Fibroblast growth factor 19-fibroblast growth factor receptor 4 axis: From oncogenesis to targeted-immunotherapy in advanced hepatocellular carcinoma

Abstract

Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality globally, with limited therapeutic progress for advanced stages. The aberrant fibroblast growth factor 19 (FGF19)-fibroblast growth factor receptor 4 (FGFR4) axis promotes oncogenesis and is linked to targeted-immunotherapy of HCC. Multi-kinase inhibitors (MKIs) enhance anti-tumor effects by targeting this axis and FGF19 overexpression upregulates programmed cell death ligand 1 in tumor microenvironment. Clinical studies have demonstrated the efficacy of selective FGFR4 inhibitors in HCC treatment, with enhanced anti-tumor effects when combined with MKIs or immune checkpoint inhibitors. Phase I clinical trials of Irpagratinib (ABSK-011) demonstrated an objective response rate of 43.5%, which increased to 55.6% combined with atezolizumab. FGF19 also serves as a biomarker for HCC. This review systematically summarizes the literature retrieved from PubMed and other databases using search terms “HCC”, “fibroblast growth factor 19”, “fibroblast growth factor receptor 4”, “FGFR4 inhibitor”, “targeted therapy”, “multi-kinase inhibitor”, “immunotherapy”, “immune checkpoint inhibitor”, and “biomarker”. It also firstly synthesizes combination strategies and underlying mechanisms between FGFR4 inhibitors and targeted-immunotherapy, addressing critical gaps in existing reviews. Additionally, we discuss the potential of FGF19 as a predictive biomarker, integrating mechanistic and clinical evidence to advance precision HCC therapeutics.

Key Words: Hepatocellular carcinoma; Fibroblast growth factor 19; Selective fibroblast growth factor receptor 4 inhibitor; Adverse events; Resistance; Targeted-immunotherapy; Tumor microenvironment; Biomarker

Core Tip: The aberrant Fibroblast growth factor 19 (FGF19)-fibroblast growth factor receptor 4 (FGFR4) axis plays a crucial role in progression of hepatocellular carcinoma. Multi-kinase inhibitors enhance anti-tumor effects by targeting the FGF19-FGFR4 axis and FGF19 overexpression contributes to the development of immune suppression. These findings highlight the association between FGF19 and targeted-immunotherapy in hepatocellular carcinoma (HCC). Selective FGFR4 inhibitors, either alone or combining with targeted-immunotherapy, have demonstrated therapeutic potential for advanced HCC, although the clinical application is hindered by challenges such as adverse effects and drug resistance. Furthermore, FGF19 serves as a promising biomarker for HCC, underscoring its potential for precision treatment.

INTRODUCTION

Primary liver cancer is the sixth most common malignancy worldwide and the third leading cause of cancer-related death. Its incidence is particularly high in East Asia, with China being a representative example. Hepatocellular carcinoma (HCC) is the most prevalent form of primary liver cancer, accounting for 75%-85% of cases[1]. Most HCC patients are diagnosed at advanced stages, leaving them ineligible for curative surgery. For these patients, the standard treatment approach is systemic therapy, typically involving multi-kinase inhibitors (MKIs) in combination with immune checkpoint inhibitors (ICIs)[2]. While targeted-immunotherapy has shown promising results, several factors still limit its effectiveness. For example, tumor cells can resist MKIs-induced apoptosis by activating cell survival pathways, suppressing reactive oxygen species (ROS) production, and promoting epithelial-to-mesenchymal transition (EMT). Furthermore, the exclusion and dysfunction of T cells within the tumor microenvironment (TME) contribute to the low response and resistance to immunotherapy[3,4]. Therefore, ongoing research is essential to explore potential therapeutic strategies for advanced HCC.

Fibroblast growth factor 19 (FGF19) is a secreted protein primarily produced in the small intestine, where it interacts with fibroblast growth factor receptor 4 (FGFR4) to form a signaling pathway that plays a critical role in regulating hepatic bile acid (BA) synthesis[5]. Shortly after the discovery of FGF19, its involvement in liver tumor formation was identified through studies in transgenic mice overexpressing FGF19 in skeletal muscle, which highlighted its role in HCC[6]. Research has shown that FGF19 overexpression contributes to sustained proliferation, metastasis, and resistance to apoptosis in HCC cells, and is associated with poor prognosis and a high risk of recurrence[7-9]. Targeted inhibition of vascular endothelial growth factor receptor (VEGFR) and epidermal growth factor receptor (EGFR) is a key anti-tumor mechanism of MKIs. However, the resistance due to heterogeneity of HCC and secondary mutations in the targeted tyrosine kinases has driven research into FGFR inhibitors[10]. Given the critical role of the FGF19-FGFR4 axis in HCC development, inhibiting this pathway offers new therapeutic opportunities. In recent years, several FGFR4 inhibitors (e.g., H3B-6527, BLU-554, FGF401) have entered clinical trials, offering hope for the treatment of advanced HCC[11]. However, the clinical benefits of these monotherapies remain limited, suggesting that combination therapies may be necessary to enhance efficacy. This review summarizes the mechanisms by which FGF19 overexpression promotes HCC, clinical studies on FGFR4 inhibitors, and recent progress in targeting the aberrant FGF19-FGFR4 axis in targeted-immunotherapy of advanced HCC. Furthermore, although several potential biomarkers such as alpha-fetoprotein (AFP) and Des-γ-carboxy prothrombin (DCP) have been proposed, they remain insufficient to fully predict the response to targeted-immunotherapy of advanced HCC[12]. Thus, we also discuss the studies on combining FGF19 with traditional biomarkers for HCC diagnosis and treatment, with the goal of providing new perspectives on more precise and effective treatment options.

METHODOLOGY

We conducted searches using the following keywords in PubMed, EMBASE, Web of Science, Google Scholar and Cochrane Library databases, as well as in the reference lists of related studies: Hepatocellular carcinoma, fibroblast growth factor 19, fibroblast growth factor receptor 4, FGFR4 inhibitor, targeted therapy, multi-kinase inhibitor, immunotherapy, immune checkpoint inhibitor, biomarker[13]. The literature was independently screened by two professionals, and any discrepancies were resolved through discussion to achieve consensus. The relevant ongoing clinical trials registered at ClinicalTrials.gov are included. The specific search strategy summary is shown in Table 1.

Table 1 Search strategy summary.

Items	Specification
Date of search	Jun 5, 2025
Databases and other sources searched	PubMed, EMBASE, Web of Science, Google Scholar and Cochrane Library
Search terms used	Hepatocellular carcinoma, fibroblast growth factor 19, fibroblast growth factor receptor 4, FGFR4 inhibitor, targeted therapy, multi-kinase inhibitor, immunotherapy, immune checkpoint inhibitor, biomarker
Timeframe	1990-2025
Inclusion criteria	(1) Study type: All types; and (2) Language: English
Selection process	The literature was independently screened by two professionals, and any discrepancies were resolved through discussion to achieve consensus

FGFR4: Fibroblast growth factor receptor 4.

FGF19-FGFR4 AXIS IN HCC

FGF19 is localized on chromosome 11q13, while FGFR4 is located on chromosome 5q35.1[14]. FGF19 is primarily secreted by the ileal villi and the gallbladder epithelial cells. Its binding to FGFR4 requires cofactors, including heparan sulfate proteoglycans and β-Klotho, which facilitate dimerization and phosphorylation of the receptor's intracellular tyrosine kinase domains and initiate downstream signalings[15-18]. The FGF19-FGFR4 axis plays a critical role in maintaining BA-related metabolic homeostasis under physiological conditions. It regulates BA synthesis by inhibiting cholesterol 7α-hydroxylase (CYP7A1), the rate-limiting enzyme in BA synthesis[19]. Additionally, FGF19 regulates nutrient metabolism by stimulating glycogen and protein synthesis, and inhibiting fatty acid synthesis[20-22].

The aberrant activation of FGF19-FGFR4 axis can influence hepatocyte proliferation and migration, leading to the development of malignant tumors such as HCC and hepatoblastoma[23,24]. FGF19 gene amplification occurs in up to 20% of cases in HCC and is associated with adverse prognostic factors, including high AFP levels, microvascular invasion, and a higher risk of recurrence[25,26]. Binding of FGF19 to FGFR4 induces FGFR4 phosphorylation, which recruits adaptor proteins such as FGFR substrate-2 (FRS2), growth factor receptor-bound protein 2, and guanine nucleotide exchange factors son of sevenless homolog 1/2. This leads to the activation of the RAS-RAF-ERK1/2-MAPK and PI3K-AKT pathways, promoting cell proliferation and resistance to apoptosis[27]. Phosphorylated FGFR4 can also activate phospholipase Cγ, leading to the phosphorylation of protein kinase C and further activation of the RAS-RAF-MAPK signaling pathway[28]. Studies have also indicated that EMT plays a crucial role in FGF19-driven HCC. The FGF19-β-Klotho-FGFR4 complex can induce endoplasmic reticulum (ER) stress, leading to glycogen synthase kinase 3β (GSK3β) phosphorylation and increased nuclear translocation of β-catenin. Additionally, it can activate the JAK2/STAT3 pathway, thereby promoting EMT and enhancing HCC metastasis[29-31]. Figure 1 illustrates the mechanism of the aberrant FGF19-FGFR4 axis promoting HCC[32].

Figure 1 The aberrant fibroblast growth factor 19-fibroblast growth factor receptor 4 axis drives hepatocellular carcinoma. Overexpression of fibroblast growth factor 19 (FGF19) binds to fibroblast growth factor receptor 4 (FGFR4), leading to its phosphorylation and recruitment of FGFR substrate-2 (FRS2), GRB2, and SOS, which subsequently trigger a cascade of downstream effector proteins. The aberrant FGF19-FGFR4 axis primarily promotes tumor growth through the RAS-RAF-MAPK and PI3K-AKT signaling pathways. Additionally, it facilitates the phosphorylation of GSK3β, preventing the inactivation of downstream β-catenin. Activated FGFR4 can also activate the JAK2/STAT3 pathway. Together, these mechanisms promote hepatocellular carcinoma cell proliferation and resistance to apoptosis. Additionally, the FGF19-FGFR4 axis shares similar downstream molecules with the vascular endothelial growth factor receptor pathway. FGF19: Fibroblast growth factor 19; FGFR4: Fibroblast growth factor receptor 4; VEGF: Vascular endothelial growth factor; KLB: Β-Klotho; GRB2: Growth Factor receptor-bound protein 2; SOS: Son of sevenless homolog; FRS2: FGFR substrate-2; SHC: Src homology domain protein C1; JAK: Janus kinase; STAT: Signal transducer and activator of transcription; RAS: Rat sarcoma; RAF: Rapidly accelerated fibrosarcoma; MEK: Mitogen-activated protein kinase; ERK: Extracellular signal-regulated kinase; PI3K: Phosphoinositide 3-kinase; AKT: Protein kinase B; mTOR: Mechanistic target of rapamycin; GSK3β: Glycogen synthase kinase 3β; C-MYC: Cellular myelocytomatosis; EMT: Epithelial-mesenchymal transition; Bcl-2: B-cell lymphoma 2; IL-6: Interleukin-6; PD-L1: Programmed death-ligand 1. This figure was created with BioGDP.com[32].

In recent years, researchers have continued to investigate the mechanisms of hepatocarcinogenesis associated with the FGF19-FGFR4 axis, with a particular focus on signaling molecules upstream and downstream of this pathway. Studies have shown that upregulation of small nucleolar RNA host gene 16 (SNHG16) increases FGF19 expression via the SNHG16/miR-302a-3p axis[33]. Additionally, an increasing number of transcription factors have been implicated in FGF19-driven HCC. For example, overexpression of the sex-determining region Y box transcription factor 18 (SOX18) upregulates FGFR4, and the upregulation of FGF19-FGFR4 axis enhances SOX18 expression through GSK3β-β-catenin signaling in turn[34]. These three factors form a positive feedback loop and promote HCC invasion. Furthermore, FGF19 upregulates the expression of Homeobox B5, a member of the HOX family of transcription factors, through the PI3K/AKT/hypoxia-inducible factor (HIF) 1α pathway. This is positively correlated with poor tumor differentiation and advanced tumor-node-metastasis stages in HCC[35]. These findings further elucidate the link between the dysregulated FGF19-FGFR4 axis and HCC, offering new perspectives for early diagnosis and treatment. However, further exploration of FGF19-driven HCC is necessary.

DEVELOPMENT OF SELECTIVE FGFR4 INHIBITORS

Selective FGFR4 inhibitors and related clinical trials

Small molecule inhibitors and biologics targeting FGFR have been developed for the targeted therapy of patients with FGFR mutations[36]. The FGF19-FGFR4 axis has emerged as a promising therapeutic target for HCC, and small molecule inhibitors directed at this axis are under extensive investigation as a key treatment approach. Due to the structural similarity between the VEGFR and FGFR kinase domains, most MKIs exhibit FGFR inhibitory activity. However, they generally have a significantly higher affinity for VEGFR than FGFR[37]. Additionally, the intrinsic flexibility of the FGFR4 kinase domain results in relatively low binding affinity of pan-FGFR inhibitors for FGFR4[38]. Furthermore, inhibition of FGFR1-3 may lead to dose-limiting toxicities, such as hyperphosphatemia, which further diminishes the effectiveness of pan-FGFR inhibitors targeting FGFR4[39]. These factors hinder the targeted inhibition of the FGF19-FGFR4 axis, highlighting the importance of developing selective FGFR4 inhibitors to enhance the therapeutic efficacy in HCC treatment.

The Cys 552 residue in the hinge region of the ATP-binding site in FGFR4 can modify the size of the ATP binding pocket and facilitate covalent binding with the ligand, providing a structural basis for the development of selective FGFR4 inhibitors[40]. Based on the selective FGFR inhibitor PD173074[41], Hagel et al[42] developed BLU-9931 (FGFR1 IC₅₀ = 591 nm, FGFR2 IC₅₀ = 493 nm, FGFR3 IC₅₀ = 150 nm, FGFR4 IC₅₀ = 3 nm) as the first irreversible selective FGFR4 inhibitor in 2015. They used the aminoquinazoline core as a scaffold, and introduced a 2,6-dichloro-3,5-dimethoxyphenyl group to substitute the original dimethoxyphenyl and acrylamide moiety. BLU-9931 demonstrated dose-dependent anti-tumor activity in FGF19-overexpressing Hep 3B cell lines and xenograft models. However, the planarity of the phenyl ring in the compound limited its further development. To reduce the dose-limiting toxicity when binding to FGFR1-3 and enhance selectivity for FGFR4, Joshi et al[43] chose an N-aryl-N0-pyrimidin-4-yl urea scaffold and added an acrylamide to the ortho-position of the phenyl ring, and then created a highly selective FGFR4 inhibitor H3B-6527 (FGFR1 IC₅₀ = 320 nm, FGFR2 IC₅₀ = 1290 nm, FGFR3 IC₅₀ = 1060 nm, FGFR4 IC₅₀ < 1.2 nm) in 2017. H3B-6527 inhibited the proliferation of the HCC cell line Hep3B and induced apoptosis. It also induced tumor regression in the Hep3B human HCC xenograft models at doses of 100 mg/kg and 300 mg/kg. Furthermore, combining H3B-6527 with the CDK4/6 inhibitor palbociclib enhanced its anti-tumor activity in FGF19-overexpressing JHH-7 cell line xenograft models. The Phase I clinical trial of H3B-6527 showed the most common adverse events (AEs) were diarrhea (58.6%), nausea (25%), and fatigue (20.3%). Among the serious AEs (SAEs), the highest incidences were fatigue (3.9%) and tumor bleeding (3.1%)[44].

In 2019, Kim et al[45] tried to improve the planarity of the benzene ring by converting it into the pyran ring and introduced Fisogatinib (BLU-554) (FGFR1 IC₅₀ = 591 nm, FGFR2 IC₅₀ = 493 nm, FGFR3 IC₅₀ = 150 nm, FGFR4 IC₅₀ = 3 nm), an optimized version of BLU-9931. This compound effectively induced dose-dependent tumor regression in FGF19 immunohistochemistry (IHC)-positive xenograft models. Compared to pan-FGFR inhibitors, BLU-554 treatment at all dose levels did not cause hyperphosphatemia, confirming its selective inhibition of FGFR4. In terms of clinical activity, an analysis of 66 FGF19 IHC-positive HCC patients showed an overall response rate (ORR) of 17% (11/66). The most common AEs to BLU-554 were nausea (74%), vomiting (42%), and diarrhea (35%), all of which were associated with increased BA synthesis and could be managed through symptomatic treatment. A total of 43% of patients experienced grade 3 or higher AEs, most notably elevated aspartate aminotransferase (AST) (15%) and alanine aminotransferase (ALT) (11%) levels[46]. To address potential resistance to monotherapy, a subsequent Phase Ib/II clinical trial (NCT04194801) was conducted to evaluate BLU-554 in combination with the anti-programmed death-ligand 1 (PD-L1) monoclonal antibody CS1001 in patients with locally advanced or metastatic HCC[47]. In the published results, four FGF19 IHC-positive patients received the combination treatment, yielding an ORR of 50% and a disease control rate of 100%. Only one patient experienced an immune-related AE[48]. This clinical trial was completed in February 2023.

In 2020, Roblitinib (FGF401) was introduced as the first reversible FGFR4 inhibitor (FGFR1/2/3 IC₅₀ > 10000 nm, FGFR4 IC₅₀ = 1.9 nm), with a reversible covalent hemiacetal structure identified to form between Cys552 and the aldehyde group. Compared to irreversible FGFR4 inhibitors, FGF401 demonstrates more sustained and complete inhibition of FGFR4 activity[49,50]. Studies in HCC models with high FGF19 expression have shown that FGF401 induced G1 phase cell cycle arrest and apoptosis in tumor cells, while also promoting blood vessel normalization in the TME. Moreover, the vascular normalization induced by Roblitinib sensitized tumors to mitotic drugs, and evidence suggested it could synergize with the anti-tumor drug vincristine[51]. A Phase I/II clinical trial (NCT02325739) investigating FGF401 alone and in combination with spartalizumab [a programmed cell death protein 1 (PD-1) monoclonal antibody] in HCC and biomarker-selected solid tumors revealed that 72.5% (116/160) of patients experienced ≥ grade 3 AEs in FGF401 monotherapy group. The most common AEs were diarrhea (73.8%), elevated AST (47.5%), and elevated ALT (43.8%) levels. In combination therapy group, all patients experienced AEs, with diarrhea (58.3%), elevated AST (50%), and hyperphosphatemia (41.7%) being the most frequent. Additionally, 50% (6/12) of patients in combination group experienced AEs of ≥ grade 3 severity[52].

In recent years, the oral drug Irpagratinib (ABSK-011), a highly selective and potent FGFR4 inhibitor, has emerged as a promising treatment option for advanced HCC[53]. A Phase I clinical trial (NCT04906434) aimed at evaluating the safety, pharmacokinetics (PK), pharmacodynamics, and preliminary efficacy of ABSK-011 found that it demonstrated a promising ORR of 43.5%. The most common treatment-related AEs were diarrhea (72.0%), elevated ALT (70.7%), and elevated AST level (57.3%), with a 29.3% incidence of Grade 3/4 treatment-related AEs (TRAEs). These preliminary results suggested that ABSK-011 had a favorable safety profile and promising clinical potential. A recent Phase II open-label study (NCT05441475) assessed the preliminary efficacy and safety of ABSK-011 in combination with atezolizumab for the treatment of FGF19 IHC-positive HCC. Among 28 patients assessable for efficacy, seven patients achieved partial remission (6 FGF19+ and 1 FGF19-). In patients receiving ABSK-011 220 mg twice daily, the ORR reached 55.6% (5/9). Regarding safety, thirty patients (83.3%) experienced TRAEs, most of which were transient and improved after dose interruption or supportive treatment and 4 patients experienced severe TRAEs[54]. A Phase III clinical trial for ABSK-011 is currently being planned. In addition to drugs mentioned before, researchers are exploring various strategies to enhance the anti-tumor activity of FGFR4 inhibitors, including structural modifications to the solvent-binding regions of existing selective FGFR4 inhibitors[55,56], substituting classic electrophilic reagents with nucleophilic-substituted covalent warheads[57], and targeting multiple cysteine residues on FGFR4[58,59]. These approaches offer new avenues for the development of future anti-tumor drugs. Table 2 summarizes the clinical trials related to selective FGFR4 inhibitors.

Table 2 Clinical trial of selective fibroblast growth factor receptor 4 inhibitors.

Drug	FGFR4 IC50	Participants	Key outcomes	Adverse events	ClinicalTrial.gov ID (phase)
H3B-6527[44]	< 1.2 nM	128	Part II H3B-6527 1000 mg QD	AEs (All)	NCT02834780 (I)
			ORR: 5.6% mOS: 9.5 months mPFS: 2.6 months	Diarrhea: 58.6% Nausea: 25% Fatigue: 20.3%
			Part II H3B-6527 500 mg BID	sAEs
			ORR: 0% mOS: 7.7 months mPFS: 3.0 months	Asthenia: 3.9% Tumor haemorrhage: 3.1%
Fisogatinib (BLU-554)[46]	3 nM	115	FGF19 IHC positive	AE (All grades): 96%	NCT02508467 (I)
			ORR: 17% DCR: 62% mPFS: 3.3 months	Diarrhea: 74% Nausea: 42% Vomitting: 35%
			FGF19 IHC negative	AE (grade ≥ 3): 43%
			ORR: 0% DCR: 50% mPFS: 2.3 months
Fisogatinib + CS1001[47]		26	Phase Ib	AEs (All)	NCT04194801 (Ib/II)
			Fisogatinib 400 mg + CS1001 1200 mg: ORR: 0% mOS: 16.2 months mPFS: 2.3 months	Blood bilirubin increased: 76.9%
			Fisogatinib 600 mg + CS1001 1200 mg	ALT increased: 76.9% AST increased: 76.9%
			ORR: 14.3% DCR: 42.9% mPFS: 2.1 months	sAEs
			Phase II	ALT increased: 7.7% AST increased: 7.7%
			Fisogatinib 600 mg + CS1001 1200 mg
			ORR: 20% DCR: 60% mPFS: 4.1 months
Roblitinib (FGF401)[52]	3 nM	160	Phase I	AEs (All)	NCT02325739 (I/II)
			ORR: 6.6% DCR: 59.0%	Diarrhea: 74.4% AST increased: 49.4% ALT increased: 45%
			Phase II	sAEs
			ORR: 6.1% DCR: 53%	Abdominal pain: 5.6% Ascites: 3.1% Haemoptysis: 2.5%
Roblitinib + PDR001[52]		12	FGF401 80 mg + PDR001 300 mg	AEs	NCT02325739 (I/II)
			ORR: 16.7% DCR: 50%	Diarrhea: 75% AST increased: 50% Hyperphosphataemia: 41.7%
			FGF401 120 mg + PDR001 300 mg	sAEs
			ORR: 16.7% DCR: 50%	Diarrhea: 16.7% Pyrexia: 16.7%
ABSK-011[53]	4.35 nM	75	CR: 1	AEs (All grades)	NCT04906434 (I)
			PR: 12	Diarrhea: 72.0% ALT increased: 70.7% AST increased: 57.3%
			BID dosing ORR: 43.5%	AEs (Grade ≥ 3)
				ALT increased: 10.7% AST increased: 9.3% Diarrhea: 5.3%
ABSK-011 + Atezolizumab[54]		36	PR: 7	AEs: 83.3%	NCT05441475 (II)
			DOR: 15.4 months	sAEs
			ABSK-011 220 mg BID + Atezolizumab 1200 mg IV Q3W: ORR: 55.6%	ALT increased: 8.3% AST increased: 5.6%; Blood bilirubin increased: 5.6% haematochezia: 2.8%

FGF19: Fibroblast growth factor 19; FGFR4: Fibroblast growth factor receptor 4; IHC: Immunohistochemistry; ORR: Objective response rate; DCR: Disease control rate; mPFS: Median progression-free survival; AE: Adverse events; sAE: Serious adverse events; mOS: Median overall survival; ALT: Alanine aminotransferase; AST: Aspartate transaminase; CR: Complete response; PR: Partial response; DOR: Duration of response; IV: Intravenous.

Limitations and solutions in clinical application of selective FGFR4 inhibitors

Significant progress has been made in the development of FGFR4 inhibitors. However, several issues such as AEs and drug resistance still limit their clinical application. Therefore, this section will examine the challenges associated with selective FGFR4 inhibitors and propose corresponding management strategies to offer effective solutions for their broader clinical application.

AEs

While Selective FGFR4 inhibitors exhibit anti-tumor effects, they are frequently associated with AEs that can impact patient adherence to treatment and quality of life. Current studies indicate that diarrhea, nausea, and elevated ALT or AST levels are the most common AEs linked to FGFR4 inhibitors, which present challenges for their clinical application. Diarrhea may be associated with the upregulation of cholesterol conversion to BAs, resulting from the inhibition of the FGF19-FGFR4 axis. This BA metabolic imbalance leads to diarrhea by increasing intestinal water secretion, enhancing mucosal permeability, and stimulating intestinal motility[60]. If grade ≤ 2 diarrhea occurs, symptoms can be relieved by taking anti-diarrheal drugs like loperamide. When grade ≥ 3 diarrhea occurs, FGFR4 inhibitors should be suspended immediately and active supportive treatment should be initiated[61]. Cholestyramine, a BA sequestrant, has been reported to be effective in treating high BA diarrhea caused by selective FGFR4 inhibitors. Cholestyramine is a strong ion-exchange resin that binds BAs in the intestine and forms insoluble complexes, which prevents the reabsorption of BAs and promotes their excretion through feces[62]. AEs like nausea and elevated transaminase levels are also predictable consequences of BA upregulation following FGFR4 inhibition, and symptomatic treatment generally resolves these issues[45]. Schadt et al[63] found that cholestyramine could also mitigate the elevation of ALT level caused by FGFR4 inhibitors. Given the close relationship between BA and these AEs, monitoring serum BA levels during the use of FGFR4 inhibitors could be an essential strategy for managing related side effects. However, due to individual variations in tolerance, as well as the effects of underlying liver disease and cirrhosis on the BA pool in HCC patients[64], there is currently no clear standard for risk stratification based on BA levels.

Compared to digestive system, AEs involving other systems are not common after using selective FGFR4 inhibitors and symptomatic treatment can usually reduce their impact. Stomatitis and dry mouth can be relieved by baking soda rinses, stimulating salivary, and using intraoral topical agents. Dermatologic AEs such as alopecia can be alleviated with minoxidil, and dry skin can be treated with preparations containing urea or salicylic acid. Hand-foot syndrome can be relieved with urea, creams, or topical steroids[61]. To minimize the risk of gastrointestinal bleeding, a gastrointestinal endoscopy can be performed before starting selective FGFR4 inhibitor treatment. In addition to systematically managing AEs with symptomatic treatment, future research should focus on exploring how to minimize the occurrence of drug-related AEs based on the anti-tumor effects, such as determining the optimal drug dosage and developing relevant predictive biomarkers.

Drug resistance

Additionally, the potential for drug resistance in HCC limits the application of FGFR4 inhibitors. Mutations in specific FGFR4 sites during treatment contribute to drug resistance. Studies have demonstrated that two HCC patients treated with BLU-554 experienced disease progression due to mutations in the FGFR4 gatekeeper site (V550M, V550 L) and hinge-1 site (C552R)[65]. These mutations confer resistance to BLU-554 both in vivo and in vitro. Since gatekeeper site mutations are more prevalent than hinge-1 mutations, Shao et al[66] designed and synthesized a series of amino-indazole derivatives to target this resistance mechanism. The representative compound 7v exhibited excellent activity against FGFR4, FGFR4V550 L, and FGFR4V550M in vitro. However, due to its poor PK, compound 7v displayed limited anti-tumor activity in the Huh7 xenograft mouse model[66]. Nevertheless, this study provided new insights for optimizing existing selective FGFR4 inhibitors. Subsequent research explored a series of derivatives based on 5-Formyl-pyrrolo[3,2-b]pyridine-3-carboxamides and 1H-indazole to target FGFR4 gatekeeper mutations and overcome tumor resistance[67,68]. In addition to resistance mutations, bypass signaling has been identified as another mechanism contributing to resistance against MKIs[69]. The EGFR shares common downstream signaling pathways with FGFR in the RAS-RAF-ERK and PI3K-AKT pathways and has been found to synergize in oncogenic effects in lung adenocarcinoma and colorectal cancer[70,71]. In 2021, Jin et al[72] observed that lenvatinib treatment for HCC led to feedback activation of the EGFR-PAK2-ERK5 pathway due to FGFR inhibition, which limited the anti-tumor efficacy. This suggests that EGFR activation may be a contributing factor in FGFR inhibitor resistance[72]. In 2023, Shen et al[73] developed an FGF19-overexpressing resistant cell line by long-term exposure of the Huh7 cell line to the selective FGFR4 inhibitor BLU-554. They found significant enrichment of multiple gene sets activated by EGF or EGFR and demonstrated that combined inhibition of the EGFR signaling pathway restored HCC sensitivity to FGFR4 inhibitors. This study revealed that EGFR-mediated bypass activation contributed to FGFR4 inhibitor resistance and indicated that combining EGFR inhibitors might provide a strategy to overcome this resistance. Additionally, dual blockade of EGFR and FGFR4 may be a promising future therapeutic strategy for enhanced management of HCC. As early as 2020, Chen et al[74] highlighted the potential of machine learning strategies to discover dual-target EGFR and FGFR inhibitors, and they selected compound 1 exhibiting inhibitory activity against FGFR4 (IC₅₀ = 86.2 nm) and EGFR (IC50 = 83.9 nm) kinase respectively. In 2024, Xue et al[75] investigated five natural compounds with electrophilic Michael-acceptor warhead for covalent bonding and anti-cancer properties, and targeted both EGFR and FGFR4 covalently via cysteine residues Cys797 and Cys552. By comparing the X-ray crystal structures with the known potent inhibitors of EGFR (Afatinib) and FGFR4 (BLU-9931), they found that curcumin, syringolin A and andrographolide may serve as dual EGFR and FGFR4 covalent irreversible inhibitors, which could be used alone or in combination with other drugs for HCC treatment. These studies offer directions for overcoming resistance to FGFR4 inhibitors. However, further exploration is needed to understand the mechanisms of drug resistance and identify solutions. For instance, TP53 mutations have been identified in patients with advanced solid tumors treated with selective FGFR1-3 inhibitors, though it remains uncertain whether these mutations contribute to resistance against FGFR4 inhibitors[76].

FGF19-FGFR4 AXIS IN TARGETED THERAPY

MKIs have demonstrated broad efficacy in the treatment of advanced HCC. Sorafenib was approved by the Food and Drug Administration (FDA) as the first-line treatment for advanced HCC in 2007, followed by lenvatinib in 2018[77,78]. Additionally, second-line treatment options, such as regorafenib and cabozantinib, have become available[79,80]. As mentioned above, the aberrant FGF19-FGFR4 axis promotes the development of HCC by activating a series of downstream signals and FGFR4 is also a target of MKIs, such as lenvatinib[81]. Given the critical role of this signaling pathway in HCC, its relationship with targeted therapy for advanced HCC has become a hot topic in current research.

Sorafenib inhibits multiple signaling pathways, including RAF-MEK-ERK, VEGFR, platelet-derived growth factor receptor, FMS-like tyrosine kinase 3 and rearranged during transfection[77]. Although there is no direct evidence that sorafenib targets FGFR in vivo or in vitro, recent studies have begun to explore the interaction between the FGF19-FGFR4 axis and sorafenib in the treatment of advanced HCC. In 2017, Gao et al[82] reported that overexpression of FGF19 in the MHCC97 L HCC cell line prevented sorafenib-induced apoptosis. Sorafenib induces oxidative stress in HCC cells by increasing intracellular ROS, whereas FGF19 overexpression mitigates this effect. Conversely, knockout of FGF19 or FGFR4 enhances oxidative stress in HCC cells. These findings suggest that combining FGFR4 inhibition with sorafenib may enhance its anti-tumor efficacy. They subsequently co-administered the FGFR4 inhibitor ponatinib with sorafenib, demonstrating that the combination therapy resulted in superior antitumor effects compared to monotherapy[82]. Similarly, Kanzaki et al[83] observed a greater degree of tumor regression in non-obese diabetic/severe combined immunodeficiency mice treated with both sorafenib and selective FGFR4 inhibitors compared to those treated with sorafenib alone. These results indicate a promising new targeted therapeutic approach for advanced HCC and provide a foundation for improving the efficacy of sorafenib.

Angiogenesis is a critical characteristic of HCC, and the targeted inhibition of the VEGF-VEGFR signaling pathway plays a critical role in anti-tumor therapies[84]. Both the VEGFR and FGFR signaling pathways share common downstream molecules, which contribute to HCC progression through non-autonomous mechanisms[85]. Elevated levels of FGFs have been shown to enhance VEGF-mediated angiogenesis, thereby promoting resistance to anti-VEGFR therapies[86]. As a result, numerous studies have focused on exploring the relationship between anti-angiogenic agents and the FGF19-FGFR4 axis. Both sorafenib and lenvatinib inhibit VEGFR, but lenvatinib is distinguished by its potent inhibition of FGFR. In sorafenib-resistant HCC cells, lenvatinib inhibits cell growth, likely through suppression of the FGF19-FGFR4-ERK signaling pathway[82]. Inhibition of FGFR by lenvatinib may contribute to its superior progression-free survival (PFS) and ORR in advanced HCC compared to sorafenib. Several studies have examined how lenvatinib exerts anti-tumor effects in advanced HCC by targeting FGFR. For instance, Matsuki et al[87] demonstrated in preclinical HCC models that lenvatinib inhibited FRS2 phosphorylation in Hep3B2.1-7 cells expressing FGF19 in a concentration-dependent manner. In Hep3B2.1-7 and SNU-398 xenografts, lenvatinib also exerted anti-tumor effects by inhibiting both FRS2 and ERK1/2 phosphorylation. Iseda et al[88] reported that lenvatinib suppressed the expression of Cystine/Glutamate Transporter and glutathione peroxidase 4 by inhibiting FGFR4, leading to the accumulation of ROS and ferroptosis in HCC cells. Furthermore, Yi et al[89] investigated whether lenvatinib could enhance the anti-tumor immune response in HCC by inhibiting FGFR4. They found that lenvatinib inhibited GSK3β phosphorylation, promoting PD-L1 degradation via ubiquitination, and interfered with the differentiation of naive T cells into regulatory T cells (Tregs) by inhibiting STAT5 phosphorylation. These findings suggest that combining lenvatinib with ICIs could potentially improve the efficacy of advanced HCC treatment. Additionally, the inhibition of FGFR by lenvatinib extends beyond FGFR4. Hoshi et al[90] demonstrated that lenvatinib could exert anti-HCC effects under nutrient-deprived conditions by targeting MAPK to inhibit FGFR1-3 and promote proline-rich acidic protein cleavage. Previous studies have demonstrated that FGFR redundancy, particularly FGFR3, can reduce the anti-tumor efficacy of FGFR4 inhibitors and induce drug resistance[91]. Consequently, the dual inhibition of both FGFR3 and FGFR4 by lenvatinib may further enhance its anti-tumor activity. However, the underlying mechanisms remain to be fully elucidated. In conclusion, lenvatinib inhibits FGFR through multiple mechanisms, contributing to its antitumor effects in advanced HCC. Building on previous research, scholars have explored whether the combination of FGFR4 inhibitors and lenvatinib can enhance its antitumor efficacy. In a Patient-Derived Xenograft model using the FGF19-positive Hep3B HCC cell line, Zhao et al[85] combined lenvatinib with H3B-6527 and demonstrated that the combination resulted in a significant regression of tumors compared to either agent alone. Notably, the improved efficacy was not due to enhanced FGFR4 inhibition, but rather to the modulation of hypoxia and glycolytic pathways within the TME. These results suggest that H3B-6527 potentiates the anti-angiogenic effects of lenvatinib. Ito et al[92] found that class IIa histone deacetylase inhibitors (HDACIs) reduced FGFR4 expression in HCC cells by inhibiting FGFR4 transcriptional activity. In vitro, the combination of HDACIs and lenvatinib exhibited synergistic anti-tumor effects in FGF19-expressing JHH-7 and HuH-7 cells. This preclinical evidence supports the hypothesis that FGFR4 inhibition enhances the anti-tumor activity of lenvatinib in FGF19-driven HCC. These findings support the need for further clinical trials to evaluate the efficacy of this combination regimen in patients with advanced HCC.

In summary, inhibiting the FGF19-FGFR4 axis has demonstrated substantial potential in the treatment of advanced HCC. Combining FGFR4 inhibitors with MKIs enhances the anti-tumor effect, particularly in suppressing tumor angiogenesis. However, there are several limitations in current research. While previous studies have highlighted the role of the FGF19-FGFR4 axis in HCC, the interaction mechanisms between this pathway and others, such as VEGF and PDGF, remain poorly understood. The use of FGFR4 inhibitors may lead to target resistance mutations or activation of oncogenic bypass pathways, contributing to resistance to targeted therapies. Additionally, most preclinical studies focus on cell and animal models, with insufficient large-scale clinical data to confirm their efficacy and safety, which hinders the translation of combination therapies into clinical practice. Future research should delve deeply into the interaction mechanisms between the FGF19-FGFR4 axis and other targets. Moreover, liquid biopsy technologies could be used to identify relevant biomarkers and monitor potential resistance mutations dynamically, enabling timely adjustments to treatment strategies. Finally, large-scale and multi-center clinical trials should be conducted to assess the safety and clinical benefits of combining FGFR4 inhibitors with MKIs in patients with advanced HCC.

FGF19-FGFR4 AXIS IN IMMUNOTHERAPY

HCC is a common malignant tumor characterized by a highly complex and immunosuppressive TME and multiple immune evasion mechanisms[93,94]. These abnormal immune states enable tumor cells to evade host immune surveillance, promoting HCC metastasis and contributing to resistance to ICIs[95]. In recent years, immunotherapy for HCC has become a focal point of research. ICIs, by blocking immune checkpoints such as PD-1/PD-L1, restore the host's anti-tumor immune response and transform the treatment paradigm for advanced HCC[96]. In 2017, the FDA approved the PD-1-targeting ICIs, nivolumab and pembrolizumab, as second-line treatments for advanced HCC[97,98]. In 2020, the IMBRAVE-150 Phase III study (NCT03434379) demonstrated that the combination of atezolizumab (anti-PD-L1) and bevacizumab (anti-VEGF) exhibited superior efficacy in treating advanced HCC compared to sorafenib, leading to its approval as the first-line treatment[99]. Despite these encouraging results, the ORR of HCC patients to immunotherapy remains low. Given the critical role of the FGF19-FGFR4 axis in the progression and treatment of HCC, exploring the effect of FGF19 overexpression on the efficacy of immunotherapy, as well as investigating whether combining FGFR4 inhibitors with ICIs can enhance anti-tumor activity is significant.

FGF19-FGFR4 axis with TME

The TME encompasses the complex and dynamic surroundings of tumor cells, including immune cells, the vascular system, extracellular matrix, and various signaling molecules[100]. The TME plays a critical role in tumor immune evasion, drug resistance, and treatment responses.

Overexpression of FGF19 can deplete CD8+ T cells and recruit immunosuppressive cells. CD8+ T cells are considered the primary immune cells responsible for antitumor activity, as they participate in immune surveillance by recognizing and killing tumor cells[101]. The PD-1 is an inhibitory receptor primarily expressed on activated T cells, B cells, and certain natural killer (NK) cells. PD-1 is significantly upregulated on CD8+ T cells in HCC patients, where it impairs anti-tumor immune responses[102,103]. In studies by Guo et al[104], exogenous FGF19 was applied to HepG2 and Hep3B cells, resulting in a significant increase in PI3K/AKT phosphorylation and PD-L1 expression in HCC cells. This effect was attenuated upon FGFR4 knockdown, suggesting that FGF19 overexpression may suppress CD8+ T cell infiltration by regulating PD-L1 expression, thereby exerting an immunosuppressive effect. In addition to cytotoxic cells, there are various immunosuppressive cells in the TME. Tregs inhibit CD8+ T cell activity through mechanisms such as sustained expression of Cytotoxic T-Lymphocyte Antigen 4 and the secretion of immunosuppressive factors, including IL-10 and TGF-β, thus hindering anti-tumor immune responses[105,106]. Tumor-associated macrophages (TAMs) can recruit Tregs and promote angiogenesis in HCC. Myeloid-derived suppressor cells (MDSCs) inhibit the proliferation and activation of CD8+ T cells and NK cells, as well as induce Treg expansion, exhibiting notable immunosuppressive functions[107-111]. These immunosuppressive cells contribute to immune dysfunction, facilitating immune evasion by tumor cells, and play a critical role in tumor initiation and progression. In research on the E-twenty-six-specific sequence variant 4 (ETV4), Xie et al[112] found that overexpression of ETV4 increased PD-L1 Levels, inhibited CD8+ T cell infiltration, and promoted the infiltration of TAMs and MDSCs. FGF19 upregulates ETV4 expression via the FGFR4-ERK-ELK1 pathway, and overexpressed ETV4 positively feedbacks to enhance FGFR4 expression. These findings suggest that FGF19 influences the TME by modulating immune cell infiltration, thereby promoting HCC.

FGF19 overexpression can also indirectly impact the TME through various mechanisms. The aberrant FGF19-FGFR4 axis may synergize with VEGF, contributing to angiogenesis in HCC[113]. Tumor endothelial cells express glycoprotein nonmetastatic melanoma protein B, which induces exhaustion of CD8+ T cells. Under hypoxic conditions, these cells may also secrete diacylglycerol kinase gamma, promoting Treg differentiation via the ZEB2/TGF-β1 axis[114,115]. FGF19 overexpression promotes the proliferation of HCC cells and abnormal angiogenesis, which can create an aberrant blood flow state locally and lead to a hypoxic microenvironment. Over time, the persistent hypoxic microenvironment provides a foundation for the immune escape of HCC. Firstly, tumor cells can escape the surveillance and killing by cytotoxic T lymphocytes and NK cells through the degradation of granzyme B and the extracellular vesicles containing TGF-β[116]. Additionally, the accumulation of lactate in cytoplasm caused by hypoxia can weaken the activity of CD8+ T cells and induce apoptosis of NK cells[117,118]. The continuous activation of HIF can further promote immune suppression through multiple pathways, including EMT and upregulating PD-1/PD-L1 in immune cells through the HIF-1α - CD39/CD73-adenosine pathway[119]. Overall, FGF19 overexpression is closely linked to immune suppression in HCC. Targeting the FGF19-FGFR4 axis in combination with anti-tumor immunotherapy may offer a promising strategy for treating advanced HCC. However, more mechanisms by which FGF19 overexpression induces the immunosuppressive microenvironment require further investigation.

Inhibition of FGF19-FGFR4 axis and its impact on immunotherapy in HCC

Given the regulatory role of the FGF19-FGFR4 axis in the TME, researchers have begun exploring the potential of combining FGFR4 inhibition with PD-1/PD-L1 blockade for the treatment of advanced HCC. Studies have demonstrated that simultaneous FGFR4 knockdown and PD-1 monoclonal antibody administration significantly increase CD8+ T cell infiltration and serum levels of TNF-α and IFN-γ, further delaying tumor progression compared to monotherapy[104]. Moreover, combining the FGFR4 inhibitor BLU-554 with anti-PD-L1 treatment in HCC mice substantially inhibits the accumulation of TAMs and MDSCs, while enhancing the proportion of CD8+ T cells. This combination therapy not only further suppresses lung metastasis in HCC but also extends the OS of the mice compared to monotherapy[112]. Preclinical trials evaluating the selective FGFR4 inhibitor ABSK-011 have shown that targeted inhibition of the FGF19-FGFR4 pathway results in sustained enrichment of the IFN pathway and significant infiltration of CD8+ T cells in animal models, suggesting an enhanced immune response in HCC[120].

In addition to preclinical studies, several clinical trials are currently underway to explore the combination of FGFR4 inhibitors and ICIs for the treatment of advanced HCC. A Phase I/II clinical trial (NCT02325739) evaluated the safety and efficacy of combining Roblitinib with PDR001 (anti-PD-1) and found that the combination therapy significantly increased the ORR compared to Roblitinib monotherapy[52]. A Phase Ib/II trial (NCT04194801) investigating the combination of BLU-554 and the PD-L1 antibody CS1001 in patients with locally advanced or metastatic HCC demonstrated improvements in both OS and PFS compared to monotherapy[47]. In another Phase I clinical trial (NCT04906434), researchers assessed the efficacy of combining ABSK011 with various drugs, including anti-PD-(L)1, across multiple models. The study revealed that ABSK-011 exhibited significant synergistic effects, leading to inhibition of tumor progression, surpassing the effects of single-agent treatments[121].

In summary, numerous studies have demonstrated that the FGF19-FGFR4 axis plays a critical role in the TME and immune evasion mechanisms of HCC and inhibiting this pathway may improve the efficacy of immunotherapy. Figure 2 illustrates the mechanisms of aberrant FGF19-FGFR4 axis inducing immune suppression and the strategies for combination therapy. However, there are still related issues that need to be further explored in future studies, such as the relationship between FGF19 and immune hyperprogression. Hyperprogression refers to the unexpected acceleration of tumor growth during anti-tumor treatment, with particularly high incidence rates in patients receiving ICIs. Several mechanisms have been proposed to explain immune hyperprogression, including the expansion of suppressive immune cells, compensatory upregulation of immune checkpoints leading to T-cell exhaustion, and inflammation-induced oncogene activation[122]. However, whether immune hyperprogression is a natural process of immunotherapy remains controversial, and no reliable predictive biomarkers have been identified to recognize patients at risk. Previous studies have identified that amplification of MDM2/MDM4, EGFR, and genes located in the 11q13 chromosomal region (CCND1, FGF3, FGF4, FGF19) is associated with immune hyperprogression[123]. Wei et al[124] analyzed the genomic alterations associated with immune hyperprogression in 1292 Chinese patients with HCC and identified that 118 (9.13%) of the patients harbored amplifications in 11q13 region. In 2022, Zhang et al[125] investigated 30 patients with unresectable HCC who had received immunotherapy or targeted therapy. They found that 5 patients experienced hyperprogression after receiving ICIs and all of them had 11q13 amplification. They also found that among patients with 11q13 amplification, those receiving targeted therapy had significantly longer PFS compared to those receiving ICIs (P = 0.025). Wang et al[126] explored potential biomarkers for predicting immune hyperprogression. They performed whole-exome sequencing on 102 patients with HCC and found that 6% of them had FGF19 gene mutations. These studies have indicated that FGF19 may serve as a biomarker for predicting immune hyperprogression in HCC patients, and it may help guide the selection of different therapeutic approaches.

Figure 2 The aberrant fibroblast growth factor 19-fibroblast growth factor receptor 4 axis induces immune suppression. The aberrant fibroblast growth factor 19 (FGF19)-fibroblast growth factor receptor 4 (FGFR4) axis promotes IGF2BP1 expression by activating the PI3K-AKT pathway and can also increase ETV4 expression via the ERK 1/2-ELK 1 pathway. ETV4 can upregulate the expression of FGFR4, forming a positive feedback loop. Overexpression of FGF19 can promote tumor cell proliferation and angiogenesis, leading to a hypoxic microenvironment. Hypoxia-inducible factors can induce upregulation of programmed death-ligand 1 through pathways such as lactic acid, hypoxia-inducible factor-1α-CD39/CD73-adenosine. This ultimately induces immune suppression by reducing infiltration of CD8+ T cells and natural killer cells, polarization of macrophages from the M1 to M2, and increasing infiltration of MDSCs and regulatory T cells. The combination of FGFR4 inhibitors with anti-angiogenic drugs or immune checkpoint inhibitors can enhance the anti-tumor effect. FGF19: Fibroblast growth factor 19; FGFR4: Fibroblast growth factor receptor 4; KLB: Β-klotho; PI3K: Phosphoinositide 3-kinase; AKT: Protein kinase B; IGF2BP1: Insulin-like growth factor 2 mRNA-binding protein 1; RAS: Rat sarcoma; RAF: Rapidly accelerated fibrosarcoma; MEK: Mitogen-activated protein kinase; ERK: Extracellular signal-regulated kinase; ELK: ETS-like gene; ETV4: E-twenty-six-specific sequence variant 4; MCT4: Monocarboxylate transporter 4; ATP: Adenosine triphosphate; NK cell: Natural killer cell; ICI: Immune checkpoint inhibitor; MDSC: Myeloid-derived suppressor cell; Treg: Regulatory T cell. This figure was created with BioGDP.com[32].

Additionally, while combination of FGFR4 inhibitors and ICIs holds promise as a potential therapeutic approach for patients with advanced HCC, no FGFR4 inhibitor has yet been approved for clinical use. Further research is necessary to optimize this combination therapy and identify the patients most likely to benefit from it.

FGF19 AS A BIOMARKER IN HCC

HCC is a highly malignant disease marked by considerable molecular heterogeneity and complex pathogenesis, often diagnosed at advanced stages. While systemic therapies can substantially improve long-term outcomes in select patient populations, drug responsiveness remains relatively low in those with advanced HCC. AFP and DCP are commonly used serum biomarkers for the early diagnosis of HCC. However, their elevation in other benign and malignant conditions reduces their specificity[127]. Therefore, identifying new biomarkers, either individually or in combination, is crucial to enhance the sensitivity and specificity of early HCC diagnosis. Furthermore, it is essential to identify patient populations that are highly responsive to systemic therapies to optimize therapeutic efficacy. Given the strong association between the FGF19-FGFR4 axis and HCC, several studies have focused on identifying biomarkers related to this signaling pathway.

Serum FGF19 levels for early diagnosis of HCC

FGF19 has shown potential as a biomarker for the early diagnosis of HCC. Li et al[128] demonstrated that serum FGF19 Levels were significantly higher in HCC patients compared to the control group (145.57 ± 118.72 pg/mL vs 90.18 ± 13.88 pg/mL, P = 0.044. Furthermore, both FGF19 Levels and mRNA expression in tumor tissues of the HCC group were significantly higher than in paired peri-tumoral tissues. A strong positive correlation was also observed between FGF19 expression and the histological severity of liver disease. Maeda et al[129] used serum FGF19 Levels as a predictive factor for HCC diagnosis. They found that, at a cut-off value of 200 pg/mL, serum FGF19 demonstrated comparable sensitivity (53.2%), specificity (95.1%), positive predictive value (95.9%), and negative predictive value (48.7%) to AFP and DCP. Notably, the sensitivity of serum FGF19 Levels for small HCC (tumor diameter < 20 mm) was significantly higher than that of other markers (FGF19: 55.0%, AFP: 30.4%, DCP: 33.3%. Combining serum FGF19 with AFP increased sensitivity from 44.4% to 76.0%, while combining serum FGF19 with DCP increased sensitivity from 62.2% to 81.3%. The combination of all three markers further raised sensitivity from 73.7% to 87.5%. Additionally, they explored the potential of serum FGF19 in predicting treatment efficacy in HCC patients. After radiofrequency ablation, patients with serum FGF19 Levels ≥ 200 pg/mL had lower recurrence-free survival (RFS), although this result was not statistically significant (P = 0.106). Mohamed et al[130] adjusted the cut-off value for serum FGF19 and found that, at a cut-off value of 180 pg/mL, sensitivity reached 100% and specificity was 90.0%. However, this study was limited by a small sample size and a high proportion of patients with advanced HCC. In 2024, Rashad et al[131] determined that, at a cut-off value of 140.8 pg/mL, serum FGF19 effectively distinguished between the HCC and cirrhosis groups, with sensitivity and specificity of 81.8% and 87.9%, respectively. Combining serum FGF19 Levels ≥ 140.8 pg/mL with AFP ≥ 13.1 ng/mL or DCP ≥ 0.725 ng/mL increased the sensitivity for distinguishing HCC from cirrhosis to 93.9% and 90.9%, respectively, but reduced specificity to 66.7% and 45.5%. The combination of all three biomarkers increased diagnostic sensitivity to 93.9%, although specificity decreased further to 27.3%.

These studies suggest that serum FGF19 Levels hold diagnostic potential for HCC. However, due to limitations in sample size and the heterogeneity of HCC patients, no established cut-off value for HCC detection currently exists. Moreover, the diagnosis of HCC with serum FGF19 still depends on its combination with traditional biomarkers, such as AFP and DCP. While this combination improves sensitivity, it also reduces specificity. Therefore, further optimization is necessary to achieve an appropriate balance between sensitivity and specificity in clinical decision-making. Table 3 summarizes the diagnostic performance of serum FGF19 Levels alone or in combination with other biomarkers for HCC under different cut-off points.

Table 3 Diagnostic performance of serum fibroblast growth factor 19 Levels and other existing markers alone and in combination for differentiation of hepatocellular carcinoma.

Cut-off value of FGF19	Biomarker	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	AUC value
FGF19 cut-off value: 200 pg/mL[129]	FGF19 ≥ 200 pg/mL	53.2	95.1	95.9	48.7	0.795
	FGF19 ≥ 200 pg/mL and AFP ≥ 20 ng/mL	76.0	91.5	95.1	64.0
	FGF19 ≥ 200 pg/mL and DCP ≥ 40 mAU/mL	81.3	91.0	95.4	68.8
	FGF19 ≥ 200 pg/mL and AFP ≥ 20 ng/mL, DCP ≥ 40 mAU/mL	87.5	87.2	94.0	76.3
FGF19 cut-off value: 180 pg/mL[130]	FGF19 > 180 pg/mL	100	90	90	100	0.98
FGF19 cut-off value: 140.8 pg/mL[131]	FGF19 ≥ 140.8 pg/mL	81.8	87.9	87.1	82.9
	FGF19 ≥ 140.8 pg/mL and AFP ≥ 13.1 ng/mL	93.9	66.7	75.6	90.9
	FGF19 ≥ 140.8 pg/mL and DCP ≥ 0.725 ng/mL	90.9	45.5	62.5	83.3
	FGF19 ≥ 140.8 pg/mL and AFP ≥ 13.1 ng/mL, DCP ≥ 0.725 ng/mL	93.9	27.3	56.4	81.8

FGF19: Fibroblast growth factor-19; AFP: Alpha-fetoprotein; DCP: Des-γ-carboxy prothrombin; PPV: Positive predictive value; NPV: Negative predictive value; AUC: Area under the curve.

The predictive effect of FGF19 on efficacy and response to systemic therapy

In addition to the role in the early diagnosis, studies have explored the potential of FGF19 in predicting the efficacy and response to targeted-immunotherapy in HCC. Kanzaki et al[83] measured serum FGF19 Levels in 213 HCC patients before treatment with MKIs. Using a cut-off value of 200 pg/mL, they divided the patients into two groups: High FGF19 (FGF19^high) and low FGF19 (FGF19^low). In the 173 patients treated with sorafenib, they found that the PFS and OS were significantly longer in the FGF19^low group compared to the FGF19^high group (median PFS: 139 days vs 86 days, P = 0.003; median OS: 494 days vs 353 days, P = 0.039). However, no significant differences in OS and PFS were observed among the 40 patients treated with lenvatinib. Chuma et al[132] identified serum FGF19 and angiopoietin-2 (Ang-2) levels as independent predictors of PFS in HCC patients. Zhang et al[125] reported that amplification of chromosome 11q13, which includes FGF19, was associated with a significant reduction in PFS for HCC patients receiving immunotherapy. However, as 11q13 amplification also involves genes such as cyclin D1 (CCND1), FGF3, and FGF4, the impact of FGF19 in this context warrants further investigation. Cheng et al[133] utilized IHC to assess FGF19 expression in 68 HCC patients undergoing ICIs treatment. They found that FGF19 IHC-positive patients had significantly shorter median PFS compared to FGF19 IHC-positive patients (6.1 m vs 11.4 m, P < 0.05; HR = 0.45, 95%CI: 0.22-0.89). These findings suggest that FGF19 may serve as a biomarker for predicting the prognosis of HCC patients receiving targeted-immunotherapy. Future research should aim to expand sample sizes and refine patient stratification based on their treatment with MKIs and ICIs, as well as establish standardized methods for FGF19 detection.

Studies on the role of FGF19 in response to sorafenib have yielded conflicting results. Gao et al[82] reported that knockdown of FGF19 in sorafenib-resistant HCC cells reduced cell viability and increased ROS-related apoptosis, suggesting that FGF19 knockdown enhances sensitivity to sorafenib. However, Kaibori et al[134] found that among 45 HCC patients treated with sorafenib, an increased frequency of FGF19 copy number gains was observed in cases with complete response (2/6, 33.3%) compared to non-responders (2/39, 5.1%) (P = 0.024), suggesting that FGF19 copy number increase may predict a better response to sorafenib treatment. This discrepancy may be influenced by factors such as liver function, tumor stage, and the microenvironment of tumor. Research on lenvatinib response suggests that while FGF19 overexpression is associated with poor prognosis, it may also serve as a biomarker for predicting a high response. Makoto Chuma et al[132] observed that serum FGF19 Levels significantly increased in responders to lenvatinib at 4 and 8 weeks of treatment, accompanied by a decrease in Ang-2 Levels, indicating that elevated FGF19 and reduced Ang-2 could serve as early indicators of treatment response. Myojin et al[135] found that HCC cells expressing FGF19 were more sensitive to lenvatinib, but resistance developed upon FGF19 knockout. They also identified ST6GAL1, a tumor-secreted protein regulated by FGF19, as a potential marker for identifying FGF19-driven and highly malignant HCC sensitive to lenvatinib. This suggests that biomarker exploration may extend beyond FGF19 itself, including signal molecules upstream or downstream of the FGF19-FGFR4 axis, which could serve as targets for future predictive biomarker screening in HCC.

These studies suggest that FGF19 holds promise as a biomarker for the early diagnosis of HCC and for predicting the efficacy and response to systemic therapy. However, several challenges still limit its clinical application. First, no correlation was found between FGF19 amplification, FGF19 expression in tumor tissues, and serum FGF19 Levels. Additionally, combining FGF19 with traditional HCC biomarkers can enhance diagnostic sensitivity but reduce specificity. Future research may focus on integrating FGF19 with other biomarkers through various models, such as artificial intelligence (AI), to create predictive models that balance sensitivity and specificity. Furthermore, studies investigating the use of FGF19 to predict responses to systemic therapy for HCC are largely confined to laboratory settings, and further multi-center, large-sample prospective studies are required for clinical validation.

DISCUSSION AND FUTURE DIRECTIONS

The overexpression of FGF19 plays a critical role in the development of HCC. The FGF19-FGFR4 signaling pathway promotes tumor cell proliferation and inhibits apoptosis through the activation of downstream RAS-RAF-ERK, PI3K-AKT, and JAK2-STAT3 pathways. This complex role of FGF19 in HCC provides a theoretical foundation for targeting the FGF19-FGFR4 axis as a novel therapeutic strategy for advanced HCC. Currently, FGFR4 inhibitors, such as BLU-9931, H3B-6527, BLU-554, FGF401, and ABSK-011, achieve high selectivity for FGFR4 by targeting kinase domains (e.g., Cys552) and preclinical studies have shown significant tumor regression. In clinical translation, phase I clinical trial of ABSK-011 has reported an ORR of 43.5%, which increased to 55.6% (220 mg, BID) when combined with the PD-L1 inhibitor atezolizumab, suggesting potential synergy in reshaping the immune microenvironment. However, the clinical application of FGFR4 inhibitors is limited by resistance mutations (e.g., V550M/L) and BA metabolism-related toxicity, necessitating optimization of efficacy through combination with EGFR inhibitors or strategies to address toxicity.

The FGF19-FGFR4 axis exerts multi-dimensional regulatory effects in systemic therapy, with its overexpression counteracting the pro-apoptotic effects of sorafenib by reducing oxidative stress. Lenvatinib enhances ferroptosis and PD-L1 degradation by inhibiting FGFR4, significantly improving its anti-tumor efficacy. Meanwhile, FGF19 overexpression upregulates PD-L1, recruits suppressive immune cells and may induce hypoxia-related immune evasion, thereby promoting an immunosuppressive microenvironment. Additionally, FGF19 is not only a therapeutic target for HCC but also a potential early diagnostic biomarker for predicting the efficacy and response to targeted-immunotherapy. However, its clinical application faces challenges, including the lack of standardized detection methods, insufficient validation, and an imbalance between sensitivity and specificity.

Although the relationship between the FGF19-FGFR4 axis and HCC has been extensively studied, certain limitations also remain. First, the mechanisms through which FGF19 overexpression promotes HCC progression require further investigation. Studies have demonstrated that the FGF19-FGFR4 axis can inhibit liver fibrosis by downregulating the activation of hepatic stellate cells (HSCs), potentially serving as a protective mechanism to delay HCC progression[136]. Zhou et al[137] discovered that liver progenitor-like cells secrete amino-regulators, which synergize with FGF19 in anti-fibrosis. These factors activate the JAK-STAT1 pathway to induce HSCs apoptosis, thereby exerting anti-fibrotic effects and offering potential therapeutic targets for liver cirrhosis. This suggests that FGF19 may play a dual role in regulating HCC progression. Additionally, there is a lack of standardized methods for detecting FGF19, and the heterogeneity of detection criteria may potentially influence the outcomes of clinical trials. Although the relationship between serum FGF19 Levels and HCC has been extensively studied, it can be influenced by factors such as BA levels or the patient's feeding status. The level of IHC positivity was not correlated with genomic amplification of the FGF19 Locus, which could relate to a limited dynamic range of the assay or heterogeneity of FGF19 expression across different tumor lesions[45]. Therefore, establishing a standardized FGF19 detection criteria is essential for its clinical application as a predictive biomarker.

These following issues may become directions for future research. First, the FGF19-FGFR4 axis is a critical regulator of BA metabolism in body. FGF19 overexpression can reduce BA synthesis by inhibiting CYP7A1 activity. Native FGF19 or its analogs have shown efficacy in the treatment of cholestatic diseases such as primary biliary or sclerosing cholangitis[138]. Zhou et al[137] have developed FGF19 analogs that can ameliorate BA toxicity while reducing its promotional effects on HCC. Different BA types have distinct effects on HCC development. For instance, deoxycholic acid can induce HSCs senescence and exhibit behaviors characteristic of liver malignancy, while lithocholic acid (LCA) disrupts liver lipid homeostasis and induces cholestatic injury[139,140]. Both of them promote the development of HCC. In contrast, ursodeoxycholic acid can induce tumor cell apoptosis and inhibit abnormal angiogenesis, thus suppressing HCC progression[141]. Future studies may focus on developing FGF19 analogs with BA subtype selectivity, identifying the relevant BA subtypes and inhibiting the synthesis of BAs that promote the progression of HCC. Furthermore, under the context of rapid development in computer science and AI, it has become possible to use AI to screen for predictive biomarkers related to FGF19 in HCC. AI has been proposed for application in the clinical management of HCC, including improving the prediction of future HCC risk in patients with chronic liver disease, improving the accuracy of HCC diagnosis in patients undergoing surveillance imaging or liver biopsies and improving prognostication in patients with established HCC[142]. Zeng et al[143] developed a model of atezolizumab-bevacizumab response signature prediction, which leverages AI to predict the sensitivity of HCC patients to targeted-immunotherapy. Hsu et al[144] measured serum biomarkers, including AFP, albumin-bilirubin grade, FGF19, FGF21, and used AI to predict the efficacy of lenvatinib in unresectable HCC. Although their study did not find a predictive effect of baseline serum FGF19 Levels, it provided direction for using AI to assess the predictive role of FGF19 in HCC.

CONCLUSION

In summary, our review discusses the mechanisms by which the aberrant FGF19-FGFR4 axis contributes to HCC, the research on FGFR4 inhibitors and their limitations, the role of FGF19 in targeted-immunotherapy, and the potential of FGF19 as a biomarker for early diagnosis and treatment efficacy prediction in HCC. Challenges such as the complex interplay between FGF19 and the TME, the absence of standardized detection criteria, drug resistance and AEs associated with FGFR4 inhibitors hinder the clinical translation. However, the FGF19-FGFR4 axis remains a promising target for precision therapy in HCC. Possible future research directions include the development of functionally selective FGF19 variants, the screening of BA subtype-specific inhibitors, or the development of an AI-based FGF19 predictive biomarker. These approaches may provide potential solutions to overcome the current treatment bottlenecks.

ACKNOWLEDGEMENTS

We sincerely thank our colleagues for their valuable suggestions and technical assistance for this research.