ABSTRACT
-
Background/Aims
Bezafibrate (BZF), a dual peroxisome proliferator-activated receptor/pregnane X receptor agonist, has demonstrated efficacy in combination with ursodeoxycholic acid (UDCA) for primary biliary cholangitis (PBC) and primary sclerosing cholangitis (PSC). Although one of the therapeutic effects of BZF is suppression of bile acid synthesis, its specific impact on bile acid synthesis pathways has not been thoroughly explored. This study investigated bile acid profiles, synthesis intermediates, and their associations with liver biochemistries in patients with PBC and PSC, and evaluated the impact of BZF treatment on these associations.
-
Methods
We enrolled 30 patients with PBC, 10 with PSC, and 30 control subjects. We measured total bile acids, bile acid components, plasma levels of 7α-hydroxycholesterol (7α-OH-C), 7α-hydroxy-4-cholesten-3-one (C4), and 27-hydroxycholesterol (27-OH-C) to assess the classic and alternative bile acid synthesis pathways and analyzed the association with liver biochemistries with and without BZF treatment.
-
Results
Total bile acid levels were elevated in PBC and PSC compared to controls, correlating significantly with liver biochemistries. BZF treatment significantly suppressed the classic pathway, as evidenced by reduced 7α-OH-C and C4 levels. However, 27-OH-C levels, possibly reflecting the alternative pathway activity, were not reduced in those with elevated liver biochemistries despite BZF treatment, suggesting incomplete suppression of alternative pathway in patients with suboptimal BZF response.
-
Conclusions
These findings indicate that while BZF effectively suppresses the classic pathway, alternative pathway activity may compromise its therapeutic efficacy in treatment-resistant cases, highlighting the need for novel therapies inhibiting the alternative pathway in patients with inadequate response to BZF.
-
Keywords: Primary biliary cholangitis; Primary sclerosing cholangitis; Bile acid; Ursodeoxycholic acid
Study Highlights
• This study comprehensively investigated bile acid profiles and synthesis pathways in patients with chronic cholestatic liver disease, primary biliary cholangitis and primary sclerosing cholangitis. BZF effectively suppressed the classic pathway, reflected by reduced 7α-hydroxycholesterol and C4, but failed to sufficiently inhibit the alternative pathway, as indicated by persistent elevation of 27-hydroxycholesterol in patients with inadequate biochemical response. These results reveal a mechanistic explanation for variable BZF efficacy and emphasize the need for novel therapeutic agents that can more effectively target the alternative pathway to improve outcomes in refractory cholestatic liver diseases.
Graphical Abstract
INTRODUCTION
Primary biliary cholangitis (PBC) and primary sclerosing cholangitis (PSC) are both chronic intrahepatic cholestatic liver diseases [
1,
2]. In PBC, the intrahepatic small bile ducts are affected, whereas in PSC, both the intra- and extrahepatic large bile ducts are involved, and both conditions lead to chronic cholestasis. The etiology of both diseases remains unknown, and no curative treatment has been identified to date. In PBC, although an aberrant autoimmune response against biliary epithelial cells is thought to play a crucial role, immunosuppressive agents have shown limited efficacy, and current treatment primarily focuses on cholestasis-improving agents such as ursodeoxycholic acid (UDCA) and peroxisome proliferator-activated receptor (PPAR) agonists, such as bezafibrate (BZF), seladelpar and elafibranor [
1]. Real-world data have demonstrated that both UDCA and BZF, a dual PPAR/pregnane X receptor (PXR) agonist, improve long-term outcomes in patients with PBC [
3,
4]. In PSC, genome-wide association studies have suggested an autoimmune nature of the disease [
5], and the high prevalence of comorbid inflammatory bowel disease implicates a significant role for gut microbiota in pathogenesis [
6], which has recently become a focus for therapeutic strategies [
7]. Nevertheless, UDCA remains the cornerstone of medical treatment in PSC, with reports of its beneficial effects on long-term prognosis [
8], while ongoing clinical trials of nor-UDCA (norcholic acid) [
9] are awaited for further results. Additionally, the short-term biochemical improvement with a combination of UDCA and BZF was demonstrated in patients with PSC [
10].
The efficacy of UDCA and BZF in treating chronic cholestatic liver diseases such as PBC and PSC is attributed to their ability to mitigate hepatocellular and cholangiocyte damage caused by the accumulation of hydrophobic and toxic bile acids within the liver [
11-
13]. In the context of cholestasis, hydrophobic bile acids such as lithocholic acid can inflict cellular injury. UDCA, a hydrophilic bile acid, exerts choleretic effects that alleviate cholestasis and increase the proportion of UDCA within the bile acid pool, thereby reducing cell damage mediated by hydrophobic bile acids [
14]. On the other hand, BZF ameliorates cholestasis by downregulating bile acid synthesis [
15]. It achieves this through the activation of PPARα, which downregulates cholesterol 7α-hydroxylase (CYP7A1), the rate-limiting enzyme of the classic pathway of bile acid synthesis, as well as sterol 27-hydroxylase (CYP27A1), the rate-limiting enzyme of the alternative pathway [
16,
17].
However, in clinical practice, there are not infrequent cases where the effects of BZF in addition to UDCA are inadequate. Our investigation suggests that in advanced cases presenting with hypoalbuminemia or hyperbilirubinemia at diagnosis, BZF may have limited efficacy, potentially leading to poorer long-term outcomes [
18]. Nevertheless, there are no studies evaluating the effects of BZF from the perspective of bile acid synthesis. In this study, we first measured the total bile acids, individual bile acid compositions, and intermediary products in the classic and alternative pathway in the bile acid biosynthesis pathway in patients with PBC and PSC, comparing these measurements with disease controls. Then we examined the relationships between these parameters and liver biochemistries. Finally, we further investigated the association between liver biochemistries and intermediary products in bile acid synthesis with the presence or absence of BZF treatment, in order to assess the role of the classical and alternative pathways during BZF therapy.
MATERIALS AND METHODS
Plasma preparation from patients with PBC, PSC or other liver diseases
We prospectively and consecutively enrolled a total of 70 patients in this study at the outpatient clinic of Teikyo University Hospital, comprising 30 patients with PBC, 10 with PSC, and 30 with other liver diseases (including hepatitis B virus [HBV] infection, drug-induced liver injury, and hepatitis C virus [HCV] infection) as control subjects. In patients with PBC or PSC, we excluded treatment-naïve patients and those in which the length of treatment (UDCA or BZF) was less than one year. From patients who had given their consent, 10 milliliters of blood were collected, and the plasma was separated by the centrifugation at 10,000 g for 10 minutes. The collected plasma was stored at –80°C until analysis. In addition, liver biochemistries at enrollment were measured as a routine blood test. The study was approved by the Ethics Committee of the Teikyo University Hospital (19-110) and Chugai Pharmaceutical Co., Ltd. (E19049).
Bile acid analysis and quantification
Analysis was performed using the Bile Acids Kit (Biocrates Life Sciences AG, Innsbruck, Austria) as previously reported [
19]. Reference standards of 20 bile acids were purchased from Sigma-Aldrich (St. Louis, MO, USA), TRC Chemicals (Toronto, Canada), Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan) and Cayman Chemical Co. (Ann Arbor, MI, USA). Seven stable isotope-labeled bile acids (CA-D5, βHDCA-D4, DCA-D5, TCA-D4, GCDCA-D4, TCDCA-D5, and αMCA-D5) were obtained as internal standards (IS) from TRC Chemicals, IsoSciences LLC. (King of Prussia, PA, USA), and Cayman Chemical Co. The measurements were performed on an ACQUITY UPLC system (Waters Co., Milford, MA, USA) coupled to a QTRAP 5500 mass spectrometer (AB Sciex Pte. Ltd, Milford, MA, USA) or using a system consisting of a Nexera ultra HPLC system (Shimadzu Co., Kyoto, Japan) and a QTRAP 6500 mass spectrometer (AB Sciex Pte. Ltd.) equipped with an ESI ion source using scheduled MRM. Chromatographic separations were performed on an RRHD Eclipse Plus C18 column (1.8 μm, 2.1 × 50 mm, Agilent, Santa Clara, CA, USA) with a column temperature of 50°C. The HPLC systems were used with a flow rate of 0.5–1.0 mL/minute and a run time of 5 minutes. The mobile phases were water containing 10 mmol/L ammonium formate and 0.015% formic acid (A) and acetonitrile/methanol/water (v/v/v=65/30/4) containing 10 mmol/L ammonium formate and 0.015% formic acid (B). The conditions of the elution gradient were as follows: 0–0.25 minutes (1–35% B at 0.5 mL/minute), 0.25–0.35 minutes (35–45% B) at 0.5 mL/minute, 0.35–1.90 minutes (40–45% B at 0.5–0.8 mL/minute), 1.9–2.1 minutes (45–55% B at 0.8 mL/minute), 2.1–3.3 minutes (55–65% B at 0.8–1.0 mL/minute), 3.3–3.5 minutes (65–100% B at 1.0 mL/minute) and 3.5–4.0 minutes (100% B at 1.0 mL/minute) and 4.0–5.0 minutes (100–1% B at 1.0–0.5 mL/minute). Sample preparation for liquid chromatographtandem mass spectrometer (LC-MS/MS) analysis was performed as follows: A 10-μL aliquot of plasma sample and 10 μL of 50% isopropanol were added to 110 μL of acetonitrile/methanol (v/v=4:6) containing 7 IS. After pipetting, the mixture was transferred to a MultiScreen Solvinert filter plate (Merck Millipore, Burlington, MA, USA), centrifuged at 560 g for 5 minutes and used for LC-MS/MS analysis.
Reference standards, 7α-hydroxycholesterol (cholest-5-en-3β,7α-diol) and 27-hydroxycholesterol (25(R)-cholest-5-en-3β,26-diol) were purchased from Avanti Polar Lipids Inc. (Birmingham, AL, USA) and C4 (7α-hydroxy-4-cholesten-3-one) was purchased from Sigma-Aldrich. Stable isotope-labeled standards as IS, C4-d7 was purchased from TRC Chemicals and 27-hydroxycholesterol-d6 was purchased from Avanti Polar Lipids Inc. Measurements were performed using a Waters ACQUITY UPLC system coupled to a QTRAP 5500 mass spectrometer or using a system consisting of a Nexera ultra HPLC system and a QTRAP 6500 mass spectrometer, equipped with an APIC ion source with MRM analysis. The chromatographic separations were performed using an RRHD Eclipse Plus C18 column (1.8 μm, 2.1 × 100 mm, Agilent) with a column temperature of 40°C. The step gradient used was as follows: after injection, 70% to 88% solvent B in 8.0 minutes, hold 88% solvent B for 2.0 minutes, 88% to 99% B in 1.0 minute, hold 99% solvent B for 5.0 minutes, 99% to 70% solvent B in 0.5 minutes, reequilibrate at 70% solvent B for 3.5 minutes. Total run time was 20 minutes. MRM transitions for 7α-hydroxycholesterol, 27-hydroxycholesterol, C4, 27-hydroxycholesterol-d6 and C4-d7 were m/z 367.3>91.0, 385.1>91.1, 401.4>177.3, 391.4>91.1 and 408.4>177.3, respectively. Sample preparation were performed as follows: A 60 μL aliquot of plasma sample and 3 μL of 70% isopropanol were added to 120 μL of acetonitrile/methanol (v/v=1/1) containing 2 IS. After pipetting, the mixture was transferred to a MultiScreen Solvinert filter plate (Merck Millipore), centrifuged at 560 g for 5 minutes, and used for LC-MS/MS analysis.
Data and statistical analysis
Analysis of bile acids and oxysterols was performed using Analyst 1.7.1 software. Dunnett’s multiple comparison test was conducted. The Wilcoxon test was used for bivariate analysis. Each statistical analysis was performed with JMP (version 15.0.0; SAS Institute Inc., Cary, NC, USA).
RESULTS
Patient profile
The patient demographics are summarized in
Table 1. All patients with PBC, 3 patients with PSC, and 14 control patients were female. The mean ages were 67, 44, and 66 years for the PBC, PSC, and control groups, respectively. The majority of patients (87% with PBC and 100% with PSC) had received treatment with UDCA (600 mg per day). BZF was used in 27% of patients with PBC and in 50% of those with PSC. Regarding liver biochemistries, alkaline phosphatase (ALP) levels were significantly elevated, and albumin levels were significantly reduced in patients with PBC and PSC compared to controls. Additionally, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyl transferase (GGT) levels were significantly elevated in the PSC group but not in the PBC group.
Total plasma bile acid levels were significantly elevated in patients with PBC and PSC compared to controls (
Table 2). Given that the concentrations of UDCA and its glycine and taurine conjugates (GUDCA and TUDCA) were substantially influenced by UDCA treatment, we assessed total bile acid levels excluding UDCA and its conjugates. In this adjusted analysis, total bile acid levels were significantly elevated in patients with PSC and increased, though not significantly, in those with PBC compared to controls (
Fig. 1A). In terms of conjugated and unconjugated bile acids, no statistically significant differences were observed in the levels of unconjugated bile acids among the groups. However, conjugated bile acids were significantly elevated in both PBC and PSC compared to controls. The median ratio of conjugated to unconjugated bile acids was 3.23 in PBC, 12.67 in PSC, and 1.72 in controls, indicating an increase in this ratio in PBC and PSC (
Table 2).
We examined the correlations between total bile acid levels (excluding UDCA and its conjugates) and various liver biochemical markers in patients with PBC and PSC (
Fig. 2). Significant positive correlations were observed between total bile acid levels and AST (
P<0.001), ALT (
P=0.004), bilirubin (
P=0.019), and GGT (
P<0.001). However, no significant correlation was found with ALP (
P=0.160).
Bile acids are synthesized via both the classic and alternative pathways. In the classic pathway, CYP7A1 and 3β-hydroxy-Δ
5-C
27-steroid dehydrogenase/isomerase (HSD3B7) catalyze the initial two steps, resulting in the formation of 7α-hydroxycholesterol (7α-OH-C) and 7α-hydroxy-4-cholesten-3-one (C4). The alternative pathway involves the synthesis of 27-OH-C by CYP27A1. To evaluate the contribution of each pathway to plasma bile acid levels in patients with PBC and PSC, plasma levels of 7α-OH-C, C4, and 27-OH-C were measured (
Table 2). In PSC, the level of 27-OH-C was significantly elevated compared to controls, whereas levels of 7α-OH-C was lower in PBC. Moreover, the ratio of 7α-OH-C to 27-OH-C was significantly reduced in both PBC and PSC compared to controls (
Fig. 1B), suggesting an increased contribution of the alternative pathway in these conditions. Additionally, 27-OH-C levels showed significant positive correlations with AST, ALT, and GGT in patients with PBC and PSC (
Fig. 3).
To evaluate the impact of BZF treatment on bile acid synthesis in these chronic cholestatic diseases, we assessed the plasma concentrations of intermediary bile acid synthesis products in patients with PBC and PSC. As illustrated in Figure 4, levels of 7α-OH-C and C4 were significantly reduced in patients receiving BZF treatment, whereas 27-OH-C levels were elevated but not significant. These findings suggest that BZF treatment significantly inhibits the classic pathway of bile acid synthesis, while the alternative pathway appears unaffected. The reduction in the levels of 7α-OH-C and C4 was not affected by ALP, GGT and bilirubin in cases treated with BZF (data not shown), indicating that the suppression of the classical pathway by BZF was not associated with its efficacy.
Next, we stratified plasma 27-OH-C levels based on ALP, GGT and bilirubin levels and BZF treatment status (
Fig. 5). We observed a trend toward statistical significance was observed in the differences in plasma 27-OH-C levels between BZF-treated patients with and without elevated GGT and bilirubin. Furthermore, 27-OH-C was significantly higher (GGT and bilirubin) or tended to be higher (ALP) in patients with elevated liver biochemistries despite BZF treatment compared to those not receiving BZF. To further investigate the state of the alternative pathway in cases with elevated liver biochemistries despite BZF treatment, we examined the proportion of conjugated and unconjugated UDCA (UDCA+GUDCA+TUDCA) in total bile acids in patients receiving UDCA. The analysis revealed that in cases with insufficient BZF efficacy, i.e., those with elevated liver biochemistries despite BZF treatment, the UDCA ratio was significantly lower compared to those without BZF administration or those with effective BZF treatment (
Fig. 6), indicating that bile acids other than UDCA are produced in significantly higher amounts in cases with insufficient BZF efficacy via the alternative pathway.
Taken together, in BZF-treated patients with PBC and PSC patients, bile acid synthesis via the classic pathway is significantly inhibited regardless of the efficacy of BZF. In contrast, suppression of bile acid synthesis via the alternative pathway remains incomplete in cases with an insufficient response to BZF compared to those with complete response, although a causal relationship between incomplete suppression of the alternative pathway and the suboptimal therapeutic response to BZF has yet to be elucidated.
DISCUSSION
In this study, we investigated the associations among total bile acids, bile acid compositions, bile acid synthesis pathways, and liver biochemistries in patients with PBC and PSC. We also examined how these associations are influenced by the administration of BZF. All subjects included in this study were currently undergoing treatment with UDCA and/or BZF. In general, PSC exhibit a poorer therapeutic response to UDCA or BZF compared to PBC. Indeed, in the present cohort, PSC cases had higher levels of ALP and GGT (
Table 1). Total bile acid levels, known as a sensitive indicator of liver injury, were significantly elevated in patients with PSC after extracting UDCA and its conjugates compared to disease controls (
Fig. 1A), and a significant correlation was observed between total bile acids and various liver biochemistries, except for ALP (
Fig. 2).
The analysis of bile acid compositions revealed that the conjugated-to-unconjugated bile acid ratio was 12.67 in PSC and 3.23 in PBC, both significantly higher than the ratio of 1.72 observed in disease controls (
Table 2). Bile acids in the bloodstream originate from two primary sources: reabsorption from the intestines and reflux from the liver. Intestinal bile acids may become partially unconjugated depending on the gut microbiota, but upon returning to the liver, they are fully conjugated [
20]. Consequently, bile acids derived from the liver are exclusively conjugated. In addition, decreased bile secretion reduces the opportunity for bile acids to come into contact with intestinal bacteria having deconjugation activity. Therefore, in the context of cholestasis, where liver bile acids reflux into the bloodstream, the proportion of conjugated bile acids increases. This likely explains the elevated conjugated-to-unconjugated bile acid ratio observed in PSC and PBC in the current analysis.
Regarding the intermediates of bile acid synthesis through the classic and alternative pathways, including 7α-OH-C, C4, and 27-OH-C, no clear association was observed between PBC or PSC and controls, unlike total bile acids or the conjugated/unconjugated (C/U) bile acid ratio (
Table 2). However, significant correlations were identified between these intermediates and liver biochemistries (
Fig. 3). Furthermore, in patients treated with BZF, levels of 7α-OH-C and C4 were significantly lower compared to BZF-untreated patients, regardless of whether they had PBC or PSC. In contrast, no significant difference was observed in 27-OH-C levels between the two groups (
Fig. 4). The reduction in 7α-OH-C and C4 clearly indicates the suppression of CYP7A1, the rate-limiting enzyme in the classic pathway, as previously reported in studies on BZF treatment [
16,
17]. On the other hand, despite reports suggesting that fibrate administration may downregulate 27-OH-C [
17], the present study found no reduction in 27-OH-C levels in BZF-treated cases.
To further investigate the relationship between BZF and its therapeutic effects, liver biochemistries, and intermediates of bile acid synthesis through the classic and alternative pathways, we analyzed the associations between ALP, GGT, and bilirubin levels, and the presence or absence of BZF treatment with these intermediates. The analysis demonstrated that, consistent with previous findings, levels of 7α-OH-C and C4 were reduced in patients receiving BZF treatment regardless of liver biochemistry values (data not shown). This suggests that the therapeutic effects of BZF are, at least in part, attributable to its suppression of the classic pathway, resulting in a reduction in bile acid synthesis [
15,
16]. In contrast, the intermediate of the alternative pathway, 27-OH-C, was not suppressed by BZF treatment, showing levels comparable to those without BZF administration. Furthermore, in cases where liver biochemistries remained elevated despite BZF treatment, 27-OH-C levels were even increased (
Fig. 5). Additionally, among cases treated with UDCA, the proportion of conjugated and unconjugated UDCA is significantly reduced in those with elevated liver biochemistries despite BZF treatment compared to those with excellent response to BZF (
Fig. 6), indicating that total BA synthesis is not reduced. These findings suggest that in cases where BZF treatment is insufficient to reduce levels of liver biochemistries, the suppression of the alternative pathway is incomplete, leading to a lack of reduction in bile acid synthesis in treatment-refractory PBC and PSC.
In addition to incomplete suppression of bile acid synthesis via the alternative pathway, several other factors may contribute to liver injury in patients with an inadequate response to BZF. First, the bile acid transporter NTCP plays a crucial role in the hepatic uptake of bile acids from portal blood, and BZF has been reported to downregulate NTCP expression [
15]. It is possible that this inhibitory effect on NTCP is attenuated in patients with inadequate responses to BZF. Second, the conjugation status of bile acids may also be relevant. Unconjugated bile acids are more hydrophobic and more likely to induce hepatotoxicity compared to conjugated bile acids. However, in the present study, no association was observed between the therapeutic efficacy of BZF and the C/U ratio, suggesting that conjugation states are unlikely to be involved. Finally, enterohepatic circulation should also be considered. The magnitude of enterohepatic circulation is largely determined by the apical sodium-dependent bile acid transporter (ASBT) located in the terminal ileum, which has been reported to be activated by PPARα [
21]. Therefore, individuals in whom ASBT expression is more readily upregulated by BZF may experience enhanced enterohepatic circulation, potentially diminishing the therapeutic efficacy of BZF.
The interpretation of these findings requires careful consideration of several points. First, the sample size in the subanalysis of BZF treatment efficacy presented in Figures 5 and 6 was very small, with only 12 patients receiving BZF therapy. Given this limited number of cases, it is challenging to obtain sufficiently robust results, and any statistical analysis must be interpreted with caution, as the findings may be subject to false positives or false negatives. Further investigation with a larger cohort is warranted to validate these observations. Additionally, the relationship between GGT or bilirubin levels and 27-OH-C in BZF-treated cases did not reach statistical significance, indicating only a trend toward an association. As pointed out above, this is likely due to the limited number of BZF-treated cases, with only 13 patients in total, including 8 with elevated GGT and 6 with elevated bilirubin. In fact, we were able to observe significant differences in comparison of the proportion of UDCA in total bile acids between sufficient and insufficient response to BZF. Second, among cholestasis markers, while associations were observed between GGT or bilirubin and total bile acids or 27-OH-C under BZF treatment, no such relationship was identified for ALP. Typically, ALP is considered a sensitive marker of cholestasis in PBC and PSC and has been used as a primary endpoint, along with bilirubin, in clinical trials for recently approved drugs such as seladelpar and elafibranor [
22,
23]. The lack of a significant association with ALP in this study is unclear. While the limited number of cases with elevated ALP (n=6) may have contributed to the absence of an association with 27-OH-C under BZF treatment, no significant correlation was observed between ALP and total bile acids across the entire cohort of 60 cases. Furthermore, no patients with comorbidities that could elevate ALP, such as bone disease, were included in the study. This also warrants further investigation. It is of note that a significant difference was observed in ALP as well in comparison of the proportion of UDCA in total bile acids between sufficient and insufficient response to BZF. Third, although PPARα is known to regulate a wide range of target genes beyond CYP7A1 and CYP27A1, such as ACOX1 and CPT1A, the present study did not evaluate the expression of these targets. A more detailed analysis of PPARα signaling could have provided further insight into the relationship between the therapeutic efficacy of BZF and the suppression of CYP7A1 and is an area for future investigation. Lastly, it is important to note that an increase in 27-OH-C does not necessarily indicate activation of the alternative pathway. While elevated 27-OH-C can result from increased activity of CYP27A1, the rate-limiting enzyme of the alternative pathway, it may also result from reduced catabolism of 27-OH-C. The degradation of 27-OHC is mediated by CYP7B1, which, along with CYP7A1, is suppressed upon PPARα activation by BZF [
15]. This suppression may contribute to increased 27-OH-C levels. However, if the elevation of 27-OH-C in cases with insufficient response to BZF is due to decreased activity of CYP7B1, the activity of the entire alternative pathway would also be expected to decrease, leading to a reduction in endogenous bile acids and an increase in the UDCA/total bile acid ratio. However, as shown in Figure 6, the UDCA/total bile acid ratio is decreased in these cases, suggesting that endogenous bile acids are not reduced.
In conclusion, in the current study we conducted a detailed analysis of bile acid profiles and intermediates of bile acid synthesis pathways in patients with PBC, PSC, and disease controls, examining their relationships with liver biochemistry parameters. Our findings revealed that while the classic pathway was significantly suppressed under BZF treatment, suppression of the alternative pathway was incomplete in cases where the efficacy of BZF treatment was suboptimal, suggesting that bile acid synthesis was not adequately reduced in these patients. It should be noted that the present findings do not demonstrate a causal relationship between incomplete suppression of bile acid synthesis via the alternative pathway, and the suboptimal therapeutic response to BZF. Nevertheless, to achieve adequate therapeutic efficacy in patients with inadequate response to BZF, the development of novel agents capable of further suppressing the alternative pathway, which remains insufficiently inhibited, is likely necessary.
FOOTNOTES
-
Authors’ contribution
M.I., S.O. and A. Tanaka designed the study. A. Takeuchi, R.M., Y.A., and A. Tanaka collected samples and clinical data. M.I., A. Higashide, S.O., N.H., A.Honda, and A. Tanaka analyzed and interpreted the data. M.I. and A. Tanaka drafted the paper; all authors critically reviewed the manuscript.
-
Acknowledgements
Chugai Pharmaceutical Co., Ltd. supported all of the research.
-
Conflicts of Interest
The authors have no conflicts to disclose.
Figure 1.Total bile acids (BA) level without UDCA and its conjugates and the ratio of 7α-OH-C to 27-OH-C in PBC, PSC and controls. Bars represent the mean and standard error, and the asterisk (*) denotes a significant difference compared to the control group (P<0.05, non-parametric Dunnett’s t-test). (A) Total BA level without UDCA and its conjugates. The value was calculated by subtracting the values of UDCA and its conjugation from total BA. (B) The ratio of 7α-OH-C to 27-OH-C. The value was shown by ratio of the level of plasma of 7α-OH-C to 27-OH-C. UDCA, ursodeoxycholic acid; PBC, primary biliary cholangitis; PSC, primary sclerosing cholangitis.
Figure 2.The association between total bile acid (BA) levels and individual liver biochemical parameters in patients with PBC/PSC (n=40) was examined. The relationships between total BA levels excluding ursodeoxycholic acid (UDCA) and its conjugates, and various biochemical markers, including aspartate aminotransferase (AST, U/L) (A), alanine aminotransferase (ALT, U/L) (B), alkaline phosphatase (ALP, U/L) (C), gamma-glutamyl transferase (GGT, U/L) (D), and bilirubin (BIL, mg/dL) (E), are depicted as scatter plots. PBC, primary biliary cholangitis;PSC, primary sclerosing cholangitis.
Figure 3.The association between plasma levels of 27-OH-C and various biochemical parameters in patients with PBC/PSC (n=40) was analyzed. The relationships between 27-OH-C levels and individual biochemical markers, including aspartate aminotransferase (AST, U/L) (A), alanine aminotransferase (ALT, U/L) (B), alkaline phosphatase (ALP, U/L) (C), gamma-glutamyl transferase (GGT, U/L) (D), and bilirubin (BIL, mg/dL) (E), are depicted as scatter plots. PBC, primary biliary cholangitis; PSC, primary sclerosing cholangitis.
Figure 4.Plasma levels of intermediary products involved in bile acid synthesis were analyzed. The values of these intermediary products in patients with PBC/PSC (n=40), with or without bezafibrate therapy, are presented. ‘-’ indicates patients without bezafibrate therapy, and ‘+’ indicates those receiving bezafibrate therapy. (A) 7a-OH-C (nmol/L). (B) C4 (nmol/L). (C) 27-OH-C (nmol/L). Bars represent the mean and standard error, with an asterisk (*) indicating a statistically significant difference between patients with and without bezafibrate therapy (P<0.05, non-parametric Dunnett’s t-test). PBC, primary biliary cholangitis; PSC, primary sclerosing cholangitis.
Figure 5.Plasma levels of 27-OH-C in the alternative pathway of bile acids synthesis stratified by BZF treatment as well as ALP (A), GGT (B) and Bil (C) levels in patients with PBC/PSC. Stratification was conducted by ULN (ALP), and median value (GGT and Bil) in all patients with PBC/PSC. Gray and black bars indicate values in patients without and with BZF therapy, respectively. A single asterisk (*) and a sharp symbol (#) denotes a significant difference (P<0.05) and a trend toward statistical significance, respectively (non-parametric Dunnett’s t-test). ALP, alkaline phosphatase; BZF, bezafibrate; ULN, upper limit of normal; GGT, gamma-glutamyl transpeptidase; Bil, bilirubin; PBC, primary biliary cholangitis; PSC, primary sclerosing cholangitis.
Figure 6.The proportion of unconjugated and conjugated UDCA (UDCA+GUDCA+TUDCA) in total bile acids (BA) in cases treated with UDCA, stratified by BZF treatment as well as ALP (A), GGT (B) and Bil (C) levels in patients with PBC/PSC. Stratification was conducted by ULN (ALP), and median value (GGT and Bil) in all patients with PBC/PSC. Gray and black bars indicate values in patients without and with BZF therapy, respectively. A double asterisk (**), a single asterisk (*), and a sharp symbol (#) denotes a significant difference (P<0.01, and P<0.05), and a trend toward statistical significance, respectively (non-parametric Dunnett’s t-test). ALP, alkaline phosphatase; BZF, bezafibrate; ULN, upper limit of normal; GGT, gamma-glutamyl transpeptidase; Bil, bilirubin; UDCA, ursodeoxycholic acid; GUDCA, glycoursodeoxycholic acid; TUDCA, tauroursodeoxycholic acid; PBC, primary biliary cholangitis; PSC, primary sclerosing cholangitis.
Table 1.Profiles and biochemistries of the participating patients
Table 1.
|
Variable |
PBC (n=30) |
PSC (n=10) |
Controls (n=30)†
|
|
Demographics |
|
|
|
|
Male/Female |
0/30 |
7/3 |
16/14 |
|
Age, yr (mean) |
67 |
44 |
66 |
|
Concomitant IBD (%) |
0 |
40 |
0 |
|
Treatment |
|
|
|
|
UDCA (%) |
87 |
100 |
3 |
|
BZF (%) |
27 |
50 |
0 |
|
Liver biochemistries‡
|
|
|
|
|
AST (U/L) |
25 (19,37) |
54 (41, 73)* |
24 (19,28) |
|
ALT (U/L) |
19 (15, 28) |
74(31, 102)* |
22 (15, 28) |
|
ALP (U/L) |
306 (233, 346)* |
437 (322, 556)* |
198 (161, 229) |
|
GGT (U/L) |
37 (18, 85) |
129 (93, 325)* |
23 (16, 38) |
|
Bilirubin (mg/dL) |
0.64 (0.55, 0.76)* |
0.87 (0.66, 1.06) |
0.88 (0.65, 1.00) |
|
Albumin (g/dL) |
4.0 (3.8, 4.3)* |
4.1 (3.8, 4.3)* |
4.4 (4.1, 4.5) |
Table 2.Plasma bile acid compositions and plasma 7α-OH-C, C4, and 27-OH-C
Table 2.
|
PBC (n=30) |
PSC (n=10) |
Controls (n=30) |
|
Total BAs |
15,425 (3,853, 36,981)* |
28,024 (19,608, 34,250)* |
2,752 (1,695, 4,827) |
|
CA |
111 (32, 299) |
72 (41, 79) |
39 (25, 85) |
|
GCA |
366 (120, 835) |
2,417 (1,286, 5,767)* |
182 (100, 280) |
|
TCA |
37 (14, 123) |
739 (397, 1,363)* |
69 (36, 130) |
|
DCA |
220 (145, 338) |
114 (59, 338) |
354 (203, 622) |
|
GDCA |
218 (96, 902) |
551 (375, 730) |
293 (168, 484) |
|
TDCA |
37 (16, 100) |
87 (43, 297) |
39 (18, 100) |
|
CDCA |
311 (124, 669) |
131 (90, 281) |
186 (102, 408) |
|
GCDCA |
1,203 (618, 2,988) |
2,970 (2,050, 4,457)* |
906 (458, 1,449) |
|
TCDCA |
132 (63, 373) |
1,018 (535, 1,616)* |
99 (42, 230) |
|
LCA |
95 (76, 114)* |
232 (106, 344) |
143 (106, 192) |
|
GLCA |
13 (5, 62) |
31 (17, 74)* |
11 (4, 26) |
|
TLCA |
9 (2, 20) |
13 (11, 25) |
7 (4, 18) |
|
UDCA |
2,675 (557, 7,301)* |
4,627 (608, 9,262)* |
62 (38, 299) |
|
GUDCA |
8,180 (1,395, 17,427)* |
10,380 (7,983, 16,963)* |
139 (58, 360) |
|
TUDCA |
338 (56, 1,104)* |
1,308 (956, 2,401)* |
29 (20, 50) |
|
Conjugated†
|
2,639 (1,235, 4,817)* |
8,545 (4,885, 12,383)* |
1,604 (950, 2,921) |
|
Unconjugated‡
|
917 (527, 1,220) |
559 (466, 797) |
771 (545, 1,122) |
|
C/U ratio§
|
3.23 (1.13, 5.56)* |
12.67 (9.67, 22.96)* |
1.72 (1.08, 3.68) |
|
7α-OH-C |
10 (6, 17)* |
9 (5, 17) |
19 (11, 33) |
|
C4 |
39 (28, 59) |
35 (11, 53) |
68 (39, 116) |
|
27-OH-C |
54 (44, 62) |
63 (48, 78)* |
48 (40, 55) |
Abbreviations
apical sodium-dependent bile acid transporter
aspartate aminotransferase
gamma-glutamyl transferase
glycoursodeoxycholic acid
liquid chromatograph-tandem mass spectrometer
primary biliary cholangitis
peroxisome proliferator-activated receptor
primary sclerosing cholangitis
taurochenodeoxycholic acid
REFERENCES
- 1. Tanaka A, Ma X, Takahashi A, Vierling JM. Primary biliary cholangitis. Lancet 2024;404:1053-1066.
- 2. Dyson JK, Beuers U, Jones DEJ, Lohse AW, Hudson M. Primary sclerosing cholangitis. Lancet 2018;391:2547-2559.
- 3. Harms MH, van Buuren HR, Corpechot C, Thorburn D, Janssen HLA, Lindor KD, et al. Ursodeoxycholic acid therapy and liver transplant-free survival in patients with primary biliary cholangitis. J Hepatol 2019;71:357-365.
- 4. Tanaka A, Hirohara J, Nakano T, Matsumoto K, Chazouillères O, Takikawa H, et al. Association of bezafibrate with transplant-free survival in patients with primary biliary cholangitis. J Hepatol 2021;75:565-571.
- 5. Karlsen TH, Kaser A. Deciphering the genetic predisposition to primary sclerosing cholangitis. Semin Liver Dis 2011;31:188-207.
- 6. Karlsen TH. Primary sclerosing cholangitis: 50 years of a gut-liver relationship and still no love? Gut 2016;65:1579-1581.
- 7. Nakamoto N, Sasaki N, Aoki R, Miyamoto K, Suda W, Teratani T, et al. Gut pathobionts underlie intestinal barrier dysfunction and liver T helper 17 cell immune response in primary sclerosing cholangitis. Nat Microbiol 2019;4:492-503.
- 8. Arizumi T, Tazuma S, Isayama H, Nakazawa T, Tsuyuguchi T, Takikawa H, et al.; Japan PSC Study Group (JPSCSG). Ursodeoxycholic acid is associated with improved long-term outcome in patients with primary sclerosing cholangitis. J Gastroenterol 2022;57:902-912.
- 9. Fickert P, Hirschfield GM, Denk G, Marschall HU, Altorjay I, Färkkilä M, et al.; European PSC norUDCA Study Group. norUrsodeoxycholic acid improves cholestasis in primary sclerosing cholangitis. J Hepatol 2017;67:549-558.
- 10. Mizuno S, Hirano K, Isayama H, Watanabe T, Yamamoto N, Nakai Y, et al. Prospective study of bezafibrate for the treatment of primary sclerosing cholangitis. J Hepatobiliary Pancreat Sci 2015;22:766-770.
- 11. Beuers U, Trauner M, Jansen P, Poupon R. New paradigms in the treatment of hepatic cholestasis: from UDCA to FXR, PXR and beyond. J Hepatol 2015;62(1 Suppl):S25-S37.
- 12. Colapietro F, Gershwin ME, Lleo A. PPAR agonists for the treatment of primary biliary cholangitis: old and new tales. J Transl Autoimmun 2023;6:100188.
- 13. Fuchs CD, Simbrunner B, Baumgartner M, Campbell C, Reiberger T, Trauner M. Bile acid metabolism and signalling in liver disease. J Hepatol 2025;82:134-153.
- 14. Ikegami T, Matsuzaki Y. Ursodeoxycholic acid: mechanism of action and novel clinical applications. Hepatol Res 2008;38:123-131.
- 15. Honda A, Ikegami T, Nakamuta M, Miyazaki T, Iwamoto J, Hirayama T, et al. Anticholestatic effects of bezafibrate in patients with primary biliary cirrhosis treated with ursodeoxycholic acid. Hepatology 2013;57:1931-1941.
- 16. Marrapodi M, Chiang JY. Peroxisome proliferator-activated receptor alpha (PPARalpha) and agonist inhibit cholesterol 7alpha-hydroxylase gene (CYP7A1) transcription. J Lipid Res 2000;41:514-520.
- 17. Post SM, Duez H, Gervois PP, Staels B, Kuipers F, Princen HM. Fibrates suppress bile acid synthesis via peroxisome proliferator-activated receptor-alpha-mediated downregulation of cholesterol 7alpha-hydroxylase and sterol 27-hydroxylase expression. Arterioscler Thromb Vasc Biol 2001;21:1840-1845.
- 18. Matsumoto K, Hirohara J, Takeuchi A, Miura R, Asaoka Y, Nakano T, et al. Determinants of the effectiveness of bezafibrate combined with ursodeoxycholic acid in patients with primary biliary cholangitis. Hepatol Res 2023;53:989-997.
- 19. Pham HT, Arnhard K, Asad YJ, Deng L, Felder TK, St John-Williams L, et al. Inter-laboratory robustness of next-generation bile acid study in mice and humans: international ring trial involving 12 laboratories. J Appl Lab Med 2016;1:129-142.
- 20. Hofmann AF. The enterohepatic circulation of bile acids in mammals: form and functions. Front Biosci (Landmark Ed) 2009;14:2584-2598.
- 21. Jung D, Fried M, Kullak-Ublick GA. Human apical sodium-dependent bile salt transporter gene (SLC10A2) is regulated by the peroxisome proliferator-activated receptor alpha. J Biol Chem 2002;277:30559-30566.
- 22. Kowdley KV, Bowlus CL, Levy C, Akarca US, Alvares-da-Silva MR, Andreone P, et al.; ELATIVE Study Investigators’ Group; ELATIVE Study Investigators’ Group. Efficacy and safety of elafibranor in primary biliary cholangitis. N Engl J Med 2024;390:795-805.
- 23. Hirschfield GM, Bowlus CL, Mayo MJ, Kremer AE, Vierling JM, Kowdley KV, et al.; RESPONSE Study Group. A phase 3 trial of seladelpar in primary biliary cholangitis. N Engl J Med 2024;390:783-794.
