Diabetes (no vs yes) | 3.291 (1.362-7.952) | 0.008 | 4.319 (0.783-23.801) | 0.093 |
| Hypertension (no vs yes) | 1.267 (0.646-2.483) | 0.491 | ||
| Smoking (no vs yes) | 1.042 (0.516-2.101) | 0.909 | ||
| Drinking (no vs yes) | 1.120 (0.531-2.360) | 0.766 | ||
| Tumor size (≤ 5 vs > 5 cm) | 1.626 (0.728-3.631) | 0.235 | ||
| Child-Pugh class (A vs B + C) | 2.456 (1.127-5.352) | 0.024 | 1.100 (0.349-53.465) | 0.871 |
| ECOG PS (0 vs ≥ 1) | 1.002 (0.454-2.212) | 0.996 | ||
| Portal vein invasion tumor (no vs yes) | 1.473 (0.688-3.153) | 0.319 | ||
| CNLC (IIa + IIb vs IIIa + IIIb) | 1.347 (0.685-2.649) | 0.388 | ||
| ALT (≤ 50 vs > 50 IU/L) | 2.044 (0.993-4.207) | 0.052 | 2.221 (0.984-5.011) | 0.055 |
| AST (≤ 40 vs > 40 IU/L) | 1.353 (0.656-2.793) | 0.413 | ||
| ALB (< 35 vs ≥ 35 g/L) | 0.659 (0.313-1.389) | 0.273 | ||
| TBIL (≤ 23 vs > 23 μmol/L) | 0.622 (0.238-1.622) | 0.331 | ||
| AFP (< 400 vs ≥ 400 ng/mL) | 2.332 (1.140-4.769) | 0.020 | 1.805 (0.743-4.383) | 0.192 |
| INR (≤ 1.2 vs > 1.2) | 0.720 (0.098-5.294) | 0.747 | ||
| First-line ICI therapy (no vs yes) | 1.891 (0.921-3.883) | 0.083 | 2.159 (0.794-5.868) | 0.132 |
| Initial daily dose of regorafenib (< 160 vs 160 mg) | 1.440 (0.594-3.491) | 0.419 | ||
| Second-line PD-1 inhibitors therapy (no vs yes) | 0.505 (0.260-0.983) | 0.044 | 0.502 (0.203-1.239) | 0.135 |
For PFS, univariate Cox regression analysis identified sex (P = 0.087), Child-Pugh class (P = 0.052), albumin (P = 0.068), AFP (P = 0.001), and second-line PD-1 inhibitor therapy (P = 0.004) as candidate variables for multivariable analysis (Table 4). In the multivariable analysis, AFP (HR = 1.749, 95%CI: 1.027-2.980, P = 0.040) and second-line PD-1 inhibitor therapy (HR = 0.549, 95%CI: 0.316-0.953, P = 0.033) remained independently associated with PFS (Table 4).
| Characteristics | Univariate analysis | Multivariate analysis | ||
| HR (95%CI) | P value | HR (95%CI) | P value | |
| Gender (female vs male) | 1.552 (0.939-2.566) | 0.087 | 1.800 (0.891, 3.635) | 0.101 |
| Age (< 65 vs ≥ 65 years) | 0.909 (0.606-1.364) | 0.645 | ||
| Hepatitis B infection (no vs yes) | 1.246 (0.793-1.957) | 0.340 | ||
| Hepatitis C infection (no vs yes) | 0.923 (0.128-6.635) | 0.936 | ||
| Diabetes (no vs yes) | 1.484 (0.794-2.772) | 0.216 | ||
| Hypertension (no vs yes) | 0.927 (0.627-1.372) | 0.707 | ||
| Smoking (no vs yes) | 0.748 (0.501-1.117) | 0.156 | ||
| Drinking (no vs yes) | 0.762 (0.488-1.192) | 0.234 | ||
| Tumor size (≤ 5 vs > 5 cm) | 1.021 (0.618-1.688) | 0.935 | ||
| Child-Pugh class (A vs B + C) | 1.680 (0.995-2.837) | 0.052 | 1.260 (0.658-2.414) | 0.485 |
| ECOG PS (0 vs ≥ 1) | 1.171 (0.662-2.072) | 0.587 | ||
| Portal vein invasion tumor (no vs yes) | 1.210 (0.762-1.919) | 0.419 | ||
| CNLC (IIa + IIb vs IIIa + IIIb) | 1.139 (0.787-1.649) | 0.490 | ||
| ALT (≤ 50 vs > 50 IU/L) | 1.037 (0.689-1.560) | 0.863 | ||
| AST (≤ 40 vs > 40 IU/L) | 1.233 (0.840-1.810) | 0.284 | ||
| ALB (< 35 vs ≥ 35 g/L) | 0.692 (0.466-1.027) | 0.068 | 0.698 (0.425-1.176) | 0.177 |
| TBIL (≤ 23 vs > 23 μmol/L) | 0.982 (0.641-1.505) | 0.934 | ||
| AFP (< 400 vs ≥ 400 ng/mL) | 1.908 (1.285-2.834) | 0.001 | 1.749 (1.027-2.980) | 0.040 |
| INR (≤ 1.2 vs > 1.2) | 0.878 (0.380-2.026) | 0.760 | ||
| First-line ICI therapy (no vs yes) | 1.344 (0.915-1.974) | 0.132 | ||
| Initial daily dose of regorafenib (< 160 vs 160 mg) | 1.020 (0.593-1.755) | 0.942 | ||
| Second-line PD-1 therapy (no vs yes) | 0.561 (0.379-0.830) | 0.004 | 0.549 (0.31-0.953) | 0.033 |
This multicenter real-world study showed favorable clinical outcomes with regorafenib plus PD-1 inhibitors as second-line therapy for advanced HCC. The RP group achieved a DCR of 69.31% and a median PFS of 5.4 months, compared with 38.46% and 2.7 months in the R group. These findings suggest that the RP regimen offers tangible clinical benefits in advanced HCC. Previous preclinical studies have shown that regorafenib enhances PD1 blockade in a dosedependent manner in HCC models[13]. A multicenter study also found that regorafenib combined with sintilimab achieved better ORR (36.2% vs 16.4%), PFS (5.6 months vs 4.0 months) and OS (13.4 months vs 9.9 months) in advanced HCC compared with regorafenib monotherapy[14]. The combination therapy inhibits STAT3 activity and increases the expression of chemokine CXCL10, thereby enhancing the infiltration of tumor-penetrating and activated CD8+ T cells and improving their survival[13]. Collectively, these preclinical and clinical studies support the therapeutic rationale for regorafenib plus PD-1 inhibitors in advanced HCC. Our findings add to the growing clinical evidence supporting the use of the combination regimen as second-line therapy for advanced HCC.
Nevertheless, the ORR observed in the present study was lower than that reported in several prospective clinical trials. This discrepancy is more likely attributable to differences in response assessment and patient characteristics than to reduced therapeutic activity. Because retrospective imaging data were insufficient for standardized mRECIST- or iRECIST-based reassessment, tumor response was evaluated using RECIST version 1.1, which may underestimate treatment response by failing to capture treatment-related tumor necrosis and atypical immune-related response patterns[15]. In addition, all enrolled patients had experienced first-line treatment failure and generally had more advanced disease and greater clinical heterogeneity than those included in prospective trials. Therefore, direct comparisons of ORR across studies should be interpreted cautiously. Taken together, the observed improvement in DCR and PFS suggests that the clinical benefit of regorafenib plus PD-1 inhibitors may be primarily reflected by prolonged disease control rather than increased objective tumor response.
Moreover, the phase 3 non-inferiority trial by Kudo et al[16] reported that lenvatinib yielded non-inferior OS relative to sorafenib, which is consistent with our finding that survival outcomes were comparable between patients who had received lenvatinib or sorafenib as first-line treatment. Multivariable Cox analysis demonstrated that AFP ≥ 400 ng/mL and second-line PD-1 inhibitor therapy were independently associated with PFS. Elevated AFP has been widely reco
The survival benefit of continuing immunotherapy after progression on first-line ICI-based therapy remains controversial. Previous studies have suggested that for patients who experienced PD after first-line TKI plus PD-1 inhibitor combination therapy, continuous second-line immunotherapy does not confer additional survival benefits[18]. In contrast, Cheng et al[19] reported that the combination of an ICI with regorafenib was associated with improved OS in patients with advanced HCC following progression on first-line ICI-based therapy. In line with these latter findings, our subgroup analysis showed that patients with prior first-line ICI exposure achieved longer OS in the RP group than in the R group, whereas no significant difference was observed in those without prior ICI exposure. Notably, among patients treated with the two PD-1 inhibitors for which adequate data were available, camrelizumab and tislelizumab achieved comparable PFS and OS, consistent with the limited available evidence in advanced HCC[20]. These findings further suggest that the observed benefit may be attributable to continued PD-1 blockade rather than to a specific PD-1 inhibitor.
One plausible explanation for this finding is the immunologic priming effect induced by prior ICI therapy[21,22]. Initial exposure to ICIs may enhance tumor immunogenicity and sustain antitumor immune activation, thereby in
Although the proposed mechanisms are biologically plausible and supported by accumulating experimental evidence, they could not be directly evaluated in the present study because tumor microenvironment-related biomarkers, including baseline CD8+ T-cell density, PD-L1 expression, and immune-related gene signatures were unavailable. Consequently, whether immune remodeling contributed to the observed therapeutic effects could not be determined. Prospective biomarker-driven studies incorporating paired pre- and on-treatment tumor biopsies, serial peripheral blood sampling, and comprehensive immune profiling are warranted to validate these proposed mechanisms and identify predictive biomarkers to guide patient selection in combined therapy following prior ICI exposure.
This study has several limitations. As a retrospective observational study, selection bias and residual confounding cannot be completely excluded. Furthermore, the relatively small regorafenib monotherapy group may have reduced the precision of treatment effect estimates, particularly for OS. Incomplete information on portal vein tumor thrombus classification and regorafenib treatment exposure limited evaluation of their potential impact on clinical outcomes. Although comparable outcomes were found between camrelizumab and tislelizumab, information on specific PD-1 inhibitors and dosing schedules was incomplete for a proportion of patients. Therefore, treatment heterogeneity could not be fully evaluated, and these subgroup findings should be interpreted cautiously because of the limited subgroup sample size and incomplete treatment information. Tumor response was assessed using RECIST version 1.1 because retrospective imaging data were insufficient for standardized mRECIST- or iRECIST-based reassessment. Consequently, the antitumor activity and the combination regimen may have been underestimated. Finally, treatment-related adverse events were not systematically documented or graded according to Common Terminology Criteria for Adverse Events, limiting a comprehensive comparison of safety between the two groups. Accordingly, adequately powered prospective studies incor
This multicenter real-world study confirms that regorafenib combined with PD-1 inhibitors is more effective than regorafenib monotherapy in the treatment of advanced unresectable HCC.
The authors thank all participating hospital staff for their efforts and all patients for their participation.
| 1. | Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74:229-263. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 16785] [Cited by in RCA: 18093] [Article Influence: 9046.5] [Reference Citation Analysis (55)] |
| 2. | Zhou M, Wang H, Zeng X, Yin P, Zhu J, Chen W, Li X, Wang L, Wang L, Liu Y, Liu J, Zhang M, Qi J, Yu S, Afshin A, Gakidou E, Glenn S, Krish VS, Miller-Petrie MK, Mountjoy-Venning WC, Mullany EC, Redford SB, Liu H, Naghavi M, Hay SI, Wang L, Murray CJL, Liang X. Mortality, morbidity, and risk factors in China and its provinces, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2019;394:1145-1158. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3067] [Cited by in RCA: 2783] [Article Influence: 397.6] [Reference Citation Analysis (8)] |
| 3. | Villanueva A. Hepatocellular Carcinoma. N Engl J Med. 2019;380:1450-1462. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3748] [Cited by in RCA: 3547] [Article Influence: 506.7] [Reference Citation Analysis (24)] |
| 4. | Zhou J, Sun H, Wang Z, Cong W, Zeng M, Zhou W, Liu L, Wen T, Kuang M, Zhang B, Tao K, Han G, Yan Z, Wang M, Liu R, Guo J, Zeng Z, Liang P, Ren Z, Hou J, Zhang Y, Liu X, Pan H, Bi F, Liang C, Chen M, Yan F, Xu H, Xie X, Ju S, Ji Y, Yun J, Li Z, Bai X, Cai D, Chen W, Chen Y, Chen Y, Cheng W, Cheng S, Dai Z, Dai C, Gao Q, Guo R, Guo W, Guo Y, Hua B, Huang X, Jiang H, Jia W, Li Q, Li T, Li X, Li X, Li Y, Li Y, Liang J, Liang X, Ling C, Liu H, Liu T, Lu S, Lv G, Mao Y, Meng Z, Peng T, Ren W, Shi G, Shi H, Shi M, Song T, Tan G, Wang J, Wang K, Wang L, Wang W, Wang X, Wang Z, Xiang B, Xia J, Xing B, Xu J, Xu J, Yang J, Yang X, Yang Y, Yang Y, Yao X, Yin Z, Yuan Z, Zeng Y, Zeng Y, Zhang B, Zhang L, Zhang S, Zhang T, Zhang Z, Zhao M, Zhao Y, Zheng H, Zhou L, Zhu J, Zhu K, Shi Y, Liu R, Zhang L, Xiao Y, Yang C, Wu Z, Ding Z, Zhu X, Tang Z, Huang X, Han H, Wu H, Chen M, Wang W, Li Q, Cai J, Shen F, Cai X, Qin S, Teng G, Fan J. China Liver Cancer Guidelines for the Diagnosis and Treatment of Hepatocellular Carcinoma (2024 Edition). Liver Cancer. 2025;14:779-835. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 4] [Cited by in RCA: 51] [Article Influence: 51.0] [Reference Citation Analysis (5)] |
| 5. | Kuo HY, Khan KA, Kerbel RS. Antiangiogenic-immune-checkpoint inhibitor combinations: lessons from phase III clinical trials. Nat Rev Clin Oncol. 2024;21:468-482. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 96] [Reference Citation Analysis (0)] |
| 6. | Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim TY, Kudo M, Breder V, Merle P, Kaseb AO, Li D, Verret W, Xu DZ, Hernandez S, Liu J, Huang C, Mulla S, Wang Y, Lim HY, Zhu AX, Cheng AL; IMbrave150 Investigators. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. N Engl J Med. 2020;382:1894-1905. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 6170] [Cited by in RCA: 5990] [Article Influence: 998.3] [Reference Citation Analysis (15)] |
| 7. | Song YG, Yoo JJ, Kim SG, Kim YS. Complications of immunotherapy in advanced hepatocellular carcinoma. J Liver Cancer. 2024;24:9-16. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 24] [Reference Citation Analysis (0)] |
| 8. | Jindal A, Thadi A, Shailubhai K. Hepatocellular Carcinoma: Etiology and Current and Future Drugs. J Clin Exp Hepatol. 2019;9:221-232. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 115] [Cited by in RCA: 207] [Article Influence: 29.6] [Reference Citation Analysis (5)] |
| 9. | Bruix J, Qin S, Merle P, Granito A, Huang YH, Bodoky G, Pracht M, Yokosuka O, Rosmorduc O, Breder V, Gerolami R, Masi G, Ross PJ, Song T, Bronowicki JP, Ollivier-Hourmand I, Kudo M, Cheng AL, Llovet JM, Finn RS, LeBerre MA, Baumhauer A, Meinhardt G, Han G; RESORCE Investigators. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;389:56-66. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3029] [Cited by in RCA: 2923] [Article Influence: 324.8] [Reference Citation Analysis (12)] |
| 10. | Liu J, Tao H, Yuan T, Li J, Li J, Liang H, Huang Z, Zhang E. Immunomodulatory effects of regorafenib: Enhancing the efficacy of anti-PD-1/PD-L1 therapy. Front Immunol. 2022;13:992611. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 17] [Cited by in RCA: 16] [Article Influence: 4.0] [Reference Citation Analysis (2)] |
| 11. | Zhao J, Guo Y, Feng T, Rong D, Kong X, Huang T, Lopez-Lopez V, Yarmohammadi H, Sakamoto Y, Zhu D, Yao A, Xia Y. Efficacy and safety of regorafenib in combination with immune checkpoint inhibitor therapy as second-line and third-line regimen for patients with advanced hepatocellular carcinoma: a retrospective study. J Gastrointest Oncol. 2023;14:2549-2558. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 9] [Reference Citation Analysis (0)] |
| 12. | Bi H, Pei J, Ma K, Zhang Y, Aprile G, Moris D, Méndez-Sánchez N, Samant H, Sun X, Xia F. Regorafenib combined with a PD-1 inhibitor in the second-line setting for unresectable hepatocellular carcinoma in real-world practice. J Gastrointest Oncol. 2024;15:368-376. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1] [Cited by in RCA: 5] [Article Influence: 2.5] [Reference Citation Analysis (0)] |
| 13. | Shigeta K, Matsui A, Kikuchi H, Klein S, Mamessier E, Chen IX, Aoki S, Kitahara S, Inoue K, Shigeta A, Hato T, Ramjiawan RR, Staiculescu D, Zopf D, Fiebig L, Hobbs GS, Quaas A, Dima S, Popescu I, Huang P, Munn LL, Cobbold M, Goyal L, Zhu AX, Jain RK, Duda DG. Regorafenib combined with PD1 blockade increases CD8 T-cell infiltration by inducing CXCL10 expression in hepatocellular carcinoma. J Immunother Cancer. 2020;8:e001435. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 129] [Cited by in RCA: 128] [Article Influence: 21.3] [Reference Citation Analysis (1)] |
| 14. | Huang J, Guo Y, Huang W, Hong X, Quan Y, Lin L, Zhou J, Liang L, Zhang Y, Zhou J, Cai M, Zhu K. Regorafenib Combined with PD-1 Blockade Immunotherapy versus Regorafenib as Second-Line Treatment for Advanced Hepatocellular Carcinoma: A Multicenter Retrospective Study. J Hepatocell Carcinoma. 2022;9:157-170. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 4] [Cited by in RCA: 46] [Article Influence: 11.5] [Reference Citation Analysis (1)] |
| 15. | Zand KA, Kim TH, Mukherji R, Ismail M, El Homsi M, Imanzadeh A, Kamath A, Singh C, Yacoub JH. Review of systemic therapy in HCC with an approach to response assessment in clinical practice. Abdom Radiol (NY). 2026. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 1] [Reference Citation Analysis (4)] |
| 16. | Kudo M, Finn RS, Qin S, Han KH, Ikeda K, Piscaglia F, Baron A, Park JW, Han G, 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;391:1163-1173. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 4538] [Cited by in RCA: 4392] [Article Influence: 549.0] [Reference Citation Analysis (19)] |
| 17. | Jiang S, Wang Y, Fu X, Lou J, Gong B, Liu W, Li Y, Sun Y, Liu C, Song Z, Zhou G. Clinical significance of AFP and PIVKA-II for predicting prognosis in hepatocellular carcinoma patients treated with immune checkpoint inhibitors. Ther Adv Med Oncol. 2025;17:17588359251386801. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 18. | Ma W, Li Y, Lu Y, Liang Z, Yu H, Han J, Liu J, Wang W, Peng C, Cheng J. A Multicenter Retrospective Study: Impact of First-Line Treatment Strategies on second-Line Efficacy and Safety of Regorafenib with or Without PD-1 Inhibitors in Unresectable Hepatocellular Carcinoma. J Hepatocell Carcinoma. 2025;12:2123-2137. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 10] [Reference Citation Analysis (0)] |
| 19. | Cheng Y, Wang JY, Lu Y, Xia YX, Zhao H, Wang Q, Zhu XL, Zu QQ, Li HK, Chen Z, Li XC; China Liver Cancer Study Group Young Investigators (CLEAP). Efficacy of regorafenib following first-line immune checkpoint inhibitor failure in patients with advanced hepatocellular carcinoma: A multicenter study. Hepatobiliary Pancreat Dis Int. 2025;24:656-665. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 20. | Zou H, Ge Y, Chen W, Yao D, Oi Lam Ung C, Lai Y, Hu H. Real-world treatment patterns and outcomes for patients with advanced hepatocellular carcinoma initially treated with PD-1 inhibitors. Int Immunopharmacol. 2024;132:111947. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 21. | Kikuchi H, Matsui A, Morita S, Amoozgar Z, Inoue K, Ruan Z, Staiculescu D, Wong JS, Huang P, Yau T, Jain RK, Duda DG. Increased CD8+ T-cell Infiltration and Efficacy for Multikinase Inhibitors After PD-1 Blockade in Hepatocellular Carcinoma. J Natl Cancer Inst. 2022;114:1301-1305. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 54] [Cited by in RCA: 50] [Article Influence: 12.5] [Reference Citation Analysis (1)] |
| 22. | Chuang CH, Chen CT, Hsu CH, Shao YY. Carry-over effect of immunotherapy in patients with advanced hepatocellular carcinoma. Cancer Immunol Immunother. 2025;74:208. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 23. | Xie L, Liu M, Cai M, Huang W, Guo Y, Liang L, Cai W, Liu J, Liang W, Tan Y, Lai M, Lin L, Zhu K. Regorafenib enhances anti-tumor efficacy of immune checkpoint inhibitor by regulating IFN-γ/NSDHL/SREBP1/TGF-β1 axis in hepatocellular carcinoma. Biomed Pharmacother. 2023;159:114254. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 20] [Reference Citation Analysis (0)] |
| 24. | Arai H, Battaglin F, Wang J, Lo JH, Soni S, Zhang W, Lenz HJ. Molecular insight of regorafenib treatment for colorectal cancer. Cancer Treat Rev. 2019;81:101912. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 151] [Cited by in RCA: 143] [Article Influence: 20.4] [Reference Citation Analysis (3)] |
| 25. | Moseman JE, Rastogi I, Jeon D, McNeel DG. PD-1 blockade employed at the time CD8+ T cells are activated enhances their antitumor efficacy. J Immunother Cancer. 2025;13:e011145. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 14] [Cited by in RCA: 16] [Article Influence: 16.0] [Reference Citation Analysis (0)] |
| 26. | Liu WR, Fisher DE. Epitope Spreading and the Efficacy of Immune Checkpoint Inhibition in Cancer. Int J Oncol Res. 2021;4:029. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 6] [Reference Citation Analysis (0)] |