The human HCC cell lines, SMMC-7721 and MHCC97H, were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 100 units/mL penicillin, and 100 mg/mL streptomycin sulfate. They were maintained in a humidified incubator at 37°C with an atmosphere of 5% CO₂.

Acquired regorafenib-resistant HCC cells (97H-R and 7721-R) were established by treating MHCC97H and SMMC-7721 cells, respectively, with gradually increasing concentrations of regorafenib for over 6 months, as described in our previous studies (16).

Cells (10,000 cells/well) were seeded into 96-well plates and cultured for 24 h. CCK-8 solution (10 μL/well; Selleck, Wuhan, China) was added, and the plates were incubated for 1–2 h. Then, the absorbance at 450 nm was measured using a microplate reader (Bio-Rad, CA, United States), and cell viability was determined.

Cells (1 × 10⁶ cells/well) were harvested after various treatments and fixed in 70% ethanol at −20°C for 2 h. After two washes with PBS and centrifugation, the cells were stained with propidium iodide (PI; 10 μg/mL) and RNase A (100 μg/mL) at room temperature for 30 min, followed by their detection using a BD FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, United States). The distribution of cells in different phases of the cell cycle was analyzed using FlowJo software (Tree Star, San Carlos, CA, United States).

Cells (1 × 10⁶ cells/well) were collected and resuspended in 500 μL of binding buffer. Then, 5 μL of Annexin V-FITC staining solution and 5 μL of PI staining solution were added to the suspension, followed by incubation for 30 min at room temperature. Fluorescence intensity was measured using a BD FACSCalibur cytometer (Becton Dickinson), and the percentage of apoptotic cells was determined using FlowJo software (Tree Star).

Cells (1,000 cells/well) were plated in six-well plates and monitored for 7–14 days to assess colony growth. Afterward, the cells were fixed with 4% paraformaldehyde for 20 min, washed, and incubated with 0.4% crystal violet solution for 30 min. The cells were then washed with PBS, dried, and counted. Colonies containing more than 30 cells were included in the count.

Cells were lysed on ice for 5 min using 150 μL of lysis buffer (Beyotime, Shanghai, China) containing 1% protease inhibitors (Thermo Fisher Scientific, Waltham, MA, United States). The lysates were then harvested and centrifuged at 12,000 × g for 5 min at 4°C. Protein concentrations were determined using a BCA kit (Thermo Fisher Scientific). Equal amounts of protein (20 μg/lane) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred onto polyvinylidene difluoride (PVDF) membranes, and blocked with 5% nonfat dry milk for 1 h at room temperature. Immunoblotting was performed by incubating the membranes with primary antibodies overnight at 4°C, followed by incubation with secondary antibodies the next day. Protein signals were detected using an enhanced chemiluminescence reagent (EMD Millipore, Billerica, MA, United States).

Differentially expressed proteins (DEPs) were identified using the following reference thresholds: a fold change of 1.2 and a p-value of <0.05. Gene Ontology (GO) analysis was performed via the GO Database¹ using Blast2GO (version 3.3.5). The GO functions of DEPs, including biological process (BP), molecular function (MF), and cellular component (CC), were analyzed.

The study protocol was approved by the Institutional Ethics Committee of the Affiliated Drum Tower Hospital, Nanjing Medical University. We sourced six-week-old male BALB/c nude mice from the Model Animal Research Center at Nanjing University and maintained them in accordance with the institution’s guidelines. Two nude mice were subcutaneously injected with 1 × 10⁷ regorafenib-resistant MHCC-97H cells in 100 μL of PBS. Once the tumors reached a volume of 100 mm³, they were excised, dissected into 1 mm³ segments, and implanted subcutaneously on the dorsal side near the right armpit of the mice. The mice were then randomly assigned to four groups and received oral treatments of either vehicle control, regorafenib, gefitinib, or a combination of regorafenib and gefitinib. The dosages were 20 mg/kg for regorafenib and 50 mg/kg for gefitinib, administered orally once daily for an initial period of 21 days. Body weight and tumor volume were monitored every 2 days. After 4 weeks of treatment, the mice were euthanized, and the tumors were collected for further analysis.

Tissues fixed with 4% paraformaldehyde were embedded in paraffin and sectioned into 5 μm thick slices. The tissue sections were deparaffinized in xylene, rehydrated through a graded series of ethanol, and washed in PBS. They were then stained with hematoxylin, agitated for 30 s, and rinsed in water. Subsequently, they were stained with eosin, agitated for 10–30 s, and rinsed in water. After staining, the sections were dehydrated, mounted, and covered with a cover slip. Microscopic examination was performed using a microscope (BX50; Olympus, Tokyo, Japan).

PSS 19.0 statistical software (IBM Corp., Chicago, IL, United States) was used for data analysis. Data are presented as means ± SD from triplicate experiments. Comparisons between two groups were performed using the Student’s t-test. Significance among multiple groups was determined by one-way analysis of variance followed by the Student–Newman–Keuls post hoc test. A p-value of <0.05 was considered to indicate a statistically significant difference.

Regorafenib-resistant HCC cell lines were established using MHCC97H and SMMC-7721 cell lines through a stepwise dose-escalation method and were designated as 97H-R and 7721-R, respectively. The successful establishment of regorafenib-resistant HCC cells was confirmed using the CCK-8 assay. The IC₅₀ values for regorafenib (Table 1) were markedly higher in the resistant cells (97H-R: 16.85 μM; 7721-R: 12.27 μM) compared with the parental cells (97H: 6.266 μM; 7721: 5.431 μM). Based on high-throughput proteomics analysis, we identified DEPs between MHCC97H and 97H-R cells. These DEPs were used to generate a heatmap (Figure 1A), which demonstrated a significant difference in protein expression between the resistant and parental cells. The detailed list of DEPs is provided in Supplementary Table S1. GO enrichment analysis showed that the DEPs were primarily enriched in cell division under the CC subcategory (Figure 1B). EGFR, a key member of the tyrosine kinase receptor family, was the most overexpressed protein associated with RTK pathways. The protein level of EGFR in 97H-R cells was 2.46-fold higher than that in MHCC97H cells (Figure 1C). Western blotting analysis further confirmed that the protein expression of EGFR was significantly upregulated in regorafenib-resistant cells compared with parental cells (p < 0.001; Figure 1D). These results indicate that EGFR might mediate regorafenib resistance in HCC.

To assess the role of EGFR in regorafenib resistance in HCC, we used gefitinib, a selective inhibitor of EGFR. We first evaluated the effect of gefitinib on the viability of 97H-R and 7721-R cells using the CCK-8 assay. Treatment with gefitinib alone at concentrations up to 20 μM for 48 h had minimal inhibitory effect on the viability of 97H-R cells, while the inhibitory concentration for 7721-R cells was 10 μM (Figure 2A). However, when combined with regorafenib, the viability of 97H-R cells decreased after treatment with 20 μM gefitinib for 48 h (10 μM gefitinib for 7721-R cells; Figure 2B). The IC₅₀ values for regorafenib in 97H-R and 7721-R cells also decreased following gefitinib supplementation (Table 2). These results indicate that EGFR inhibition reverses regorafenib resistance in HCC cells.

We further evaluated the antitumor efficacy of the combination treatment with gefitinib and regorafenib in vitro. The results showed that both the number and cell state of 97H-R and 7721-R cells were decreased when exposed to the combination of regorafenib and gefitinib, as observed through direct microscopy. In contrast, treatment with either regorafenib or gefitinib alone had minimal impact on the cell number or cell state of these two cell lines (Figure 3A). The proportion of apoptotic cells was significantly higher in the combination treatment group compared with the other groups (97H-R: p < 0.001; 7721-R: p < 0.001; Figure 3B). Moreover, the combination treatment increased the number of cells arrested in the G1 phase (Figure 3C) and reduced colony formation (Figure 3D) by 97H-R and 7721-R cells. To further confirm the differential effects of gefitinib and regorafenib on cell apoptosis, we measured the protein levels of two apoptotic cascade-related proteins, B-cell leukemia/lymphoma 2 (Bcl2) and cleaved poly ADP-ribose polymerase (PARP). The effects on cell proliferation were verified by examining the expression levels of cyclin D1 and cyclin-dependent kinases 2 and 4 (CDK2 and CDK4). The alterations in the protein levels of Bcl2, cleaved PARP, cyclin D1, CDK2, and CDK4 further confirmed that the combination of gefitinib and regorafenib indeed promoted cell apoptosis and inhibited the proliferation of regorafenib-resistant HCC cells (Figure 3E). These results suggest that inhibition of EGFR may reduce regorafenib resistance in HCC cells.

This study further investigated whether inhibition of EGFR by gefitinib could reverse regorafenib resistance in a nude mouse xenograft model established with 97H-R cells. No significant difference in body weight was observed between the drug-treated mice and the vehicle-treated controls, and no mice died during the treatment, indicating minimal systemic toxicity from the drugs (Figure 4A). Furthermore, H&E staining of lung, kidney, liver and heart tissues indicated barely any difference between the groups (Supplementary Figure S1). As shown in Figures 4B,C, treatment with either regorafenib or gefitinib alone produced a mild tumor inhibitory effect compared with the vehicle control, while the combination of the two drugs significantly inhibited tumor growth. The volume and weight of the tumors at the end confirmed that supplementation with gefitinib sensitized regorafenib-resistant HCC tumors to regorafenib treatment (Figure 4D). These results indicate that combination therapy with gefitinib and regorafenib exerts a strong tumor-inhibitory effect in vivo.

Although we confirmed that EGFR overexpression promotes acquired regorafenib resistance in HCC, the underlying molecular mechanisms remained unclear. Previous studies have identified the RAS/RAF/ERK signaling as an important downstream pathway of EGFR (17, 18). Therefore, we first measured the protein levels of RAS, c-RAF, and ERK1/2, as well as the phosphorylation of ERK1/2, in regorafenib-resistant cells (97H-R and 7721-R) and their corresponding parental cells (MHCC97H and SMMC-7721). No significant differences in ERK1/2 protein levels were observed between MHCC97H and 97H-R cells; however, the protein levels of RAS, c-RAF, and P-ERK1/2 were markedly higher in 97H-R cells compared with those in MHCC97H cells (Figure 5A). Similar results were obtained for SMMC-7721 and 7721-R cells. Additionally, we assessed the protein levels of RAS, c-RAF, ERK1/2, and P-ERK1/2 following EGFR inhibition by gefitinib in 97H-R and 7721-R cells. EGFR inhibition significantly reduced the elevated expression of RAS, c-RAF, and P-ERK1/2. Furthermore, these protein levels were significantly decreased after combined treatment with regorafenib and gefitinib (Figure 5B). These findings indicate that EGFR overexpression promotes regorafenib resistance by activating the RAS/RAF/ERK signaling pathway.