Targeting AKR1C1 overcomes lenvatinib resistance in hepatocellular carcinoma through the STAT3-ABC transporters pathway

Introduction

Hepatocellular carcinoma (HCC) represents a major global health concern, as it is the third most prevalent cause of cancer-related deaths worldwide (1). Clinically, the majority of symptoms of early-stage HCC are not easily identifiable and HCC is often diagnosed at a late stage, by which time treatment options are extremely limited. Recent advancements, such as molecular targeted therapies (e.g., sorafenib, lenvatinib), anti-VEGF agents, immune checkpoint inhibitors (e.g., PD-1/PD-L1 inhibitors), and their various combinations, have renewed hope for patients with advanced HCC (2). Among these, lenvatinib, an oral multi-target inhibitor acting against fibroblast growth factor receptor (FGFR) 1–4, vascular endothelial growth factor receptor (VEGFR) 1–3, platelet-derived growth factor receptor (PDGFR) α/β, c-KIT, and RET, holds a significant position in the treatment of advanced HCC (3). However, despite the initial response to lenvatinib, most patients will eventually exhibit resistance to the treatment, which limits its efficacy and clinical application (4,5). Thus, exploring the mechanisms of lenvatinib resistance and identifying new therapeutic strategies to overcome it are therefore crucial for improving the prognosis of HCC patients.

The aldo-keto reductase 1 (AKR1) family of NADPH-dependent oxidoreductases has emerged as a pivotal regulator in tumor progression and therapy resistance across various cancers. A critical mechanism involves their interaction with oncogenic signaling pathways, particularly the STAT3 signaling axis. For instance, AKR1B1 promotes gastric cancer progression by interacting with STAT3 to activate SLC7A11, thereby inhibiting ferroptosis (6). Similarly, AKR1C1 confers resistance to oxaliplatin in colorectal cancer and to cisplatin in head and neck squamous cell carcinoma by activating STAT3 signaling to enhance glutathione synthesis (7,8). In lung cancer, AKR1C1 promotes metastasis via STAT3 activation, while its targeting synergizes with gefitinib to overcome EGFR-therapy resistance (9,10). AKR1B1 also contributes to EGFR-therapy and systemic therapy resistance in lung and HCCs by driving glutathione synthesis and metabolic reprogramming (11,12). Contrastingly, AKR1B10 suppresses colorectal cancer metastasis by destabilizing c-Myc (13), and AKR1D1 acts as a tumor suppressor in liver cancer by enhancing NK cell cytotoxicity (14). Nevertheless, the precise mechanisms through which AKR1 family members contribute to lenvatinib resistance in HCC require further clarification.

Cancer development and therapy resistance are heavily reliant on the ATP-binding cassette (ABC) superfamily transporters (ABCBs, ABCCs, and ABCGs) that promote drug efflux in multiple human cancers (15). Previous studies have shown that multidrug resistance protein 1 (MDR1, also known as ABCB1) and breast cancer resistance protein (BCRP, also known as ABCG2) transporters are significantly upregulated after lenvatinib resistance induction, and that inhibition of these transporters may enhance therapeutic efficacy in vitro and in vivo (16,17). Recent studies discovered that bypass activation of the HGF/c-Met axis or epidermal growth factor receptor (EGFR) axis promotes lenvatinib resistance (4,18). Similarly, HCC cells acquire resistance to lenvatinib by activating EGFR and stimulating the EGFR-STAT3-ABCB1 axis (17). Targeting EGFR and its downstream AKT signaling pathway, MDR1, and BCRP can effectively reverse this resistance (4,16,17). These results revealed that regulating EGFR and its downstream STAT3 and AKT signaling pathways, as well as ABC superfamily transporters, may be crucial for improving the clinical benefits of lenvatinib in HCC patients.

In this study, we established lenvatinib-resistant HCC cell lines through prolonged lenvatinib exposure. We identified AKR1C1 within the AKR1 family, which is highly expressed in lenvatinib-resistant HCC cells and patient plasma, as a potential driver of lenvatinib resistance. Mechanistic studies revealed that AKR1C1 may promote lenvatinib resistance in HCC by activating the STAT3-ABC transporters pathway. Finally, targeting AKR1C1 with flufenamic acid (FFA), a non-steroidal anti-inflammatory drug, effectively reversed lenvatinib resistance. This study indicates that AKR1C1 could be both a promising target for overcoming lenvatinib resistance and a potential biomarker for predicting such resistance in HCC. We present this article in accordance with the MDAR reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-892/rc).

Methods

Reagents

Antibodies against STAT3 (#4904), Phospho-STAT3(Tyr705) (#9145), Phospho-JNK (Thr183/Tyr185) (#4668), Phospho-p38 MAPK (Thr180/Tyr182) (#4511), Phospho-Akt (Ser473) (#4060), Phospho-ERK1/2 (Thr202/Tyr204) (#4370), E-cadherin (#3195), and Vimentin (#5741) were obtained from Cell Signaling Technology (Danvers, MA, USA). Antibodies against Bcl-2 (26593-1-AP), Bax (50599-2-Ig), Caspase-3 (19677-1-AP), BCRP (also known as ABCG2, 27286-1-AP), MDR1 (also known as ABCB1, 22336-1-AP), and GAPDH (60004-1-Ig) were purchased from Proteintech (Rosemont, IL, USA). MRP2 (also known as ABCC2, A5785) was purchased from Selleck (Houston, TX, USA). AKR1C1 (GTX105620) was purchased from GeneTex (Irvine, CA, USA). Secondary antibodies conjugated with HRP were obtained from Beyotime Biotechnology (Shanghai, China). Stattic (an STAT3 inhibitor, T6308), MK-571 sodium (an MRP2 inhibitor, T3148), and lenvatinib (T0520) were purchased from TargetMol (Wellesley Hills, MA, USA). FFA (S4268) was purchased from MedChemExpress (Monmouth Junction, NJ, USA). The following siRNAs were purchased from GenePharma (Shanghai, China): negative control, sense: 5'-UUCUCCGAACGUGUCACGUTT-3', antisense: 5'-ACGUGACACGUUCGGAGAATT-3'; siRNA-AKR1C1#1, sense: 5'-GGCCAGAAAGGAAAGACAATT-3', antisense: 5'-UUGUCUUUCCUUUCUGGCCAA-3'; siRNA-AKR1C1#2, sense: 5'-GUCUGCAACCAGGUGGAAUTT-3', antisense: 5'-AUUCCACCUGGUUGCAGACTT-3'. Lipofectamine 2000 was purchased from Thermo Fisher Scientific (Waltham, MA, USA).

Cell lines and cell culture

Human HCC cell lines HuH7 and Hep3B (Shanghai Institute of Cell Biology, Chinese Academy of Sciences) were cultivated in complete Dulbecco’s modified Eagle medium (DMEM) (VivaCell, Shanghai, China) with 10% fetal bovine serum (FBS) (Clark Bioscience, Webster, TX, USA), 100 U/mL penicillin and 0.1 mg/mL streptomycin (Beyotime Biotechnology) at 37 ℃ in a 5% CO₂ incubator. In order to establish lenvatinib resistance in HuH7/R and Hep3B/R cells, HuH7 and Hep3B cells were subjected to a long-term, stepwise induction protocol. The procedure was performed as follows: cells were initially cultured in medium containing 3 µM lenvatinib and routinely passaged upon reaching 70–90% confluency. After each passage, the concentration of lenvatinib was increased by 1 µM until the cells exhibited stable and rapid proliferation in medium containing 20 µM lenvatinib. The entire induction process was carried out over a period of 10 months.

Cell viability assay

For the cell viability assay, 5×10³ cells/100 µL/well were cultivated in 96-well plates and then incubated overnight in DMEM complete medium to allow adhesion. Subsequently, the cells were exposed to varying concentrations of lenvatinib, FFA, or siRNA-AKR1C1 in five wells at 37 ℃ for 48 h. The cell counting kit-8 (CCK-8) solution (Meilunbio, Dalian, China) was added to each well with a volume of 20 µL. The absorbances were measured at a wavelength of 450 nm using spectrophotometry. Cell viability was determined according to the following procedure: (experimental group − blank group)/(control group −blank group) × 100%. The combination index (CI) value was calculated based on the Chou-Talalay method. The CI value indicates the interaction between two drugs (CI <1, synergism; CI >1, antagonism).

Cell apoptosis assay

The apoptotic effect of lenvatinib, FFA, or siRNA-AKR1C1 on human HCC cell lines (HuH7, HuH7/R, Hep3B, or Hep3B/R) was assessed using a flow cytometry analysis with the FITC-Annexin V/PI apoptosis detection kit (Meilunbio). Briefly, cells were plated in a 6-well plate (5×10⁵ cells/well) and cultured overnight in complete DMEM medium to allow adhesion. Following treatment with lenvatinib, FFA, or siRNA-AKR1C1 for a further 48 hours, the cells were collected, washed using cold phosphate buffer saline (PBS), and subsequently resuspended in 1× binding buffer. Subsequently, cells were stained with 5 µL Annexin V-FITC and 10 µL PI for 15 min at room temperature away from light. Finally, binding buffer (400 µL) was added to each tube before analysis with DxFLEX flow cytometry (Beckman Coulter). The flow cytometry data were subsequently examined using FlowJo V10 software (FlowJo, Ashland, OR, USA).

Colony formation assay

In the colony formation study, HuH7/R and Hep3B/R cells were exposed to 100 µM FFA, or in combination with 5 µM lenvatinib for a period of 5 days. The cells were carefully washed with PBS twice, fixed with 4% paraformaldehyde (Servicebio, Wuhan, China) at room temperature for 10 min, and then stained with crystal violet (Beyotime Biotechnology). After rinsing the plates with distilled water and air-drying, colonies were counted microscopically.

Quantitative real-time polymerase chain reaction (PCR)

Total RNA was isolated with TRIzol reagent (Invitrogen) and reverse transcription of cDNA was performed using PrimeScript™ RT reagent Kit (Takara, Shiga, Japan), following the manufacturer’s protocols. The PCR was conducted with TB Green^® Premix Ex Taq™ (Tli RNaseH Plus) (Takara). The specific primers used in this study are provided in Table S1. GAPDH was used as an internal standard.

Western blotting analysis

The HCC cells were treated under various conditions and then lysed in radio-immunoprecipitation assay (RIPA) lysis buffer. Protein quantitation was performed by bicinchoninic acid (BCA) protein assay kit (NCM Biotech, Suzhou, China). The proteins were then separated by 10% SDS-PAGE gels (Servicebio) and subsequently transferred to a polyvinylidene fluoride (PVDF) membrane. Following the blocking of 5% milk in TBST, the membranes were then exposed to the indicated primary antibodies overnight at 4 ℃. The membranes were then washed in tris buffered saline with Tween 20 (TBST) and incubated with labelled secondary antibodies. To detect protein bands, we used an NcmECL Ultra Kit (NCM Biotech) via Chemiluminescence Instrument (ZHIHENG Intelligent Technology, Nanjing, China). All the antibodies utilized are listed in Table S2.

Co-immunoprecipitation (Co-IP)

The cells were washed with ice-cold PBS and lysed in Western/IP cell lysis buffer, supplemented with Protease/Phosphatase Inhibitor Cocktail (Beyotime Biotechnology). The lysate was then incubated for 20 minutes at 4 ℃, followed by centrifugation at 12,000 rpm also for 20 minutes at 4 ℃. Cell lysates were first exposed to the STAT3 or AKR1C1 antibodies at 4 ℃ overnight, and then to Protein A/G Immunomagnetic Beads (TargetMol) for a further 2 h at 4 ℃. The beads were thoroughly washed with the immunoprecipitation assay buffer, then lysed in RIPA buffer. Finally, the samples were boiled for five minutes and Western blotted with the specified antibodies.

Plasma AKR1C1-enzyme-linked immunosorbent assay (ELISA) detection

A total of 20 patients were enrolled in the study to evaluate AKR1C1 expression levels in liver cancer patients treated with lenvatinib. The response of the tumor was assessed using abdominal contrast-enhanced MRI or CT, in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST). Patients who exhibited a progressive disease (PD) were categorized as exhibiting clinical resistance to lenvatinib. All participants’ peripheral blood samples (EDTA-K2 anticoagulant) were centrifuged at 1,500 g for 10 min at 4 ℃ to obtain plasma. The plasma was then immediately transferred into clean polypropylene tubes, labelled and stored at −80 ℃. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee for Human Studies of Xuzhou Central Hospital (protocol No. XZXY-LK-20240108-0007), and informed consent was obtained from all patients.

Plasma AKR1C1 of liver cancer patients was quantified using ELISA kits following the manufacturer’s protocol (Byabscience, Nanjing, China). Briefly, samples (50 µL/well) and HRP-conjugated AKR1C1 antibody (100 µL/well) were added to the prepared plates. Thereafter, the plates were cultivated for 1 h at 37 ℃, thoroughly washed 5 times with PBST, and then exposed to a mixture of substrates A and B (100 µL/well) at 37 ℃ for 15 min. After adding 50 µL of stop solution, the optical density (OD) value was measured at 450 nm using an ELISA reader.

Statistical analysis

All of the experiments were carried out at least three times. The statistical analysis was conducted in GraphPad Prism 8.0 software. The data between the groups were analysed using Student’s t-test and one- or two-way analysis of variance (ANOVA). Data are presented as mean ± standard error of the mean (SEM). Statistical significance was set at P<0.05.

Results

Establishment of lenvatinib-resistant HCC cell lines

To acquire lenvatinib-resistant cell lines, we gradually increased the dose of lenvatinib (starting from 3 µM) over a 10-month exposure to HuH7 and Hep3B cells. The resulting cells were termed HuH7/R and Hep3B/R. The CCK-8 results showed that HuH7/R and Hep3B/R cells were less susceptible to lenvatinib treatment than their parental cells, with the median inhibition concentration (IC₅₀) values of lenvatinib in HuH7/R and Hep3B/R cells (22.05 and 34.49 µM, respectively) being significantly higher than in their parent cells (0.60 and 4.68 µM, respectively) (Figure 1A). Two weeks after lenvatinib withdrawal, there was no significant difference in the changes of IC₅₀ values between HuH7/R and Hep3B/R cells (Figure S1). Meanwhile, Western blotting analysis detected significantly higher expression of ABC transporters (BCRP, MRP2, and MDR1) in HuH7/R and Hep3B/R cells compared to their parental cells (Figure 1B). Consistently, quantitative reverse transcription polymerase chain reaction (qRT-PCR) results showed that the mRNA levels of BCRP, MRP2, and MDR1 were also markedly elevated in the resistant cells (Figure 1C). Additionally, in cultivation medium containing lenvatinib, HuH7/R and Hep3B/R cells showed increased anti-apoptotic activity in comparison to HuH7 and Hep3B cells (Figure 1D). The results indicated that the cells acquired drug-resistant characteristics and were successfully constructed in our study.

Figure 1 Establishment of lenvatinib-resistant HCC cell lines. (A) The IC₅₀ values for both parental (HuH7, Hep3B) and lenvatinib-resistant cells (HuH7/R, Hep3B/R) were determined by CCK-8 assay after 48 h of culturing with varying doses of lenvatinib. (B,C) Western blot and qRT-PCR analysis of the expression levels of BCRP, MRP2, and MDR1 in parental (HuH7, Hep3B) and lenvatinib-resistant cells (HuH7/R, Hep3B/R). (D) The cells were treated with indicated concentrations of lenvatinib for 48 h. The apoptosis resistance capability of HuH7/R and Hep3B/R cells was assessed through flow cytometry analysis. *, P<0.05; **, P<0.01. CCK-8, cell counting kit-8; HCC, hepatocellular carcinoma; IC₅₀, median inhibition concentration; qRT-PCR, quantitative reverse transcription polymerase chain reaction.

High expression of AKR1B10, AKR1C1, and AKR1C3 in HCC tissues is associated with poor prognosis

Utilizing the GEPIA2 (http://gepia2.cancer-pku.cn/) and Kaplan-Meier Plotter (https://kmplot.com/analysis/) databases, we evaluated the expression of AKR1 family members in HCC and their association with patient overall survival. Analysis revealed that AKR1B10, AKR1C1, AKR1C2, and AKR1C3 were significantly overexpressed in HCC tissues (Figure 2A). In contrast, AKR1D1 expression was significantly reduced (Figure 2A). Conversely, high expression levels of AKR1B10, AKR1C1, and AKR1C3 were associated with poorer overall survival, while elevated AKR1D1 expression correlated with a more favorable prognosis (Figure 2B). These results show that the expression levels of AKR1B10, AKR1C1, AKR1C3, and AKR1D1 correlate with the prognosis of patients with HCC.

Figure 2 AKR1B10, AKR1C1, AKR1C3, and AKR1D1 are implicated in HCC patients’ prognosis. (A) The expression level of AKR1 family members between HCC (n=369) and normal tissues (n=160) was evaluated using the GEPIA2 database. (B) Kaplan-Meier survival analysis for OS probability of HCC patients infected with or without hepatitis viruses, according to the level of AKR1 family members mRNA expression. *, P<0.05. CI, confidence interval; HCC, hepatocellular carcinoma; HR, hazard ratio; N, normal tissue; OS, overall survival; T, tumor tissue; TPM, transcripts per million.

AKR1C1 is upregulated in lenvatinib-resistant HCC cells

Recent studies have confirmed that AKR1B1 plays a critical role in treatment resistance in gastric, lung, and liver cancers (6,11,12). To identify potential resistance-driving genes within the AKR1 family, we used qRT-PCR to analyze the expression of AKR1B1, AKR1B10, AKR1C1, AKR1C3, and AKR1D1 in lenvatinib-resistant cells and their parental HuH7 cells. Analysis revealed differential expression of these genes between HuH7 cells and HuH7/R cells, with AKR1C1 displaying the most pronounced upregulation (Figure 3A). Consistently, elevated AKR1C1 expression was also detected in the plasma of lenvatinib-resistant liver cancer patients (Figure 3B). Furthermore, TIMER 2.0 database analysis revealed that AKR1C1 expression has been observed to be associated with multidrug resistance genes (ABCG2, ABCC2, ABCB1) (Figure 3C). These findings suggest that AKR1C1 may serve as a potential biomarker for predicting lenvatinib resistance in HCC. Moreover, analysis using UALCAN and the Kaplan-Meier Plotter revealed that elevated AKR1C1 expression was positively associated with tumor grade and inversely correlated with patient outcomes, including reduced recurrence-free survival (RFS), progression-free survival (PFS), and disease-specific survival (DSS) (Figure 3D).

Figure 3 AKR1C1 is overexpressed in lenvatinib-resistant HCC cells and predicts poor prognosis in HCC patients. (A) The heat maps through qRT-PCR analysis showed the RNA expression levels of AKR1 family genes between HuH7 and HuH7/R cells. (B) The levels of AKR1C1 expression in the plasma of lenvatinib-sensitive (n=10) and lenvatinib-resistant (n=10) HCC patients were detected by ELISA. (C) The correlation between AKR1C1 mRNA expression levels and ABCG2, ABCC2, or ABCB1 was assessed via the TIMER 2.0 database. (D) The expression level of AKR1C1 between HCC with different tumor grade was evaluated via the UALCAN website. The Kaplan-Meier survival analysis for OS, RFS, PFS, and DSS of HCC patients infected with or without hepatitis viruses, according to the level of AKR1C1 mRNA expression. **, P<0.01. CI, confidence interval; DSS, disease-specific survival; ELISA, enzyme-linked immunosorbent assay; HCC, hepatocellular carcinoma; HR, hazard ratio; LR, lenvatinib resistant; LS, lenvatinib sensitive; OS, overall survival; PFS, progression-free survival; qRT-PCR, quantitative reverse transcription polymerase chain reaction; RFS, recurrence-free survival; TPM, transcripts per million.

AKR1C1 boosts the lenvatinib resistance of HCC cells

Next, we investigated the role of AKR1C1 in driving lenvatinib resistance in HCC cells. Cell viability assays showed that knockdown of AKR1C1 in HuH7/R and Hep3B/R cells markedly enhanced their sensitivity to lenvatinib (Figure 4A). Correspondingly, the combination of lenvatinib (5 µM) and siRNA-mediated AKR1C1 knockdown significantly induced apoptosis in both HuH7/R and Hep3B/R cells, compared to treatment with either lenvatinib or siRNA-AKR1C1 alone, or the control group (Figure 4B). Intriguingly, silencing AKR1C1 alone was sufficient to promote apoptosis in lenvatinib-resistant cells (Figure 4B). Furthermore, inhibition of AKR1C1 expression also induced apoptosis in drug-sensitive HCC cells (Figure 4C). Collectively, these findings indicate that AKR1C1 not only drives lenvatinib resistance in HCC cells but also directly regulates apoptotic pathways in both wild-type and lenvatinib-resistant HCC cells, highlighting its potential as a therapeutic target for HCC.

Figure 4 AKR1C1 contributes to lenvatinib resistance in HCC cells. (A,B) The knockdown of AKR1C1 resensitized HuH7/R and Hep3B/R cells to lenvatinib. HuH7/R and Hep3B/R cells were treated with siRNA-AKR1C1 or combined with 5 µM lenvatinib for 48 h. (A) The cell viability was assessed using the CCK-8 assay. (B) The proportion of apoptotic cells was assessed with DxFLEX flow cytometry. (C) HuH7 and Hep3B cells were treated with siRNA-NC, siRNA-AKR1C1-#1, or siRNA-AKR1C1-#2 for 48 h. The proportion of apoptotic cells was assessed with DxFLEX flow cytometry using the FITC-Annexin V/PI apoptosis detection kit. *, P<0.05; **, P<0.01. CCK-8, cell counting kit-8; HCC, hepatocellular carcinoma; NC, negative control; siRNA, small interfering RNA.

AKR1C1 induces lenvatinib resistance by activating STAT3-ABC transporters pathway

We then explored the mechanism underlying AKR1C1-induced lenvatinib resistance in HCC. Western blot analysis revealed that levels of AKR1C1 and phosphorylated STAT3 were significantly elevated in HuH7/R and Hep3B/R cells compared with their parental HuH7 and Hep3B cells (Figure 5A). Furthermore, siRNA-mediated knockdown of AKR1C1 in HuH7/R and Hep3B/R cells markedly suppressed STAT3 phosphorylation and reduced the expression of multidrug-resistant proteins (MDR1, MRP2, and BCRP) (Figure 5B). In addition, treatment with the STAT3 inhibitor Stattic similarly inhibited MDR1, MRP2 and BCRP expressions, thereby restored lenvatinib sensitivity in HuH7/R and Hep3B/R cells (Figure 5C,5D). Subsequently, we hypothesized that the induction of STAT3 and the expression of drug-resistant proteins were caused by the binding of AKR1C1 and STAT3. Immunoprecipitation assays analysis confirmed that AKR1C1 and STAT3 co-precipitate in HuH7/R and Hep3B/R cells, indicating a direct protein–protein interaction (Figure 5E). A previous study has shown that inhibiting MDR1 and BCRP with elacridar can overcome lenvatinib resistance and improve therapeutic efficacy (16). Consistently, pharmacological inhibition of MRP2 using MK-571 also reversed lenvatinib resistance, suggesting that lenvatinib may be a substrate of MRP2 (Figure 5F, Figure S2). Collectively, these results suggest that activation of the AKR1C1-STAT3 pathway promotes lenvatinib resistance in HCC via upregulation of ABC transporters.

Figure 5 AKR1C1 drives lenvatinib resistance in HCC cells through STAT3-mediated ABC transporters pathway. (A) The phosphorylation levels of STAT3, Akt, JNK, p38, and ERK in parental (HuH7, Hep3B) and lenvatinib-resistant cells (HuH7/R, Hep3B/R) were assessed using Western blot analysis. (B) The impact of AKR1C1 knockdown on the expression of p-STAT3, MDR1, MRP2 and BCRP was assessed using Western blot analysis. (C) The influence of the STAT3 inhibitor Stattic on the expression of p-STAT3, MDR1, MRP2, BCRP in HuH7/R and Hep3B/R cells. (D) The HuH7/R, Hep3B/R cells were treated with 1 µM Stattic combined with 5 µM lenvatinib for 48 h. The proportion of apoptotic cells was assessed by DxFLEX flow cytometry. The left panel provides a selection of representative images illustrating the apoptosis assay. The number of apoptotic cells is calculated from the right-hand panel and is shown as the mean ± SEM. (E) The physical interaction between AKR1C1 and STAT3 was analysed using co-immunoprecipitation. (F) The HuH7/R, Hep3B/R cells were treated with 30 µM MRP2 inhibitor MK-571 combined with 10 µM lenvatinib for 48 h. The proportion of apoptotic cells was assessed by DxFLEX flow cytometry. The left panel provides a selection of representative images illustrating the apoptosis assay. The number of apoptotic cells is calculated from the right-hand panel and is shown as the mean ± SEM. **, P<0.01. ABC, ATP-binding cassette; CM, complete medium; HCC, hepatocellular carcinoma; IB, immunoblotting; IP, immunoprecipitation; NC, negative control; SEM, standard error of the mean; siRNA, small interfering RNA.

FFA induces apoptosis and inhibits AKR1C1-STAT3 signaling pathway in lenvatinib-resistant HCC cells

FFA, an AKR1C1 inhibitor, acts as a non-steroidal anti-inflammatory drug to treat rheumatic diseases (19). To ascertain whether FFA affects the viability and apoptosis of lenvatinib-resistant HCC cells, HuH7/R and Hep3B/R cells were incubated with 30, 100 and 300 µM FFA. Cell viability assays revealed that FFA significantly suppressed the activity of HuH7/R and Hep3B/R cells (Figure 6A,6B). Furthermore, flow cytometry analysis demonstrated that FFA could induce the apoptosis of HuH7/R and Hep3B/R cells at 300 µM (Figure 6C). Consistent with these findings, FFA treatment notably increased the expression of the pro-apoptotic protein Bax and promoted the cleavage of procaspase-3, while it downregulated the expression of Bcl-2 (Figure 6D). In addition, FFA was also found to inhibit the epithelial-mesenchymal transition (EMT) process in HuH7/R and Hep3B/R cells (Figure 6D). As expected, FFA markedly reduced the expression levels of AKR1C1, p-STAT3, and multidrug-resistance-related proteins (MDR1, MRP2, BCRP) (Figure 6E). Taken together, these results indicate that FFA can promote apoptosis and suppress the AKR1C1-STAT3 signaling pathway in lenvatinib-resistant HCC cells.

Figure 6 FFA can induce apoptosis and inhibit the AKR1C1-STAT3 signaling pathway in lenvatinib-resistant HCC cells. The HuH7/R and Hep3B/R were treated with control, 30 µM, 100 µM, and 300 µM FFA for 48 h. (A) Cell viability was assessed with CCK-8. (B) The morphologies were observed using an inverted microscope (magnification 40×), and images were acquired under bright-field illumination. (C) The proportion of apoptotic cells was assessed by DxFLEX flow cytometry. The left panel provides a selection of representative images illustrating the apoptosis assay. The number of apoptotic cells is calculated from the right-hand panel and is shown as the mean ± SEM. (D,E) Total cell lysates were prepared, and the levels of expression of Bax, Bcl-2, Procaspase-3, E-Cadherin, Vimentin, AKR1C1, STAT3, p-STAT3, BCRP, MRP2, MDR1, and GAPDH were analyzed by Western blot. One representative result is shown. *, P<0.05; **, P<0.01. CCK-8, cell counting kit-8; FFA, flufenamic acid; HCC, hepatocellular carcinoma; SEM, standard error of the mean.

FFA enhances the sensitivity of lenvatinib-resistant HCC cells to lenvatinib

As FFA had a notable impact on the AKR1C1-STAT3 signaling axis in lenvatinib-resistant HCC cells, we hypothesized that FFA could resensitize these cells to lenvatinib. To assess the potential efficacy of the lenvatinib-FFA combination, we treated HuH7/R and Hep3B/R cells with either agent alone or in combination for 48 h and measured cell viability by CCK-8 assay. The results demonstrated that the combination of lenvatinib and FFA exerted a stronger inhibitory effect on resistant cells than either treatment alone (Figure 7A). Using the Chou-Talalay method, we calculated the CI and found that CI <1 across tested concentrations of lenvatinib (3, 10, 30 µM) and FFA (30, 100, 300 µM), indicating significant synergy between the two agents (Figure 7B). Furthermore, flow cytometry analysis revealed that the combination of lenvatinib (5 µM) and FFA (100 µM) induced significantly more apoptosis in HuH7/R and Hep3B/R cells compared with the control group or either single-agent group (Figure 7C). Similarly, colony formation assays showed that the lenvatinib-FFA combination suppressed resistant cell growth more effectively than either drug alone (Figure 7D). Together, these findings suggest that FFA is a promising candidate for overcoming lenvatinib resistance in HCC.

Figure 7 FFA combined with lenvatinib can synergistically induce apoptosis in lenvatinib-resistant HCC cells. (A,B) HuH7/R and Hep3B/R cells were treated with the indicated concentrations of lenvatinib or FFA for 48 h. Cell viability was assessed using the CCK-8 assay, and the CI was calculated by the Chou-Talalay method, where CI <1 indicates synergism and CI >1 indicates antagonism. (C) The HuH7/R and Hep3B/R cells were exposed to 100 µM FFA in combination with 5 µM lenvatinib for 48 h. The proportion of apoptotic cells was assessed by DxFLEX flow cytometry. The left panel provides a selection of representative images illustrating the apoptosis assay. The number of apoptotic cells is calculated from the right-hand panel and is shown as the mean ± SEM. (D) The HuH7/R and Hep3B/R cells were treated with 100 µM FFA combined with 5 µM lenvatinib for 5 days. The colonies were then fixed with 4% paraformaldehyde and stained with 0.1% crystal violet to assess the treatment effect. The left panel provides a selection of representative images illustrating the colony assay. The number of cell colonies is calculated from the right-hand panel and is displayed as the mean ± SEM. **, P<0.01. CCK-8, cell counting kit-8; CI, combination index; FFA, flufenamic acid; HCC, hepatocellular carcinoma; SEM, standard error of the mean.

Discussion

The emergence of drug resistance has become a significant challenge for cancer therapy (18). Currently, sorafenib, lenvatinib, and therapies combining immune checkpoint inhibitors with anti-angiogenic agents have been widely used in the treatment of advanced HCC (2). Despite advances in alleviating advanced HCC, there remains a need to develop innovative combination therapies to overcome drug resistance and explore novel therapeutic strategies to improve outcomes for HCC patients. For example, the combination therapy of anti-PD-1 with lenvatinib can significantly enhance the response rate in HCC patients (20,21). This enhanced efficacy may be attributed to the ability of lenvatinib to block the VEGF signaling pathway, which promotes CD8⁺ T-cell infiltration and thereby augments the activity of immune checkpoint inhibitors (22). Supporting this mechanism, recent single-cell multi-omics studies in HCC patients receiving anti-PD-1 plus lenvatinib revealed that circulating effector memory T cells enriched with HBV-specific T cells are specifically linked to treatment response, whereas regulatory T cells are associated with resistance to the combination regimen (23).

Lenvatinib, as another targeted drug following sorafenib, is also susceptible to drug resistance. It has been reported that most patients with advanced HCC exhibit resistance to lenvatinib treatment after one year (5,17). To clarify the underlying molecular mechanisms of acquired lenvatinib resistance, we have established two human HCC cell lines that developed resistance through continuous exposure to lenvatinib. The study revealed that AKR1C1 within the AKR1 family was significantly overexpressed in lenvatinib-resistant HCC cells and in the plasma of lenvatinib-resistant HCC patients. Analysis based on the TIMER 2.0 database and Kaplan-Meier Plotter showed that high AKR1C1 mRNA expression was associated with multidrug resistance genes and poor outcomes in HCC patients. The mechanism of AKR1C1 involves interaction with and activation of STAT3, leading to the upregulation of multidrug-resistant proteins (MDR1, MRP2, and BCRP). Furthermore, our research indicated that FFA, a non-steroidal anti-inflammatory drug, could effectively restore the sensitivity of resistant cell lines to lenvatinib by inhibiting the AKR1C1-STAT3 axis. Collectively, our findings demonstrate that AKR1C1 plays a critical role in driving lenvatinib resistance in HCC and represents a promising therapeutic target for restoring drug sensitivity.

According to treatment efficacy, drug resistance can be categorized as primary or acquired resistance. A recent study has indicated that the co-administration of lenvatinib and an EGFR inhibitor gefitinib may offer notable benefits for patients exhibiting high basal EGFR expression. This observation suggests that initial resistance to lenvatinib in HCC patients may be a consequence of FGFR inhibition, resulting in feedback activation of the EGFR-PAK2-ERK5 signaling axis (5). In acquired resistance in HCC, a major mechanism contributing to drug resistance is limited drug absorption and increased drug secretion, often facilitated by ABC superfamily transporters that promote drug efflux. To date, 49 subtypes of the ABC transporters are known to exist in humans, which are further subdivided into 7 subfamilies (ABCA to ABCG) according to various characteristics (24). It is well established that at least 11 ABC transporters, including MDR1/ABCB1, MRP2/ABCC2, and BCRP/ABCG2, are engaged in mediating multidrug resistance (24). Previous studies have shown that both MDR1 (ABCB1) and BCRP are upregulated in lenvatinib-resistant cells, and that their inhibition by elacridar can counteract this resistance and improve treatment outcomes (16,17). This is further supported by the work of Hu et al., who used liquid chromatography-mass spectrometry (LC-MS) to demonstrate that ABCB1 directly promotes lenvatinib resistance by enhancing its efflux from HCC cells (17), confirming that lenvatinib is a substrate for both MDR1 and BCRP. Beyond these transporters, evidence also points to a role for MRP2 (encoded by ABCC2) in lenvatinib handling. Previous studies indicated that the ABCC2-24T variant, which is associated with reduced MRP2 transport activity, leads to a significantly higher dose-adjusted trough concentration (C0) of lenvatinib in carriers compared to those with the -24C/C genotype, suggesting that ABCC2 polymorphism influences lenvatinib pharmacokinetics (25,26). Corroborating this clinical observation, we found markedly elevated expression of MRP2 in lenvatinib-resistant cell lines (HuH7/R and Hep3B/R). Importantly, pharmacological inhibition of MRP2 with MK-571 synergistically reversed lenvatinib resistance, implying that lenvatinib may also be a substrate of MRP2. However, whether MRP2 directly enhances lenvatinib efflux from HCC cells awaits definitive validation by LC-MS.

AKR1C1 is an enzyme involved in steroid, prostaglandin, and xenobiotic metabolism, and its overexpression has been identified as a key player in the proliferation and metastasis of multiple cancers (12-14). In non-small cell lung cancer, AKR1C1 promotes metastasis by activating STAT3 (9); whereas in colorectal cancer, their interaction increases glutathione to mediate oxaliplatin resistance (7). Similarly, crosstalk between AKR1C1 and the STAT1/3 signaling pathways underlies cisplatin resistance in head and neck squamous cell carcinoma (8). Beyond the agents mentioned above, it is also involved in mediating resistance to pirarubicin in bladder cancer (27), platinum-based drugs in ovarian cancer (28), progestin in endometrial cancer (29), gefitinib in lung adenocarcinoma (30), and cisplatin in gastric carcinoma (31). These results demonstrate that AKR1C1 can serve as an important therapeutic target for tumor treatment. Consistent with the initial finding by Gao et al. that AKR1C1 is highly expressed in lenvatinib-resistant HCC cells (32), our results confirmed that AKR1C1 can interact with STAT3 and activate it in HCC cells, which then upregulates the multidrug resistance-associated proteins (MDR1, MRP2, and BCRP), thus conferring resistance to lenvatinib.

FFA has been identified as an anti-inflammatory agent since the 1960s due to its ability to reduce prostaglandin synthesis and has subsequently been found to function as an ion channel modulator (19). A recent study has demonstrated the potential of FFA as a sensitizing drug for colistin-resistant gram-negative bacteria (33). It also reduces cisplatin resistance and invasive potential in metastatic bladder cancer sublines by antagonizing AKR1C1, and enhances anti-tumor activity in metastatic breast tumors and HCC (34,35). The above studies suggest that FFA may have therapeutic benefits in overcoming drug resistance. As far as we are aware, this is the first study to demonstrate that FFA inhibits the MDR1, MRP2, and BCRP transporter expression via the AKR1C1-STAT3 signaling pathway, thereby inducing apoptosis and restoring lenvatinib sensitivity to lenvatinib-resistant HCC cells.

This study provides preliminary evidence for AKR1C1 as a promising therapeutic target and a potential predictive biomarker for lenvatinib resistance in HCC. However, there are certain limitations in this study. In the future, we plan to validate the role of circulating AKR1C1 as a biomarker using larger clinical cohorts, develop AKR1C1 expression-based stratification strategies for HCC patients, investigate its synergistic potential with immune checkpoint inhibitors, and advance the preclinical translation of optimized pharmacokinetic delivery approaches for FFA and its derivatives, including localized tumor administration.

Conclusions

This study reveals that AKR1C1 plays a pivotal role in the development of acquired lenvatinib resistance in HCC, as evidenced by its upregulation in both lenvatinib-resistant HCC cells and patient plasma samples. Mechanistically, AKR1C1 attenuates the cytotoxicity of lenvatinib by activating the STAT3-ABC transporters pathway. Importantly, the non-steroidal anti-inflammatory drug FFA effectively reverses lenvatinib resistance by inhibiting AKR1C1-STAT3-ABC transporters pathway and inducing the apoptosis of lenvatinib-resistant HCC cells. Moreover, both the STAT3 inhibitor Stattic and the MRP2 inhibitor MK-571 also restored sensitivity to lenvatinib in HuH7/R and Hep3B/R cells. Collectively, our study uncovers a novel mechanism of acquired resistance to lenvatinib and proposes AKR1C1 as a promising biomarker and therapeutic target for lenvatinib-resistant HCC.

Acknowledgments

The authors would like to acknowledge the UALCAN website (https://ualcan.path.uab.edu/), the TIMER database (https://cistrome.shinyapps.io/timer/), the Kaplan-Meier Plotter (https://kmplot.com/analysis/), and the GEPIA2 (http://gepia2.cancer-pku.cn/) for providing us with rapid and convenient clinical data analysis.

Footnote

Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-892/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-892/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-892/prf

Funding: This work was supported by the Key Research and Development Program of Xuzhou City, China (No. KC23179); the Medical Youth Reserve Talents Training Program of Xuzhou Municipal Health Commission, China (No. XWRCHT20220006); the Joint Fund Project of Jiangsu Provincial Immunological Society (No. JSMYLHJJ-2025-0001); the Medical Science and Technology Innovation Program of Xuzhou Municipal Health Commission, China (Nos. XWKYHT20240050, XWKYHT20240047); the Science and Technology Development Fund of Xuzhou Medical University Affiliated Hospital, China (Nos. XYFM202413, XYFM202414); and the Key Medical Disciplines of Jiangsu Province’s 14th Five-Year Plan, China (No. ZDXK202237).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-892/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was formally approved by the Ethics Committee for Human Studies of Xuzhou Central Hospital (protocol No. XZXY-LK-20240108-0007) and informed consent was obtained from all patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Zheng J, Wang S, Xia L, et al. Hepatocellular carcinoma: signaling pathways and therapeutic advances. Signal Transduct Target Ther 2025;10:35. [Crossref] [PubMed]
3. Kudo M, Finn RS, Qin S, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet 2018;391:1163-73. [Crossref] [PubMed]
4. Fu R, Jiang S, Li J, et al. Activation of the HGF/c-MET axis promotes lenvatinib resistance in hepatocellular carcinoma cells with high c-MET expression. Med Oncol 2020;37:24. [Crossref] [PubMed]
5. Jin H, Shi Y, Lv Y, et al. EGFR activation limits the response of liver cancer to lenvatinib. Nature 2021;595:730-4. [Crossref] [PubMed]
6. Yang K, Zhang X, Long F, et al. AKR1B1 Inhibits Ferroptosis and Promotes Gastric Cancer Progression via Interacting With STAT3 to Activate SLC7A11. Cell Biol Int 2025;49:374-83. [Crossref] [PubMed]
7. Fu Z, Wu T, Gao C, et al. AKR1C1 interacts with STAT3 to increase intracellular glutathione and confers resistance to oxaliplatin in colorectal cancer. Acta Pharm Sin B 2024;14:5305-20. [Crossref] [PubMed]
8. Chang WM, Chang YC, Yang YC, et al. AKR1C1 controls cisplatin-resistance in head and neck squamous cell carcinoma through cross-talk with the STAT1/3 signaling pathway. J Exp Clin Cancer Res 2019;38:245. [Crossref] [PubMed]
9. Zhu H, Chang LL, Yan FJ, et al. AKR1C1 Activates STAT3 to Promote the Metastasis of Non-Small Cell Lung Cancer. Theranostics 2018;8:676-92. [Crossref] [PubMed]
10. Chang L, Wei S, Qi X, et al. The targeting of AKR1C1 synergizes with gefitinib via the STAT3 signaling pathway in EGFR-mutated NSCLC. Genes Dis 2025;12:101633. [Crossref] [PubMed]
11. Zhang KR, Zhang YF, Lei HM, et al. Targeting AKR1B1 inhibits glutathione de novo synthesis to overcome acquired resistance to EGFR-targeted therapy in lung cancer. Sci Transl Med 2021;13:eabg6428. [Crossref] [PubMed]
12. Wang Q, Liu J, Yang M, et al. Targeting AKR1B1 inhibits metabolic reprogramming to reverse systemic therapy resistance in hepatocellular carcinoma. Signal Transduct Target Ther 2025;10:244. [Crossref] [PubMed]
13. Wu X, Huang S, Gao J, et al. AKR1B10 dictates c-Myc stability to suppress colorectal cancer metastasis via PP2A nitration. Sci Adv 2025;11:eadv6937. [Crossref] [PubMed]
14. Wei H, Suo C, Gu X, et al. AKR1D1 suppresses liver cancer progression by promoting bile acid metabolism-mediated NK cell cytotoxicity. Cell Metab 2025;37:1103-1118.e7. [Crossref] [PubMed]
15. Sun D, Bo L, Jiang C, et al. Beyond the boundary: The emerging roles of ATP-binding cassette transporters in multidrug resistance (MDR) and therapeutic targeting in cancer. Drug Resist Updat 2026;84:101310. [Crossref] [PubMed]
16. Sun D, Liu J, Wang Y, et al. Co-administration of MDR1 and BCRP or EGFR/PI3K inhibitors overcomes lenvatinib resistance in hepatocellular carcinoma. Front Oncol 2022;12:944537. [Crossref] [PubMed]
17. Hu B, Zou T, Qin W, et al. Inhibition of EGFR Overcomes Acquired Lenvatinib Resistance Driven by STAT3-ABCB1 Signaling in Hepatocellular Carcinoma. Cancer Res 2022;82:3845-57. [Crossref] [PubMed]
18. Ge M, Chen XY, Huang P, et al. Understanding and overcoming multidrug resistance in cancer. Nat Rev Clin Oncol 2025;22:760-80. [Crossref] [PubMed]
19. Guinamard R, Simard C, Del Negro C. Flufenamic acid as an ion channel modulator. Pharmacol Ther 2013;138:272-84. [Crossref] [PubMed]
20. Zhang W, Tong S, Hu B, et al. Lenvatinib plus anti-PD-1 antibodies as conversion therapy for patients with unresectable intermediate-advanced hepatocellular carcinoma: a single-arm, phase II trial. J Immunother Cancer 2023;11:e007366. [Crossref] [PubMed]
21. Wang XH, Liu CJ, Wen HQ, et al. Effectiveness of lenvatinib plus immune checkpoint inhibitors in primary advanced hepatocellular carcinoma beyond oligometastasis. Clin Transl Med 2023;13:e1214. [Crossref] [PubMed]
22. Mei Z, Gao X, Pan C, et al. Lenvatinib enhances antitumor immunity by promoting the infiltration of TCF1+ CD8+ T cells in HCC via blocking VEGFR2. Cancer Sci 2023;114:1284-1296. [Crossref] [PubMed]
23. Guo X, Nie H, Zhang W, et al. Contrasting cytotoxic and regulatory T cell responses underlying distinct clinical outcomes to anti-PD-1 plus lenvatinib therapy in cancer. Cancer Cell 2025;43:248-268.e9. [Crossref] [PubMed]
24. Amawi H, Sim HM, Tiwari AK, et al. ABC Transporter-Mediated Multidrug-Resistant Cancer. Adv Exp Med Biol 2019;1141:549-80. [Crossref] [PubMed]
25. Laechelt S, Turrini E, Ruehmkorf A, et al. Impact of ABCC2 haplotypes on transcriptional and posttranscriptional gene regulation and function. Pharmacogenomics J 2011;11:25-34. [Crossref] [PubMed]
26. Ozeki T, Nagahama M, Fujita K, et al. Influence of CYP3A4/5 and ABC transporter polymorphisms on lenvatinib plasma trough concentrations in Japanese patients with thyroid cancer. Sci Rep 2019;9:5404. [Crossref] [PubMed]
27. Nie Z, Gao Y, Chen M, et al. Genome-Wide Screening Identifies Gene AKR1C1 Critical for Resistance to Pirarubicin in Bladder Cancer. Cancers (Basel) 2023;15:2487. [Crossref] [PubMed]
28. Badmann S, Mayr D, Schmoeckel E, et al. AKR1C1/2 inhibition by MPA sensitizes platinum resistant ovarian cancer towards carboplatin. Sci Rep 2022;12:1862. [Crossref] [PubMed]
29. Hu M, Sun D, Yu J, et al. Brusatol sensitizes endometrial hyperplasia and cancer to progestin by suppressing NRF2-TET1-AKR1C1-mediated progestin metabolism. Lab Invest 2022;102:1335-45. [Crossref] [PubMed]
30. Zhen S, Jia Y, Zhao Y, et al. NEAT1_1 confers gefitinib resistance in lung adenocarcinoma through promoting AKR1C1-mediated ferroptosis defence. Cell Death Discov 2024;10:131. [Crossref] [PubMed]
31. Phoo NLL, Dejkriengkraikul P, Khaw-On P, et al. Transcriptomic Profiling Reveals AKR1C1 and AKR1C3 Mediate Cisplatin Resistance in Signet Ring Cell Gastric Carcinoma via Autophagic Cell Death. Int J Mol Sci 2021;22:12512. [Crossref] [PubMed]
32. Gao C, Chang L, Xu T, et al. AKR1C1 overexpression leads to lenvatinib resistance in hepatocellular carcinoma. J Gastrointest Oncol 2023;14:1412-33. [Crossref] [PubMed]
33. Zhang Y, Han Y, Wang L, et al. Flufenamic Acid, a Promising Agent for the Sensitization of Colistin-Resistant Gram-Negative Bacteria to Colistin. Microbiol Spectr 2023;11:e0405222. [Crossref] [PubMed]
34. Matsumoto R, Tsuda M, Yoshida K, et al. Aldo-keto reductase 1C1 induced by interleukin-1β mediates the invasive potential and drug resistance of metastatic bladder cancer cells. Sci Rep 2016;6:34625. [Crossref] [PubMed]
35. Li ZY, Yin YF, Guo Y, et al. Enhancing Anti-Tumor Activity of Sorafenib Mesoporous Silica Nanomatrix in Metastatic Breast Tumor and Hepatocellular Carcinoma via the Co-Administration with Flufenamic Acid. Int J Nanomedicine 2020;15:1809-21. [Crossref] [PubMed]