2024 Volume 31 Issue 11 Pages 1496-1511
Despite advancements in managing traditional cardiovascular risk factors, many cardiovascular diseases (CVDs) persist. Fibroblast growth factors (FGFs) have emerged as potential diagnostic markers and therapeutic targets for CVDs. FGF1, FGF2, and FGF4 are primarily used for therapeutic angiogenesis. Clinical applications are being explored based on animal studies using approaches such as recombinant protein administration and adenovirus-mediated gene delivery, targeting patients with coronary artery disease and lower extremity arterial disease. Although promising results have been observed in animal models and early-stage clinical trials, further studies are required to assess their therapeutic potential. The FGF19 subfamily, consisting of FGF19, FGF21, and FGF23, act via endocrine signaling in various organs. FGF19, primarily expressed in the small intestine, plays important roles in glucose, lipid, and bile acid metabolism and has therapeutic potential for metabolic disorders. FGF21, found in various tissues, improves glucose metabolism and insulin sensitivity, suggesting potential for treating obesity and diabetes. FGF23, primarily secreted by osteocytes, regulates vitamin D and phosphate metabolism and serves as an important biomarker for chronic kidney disease and CVDs. Thus, FGFs holds promise for both therapeutic and diagnostic applications in metabolic and cardiovascular diseases. Understanding the mechanisms of FGF may pave the way for novel strategies to prevent and manage CVDs, potentially addressing the limitations of current treatments. This review explores the roles of FGF1, FGF2, FGF4, and the FGF19 subfamily in maintaining cardiovascular health. Further research and clinical trials are crucial to fully understand the therapeutic potential of FGFs in managing cardiovascular health.
Cardiovascular diseases (CVDs) are the leading cause of mortality worldwide. In 2019, CVDs were responsible for an estimated 18.6 million deaths, accounting for 32% of all global deaths1). Major risk factors for CVDs include dyslipidemia, hypertension, diabetes mellitus, obesity, and smoking. Among these risk factors, low-density lipoprotein cholesterol (LDL-C) has emerged as a suitable target for the treatment and prevention of CVDs. Numerous lipid-lowering drugs have been developed for clinical use, with statins playing a major role in the primary and secondary prevention of CVDs. Statin administration has been shown to reduce the prevalence of cardiovascular events by 30–40%. However, despite their efficacy, a substantial number of cardiovascular events remain unprevented, thereby imposing a substantial burden on healthcare systems worldwide.
Recently, fibroblast growth factors (FGFs) have emerged as promising novel molecules with potential involvement in cardiovascular development and homeostasis. The human FGF family comprises 22 members, including 18 secreted signaling proteins and four intracellular proteins. Secreted FGFs bind to and activate the cell surface tyrosine kinase FGF receptors (FGFRs) that are encoded by four genes (FGFR1–4)2). In contrast, intracellular FGFs act as cofactors for voltage-gated sodium channels and other molecules3). The FGF family plays crucial roles in maintaining cardiovascular health and exerts its influence during embryonic development, organogenesis, cell growth and survival, angiogenesis, tissue repair and regeneration, metabolic function, and homeostasis maintenance.
The present review aims to highlight the potential of FGFs, specifically FGF1, FGF2, FGF4, and the FGF19 subfamily, as markers for cardiovascular risk assessment and therapeutic targets for CVD treatment. By exploring the intricate roles of FGFs in cardiovascular health and disease, we may pave the way for novel strategies in CVD prevention and management, ultimately addressing the substantial burden that remains despite current therapies.
Structure and Classification of FGF FamilyThe FGF family comprises structurally related growth factors that exert various biological effects by interacting with specific cell surface receptors. The FGF family includes 22 members, divided into seven subfamilies based on sequence homology and receptor specificity (Fig.1)3). Canonical FGFs are characterized by their conserved core domain of approximately 140 amino acids, essential for receptor binding4). Variations within the FGF family arise from differences in their receptor-binding specificity, leading to distinct physiological roles in various tissues and organ systems.
The figure illustrates the classification of the fibroblast growth factor (FGF) family into three main functional groups: paracrine FGFs, endocrine FGFs and intracellular FGFs.
Paracrine FGFs (Yellow): These FGFs function primarily through local cell signaling and are categorized into five subfamilies: FGF1 subfamily includes FGF1 and FGF2; FGF4 subfamily includes FGF4, FGF5, and FGF6; FGF7 subfamily includes FGF3, FGF7, FGF10, and FGF22; FGF8 subfamily includes FGF8, FGF17, and FGF18; FGF9 subfamily includes FGF9, FGF16, and FGF20.
Endocrine FGFs (Red): These FGFs are involved in systemic signaling and are part of the FGF19 subfamily. This function group includes FGF19, FGF21, and FGF23.
Intracellular FGFs (Blue): These FGFs function within the cell and are part of FGF11 subfamily. This functional group includes FGF11, FGF12, FGF13, and FGF14.
FGFs exert their biological effects by binding to a family of high-affinity cell surface receptors known as FGFRs. The FGFR family comprises four members (FGFR1–4) with distinct tissue distributions and specificities for different FGF ligands3). FGF binding induces FGFR dimerization and autophosphorylation of intracellular tyrosine residues, initiating downstream signaling cascades5).
The major intracellular signaling pathways activated by FGFRs include mitogen-activated protein kinase, phosphoinositide 3-kinase/protein kinase B (PI3K/AKT), and signal transducer and activator of transcription pathways4). These pathways regulate diverse cellular processes, such as proliferation, differentiation, survival, and migration, making FGF signaling critical for embryonic development and tissue homeostasis (Table 1)6-19).
| FGF sub family | FGF | Cofactor | Receptor specificity | Physiological role | Ref. | |
|---|---|---|---|---|---|---|
| FGF1 sub family | Paracrine FGFs | FGF1 (aFGF) | heparin, heparin sulfate | All FGFRs | cellular proliferation, survival, metabolism, morphogenesis, differentiation, embryonic development, angiogenesis, tissue repair, and regeneration | 6 |
| FGF2 (bFGF) | FGFR 1c, 3c > 2c, 1b, 4 | |||||
| FGF4 sub family | FGF4 | FGFR 1c, 2c > 3c, 4 | Cardiac valve leaflet formation, Limb development | 7 | ||
| FGF5 | Hair growth | 8 | ||||
| FGF6 | Myogenesis | 9 | ||||
| FGF7 sub family | FGF3 | FGFR 2b > 1b | Inner ear development | 10 | ||
| FGF7 (KGF) | Branching morphogenesis | 11 | ||||
| FGF10 | Branching morphogenesis | 12 | ||||
| FGF22 | Presynaptic neural organizer | 13 | ||||
| FGF8 sub family | FGF8 | FGFR 3c > 4 > 2c > 1c >>3b | Brain, eye, ear and limb development | 14 | ||
| FGF17 | Cerebral and cerebellar development | 15 | ||||
| FGF18 | Bone development | 16 | ||||
| FGF9 sub family | FGF9 | FGFR 3c > 2c > 1c, 3b >> 4 | Gonadal development, Organogenesis | 17 | ||
| FGF16 | Heart development | 18 | ||||
| FGF20 | Neurotrophic factor | 19 | ||||
| FGF19 sub family | Endocrine FGFs | FGF19/15 | βKlotho | FGFR 1c, 2c, 3c, 4 | Bile acid homeostasis, Lipolysis, Gall bladder filling | 42, 43 |
| FGF21 | Fasting response, Glucose homeostasis, Lipolysis and Lipogogenesis | 54-57 | ||||
| FGF23 | αKlotho | Phosphate homeostasis, Vitamin D homeostasis | 72-77 | |||
| FGF11 sub family | Intracellular FGFs | FGF11 | none?? | no known activity | not established | |
| FGF12 | ||||||
| FGF13 | ||||||
| FGF14 | ||||||
The FGF family consists of a variety of proteins that serve multiple biological functions. These proteins are essential for processes, such as cell growth, differentiation, migration, and tissue repair and regeneration. Similar to other growth factors, FGFs bind to specific cell surface receptors, known as FGF receptors (FGFR1–4)
FGF, isolated from the brain and pituitary extracts, was found to be involved in fibroblast proliferation, as revealed by bioassays20, 21). Furthermore, when the same extracts were fractionated, acidic FGF (FGF1) and basic FGF (FGF2) were obtained. FGF1 regulates and enhances the angiogenesis by triggering the migration, proliferation, and differentiation of endothelial cells22), ultimately leading to the development of new capillaries from existing vessels.
Following successful experimental studies, the effectiveness of the combined therapy with coronary artery bypass grafting and FGF1 has been reported in clinical practice. In a study involving 20 patients with three-vessel coronary artery disease, intramyocardial injection of FGF1 near the left anterior descending coronary artery during coronary artery bypass surgery improved collateral circulation and capillary proliferation23). Additionally, the FGF1-treated group exhibited increased blood flow compared to the control group, along with improved cardiac function and New York Heart Association classification after three years of follow-up24). Furthermore, the positive effects of FGF1 on peripheral blood flow have been demonstrated in clinical trials. In a phase I open-label trial, intramuscular injection of naked plasmid DNA encoding FGF1 (NV1FGF) into the lower leg improved perfusion in 51 patients with critical lower extremity arterial disease (LEAD)25). A subsequent, a phase II double-blind, randomized, placebo-controlled trial revealed that intramuscular injections of NV1FGF in 125 patients with LEAD, presenting with a non-healing ulcer and who were unsuitable for revascularization, led to a 50% reduction in amputation rates and a trend toward reduced mortality26). However, the TAMARIS trial, a large-scale, placebo-controlled, randomized phase III trial involving 525 patients with critical limb ischemia from 171 sites across 30 countries, provided no evidence that neovascularization therapy using NV1FGF was effective in reducing lower leg amputations or death27).
FGF2The angiogenic effects of FGF2 are also widely recognized. FGF2 promotes the migration and growth of endothelial cells in vivo28). It also stimulates mitogenesis of smooth muscle cells and fibroblasts, contributing to the development of collateral vessels29). In a phase I, randomized, dose-escalation trial involving 25 patients with coronary artery disease and stable angina, the administration of FGF2 reduced the ischemic area in the myocardium, improved exercise performance, and decreased angina frequency30). Subsequently, the FGF Initiating RevaScularization Trial (FIRST) was conducted on 337 patients with coronary artery disease to further evaluate the efficacy and safety of recombinant FGF2. The FIRST, a multicenter, randomized, double-blind, placebo-controlled trial of demonstrated that single intracoronary infusion of FGF2, led to beneficial effects in the initial months; however, these effects were not sustained31). In a randomized, double-blind, placebo-controlled study, surgical delivery of heparin beads with absorbed FGF2, implanted into ischemic and viable but ungraftable myocardial regions in 24 patients, reduced ischemic size and angina symptoms for three years32). Additionally, a double-blind, placebo-controlled, dose-escalation trial conducted on 19 patients with LEAD demonstrated that patients who received arterial FGF2 exhibited improved calf blood flow compared to those who received a placebo33). The therapeutic angiogenesis with recombinant FGF2 for intermittent claudication (TRAFFIC) trial investigated the effects of intra-arterial recombinant FGF2 on the exercise capacity of patients with intermittent claudication. In this randomized trial, 190 patients received either a placebo, a single dose, or a double dose of recombinant FGF2 (rFGF2). The primary outcome was the change in the peak walking time at 90 days. The single-dose group showed a significant improvement compared to the placebo group, but the double-dose group did not exhibit any additional benefit. The safety profiles were similar across all groups, supporting the potential of rFGF2 in therapeutic angiogenesis34). Overall, angiogenic therapies using FGF2 have shown promising results in both basic research and clinical trials, and FGF2 is expected to play a crucial role in the treatment of various diseases.
FGF4FGF4, a 206-amino-acid protein initially isolated from a human gastric tumor, shares approximately 40% homology with FGF1 and FGF2 35). Although FGF2 has primarily been studied using recombinant protein preparations, FGF4’s robust effects, stability, targeted delivery, promising clinical outcomes, and favorable safety profiles have made it a suitable candidate for gene therapy applications. The AGENT trial evaluated the efficacy and safety of angiogenic gene therapy using Ad5-FGF4 (a serotype 5 adenovirus encoding FGF4) in 79 patients with stable chronic angina36). Overall, the trial did not find a significant difference in exercise time between the treatment and placebo groups. However, a post hoc analysis excluding patients with a baseline exercise time exceeding 10 min revealed a significant improvement in exercise capacity in the treated group compared to that in the placebo group at both 4 and 12 weeks. Follow-up trials (AGENT-2, AGENT-3, and AGENT-4) conducted to explore the safety and efficacy of this therapy did not show significant overall benefits from the treatments itself37, 38). However, a subsequent analysis suggested a potential sex-specific benefit, with women appearing to respond more favorably.
An ongoing phase III open-label trial, ASPIRE, is assessing the efficacy of intracoronary adenoviral FGF4 compared to standard care using myocardial perfusion imaging in 100 patients with coronary artery disease39). Additionally, the AWARE phase III, randomized, double-blind, placebo-controlled trial is planned to evaluate the safety and efficacy of intracoronary adenoviral FGF4 delivery, specifically in women with stable angina pectoris.
The AFFIRM study, a phase III clinical trial, is also ongoing to determine whether a single intracoronary infusion of Ad5-FGF4 is effective in improving angina-limited exercise duration, angina functional class, frequency of angina attacks, frequency of nitroglycerin usage, and quality of life in patients with refractory angina.
The optimal strategy for administering angiogenic growth factors to stimulate therapeutic angiogenesis is also crucial. Intravenous, intracoronary, and intramyocardial routes of administration have been explored. Controlled and sustained delivery of angiogenic factors is critical for achieving stable and functional neovascularization, as the process of tissue regeneration and angiogenesis takes weeks or months40).
FGF19 Subfamily FGF19FGF19 was initially identified in the human brain during embryonic and fetal development in 1999. Its rodent counterpart, FGF15, is found in the central nervous system and promotes neuronal maturation. Numerous animal studies have validated the therapeutic potential of FGF19 in the managing metabolic disorders, such as obesity and diabetes41).
FGF19 plays an important role in various metabolic processes, including the regulation of glucose, lipid, and bile acid metabolism, as well as gall bladder relaxation and filling. FGF19 is highly expressed in the small intestine and is secreted into the enterohepatic circulation postprandially, stimulated by bile acids through the activation of the farnesoid X receptor (FXR). It acts on the FGFR/βKlotho complex expressed in the liver42). FGF19 negatively regulates bile acid synthesis by suppressing the expression of CYP7A1, a key enzyme in bile acid synthesis. The expression of FGF19 is regulated by the nuclear receptor FXR, a ligand for bile acids43). Bile acids are secreted into the intestine during feeding and play crucial roles in the absorption of lipid components. Similarly, the expression of FGF19 increases during feeding and has been reported to regulate hepatic refeeding responses, including the inhibition of gluconeogenesis and the promotion of glycogen and protein synthesis (Fig.2)44). Bile acids play a crucial role in cholesterol excretion, helping to regulate the body’s cholesterol balance and prevent excessive accumulation of cholesterol. Therefore, bile acids are critical in cholesterol metabolism and contribute to the prevention of CVDs.
FGF19 is primarily synthesized and secreted by the small intestine in response to bile acid stimulation. Bile acids in the intestine activates the farnesoid X receptor (FXR), inducing the expression of FGF19.
In an experimental study, NGM282, an engineered FGF19 analog, enhanced high-density lipoprotein (HDL) formation and cholesterol removal from the liver by modulating LXR signaling through ABCA1 and FGFR4 45). The study also demonstrated that the administration of an HMG-CoA reductase inhibitor or an antibody against proprotein convertase subtilisin/kexin type 9 eliminated the increase in total cholesterol, HDL cholesterol (HDL-C), and LDL-C caused by FGF19 in db/db mice. Additionally, in Apoe-/- mice on a Western diet, NGM282 treatment significantly decreased atherosclerotic lesions in the aorta. In a study involving 1,166 Chinese patients with coronary artery disease, the combined measurement of cardiometabolic biomarkers, including adipocyte fatty acid-binding protein, lipocalin-2, and FGF19, along with conventional risk factors improved the prediction of cardiovascular events in patients with stable coronary artery disease46). Rats with streptozotocin-induced diabetes demonstrated reduced serum FGF19 levels and lower FGF19 protein expression compared to control rats. Recombinant FGF19 administration via intraperitoneal injection in these rats decreased plasma glucose, plasma total cholesterol, plasma LDL-C, and serum malondialdehyde levels while increasing serum adiponectin and plasma HDL-C levels compared to that in control rats. Furthermore, FGF19 administration increased the protein levels of the glucose transporter GLUT4 and pyruvate dehydrogenase E1-alpha compared to those in control rats, suggesting that FGF19 improves energy metabolism by increasing glucose uptake and decreasing lipid levels47).
The prevalence of non-alcoholic fatty liver disease (NAFLD) is increasing in Western countries48). Patients with NAFLD not only face an increased risk of non-alcoholic steatohepatitis (NASH), but also often have comorbidities, such as obesity, diabetes mellitus, hypertension, and dyslipidemia, all of which heighten their CVD risk. FGF19 regulates bile acid metabolism and exerts insulin-like effects on glycogen synthesis and gluconeogenesis, potentially impacting multiple pathways involved in NASH development.
A multicenter, randomized, double-blind, placebo-controlled phase II showed that NGM282 administration significantly reduced fat content in patients without causing severe adverse events49). However, a subsequent randomized, double-blind, placebo-controlled phase IIb study with a larger cohort showed positive effects on many secondary endpoints but did not demonstrate improvement in liver fibrosis, the primary endpoint50). Consequently, the program was terminated.
Recent studies suggest that a communication exists between the gut and myocardium, potentially regulating cardiac hypertrophy51). The gut and heart have been shown to communicate through incretins produced by the gut after meals, as well as bile acids. Both incretins and bile acids target the myocardium and affect the cardiac function52). In a clinical study, circulating levels of FGF19 in patients with heart failure were significantly elevated compared to those in age-matched individuals without a history of CVDs. In a large cohort of 287 patients with heart failure, high FGF19 levels (HR=1.295, 95% CI: 1.035–1.619, p-value=0.024) were identified as independent predictors of all-cause mortality in Cox analysis53). Furthermore, Fgf15-knockout mice did not develop cardiac hypertrophy in response to high-fat diet, isoproterenol, or cold exposure. In vitro studies further confirmed that FGF19 directly promotes cardiomyocyte hypertrophy, revealing an inter-organ signaling pathway from the gut to the heart via FGF15/19. This gut-–heart signaling pathway regulates cardiac hypertrophy and myocardial fatty acid metabolism53) and holds promise for developing innovative diagnostic and therapeutic strategies for heart failure.
FGF21FGF21, initially isolated from mouse embryo, is composed of 210 amino acids. In contrast, human FGF21 consists of 209 amino acids and shares 75% identity with mouse FGF21 54). Recently, FGF21 has been reported to be highly expressed in adipose tissue as well as in the liver, pancreas, skeletal muscle, heart, kidneys, and testes55).
Studies employing glucose uptake assays have demonstrated that human recombinant FGF21 stimulates glucose uptake in primary human adipocytes and differentiated mouse 3T3-L1 adipocytes55). Unlike insulin, which activates glucose transporter-4 (GLUT-4), FGF21 primarily acts through GLUT-1 in adipocytes to regulate blood glucose levels. Administration of FGF-21 to obese rodents with diabetes and rhesus monkeys has been shown to improve abnormal glucose metabolism, body weight gain, and triglyceride metabolism55-58). When FGF21 was administered to mice lacking either FGFR1 or βKlotho, most of its effect on glucose metabolism and energy expenditure were lost59, 60). These findings indicate that direct action of FGF21 on adipocytes is crucial for metabolic regulation.
The pleiotropic actions of FGF21 result in several outcomes, including enhanced total-body insulin sensitivity61), improved pancreatic beta-cell function57), reduced glucagon secretion from the pancreas55), decreased hepatic lipogenesis, and induction of energy expenditure through the activation of brown fat55, 56). The functions of both adipose tissue types are regulated by sympathetic nerve stimulation. Norepinephrine released from sympathetic nerve terminals promotes lipolysis, fatty acid oxidation, and thermogenesis in brown adipose tissue, and induces lipolysis in white adipose tissue. Additionally, the administration of FGF21 to mice or primary cultured cells has been shown to activate the sympathetic nervous system, promote fatty acid oxidation in brown adipocytes, and induce the formation of beige cells within white adipocytes62). This suggests that FGF21 may promote energy expenditure and reduce body weight through its influence on brown adipose tissue and the induction of beige cells. FGFR1 is barely expressed in hepatocytes in vivo; therefore, it is unlikely that FGF21 acts directly on the liver63). Administration of FGF21 to primary adipocytes increases the amount of adiponectin, which regulates the overall glucose and lipid balance through its endocrine effects on the liver and skeletal muscles. Furthermore, when FGF21 was administered to adiponectin-knockout mice, FGF21’s ability to improve glucose metabolism, lipid metabolism, and energy expenditure was diminished. This finding suggests that FGF21’s metabolic effects may be mediated indirectly through the release of adiponectin from adipose tissue (Fig.3)64).
FGF21 is primarily synthesized and secreted in response to fasting. Fatty acids activate peroxisome proliferator-activated receptors-α (PPAR-α) in the liver and peroxisome proliferator-activated receptors-γ (PPAR-γ) in the pancreas, inducing the expression of FGF21.
Recently, several animal studies have reported that FGF21 exerts cardioprotective effects against various stressors. When treated with isoproterenol, Fgf21-knockout mice exhibited enhanced eccentric hypertrophy, activation of pro-inflammatory pathways, and reduced fatty acid oxidation compared to control mice 65. These changes were already present in newborn Fgf21-knockout mice, and conversely, FGF21 treatment reversed these detrimental effects in vivo and in cultured cardiomyocytes65). Furthermore, FGF21 administration attenuated angiotensin II-induced cardiac hypertrophy in mice, while Sirt1-knockout mice showed diminished beneficial effects of FGF21 in attenuating angiotensin II-induced hypertrophy66). FGF21 released from the liver and adipose tissue during ischemia/reperfusion injury appears to plays a crucial role in protecting the myocardium. This protection is mediated through the FGFR1/βKlotho–PI3K/AKT–BCL-XL/BCL-2-associated death signaling network67). Similarly, FGF21 contributes to the improved outcomes following experimental myocardial infarction, partly through an adiponectin-dependent mechanism68).
Recent findings from population studies must be validated in independent cohorts before FGF21 can be used as a biomarker in the clinical setting69). Two recent clinical trials have reported the promising effects of FGF21 analogs in improving the cardiometabolic profiles of patients with obesity and type 2 diabetes70, 71). These findings support the potential use of FGF21 as a therapeutic target for atherosclerosis and CVDs.
FGF23FGF23, a hormone, primarily secreted by osteoblasts and osteocytes, plays a crucial role in regulating phosphate metabolism and vitamin D homeostasis. It achieves this by suppressing renal tubular reabsorption of phosphate through interactions with FGFR1–4 and its co-receptor αKlotho72-74). The primary function of FGF23 is to downregulate the expression of the sodium phosphate cotransporters NaPi-2a and NaPi-2c in renal proximal tubules75-77). FGF23 acts as a phosphaturic hormone, increasing the excretion of excess phosphate in urine and effectively regulating blood phosphate levels. FGF23 also decreases the level of 1,25-dihydroxy vitamin D [1,25-(OH)2D] and increases the level of parathyroid hormone (PTH). This regulation influences gastrointestinal calcium/phosphate absorption and renal phosphate reabsorption75). Studies have shown that in chronic kidney disease (CKD), as kidney function declines, FGF23 levels rise early in the disease course, preceding increases in PTH levels and abnormalities in serum phosphorus and calcium concentrations78). The progression typically involves an initial increase in serum FGF23 and decrease in serum αKlotho concentration, followed by reductions in 1,25-(OH)2D levels, increases in serum PTH levels, and ultimately, elevated serum phosphate levels and decreased serum calcium levels in advanced CKD79).
High circulating FGF23 levels are associated with increased mortality in patients undergoing hemodialysis80) and those with CKD81), independent of serum phosphate levels and other traditional risk factors. Furthermore, in the Cardiovascular Health Study, high FGF23 levels were linked to all-cause mortality and incident heart failure in the general population82). Additionally, elevated FGF23 levels were positively associated with all-cause mortality in both patients with CKD and the general population83, 84).
FGF23 has emerged as a useful biomarker in patients with heart failure. Elevated FGF23 levels have been associated with worsening heart failure, increased hospitalizations, and mortality in patients, both in chronic and acute heart failure settings85, 86). Patients with acute heart failure exhibited increased plasma FGF23 levels compared to controls; however, myocardial FGF23 gene expression remained similar87). There is still a limited understanding of the specific production site of FGF23, the mechanisms underlying the elevation of FGF23 levels during decompensated heart failure episodes, and its effects on kidney function. Nonetheless, FGF23 shows promise for use in risk stratification and developing treatment guidelines for heart failure.
Previous studies have reported that high serum concentrations of FGF23 are associated with left ventricular hypertrophy (LVH) in both experimental models72) and patients, particularly those CKD88-92). Several hypotheses have been proposed regarding the development of LVH via FGF23 signaling. Grabner et al. showed that FGF23 directly induced LVH through FGFR4 activation and subsequent calcineurin/nuclear factor of activated T cells signaling, independent of its co-receptor αKlotho72, 93). A specific FGFR4 antibody was found to inhibit FGF23-induced hypertrophy in isolated cardiac myocytes and attenuate LVH in rats with CKD78). Thus, FGF23/FGFR4 pathway may be a potential therapeutic target for the treatment or prevention of LVH. In contrast, Andrukhova et al. demonstrated that FGF23 induced distal tubular sodium retention in an αKlotho-dependent manner, leading to an increase in blood volume and hypertension, thereby contributing to LVH94). The renin–angiotensin–aldosterone system (RAAS) is a neurohormonal signaling pathway that plays a crucial role in regulating blood pressure and influences the development and progression of LVH95). RAAS-blocking medications, such as angiotensin converting enzyme inhibitors and angiotensin receptor blockers, have been shown to improve LVH and reduce cardiovascular mortality96). FGF23 may stimulate renin production in the kidneys by suppressing the production of 1,25-(OH)2D97). Additionally, FGF23 may reduce the expression of ACE2 98), an enzyme that converts angiotensin-2 into vasodilatory angiotensin-(1-7), leading to cardiac hypertrophy and myocardial fibrosis. However, further research is required to fully understand the relationship between FGF23 and the RAAS.
Vascular calcification, characterized by the deposition of calcium phosphate minerals in arterial walls, is a complex pathological process influenced by various factors, including inflammation, oxidative stress, and interactions between cellular and molecular processes99). Large prospective studies have demonstrated that vascular calcification is associated with an increased risk of cardiovascular morbidity and mortality100, 101). Observational studies on patients with coronary artery disease have revealed the existence of a relationship between the FGF23/αKlotho axis and the occurrence of coronary artery disease and aortic valve calcifications102, 103).
When serum calcium phosphate levels increase because of disturbances in mineral homeostasis, serum fetuin-A binds to it, forming amorphous calciprotein particles (CPPs)104, 105). CPPs prevent calcium phosphate precipitation in the blood and helps to transport calcium phosphate from the intestines to the bones. However, under certain conditions, CPPs can transform from small, harmless, amorphous, spherical complexes (primary CPPs) into harmful, large crystalline complexes (secondary CPPs). Studies suggest that increase in either the calcium or phosphate concentration in cultured osteoblasts triggers FGF23 expression, and this response is influenced by CPP formation. For instance, mice receiving a continuous high-phosphate diet exhibited greater FGF23 and CPP levels than those receiving intermittent bolus phosphate106). This suggests that persistent hyperphosphatemia, as seen in CKD, may lead to more CPP formation than temporary dietary phosphate spikes. Clinical studies in patients with CKD have shown that serum CPP levels are associated with coronary artery calcification, as measured by calcium scores107), vascular stiffness108), and inflammation109, 110). These findings suggest that even in patients without renal dysfunction, CPPs might be an early trigger for increased FGF23 production in osteoblasts.
Previous Mendelian randomization studies have identified LDL-C and lipoprotein(a) as causal risk factors for aortic valve calcification111, 112). However, subsequent large randomized clinical trials have shown that LDL-C-lowering therapy failed to inhibit disease progression113). Nonetheless, statins retain the recognized advantages in treating patients at significant risk of atherosclerosis.
A study using the Multi-Ethnic Study of Atherosclerosis dataset showed that elevated serum phosphate levels were associated with increased prevalence of aortic valve calcification. However, other biomarkers such as FGF23, PTH, and urine phosphate were not significantly associated with aortic valve calcification. The study concluded that phosphate metabolism biomarkers were not strongly associated with aortic valve calcification, except in the highest quartile of FGF23 114). Klotho-deficient mouse models exhibit premature aging and develop calcified nodules in the aortic valve hinge region without leaflet thickening, extracellular matrix disorganization, or inflammation. In the aortic valves of Klotho-deficient mice, increased Runx2 expression was observed, indicating the presence of calcification, along with the increased expression of Sox9 and elevated levels of collagen 10a1 and osteopontin115). Studies on human calcified aortic valves have revealed increased activation of bone morphogenetic protein (BMP) signaling around the calcified aortic valves116). Moreover, enhanced BMP signaling and osteochondrogenic activation are observed in Klotho-deficient mice and calcified porcine aortic valve interstitial cells117). Thus, interactions between the FGF23 and BMP signaling pathways may contribute to the pathogenesis and progression of aortic stenosis.
A study involving 708 patients with calcific aortic stenosis analyzed the correlation between 49 biomarkers and the risk of death or heart failure-related hospital admission. Several biomarkers were found to be significant predictors of death and heart failure-related hospital admissions. Machine learning models have highlighted interleukin (IL)-6 and FGF23 as the most prominent biomarkers associated with adverse outcomes118). The progression of aortic stenosis may require not only FGF23 activation but also an inflammatory response119).
Atrial fibrillation (AF), the most prevalent cardiac arrhythmia in older adults, greatly increases the risk of stroke, heart failure, and CVD mortality120). Previous studies investigating the relationship between FGF23 and AF have reported inconsistent results. Two large, community-based cohort studies, excluding individuals with baseline CVDs, reported the existence of an association between elevated FGF23 levels and AF development, potentially shedding light on the link between CKD and AF121). Similarly, studies in patients with CKD demonstrated an independent association between elevated FGF23 levels and both prevalent and incident AF122). However, this association was not observed in the Atherosclerosis Risk in Communities Study123). Differences in study design and population characteristics, such as varying prevalence of CKD, exclusion of individuals with CVDs, and follow-up duration, might explain these inconsistencies.
FGF23 may be associated with AF via its direct effects on the myocardium. Animal and clinical studies suggest a correlation between increased FGF23 levels and high LVH and lower ejection fraction124, 125). However, the precise mechanism underlying FGF23’s potential negative cardiac effects remain unclear because of the presumed absence of αKlotho, a co-factor of FGF23, in myocardial cells. FGF23 might exert its direct cardiac effects through a non-canonical FGF23-FGFR4 axis, independent of αKlotho and FGF23-FGFR1 axis (Fig.4).
The effects of FGF23 can be classified into canonical and non-canonical effects. The canonical effects of FGF23 mainly involve the regulation of phosphate and vitamin D metabolism through specific receptors and co-receptor αKlotho. The non-canonical effects of FGF23 are mediated through signaling pathways independent of αKlotho.
Independent associations have been established between high FGF23 levels and inflammatory markers, such as IL-6, C-reactive protein, tumor necrosis factor-α, and fibrinogen126), as well as vascular dysfunction127-129). Additionally, the relationship between FGF23 and AF might be influenced by the suppression of 1,25-(OH)2D, the active form of vitamin D. The PRIMO trial investigated the effects of vitamin D therapy on the cardiac structure and function over 48 weeks130). Although no significant differences emerged in left ventricular volume index between the treatment and placebo groups, there were significant differences in the left atrial volume index, brain natriuretic peptide levels, and congestive heart failure hospitalizations between the two groups131). Additional research is needed to fully understand the mechanisms underlying these associations and determine whether FGF23 could be a useful biomarker or therapeutic target for AF.
Elevated serum phosphate levels in patients undergoing maintenance dialysis are associated with an increased risk of CVDs and mortality132). Current phosphate management approaches include dietary phosphate restriction and phosphate binder therapy. Based on molecular structures, phosphate binders can be categorized as calcium-based (calcium carbonate and calcium acetate), non-calcium-containing (sevelamer carbonate, sevelamer hydrochloride, and lanthanum carbonate), aluminum-containing (aluminum hydroxide), or iron-based (sucroferric oxyhydroxide and ferric citrate) compounds. Phosphate binder therapy has many side effects, such as constipation, diarrhea, and nausea, and requires frequent dosing, resulting in high pill burden. A recent systematic review of randomized controlled trials evaluated the efficacy of non-calcium-based phosphate-lowering treatments in patients with non-dialysis CKD133). The meta-analysis revealed that calcium-free phosphorus adsorbents reduced serum and urinary phosphate levels compared to the placebo. However, their use was associated with increased incidence of constipation and vascular calcification, and they demonstrated limited effectiveness in mitigating cardiovascular risk.
In patients undergoing maintenance hemodialysis, 12 months of strict phosphate control with sucroferric oxyhydroxide or lanthanum carbonate delayed the progression of coronary artery calcification compared to standard phosphate control134). Notably, both medications achieved similar results, suggesting minimal differences in their efficacy. Furthermore, the extent of coronary artery calcification was found to be associated with serum phosphate concentration. Interestingly, the coronary artery calcification scores decreased after a year of strict phosphate control therapy, indicating potential reversibility of established calcified lesions. Further studies are warranted to determine whether targeting phosphate levels within the normal range can improve prognosis.
Tenapanor acts as a non-absorbable inhibitor of the sodium-proton exchanger (NHE3) in the gastrointestinal tract, thereby reducing sodium absorption. This leads to increase in luminal sodium levels, which in turn reduces the phosphate absorption, ultimately lowering serum phosphate levels.
The FGF family plays an important role in the physiology and pathophysiology of cardiovascular health. The diverse functions of FGFs, from embryonic development to metabolic regulation and tissue repair, highlight their potential as therapeutic targets for CVDs. Despite the current advancements in CVD treatment, a considerable burden remains due to the limitations of existing therapies. For instance, FGF1 and FGF2 have shown potential in promoting angiogenesis and improving cardiac function in clinical studies. FGF19 is essential for metabolic regulation, whereas FGF21 and FGF23 are important regulators of cardiometabolic health and phosphate metabolism, respectively. Understanding and targeting the diverse roles of FGFs may lead to novel therapeutic strategies for mitigating the substantial burden of CVDs.
None.
No potential conflicts of interest were disclosed.
