Conclusion and Perspectives

Challenges in Targeting the FGF/FGFR Signaling

The FGF/FGFR signaling functions as a highly conserved, context‐dependent regulatory network with profound pleiotropy. Dysregulation of FGF/FGFR signaling, whether resulting from pathological hyperactivation or loss‐of‐function, is implicated in multiple diseases such as skeletal dysplasia, metabolic disorders, cancers, and neurological disorders. Despite the deepening understanding of the FGF/FGFR signaling, targeting the FGF/FGFR signaling still faces several challenges.

Structural Biology and Drug Design

In recent years, continuous breakthroughs in structural biology, drug design, and related fields have presented both profound opportunities for innovation and significant challenges in FGF/FGFR signaling pathway research [480]. With the elucidation of key structures such as the FGF23–FGFR–Klotho–HS quaternary complex [29], future efforts can leverage structural biological insights integrated with artificial intelligence (AI)‐driven virtual screening and molecular design to develop next‐generation FGFR inhibitors or FGF analogs with improved subtype selectivity and broader coverage of resistance mutations. Acquired resistance mutations, particularly those affecting gatekeeper residues (e.g., V565 in FGFR2, V550 in FGFR4) and molecular brake residues (e.g., N550), represent core barriers that limit the durability of therapeutic efficacy [511]. There is a pressing need for conformationally selective inhibitors targeting specific subtypes (e.g., FGFR2, FGFR4) or those specifically designed to overcome acquired resistance mutations (e.g., FGFR2 V565F, FGFR4 V550L). Furthermore, structure‐based engineering of FGF analogs (e.g., attenuating mitogenic activity while preserving metabolic functions) remains a critical strategy to address the safety bottleneck associated with endocrine FGF analogs.

Accordingly, the synergistic integration of high‐resolution structural biology and AI represents a pivotal strategy to address the above bottlenecks. For instance, high‐resolution x‐ray crystal structures of FGFR have established a structural framework for rational inhibitor design targeting distinct functional states. Meanwhile, AI‐driven protein structure prediction (e.g., AlphaFold2, RoseTTAFold) can perform virtual screening with demonstrated efficacy for kinase targets. Collectively, the integration of these cutting‐edge approaches will facilitate the development of next‐generation FGFR inhibitors or FGF analogs with robust activity and enhanced subtype selectivity.

Metabolic Regulation

Furthermore, building on the “threshold model” and “spatiotemporal specificity” principles established in FGF‐mediated metabolic regulation, recent work has proposed a novel concept of dose‐dependent critical points in FGF efficacy, offering important insights into the precise delivery, dosage optimization, and temporal control of future therapeutics [49]. Beyond endocrine FGF family‐dominated paradigm of metabolic regulation, recent studies have revealed that paracrine FGF4 and FGF1 also participate in metabolic control: Upon binding to FGFR1c, they activate the Ca2+/AMPK signaling axis, thereby directly promoting glucose uptake in skeletal muscle and adipose tissue to lower blood glucose levels [512]. This finding challenges the conventional view that endocrine FGFs serve as the primary mediator of metabolic function, thus providing a new perspective for the treatment of diabetes and metabolic disorders [154, 512]. Critically, the FGF/FGFR signaling system does not operate in isolation; instead, it is embedded within an intricate regulatory network involving multiple important signaling pathways, including EGFR signaling and gut microbiota metabolism. This crosstalk sustains tissue homeostasis under physiological conditions, whereas in pathological contexts, it contributes to disease initiation, progression, and drug resistance. Therefore, comprehensive mapping of these contextual interactions is indispensable for developing rational, mechanism‐informed combination strategies and predictive biomarker signatures for FGF/FGFR‐targeted therapeutics.

Gut Microbiota

In recent years, accumulating evidence has revealed the bidirectional regulatory effects between FGF/FGFR signaling and the gut microbiota along its metabolites, showing its integral role in systemic metabolic regulation, barrier integrity, and immune homeostasis. A representative example is the BA‐FXR‐FGF19 enterohepatic axis: The gut microbiota converts hepatocyte‐derived primary BAs into secondary BAs, which then activate FXR signaling in ileal epithelial cells to induce the expression of FGF19 [513]. Upon portal venous transport to the liver, FGF19 binds to FGFR4/β‐Klotho complex and represses the transcription of CYP7A1, the rate‐limiting enzyme in BA synthesis, thereby maintaining BA homeostasis [514]. Beyond BAs, short‐chain fatty acids (SCFAs), the principal products of dietary fiber fermentation by the gut microbiota (e.g., acetate, propionate, and butyrate), transduce signals through G protein‐coupled receptors (GPR)‐41 and GPR43 [515]. Notably, butyrate can indirectly modulate the transcriptional levels of FGF ligands or receptors by inhibiting histone deacetylases (HDACs). Moreover, butyrate reinforces intestinal barrier integrity and modulates local immune responses, and butyrate furnishes a favorable microenvironment for FGF‐mediated tissue repair [516]. Under dysbiotic conditions, aberrant expression of FGF19/15 can ensue, contributing to the pathogenesis of metabolic disorders such as obesity, Type 2 diabetes, and metabolic dysfunction‐associated steatotic liver disease (MASLD) [517, 518]. Collectively, these findings underscore a complex and intimate interplay between FGF/FGFR signaling and the gut microbiota, which constitutes a sophisticated bidirectional feedback network.

Nevertheless, our understanding of this network remains nascent. For instance, the precise molecular mechanisms by which SCFAs regulate FGF/FGFR signaling are still poorly defined, and the tissue‐specific differential regulation of FGF/FGFR signaling by microbial metabolites awaits further elucidation. Moreover, the gut microbiota exhibits substantial inter‐individual variability, which poses considerable challenges for precise interventions targeting the microbiota–FGF/FGFR signaling axis. Given that different individuals can respond markedly differently to the same intervention, developing individualized regulatory strategies targeting FGF/FGFR signaling and gut microbiota will be a critical direction for future research. The above limitations suggest that a multi‐omics integration strategy is required to deeply dissect the interaction between gut microbiota and FGF signaling. Combining metagenomics, metabolomics, and transcriptomics, along with longitudinal sampling analysis, can help distinguish causal mechanisms underlying correlations. These strategies are expected to provide feasible technical routes for precision interventions targeting the gut microbiota‐FGF signaling axis.

Liquid–Liquid Phase Separation

In recent years, LLPS has emerged as a fundamental biophysical mechanism governing spatiotemporal regulation of FGF signaling across multiple subcellular and physiological compartments. At the plasma membrane, FGF2 can undergo LLPS with HS to form condensates that augment FGFR interaction and ERK activation, ultimately regulating cell proliferation and differentiation [519]. Notably, high concentrations of HS can inhibit this process, suggesting that HS influences the intensity of FGF signaling by regulating LLPS efficiency. Furthermore, within BMSCs, nuclear FGF2 employs LLPS to control rRNA transcription and cell proliferation, a regulatory axis that further shapes rDNA chromatin architecture through cooperative interaction with STAT5 [520]. In addition, FGFBP1 itself is also directly involved in the LLPS process. In acute liver injury, FGF6 secreted by skeletal muscle affects the efficiency of its release from the cell membrane by regulating the LLPS dynamics of FGFBP1; the released FGFBP1 forms co‐condensates with FGF5 in the liver through LLPS, activates hepatocyte proliferation signals, and drives liver regeneration [46]. Collectively, the aforementioned findings indicate that LLPS is the core mechanism by which FGF signaling exerts its functions at the extracellular, nuclear, and inter‐organ levels.

Limitations of Current Therapeutic Strategies

Despite substantial advances in therapeutic strategies targeting the FGF/FGFR signaling, several key limitations of current approaches impede their clinical translation and widespread application. This section provides a systematic, mechanism‐informed analysis of these limitations of current therapeutic strategies from three perspectives: drug delivery and tissue selectivity, FGFR‐targeted CAR‐T‐cell therapy, and clinical detection with stratified treatment.

Drug Delivery and Tissue Selectivity

The clinical translation of FGF/FGFR‐targeted biologics and small‐molecule therapeutics is significantly constrained by pharmacokinetic and biodistribution limitations. The short half‐life of recombinant FGF proteins (necessitating frequent administration) and the poor tissue distribution selectivity of small‐molecule inhibitors collectively limit therapeutic efficacy [521, 522]. The therapeutic outcome of FGF/FGFR signaling intervention is highly dependent on achieving adequate drug concentrations and sustained exposure in target tissues. For indications requiring FGF/FGFR signaling activation, recombinant FGF proteins are rapidly degraded by proteases following local application, resulting in short‐lived effects and necessitating frequent dosing, whereas systemic administration carries the risk of off‐target effects. For indications requiring FGF/FGFR signaling inhibition, although small‐molecule inhibitors are amenable to oral or intravenous delivery, they often fail to adequately penetrate the blood–brain barrier to reach central nervous system tumors or achieve sufficient selective accumulation in tumor tissues.

To address these pharmacokinetic and biodistribution barriers, future research should prioritize the development of targeted delivery systems based on nanoparticles, hydrogels, exosomes, or in situ sustained‐release technologies. For instance, local delivery of FGF proteins for tissue repairs from biomaterial encapsulation extends their duration of action, whereas in tumor therapy, active targeting strategies, such as ADCs directed against tumor‐associated FGFR isoforms or BBB‐penetrating nanocarriers, may be employed to deliver FGFR inhibitors selectively to tumor tissues or across the blood–brain barrier. Critically, spatiotemporally precise regulation holds promise for enhancing therapeutic efficacy while substantially reducing off‐target toxicity.

FGFR‐Targeted CAR‐T‐Cell Therapy

The integration of FGFR‐targeting strategies and CAR‐T‐cell therapy represents a highly promising direction in tumor immunotherapy. Unlike conventional FGFR small‐molecule inhibitors, FGFR‐targeted CAR‐T cells function as living, self‐replicating therapeutics, enabling high‐affinity, MHC‐independent binding to FGFR proteins on tumor cell surfaces and triggering potent, serial target cell lysis [452]. To date, the development of FGFR‐targeted CAR‐T‐cell therapy has been almost exclusively focused on the FGFR4, with its primary indication being pediatric RMS [452]. In 2025, the US National Cancer Institute (NCI) launched the world's first Phase I clinical trial of FGFR4‐targeted CAR‐T‐cell therapy (NCT06865664). However, FGFR4‐monospecific CAR‐T cells display suboptimal efficacy against aggressive tumor subclones with low FGFR4 expression. To address this limitation, dual‐targeting CAR‐T has emerged as a leading optimization strategy [357]. By co‐expressing two independent CARs targeting FGFR4 and CD276 on the same T cell, this approach effectively circumvents the challenge of heterogeneous tumor antigen expression [357].

Despite its conceptual promise, the clinical application of FGFR‐CAR‐T‐cell therapy in solid tumors, particularly RMS, still faces significant hurdles, including intratumoral antigen heterogeneity (a challenge shared with small‐molecule inhibitors) and T‐cell exhaustion. Beyond RMS, the integration of FGFR targeting with CAR‐T‐cell technology holds potential for expansion to other FGFR‐driven solid malignancies, such as HCC and CRC. To overcome T‐cell dysfunction, rational combination strategies with ICIs or cytokines designed to rejuvenate exhausted T cells may further enhance efficacy.

Clinical Detection and Stratified Treatment

Accurate identification of patients whose diseases are molecularly driven by aberrant FGF/FGFR signaling is a prerequisite for realizing the therapeutic potential of targeted interventions. This necessitates robust, clinically actionable biomarker frameworks that integrate genomic, transcriptomic, and functional readouts to distinguish true oncogenic dependency from passenger alterations. To date, no standardized clinical assay exists to reliably quantify FGF/FGFR signaling dependence in routine practice. Future efforts should focus on establishing unified protocols for the detection of FGFR1–4 gene mutations, amplifications, fusions, and expression profiles, while concurrently developing dynamic monitoring technologies based on liquid biopsies (e.g., circulating tumor DNA and exosomal RNA). Moreover, exploring non‐FGFR biomarkers, such as Klotho expression levels, FGFBP concentrations, or downstream signaling pathway activity scores, will enable multidimensional identification of populations likely to benefit, thereby maximizing therapeutic index and minimizing off‐target toxicity.

Future Perspectives

The FGF/FGFR signaling system remains a high‐priority frontier in both basic and clinical research. Looking ahead, addressing the multifaceted challenges above will not come from any single approach; it will require integrating multiple complementary strategies. First, structure‐guided drug design, combined with AI‐driven lead compound discovery, will accelerate the development of next‐generation inhibitors with better selectivity and broader coverage against resistance mutations. Second, nanotechnology‐mediated targeted delivery systems, including nanoparticles, hydrogels, and exosomes, offer solutions to the short half‐life and poor tissue distribution of current therapeutics. Third, FGFR‐targeted CAR‐T‐cell therapy, though still in its early stages, represents a paradigm shift in immunotherapy for FGFR‐driven malignancies. Fourth, the establishment of standardized, regulatory‐grade biomarker assays for detecting FGF/FGFR signaling, including liquid biopsy‐based dynamic monitoring and panels of non‐FGFR biomarkers, will make true stratified treatment a reality. Finally, mechanism‐based combination regimens that co‐target the FGF/FGFR axis along with complementary pathways such as EGFR, immune checkpoints, or gut microbiota‐derived metabolites hold promise for achieving synergistic effects while keeping resistance in check.

Collectively, it is anticipated that the convergence of these multidisciplinary approaches will enable precise, context‐dependent modulation of FGF/FGFR signaling, ultimately providing safer and more effective therapeutic options for patients with aberrant FGF/FGFR signaling‐related diseases.