Pathophysiology of Cystic Fibrosis-Related Liver Disease

Abstract

Purpose of review

Cystic fibrosis liver disease (CFLD) is a significant non-pulmonary complication of cystic fibrosis, affecting approximately 5–10% of patients. It encompasses a spectrum of hepatic abnormalities ranging from mild, transient elevations in liver enzymes to advanced CFLD (aCFLD), which is marked by clinically relevant portal hypertension due to cirrhotic or non-cirrhotic liver pathology. This review focuses on aCFLD as the clinically meaningful form of the disease and summarizes recent mechanistic insights into its pathogenesis that may inform the development of targeted therapeutic strategies.

Recent findings

CFLD pathogenesis has been traditionally linked to defective bile secretion. Emerging evidence, however, highlights additional contributors, including cholangiocyte immune dysregulation, gut dysbiosis, and intestinal barrier dysfunction, which together promote hepatic inflammation. Furthermore, recent studies underscore the role of vascular alterations independent of cirrhosis, specifically non-cirrhotic portal hypertension, as the main clinical feature in aCFLD. These findings support a multifactorial, multi-hit model of disease in the pathogenesis of CFLD.

Summary

The complex interplay of these factors suggests that effective treatment for aCFLD may require a multifaceted approach. Advances in understanding the gut-liver axis and vascular contributions provide new therapeutic targets. Future research should focus on validating these findings and evaluating the efficacy of CFTR modulators and microbiome-targeted treatments in altering the course of CFLD.

Introduction

Cystic Fibrosis (CF) is a multiorgan, autosomal recessive disorder caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene which affects over 100,000 people worldwide [1], [2]. Although CF is a monogenic disorder, it manifests with substantial phenotypic variability. The classical presentation includes exocrine pancreatic insufficiency, chronic pulmonary disease, and male infertility, however other organ systems may also be affected [1]. In recent years, CFTR modulator therapies have markedly improved pulmonary outcomes and survival in patients with CF [3*], allowing a shift in attention towards extra-pulmonary manifestations of the disease, including the hepatic complications, known under the umbrella term cystic fibrosis-associated liver disease (CFLD). CFLD refers to a spectrum of hepatobiliary manifestations, ranging from mild and often transient liver enzyme elevations and liver steatosis to advanced CF-related liver disease (advanced CFLD; aCFLD), which is characterized by cirrhotic as well as non-cirrhotic portal hypertension [4**], [5**]. Effective treatment remains elusive, largely due to the poorly understood pathophysiology. In this review, we focus on the pathophysiology of advanced CFLD, rather than on other milder forms of CFLD with the aim of outlining recent mechanistic insights that may inform targeted therapeutic strategies.

Unmet needs in cystic fibrosis–associated liver disease (CFLD)

CFLD constitutes a heterogeneous spectrum of hepatobiliary manifestations observed in over 30% of patients with CF. Importantly, 10% of patients develop clinically significant, progressive liver disease (aCFLD), which accounts to up to 5% of CF–related mortality [6] and which reduces life expectancy [7], [8], [9] emphasizing the importance of early detection and management. Whereas multi-lobular cirrhosis was once thought to be the main cause of portal hypertension in CF, recent studies have demonstrated that portal hypertension in CF is mostly non-cirrhotic in origin. Whereas cirrhotic portal hypertension typically presents in childhood, non-cirrhotic portal hypertension can arise later during adulthood [10], [11], [12].

Despite the significant morbidity and mortality associated with aCFLD, no specific and effective treatment is currently available, largely due to a limited understanding of its pathophysiology. Patients with (mild as well as advanced) CFLD are typically treated with ursodeoxycholic acid (UDCA) [13], [14], a hydrophilic bile acid that stimulates bile flow, reduces bile hydrophobicity and thus cytotoxicity, and that exerts anti-inflammatory effects [15]. Although UDCA can improve liver markers, its long-term efficacy remains unproven [16], [17] and the CF Foundation recommends against its routine use to prevent advanced liver disease in individuals with CF [4]. CF patients are monitored for the development of cirrhosis and treated with standard (non-CFLD specific) care in case of portal hypertension; liver transplantation is reserved for severe cases [18]. While CFTR modulators have demonstrated benefits in pulmonary outcomes, their efficacy in CFLD remains inconclusive [19*], [20*].

Pathophysiological mechanisms of CFLD

CFTR is a low-conductance, cyclic AMP-activated chloride channel expressed mostly in epithelial cells of multiple organs. Among the parenchymal liver cell types, CFTR is only expressed in cholangiocytes, the epithelial cells lining the bile ducts [21], [22]. Although some evidence suggests that CFTR is expressed in endothelial cells from various organs, including human umbilical vein endothelial cells (HUVEC), its expression in liver endothelial cells has not been directly confirmed [23], [24], [25]. While CFTR dysfunction in cholangiocytes or other cell types may initiate the disease process, the mechanisms underlying CFLD are likely multifactorial, and several hypotheses have been proposed to explain its etiology and pathophysiology (Figure 1).

Figure 1. Hypothetical pathophysiological mechanisms underlying the development of CFLD.

The cartoon summarizes the current hypotheses explaining the etiology and pathophysiology of CFLD that were discussed in this review article. The illustration was created with BioRender.com.

Pathogenetic models based on CFTR expression and function in the liver

Defective CFTR in cholangiocytes: the first hit?

The secretory defect hypothesis.

The most long-standing pathogenetic hypothesis of CFLD, the secretory defect hypothesis, is directly related to CFTR dysfunction in cholangiocytes. In cholangiocytes, CFTR is localized at the apical membrane, where it mediates chloride efflux [21]. The resulting ionic and osmotic gradients induce water movement through aquaporin channels, promoting bile flow [21], [22], [26]. According to the secretory defect hypothesis, CFTR dysfunction in cholangiocytes would thus lead to reduced bile flow and consequently ductular cholestasis, which would expose cholangiocytes to cytotoxic bile acids, leading to inflammation, fibrosis, and eventually cirrhosis [27] (Figure 1). However, bile inspissation is an uncommon histological finding and clinical signs of cholestasis, such as hyperbilirubinemia, are rarely observed in CFLD patient and do not correlate with the severity of liver fibrosis [28], [29]. These findings, together with the frequent occurrence of non-cirrhotic portal hypertension, challenge the secretory defect hypothesis.

The “bicarbonate umbrella”.

Besides contributing to bile flow by mediating chloride and thus water efflux into bile ducts, CFTR in cholangiocytes also drives bicarbonate secretion through the Cl⁻/HCO₃⁻ exchanger (AE2). It is thus expected that CFTR-deficient cholangiocytes secrete less HCO₃⁻, and thus bile with decreased alkalinity (Figure 1). Bile pH in CF piglet was found to be slightly reduced in the gallbladder, but not in bile secreted in the intestine [30]. Defective biliary HCO₃⁻ secretion in the CF biliary tree could be pathogenically relevant, as biliary HCO₃⁻ secretion is essential for the creation of an alkaline mucus barrier, termed the “bicarbonate umbrella”, that shields the apical membrane of cholangiocytes from protonation of apoptosis-inducing hydrophobic bile acids and that limits their cytotoxic uptake [31], [32]. However, whether the “bicarbonate umbrella” is disrupted in CF biliary tracts has not been studied. Since HCO₃⁻ secretion is also essential for mucus secretion, it could be hypothesized that mucus secretion is also negatively affected in the CF “bicarbonate umbrella”, however this has also not been studied.

Role of CFTR in cholangiocyte innate immunity and cytoskeleton maintenance.

Cholangiocytes, besides their role in bile secretion, also represent the primary defensive barrier of the biliary system against exogenous agents: they actively participate in mucosal defence by sensing microbial products and by secreting inflammatory mediators [33], [34]. In vitro studies in murine cholangiocytes have shown that under physiological conditions, CFTR forms part of a multiprotein complex that restrains Src tyrosine kinase activity. In CF cells, defective CFTR leads to Src hyperactivation, resulting in TLR4 phosphorylation, activation of NF-κB, and increased production of proinflammatory cytokines in response to lipopolysaccharide (LPS), a component of Gram-negative bacteria contributing to biliary inflammation [35] (Figure 1). These events also affect the F-actin cytoskeleton organization causing disruption of apical junctions and compromising barrier integrity of the biliary epithelium (Figure 1). In vivo, pharmacological inhibition of Src activity in CFTR-deficient mice ameliorates biliary damage, while in vitro, it reduces inflammatory cytokine production and restores junctional integrity [35]. These results demonstrate that CFTR regulates biliary epithelial inflammation and barrier function by controlling Src tyrosine kinase activity and provide a rationale for targeting Src kinase activity as a therapeutic approach for CFLD and other cholangiopathies. Similar findings were confirmed in patient-derived human cholangiocytes (ΔF508/ΔF508), where Src inhibition not only reduces inflammation and normalizes cytoskeletal organization, but also improves the efficiency of CFTR modulators on biliary secretory function, further supporting the potential of Src as a therapeutic target for CFLD [36].

Two or multiple-hit models

Role of CFTR polymorphisms and gene modifiers.

Early genome-wide association studies (GWAS) identified genetic variants predisposing to CFLD. These studies led to the identification of SERPINA1, which encodes α-1-antitrypsin, genes whose variants cause hemochromatosis [hemochromatosis (HFE), transferrin receptor 2 (TFR2), and ferroportin 1 (FPN1)], and genes encoding glutathione S-transferase P1 (GSTP1), mannose binding lectin 2 (MBL2), angiotensin I converting enzyme (ACE), TGFβ1, interleukin 8 receptor (CXCR1), and ATP-binding cassette subfamily B member 4 (ABCB4), which is defective in progressive familial intrahepatic cholestasis type 3 [37], [38], [39], [40], [41], [42]. However, most of these studies were conducted on small cohorts and included patients with various degree of liver disease. Consequently, follow up studies only confirmed SERPINA1 as consistently associated with increased risk of developing severe CFLD. More recently, a comprehensive study involving a large cohort of patients with advanced CFLD employed GWAS, transcriptome-wide association (TWAS), and pathway analyses, comparing individual with severe CFLD (with portal hypertension) to those with CF without liver disease [43**]. This study identified several genes significantly associated with severe CFLD including, PKD1, FNBP1, DUSP6, ANKUB1 and CXCR1, AAMP and TRBV24 that are implicated in disease mechanisms such as hepatic fibrosis, inflammation, innate immune responses, vascular pathology, intracellular signaling, actin cytoskeleton and tight junction regulation. The study also confirmed the involvement of SERPINA1. These findings highlight genetic loci and pathways of clinical relevance, which warrant confirmation through mechanistic and functional studies (Figure 1).

Bile acid toxicity in CFLD.

The presence of altered bile acid (BA) homeostasis in CF suggests a potential involvement in the pathogenesis of CFLD. CF patients and animal models have a higher proportion of more hydrophobic and therefore more cytotoxic primary plasma BAs (cholate and chenodeoxycholate) [44], [45], [46], [47]. A study found that in CFLD, increased biliary taurocholate correlates with the extent of ductular reaction and fibrosis, and in vitro exposure of liver progenitor cells to taurocholate promotes biliary differentiation and chemokine secretion (MCP-1, MIP1α, RANTES), which recruits hepatic stellate cells, linking BA-driven ductular reactions to fibrogenesis [45]. Dysregulated BA metabolism in CF may involve disrupted BA synthesis, transport, and enterohepatic circulation, contributing to cholestasis, liver disease, and gastrointestinal symptoms [48]. Additionally, intestinal FXR signalling is impaired in CF, reducing Fgf15/FGF19 expression, which is known to be anti-inflammatory and to promote liver regeneration after injury [46], [48], [49*] (Figure 1). Altered microbial BA transformation in CF could contribute to the accumulation of hepatotoxic BA species, promoting a pro-fibrogenic environment in the liver [48], [50], [51]. In turn, the increased ratio of primary to secondary BAs that is observed in CF is expected to alter the microbial balance, as higher levels of primary BAs promote BA-tolerant species while suppressing others, reducing microbial diversity. Together with the altered BA profile, this dysbiosis may in turn also compromise the integrity of the intestinal barrier, potentially facilitating the translocation of bacterial byproducts to the liver and thus triggering local inflammation [48], [50], [51].

The gut-liver axis.

The gut–liver axis describes the bidirectional interactions between the intestine, its resident microbiota, and the liver. This crosstalk is essential for maintaining hepatic homeostasis through the entero-hepatic circulation and plays a key role in modulating liver disease mechanisms [52]. In the context of CFLD, this axis can provide a unifying hypothesis linking CFTR dysfunction in the gut and biliary epithelium with hepatic inflammation and fibrosis. Beyond the primary genetic defect in cholangiocytes, gut dysbiosis has emerged as a critical “second hit” in the development of CFLD.

CFTR, expressed on the apical membrane of enterocytes, regulates salt and water fluxes and contributes to maintaining luminal fluidity and pH [51], [53], [54], [55]. Several studies in CF patients, including a systematic review of 38 studies, have reported intestinal inflammation, altered microbiota composition, and dysregulated fatty acid metabolism [56].

Experimental data from CFTR-deficient mice reveal intestinal dysbiosis characterized by enrichment of Enterobacteriales, increased permeability, and mucosal inflammation, possibly leading to the translocation of microbial products (i.e. LPS) into the portal circulation (Figure 1). The livers of these mice exhibit hepatic inflammation and ductular proliferation, while single cell transcriptomic analyses confirm activation of innate immune pathways in CF cholangiocytes consistent with a response to gut derived bacteria signals [57*]. Importantly, as reviewed above, CFTR-deficient cholangiocytes are more reactive to LPS, mounting exaggerated inflammatory responses due to Src/NF-kB pathway activation and impaired biliary barrier function [36]. Thus, intestinal leakiness and cholangiocyte immune dysregulation likely act in concert to amplify liver injury. This gut-liver link is further supported by experimental evidence showing that intestine-specific rescue of CFTR expression or modulation of the gut microbiota using antibiotics prevents hepatic damage in CF mice, whereas experimentally induced colitis, which further increases intestinal permeability, exacerbate it [34], [57*]. Notably, a recent study evaluating the effects of CFTR modulator therapy elexacaftor/tezacaftor/ivacaftor (ETI) in patients with aCFLD versus those without liver diseases before and after treatment found that ETI modulated the gut microbiota in all patients and decreased fecal calprotectin levels. Noteworthy, ETI treatment specifically decreased Streptococcus salivarium and Veilonella parvula only in patients with aCFLD [58**]. Although the microbial changes observed differ from those seen in mouse models and the number of patients with aCFLD was limited, these findings support the link between dysbiosis and intestinal inflammation in aCFLD with potential improvement after ETI treatment. Overall, while gut dysbiosis is a consistent feature in CF, its impact on CFLD progression and potential modulation through diet or pharmacological interventions warrants further investigation [56].

These findings support a multifactorial mode in which intestinal factors, including dysbiosis and barrier dysfunction may act synergistically with the intrinsic cholangiocyte CFTR defect (‘first hit’ as described above) to drive CFLD. This integrated view not only enhances our understanding of CFLD pathogenesis but also provide a rationale for future therapeutic strategies targeting the gut-liver axis.

Vascular disease in CFLD.

Over the past decade, several studies have recognized that portal hypertension in aCFLD can be present in the absence of cirrhosis. References [10], [11], [12] studied liver explants from a total of 30 patients with aCFLD and portal hypertension and only three explants had histological features confirming cirrhosis. These vascular alterations, described as portal venopathy, are characterized by narrowing or absence of portal veins within the hepatic triads and a marked reduction in the dimensions of terminal branches [12]. The presence of this vascular disease challenges the traditional view that CFLD progression is primarily driven by biliary epithelial dysfunction, instead supporting a multifactorial pathogenesis [59]. The aetiology of NCPH is recognized as multifactorial and may involve immune-mediated mechanisms, chronic infections, exposure to certain medications or toxins, genetic predisposition, and prothrombotic conditions [60]. In CF, however, the underlying mechanisms and causes of non-cirrhotic portal hypertension remain largely undefined. The presence of a pro-thrombotic state observed in CF patients could contribute to microscopic portal vein obstruction and subsequent vascular remodeling [61]. An emerging hypothesis proposes that vascular dysfunction in CFLD may arise from exaggerated inflammatory responses of the portal endothelium to gut derived pathogen-associated molecular patterns (PAMPs), or alternatively from CFTR-related defects in endothelial cells themselves, possibly compounded by inflammatory signals originated from the inflamed biliary epithelium (Fiorotto Romina, unpublished data) (Figure 1). A recent single cell transcriptomic study explored the endothelial landscape of CFLD using single-cell RNA sequencing (scRNA-seq) on explanted livers from four CFLD patients, comparing endothelial cell (EC) populations with those in cirrhotic and healthy livers to identify CF-specific alterations [62**]. Results showed that CFLD endothelial cells resemble those of healthy livers more than cirrhotic ones, supporting the hypothesis that CFLD represents a distinct non-cirrhotic entity, often termed porto-sinusoidal vascular disorder [62]. Notably, a distinct population of hepatic sinusoidal endothelial cells, termed CF LSECs, was identified. These cells mainly localized in the periportal region, expressed genes involved in complement activation and coagulation, suggesting activation of pro-inflammatory and pro-thrombotic pathways [62]. However, the study focused only on ECs, without examining other liver-resident cells or cell–cell interactions. Functional validation of CF LSECs pro-coagulant or complement-activating roles was lacking, and being cross-sectional, the study provides no insight into early disease stages or effects of CFTR modulators.

Overall, these findings provide new insights into vascular disease in CFLD and suggest a potential role for endothelial cells in its pathogenesis, particularly in the development of vascular complications such as NCPH. Further studies and validation in experimental models are required to elucidate the mechanisms linking CFTR dysfunction, endothelial injury, and vascular remodelling in CFLD.

Conclusion

CFLD arises from a complex interplay of epithelial, immune, microbial, and vascular factors extending beyond the classical secretory defect hypothesis. CFTR dysfunction perturbs bile composition, cholangiocyte innate immunity, and gut-liver communication, ultimately promoting inflammation, fibrosis, and portal hypertension. These findings support a multifactorial, multi-hit model of disease. Advances in experimental models, including NGS technologies and patient-derived cell systems now offer opportunities to clarify pathogenic mechanisms and identify novel therapeutic targets. In parallel, the clinical advent of CFTR modulators and microbiome-targeted interventions opens new perspectives to assess whether correction of CFTR function or modulation of the gut-liver axis can alter the natural history of CFLD. A deeper understanding of the epithelial, immune, and vascular components of CFLD will be essential to develop effective strategies for prevention and treatment.

Key points.

• Advanced cystic fibrosis liver disease (aCFLD) involves a multifactorial pathogenesis with contributions from cholangiocyte immune responses, gut dysbiosis, and vascular changes.

• The non-cirrhotic portal hypertension observed in many CFLD patients highlights the importance of vascular contributions to the disease.

• Recent genetic studies identify specific loci and pathways critical to CFLD development, offering potential new therapeutic targets.

• Gut-liver interactions are significant in CFLD, suggesting the potential for microbiome-targeted treatments.