ABSTRACT
Aim
Bile acids accumulation in hepatocytes causes liver damage and contributes to the development of hepatocellular carcinoma. However, the association between serum bile acid levels and postoperative intrahepatic recurrence in hepatocellular carcinoma remains unclear. We investigated the effect of preoperative serum bile acid levels on intrahepatic recurrence in patients with hepatocellular carcinoma following initial curative hepatectomy.
Methods
This retrospective study included 249 patients with hepatocellular carcinoma who had preoperative serum bile acid levels measured and underwent initial curative hepatectomy between April 2010 and December 2018. We investigated the associations among serum bile acid levels, liver fibrosis, and postoperative outcomes in these patients.
Results
Patients with high serum bile acid levels had a high prevalence of hepatitis C virus‐positive, a high proportion of preoperative ursodeoxycholic acid use, and poor liver reserve function. These patients had more severe pathological liver fibrosis and necro‐inflammatory reaction and blood‐based markers for liver fibrosis were higher than those in patients with low serum bile acid levels. Compared with blood‐based markers for liver fibrosis, serum bile acids were more accurate prognostic indicators of intrahepatic recurrence. High serum bile acid levels were associated with poor overall survival and high intrahepatic recurrence rates in patients with hepatocellular carcinoma following initial curative hepatectomy. High serum bile acid levels represented an independent risk factor for postoperative intrahepatic recurrence.
Conclusions
Serum bile acid levels reflect both the degree of liver fibrosis and necro‐inflammatory reaction and are associated with postoperative intrahepatic recurrence in patients with hepatocellular carcinoma following initial curative hepatectomy.
Keywords: hepatocellular carcinoma, initial curative hepatectomy, intrahepatic recurrence, liver fibrosis, serum bile acids
Short abstract
This study demonstrated that serum bile acid levels reflect both the degree of liver fibrosis and necro‐inflammatory reactions, serving as precise indicators of the carcinogenic environment of the liver. High serum bile acid levels are associated with postoperative intrahepatic recurrence in patients with hepatocellular carcinoma following initial curative hepatectomy.
1. Introduction
Primary liver cancer was the sixth most commonly diagnosed cancer type and third most common cause of cancer‐associated deaths by organ site worldwide in 2022 [1]. Hepatocellular carcinoma (HCC), which accounts for 75%–80% of all liver cancers [1], has a poor prognosis even after curative treatment [2, 3]. Specifically, the recurrence rate exceeds 70% within 5 years after hepatectomy for HCC [4]. Because there is no established postoperative adjuvant chemotherapy for HCC, identifying the risk factors for intrahepatic recurrence and providing appropriate follow‐up strategies to enable early detection are extremely important.
Various risk factors for postoperative HCC recurrence have been reported, including pathological, biological, host, and surgical factors [5]. Host factors identified as major contributors to HCC recurrence after hepatectomy include liver fibrosis and liver inflammation [6, 7], and bile acids have recently attracted attention as factors associated with these conditions. Hepatic accumulation of hydrophobic bile acids under pathophysiological conditions plays an important role in liver injury by inducing apoptosis or necrosis of hepatocytes [8], potentially leading to liver fibrosis, cirrhosis [9], and even liver cancer [10]. Abnormal bile acid–microbiota crosstalk via the gut–liver axis, which has attracted increasing attention, has also been shown to promote HCC [11, 12].
Several clinical studies have indicated that elevated serum bile acid levels are a risk factor for HCC [13, 14, 15]. Even in patients with nonviral hepatitis, bile acid levels were significantly elevated in patients with HCC [14]. Elevated bile acid levels have also been reported as a risk factor for HCC, particularly in patients with diabetes mellitus (DM) or a high body mass index (BMI) [15]. Although some studies have revealed an association between serum bile acids and the development of HCC, no studies to date have examined the impact of serum bile acids on prognosis after hepatectomy for HCC.
This retrospective study aimed to investigate whether serum bile acid levels are associated with long‐term prognosis, including the incidence of intrahepatic HCC recurrence after initial curative hepatectomy.
2. Materials and Methods
2.1. Study Population
This retrospective study evaluated 805 patients with HCC who underwent hepatectomy between April 2010 and December 2018 at Hiroshima University Hospital. Patients who underwent repeat hepatectomy, had distant metastasis, pathological surgical margin positivity, obstructive jaundice, no preoperative serum bile acid data available, after liver transplantation, death, or recurrence within 30 days of surgery were excluded. Finally, 249 patients who underwent curative hepatectomy for the first time were enrolled after excluding 556 patients who met the exclusion criteria (Figure 1). No patient was taking elobixibat hydrate.
FIGURE 1.
Flowchart of the study design.
This study was conducted in accordance with the guidelines of the Declaration of Helsinki (Fortaleza, Brazil; October 2013) and approved by the Institutional Review Board of Hiroshima University Hospital (approval no. E2019‐1580).
2.2. Definition of Serum Bile Acids Classification
Patients were classified into two groups based on their serum bile acid levels. The two groups were defined by whether the levels were within normal limits (10 μmol/L). Patients with serum bile acid levels ≥ 10 μmol/L were defined as the high‐serum bile acids group, and those with levels < 10 μmol/L were defined as the low‐serum bile acids group.
2.3. Definition of Fibrosis Stage, Necro‐Inflammatory Reaction Grade, and Blood‐Based Markers for Liver Fibrosis
The stage of liver fibrosis (F0–F4) and degree of necro‐inflammatory reaction (A0–A3) were diagnosed based on pathological findings using the scoring system proposed by Ichida et al. [16]. The definitions of each fibrosis stage and necro‐inflammatory reaction grade were as follows: F0, no fibrosis; F1, fibrous portal expansion; F2, bridging fibrosis (portal–portal or portal–central linkage); F3, bridging fibrosis with lobular distortion; F4, cirrhosis; A0, no necro‐inflammatory reaction; A1, mild necro‐inflammatory reaction; A2, moderate necro‐inflammatory reaction; and A3, severe necro‐inflammatory reaction [16]. Each blood‐based marker for liver fibrosis was calculated using preoperative blood examination results. The steatosis‐associated fibrosis estimator (SAFE) score, fibrosis‐4 (FIB‐4) index, and aspartate aminotransferase (AST)‐to‐platelet ratio index (APRI) were calculated using the following formulas [17, 18, 19]:
Age was expressed in years, BMI in kg/m2, platelets (PLT) in ×104/μL, AST and alanine aminotransferase (ALT) in U/L, and globulin in g/dL. Globulin levels were calculated as serum total protein (g/dL) minus albumin (ALB; g/dL). For the APRI, the upper limit of the normal range of AST levels was set at 40 U/L.
2.4. Treatment and Follow‐Up
A follow‐up blood examination to identify tumor markers was performed every 3 months after surgery, for 5 years. Enhanced abdominal computed tomography was performed to rule out recurrence for 6 months. When HCC recurrence was suspected, magnetic resonance imaging was performed.
2.5. Statistical Analysis
Continuous variables were presented as medians and interquartile ranges. Nominal variables were expressed as numbers (%). Nonparametric quantitative data were analyzed using the Mann–Whitney U test. The chi‐square test or Fisher's exact test was used to determine the relations among nominal variables. For continuous variables, such as operative time and intraoperative blood loss, median values were used as cutoff values. Blood chemistry parameters and tumor markers were evaluated using their normal ranges as cutoff values. The prognostic performance of serum bile acids and blood‐based markers for liver fibrosis was assessed separately using the area under the receiver operating characteristic curve (AUROC) derived from receiver operating characteristic (ROC) curve analysis. The Kaplan–Meier method was used to analyze overall survival (OS) and intrahepatic recurrence rates, and the log‐rank test was used to compare groups. To evaluate the effects of serum bile acids and blood‐based markers for liver fibrosis levels on late recurrence, a landmark analysis was performed 2 years after hepatectomy. Only patients who were recurrence‐free at this landmark time point were analyzed for subsequent recurrence. The cutoff value for blood‐based markers for liver fibrosis was set at a level classified as high‐risk in the existing literature [17, 18, 19]. As for APRI, since few patients in the high‐risk group (APRI ≥ 1.5) were present, the cutoff value was set to 1.0. Multivariate analyses were performed to assess factors influencing OS and postoperative intrahepatic recurrence using the Cox regression model. Variables that were statistically significant in univariate analysis were included in the multivariate analysis.
To address bias caused by different distributions of covariates between the high‐ and low‐serum bile acids groups, propensity score‐matched (PSM) analysis was performed using a multiple logistic regression model based on clinicopathological variables. PSM analysis was conducted according to baseline characteristics, such as sex, smoking status, underlying liver disease, preoperative ursodeoxycholic acid (UDCA) use, PLT, prothrombin time (PT), ALB, liver enzymes, indocyanine green dye retention rate at 15 min (ICG‐R15), intrahepatic metastasis (IM), liver fibrosis stage, and necro‐inflammatory reaction, using the new Inuyama Classification.
A restricted cubic spline (RCS) curve with five knots was constructed to evaluate the risk of intrahepatic recurrence according to serum bile acid levels. Calculations were performed using JMP Student Edition v18 (SAS Institute, Cary, NC, USA). RCS analysis and visualization were performed using R version 4.5.1 (2025; R Foundation for Statistical Computing) and the R packages “rms,” “survival,” and “ggplot2.” Survival curves were generated using GraphPad Prism version 10.2.1. Statistical significance was set at p < 0.05.
3. Results
3.1. Comparison of Patient Characteristics Between High‐ and Low‐Serum Bile Acids Groups
Table 1 summarizes the preoperative characteristics of patients according to their serum bile acid levels. Overall, 148 (59.4%) and 101 (40.6%) patients belonged to the high‐ and low‐serum bile acids groups, respectively. The high‐serum bile acids group had a higher proportion of females (p = 0.001), a higher prevalence of preoperative UDCA use (p < 0.001), a higher prevalence of hepatitis C virus (HCV)‐positive liver background (p < 0.001), a lower proportion of smokers (p = 0.007), and a lower prevalence of hepatitis B virus (HBV)‐positive liver background (p = 0.001) compared with those of the low‐serum bile acids group. The high‐serum bile acids group exhibited low levels of PLT (p < 0.001), PT (p = 0.027), and ALB (p = 0.001) and a low prevalence of IM (p = 0.021). AST (p < 0.001), ALT (p = 0.001), and ICG‐R15 (p < 0.001) levels were higher in the high‐serum bile acids group. No significant differences were observed between the groups in tumor number, tumor histology, pathological vascular invasion, or operative factors.
TABLE 1.
Comparison of clinical characteristics of patients according to serum bile acid levels.
| Variables | All patients (n = 249) | High‐serum bile acids group (n = 148) | Low‐serum bile acids group (n = 101) | p‐value |
|---|---|---|---|---|
| Sex: M/F (M ratio) | 197/52 (79.1) | 107/41 (72.3) | 90/11 (89.1) | 0.001 |
| Age (year) | 71 (64–77) | 71 (65–78) | 71 (63–77) | 0.143 |
| BMI (kg/m2) | 23.0 (20.7–25.0) | 22.7 (20.6–24.7) | 23.3 (21.1–25.5) | 0.282 |
| DM (+) | 102 (41.0) | 58 (39.2) | 44 (43.6) | 0.491 |
| Smoking (+) | 162 (65.3) | 86 (58.5) | 76 (75.3) | 0.007 |
| Liver background: HCV | 123 (49.3) | 89 (60.1) | 34 (33.7) | < 0.001 |
| Liver background: HBV | 40 (16.1) | 14 (9.5) | 26 (25.7) | 0.001 |
| Liver background: Alcohol | 27 (10.8) | 17 (11.5) | 10 (9.9) | 0.693 |
| Preoperative UDCA use (+) | 97 (39.0) | 81 (54.7) | 16 (15.8) | < 0.001 |
| PLT (×104/μL) | 14.7 (10.8–19.3) | 12.8 (9.4–18.2) | 16.8 (12.7–21.2) | < 0.001 |
| PT (%) | 87 (79–96) | 87 (77–94) | 89 (81–99) | 0.027 |
| T‐Bil (mg/dL) | 0.7 (0.6–1.0) | 0.7 (0.6–1.0) | 0.8 (0.6–1.0) | 0.261 |
| ALB (g/dL) | 4.1 (3.8–4.5) | 4.0 (3.7–4.4) | 4.2 (3.9–4.6) | 0.001 |
| AST (U/L) | 33 (25–52) | 40 (29–62) | 28 (21–38) | < 0.001 |
| ALT (U/L) | 32 (21–51) | 37 (22–59) | 27 (17–39) | 0.001 |
| ICG‐R15 (%) | 12.8 (8.6–18.9) | 15.5 (9.5–20.7) | 10.8 (7.2–15.9) | < 0.001 |
| AFP (ng/mL) | 9.1 (4.8–54.5) | 11.2 (5.0–56.8) | 7.3 (3.3–51.1) | 0.097 |
| DCP (mAU/mL) | 72.0 (25.0–493.0) | 52.0 (23.0–583.3) | 95.0 (27.0–471.0) | 0.244 |
| Tumor number | 1 (1–2) | 1 (1–2) | 1 (1–2) | 0.199 |
| Tumor diameter (mm) | 25.0 (17.0–43.5) | 24.5 (16.0–40.0) | 30.0 (18.0–50.5) | 0.051 |
| Tumor capsule (+) | 189 (77.1) | 116 (79.5) | 73 (73.4) | 0.296 |
| Histology: poorly differentiation | 27 (10.8) | 16 (10.8) | 11 (10.9) | 0.984 |
| Vascular invasion: Vp (+) | 38 (15.3) | 22 (15.0) | 16 (15.8) | 0.851 |
| Vascular invasion: Vv (+) | 14 (5.7) | 9 (6.2) | 5 (5.0) | 0.676 |
| IM (+) | 18 (7.3) | 6 (4.1) | 12 (11.9) | 0.021 |
| Open/laparoscopic (open ratio) | 223/26 (89.6) | 135/13 (91.2) | 88/13 (87.1) | 0.300 |
| Anatomical resection or major hepatectomy | 188 (75.5) | 106 (71.6) | 82 (81.2) | 0.085 |
| Operative time (min) | 317 (252–386) | 314 (240–379) | 323 (270–403) | 0.320 |
| Intraoperative blood loss (mL) | 361 (170–623) | 400 (183–641) | 333 (160–605) | 0.322 |
| Intraoperative blood transfusion (+) | 17 (6.8) | 13 (8.8) | 4 (4.0) | 0.201 |
| Postoperative complications CD ≥ 3 | 33 (13.3) | 20 (13.5) | 13 (12.9) | 0.883 |
Note: The variables in bold are statistically significant (p < 0.05). Continuous variables are expressed as medians (interquartile ranges). Qualitative variables are expressed as numbers (%).
Abbreviations: AFP, α‐fetoprotein; ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; CD, Clavien‐Dindo; DCP, des‐gamma‐carboxy prothrombin; DM, diabetes mellitus; F, female; HBV, hepatitis B virus; HCV, hepatitis C virus; ICG‐R15, indocyanine green dye retention rate at 15 min; IM, intrahepatic metastasis; M, male; PLT, platelet; PT, prothrombin time; T‐Bil, total‐bilirubin; UDCA, ursodeoxycholic acid; Vp, portal vein invasion; Vv, hepatic venous invasion.
3.2. Relation Among Liver Fibrosis, Necro‐Inflammatory Reaction, and Serum Bile Acids, and Comparison of Predictive Ability for Intrahepatic Recurrence
Figure 2a,b summarize the relation between serum bile acid levels and liver fibrosis stage or necro‐inflammatory reaction, using the new Inuyama Classification. The proportions of patients with severe liver fibrosis and severe necro‐inflammatory reaction were significantly higher in the high‐serum bile acids group than in the low‐serum bile acids group (liver fibrosis: p < 0.001; Figure 2a; necro‐inflammatory reaction: p < 0.001; Figure 2b). Figure 2c–e summarizes the relation between blood‐based markers for liver fibrosis and serum bile acid levels. The values of these blood‐based markers for liver fibrosis were significantly higher in the high‐serum bile acids group than in the low‐serum bile acids group (FIB‐4 index: 2.11 vs. 3.91, p < 0.001; SAFE score: 122.8 vs. 214.5, p < 0.001; APRI: 0.40 vs. 0.83, p < 0.001; Figure 2c–e). In this study, 162 (65.1%) patients experienced recurrence, of whom 155 (95.7%) had intrahepatic recurrence. Figure 2f presents a comparison of ROC curve analyses for serum bile acids and blood‐based markers for liver fibrosis in predicting intrahepatic recurrence after surgery. The AUROC value of serum bile acids was 0.649, which was higher than that of the other indices (FIB‐4 index, 0.576; SAFE score, 0.580; APRI, 0.602).
FIGURE 2.
(a–e) Association between serum bile acids and the new Inuyama Classification or blood‐based markers for liver fibrosis. (f) Comparison of the areas under the receiver operating characteristic curves for intrahepatic recurrence prediction between serum bile acids and blood‐based marker for liver fibrosis.
3.3. Association of Serum Bile Acids With OS and Intrahepatic Recurrence in Patients With HCC Following Initial Curative Hepatectomy
Figure 3 summarizes the Kaplan–Meier analyses showing the association between serum bile acid levels and OS or postoperative intrahepatic recurrence rate. The high‐serum bile acids group was associated with poorer OS (p = 0.025, log‐rank test; Figure 3a) and a higher postoperative intrahepatic recurrence rate (p < 0.001, log‐rank test; Figure 3b). For early intrahepatic recurrence within 2 years after hepatectomy, the high‐serum bile acids group was not significantly associated with a high intrahepatic recurrence rate (p = 0.088, log‐rank test; Figure 3c); however, they were significantly associated with a high intrahepatic recurrence rate at > 2 years postoperatively (p < 0.001, log‐rank test; Figure 3d). Figure S1 summarizes the Kaplan–Meier curves for intrahepatic recurrence rates after landmark time (2 years after hepatectomy), stratified by blood‐based markers for liver fibrosis. Although not significant, the high‐risk group tended to have a high intrahepatic recurrence rate after 2 years postoperatively (SAFE score: p = 0.078, log‐rank test; Figure S1a, FIB‐4 index: p = 0.058, log‐rank test; Figure S1b, APRI: p = 0.051, log‐rank test; Figure S1c). Figure S2 summarizes the Kaplan–Meier analyses of intrahepatic recurrence rates stratified by underlying liver disease. The high‐serum bile acids group was associated with a high postoperative intrahepatic recurrence rate in patients with both non‐viral and viral liver background. (Non‐viral: p = 0.037, log‐rank test; Figure S2a; viral: p < 0.001, log‐rank test; Figure S2b). Regarding viral liver background, the high‐serum bile acids group was associated with a higher postoperative intrahepatic recurrence rate only in patients who were HCV‐positive (HBV‐positive: p = 0.366, log‐rank test; Figure S2c; HCV‐positive: p = 0.004, log‐rank test; Figure S2d). For HCV‐positive patients, the treatment period was divided into two groups (2010–2014 and 2015–2018) for analysis. Patients who underwent hepatectomy between 2015 and 2018 achieved a significantly higher sustained virological response (SVR) rate than those who underwent hepatectomy between 2010 and 2014 (25.0 vs. 48.6%, p = 0.011; Table S1). The high‐serum bile acids group was associated with a higher postoperative intrahepatic recurrence rate in both groups. (2010–2014 group: p = 0.049, log‐rank test; Figure S3a; 2015–2018 group: p = 0.023, log‐rank test; Figure 3b). PSM analysis revealed no significant differences in clinical characteristics between the two groups (Table S2 and Figure S4). Figure 4 summarizes the Kaplan–Meier analyses showing the association between serum bile acid levels and OS or postoperative intrahepatic recurrence rate after PSM analysis. Although there was no significant difference in OS between the two groups (p = 0.384, log‐rank test; Figure 4a), the postoperative intrahepatic recurrence rate in the high‐serum bile acids group was significantly higher than in the low‐serum bile acids group (p = 0.036, log‐rank test; Figure 4b). Table S3 summarizes the results of the univariate and multivariate analyses of prognostic factors for OS. Although high serum bile acid levels (p = 0.027) were associated with poorer OS in the univariate analysis, they were not independent prognostic factors in the multivariate analysis (p = 0.248). Table 2 summarizes the results of the univariate and multivariate analyses of prognostic factors for postoperative intrahepatic recurrence. In the univariate analysis, high serum bile acid levels (p < 0.001) were identified as poor prognostic factors, along with liver fibrosis, PLT, tumor number, and surgical factors. In the multivariate analysis, three factors were identified as prognostic for postoperative intrahepatic recurrence: serum bile acids ≥ 10 μmol/L (hazard ratio [HR] = 1.73; 95% confidence interval [CI] = 1.16–2.58, p = 0.007), PLT < 15 × 104/μL (HR = 1.94; 95% CI = 1.34–2.81, p < 0.001), and tumor number ≥ 3 (HR = 2.56; 95% CI = 1.60–4.07, p < 0.001).
FIGURE 3.
(a,b) Kaplan–Meier curves for overall survival and intrahepatic recurrence rates comparing high‐ and low‐serum bile acids groups. (c) Kaplan–Meier curves for 2‐year intrahepatic recurrence rates comparing high‐ and low‐serum bile acids groups. (d) Kaplan–Meier curves for intrahepatic recurrence rates at the landmark time (2 years after hepatectomy) comparing the high‐ and low‐serum bile acids groups.
FIGURE 4.
(a, b) Kaplan–Meier curves for overall survival and intrahepatic recurrence rates comparing high‐ and low‐serum bile acids groups after propensity score matching.
TABLE 2.
Univariate and multivariate analyses of prognostic factors for postoperative intrahepatic recurrence.
| Variables | N = 249 | Univariate | 95% CI | p‐value | Multivariate | 95% CI | p‐value |
|---|---|---|---|---|---|---|---|
| HR | HR | ||||||
| Male | 197 | 0.89 | 0.59–1.33 | 0.559 | |||
| Age (year) ≥ 75 | 90 | 0.99 | 0.70–1.39 | 0.949 | |||
| BMI (kg/m2) ≥ 25 | 63 | 1.09 | 0.77–1.55 | 0.614 | |||
| Liver background: HCV | 123 | 1.30 | 0.95–1.78 | 0.104 | |||
| Liver background: HBV | 40 | 0.72 | 0.46–1.11 | 0.135 | |||
| New Inuyama classification: F ≥ 3 | 127 | 1.97 | 1.39–2.77 | < 0.001 | 1.29 | 0.88–1.88 | 0.186 |
| New Inuyama classification: A ≥ 2 | 172 | 1.46 | 0.94–2.26 | 0.090 | |||
| DM (+) | 102 | 1.03 | 0.75–1.43 | 0.841 | |||
| Smoking (+) | 162 | 1.07 | 0.76–1.49 | 0.704 | |||
| Preoperative UDCA use (+) | 97 | 1.47 | 1.06–2.02 | 0.019 | 1.05 | 0.72–1.53 | 0.784 |
| Bile acids (μmol/L) ≥ 10 | 148 | 2.02 | 1.43–2.85 | < 0.001 | 1.73 | 1.16–2.58 | 0.007 |
| PLT (×104/μL) < 15 | 129 | 1.81 | 1.31–2.50 | < 0.001 | 1.94 | 1.34–2.81 | < 0.001 |
| PT (%) ≤ 70 | 23 | 1.17 | 0.67–2.02 | 0.584 | |||
| T‐Bil (mg/dL) ≥ 1.5 | 12 | 1.15 | 0.57–2.35 | 0.696 | |||
| ALB (g/dL) ≤ 3.5 | 27 | 1.60 | 0.95–2.69 | 0.078 | |||
| ICG‐R15 (%) ≥ 10 | 164 | 1.31 | 0.93–1.84 | 0.122 | |||
| AFP (ng/mL) > 20 | 96 | 1.02 | 0.73–1.41 | 0.914 | |||
| DCP (mAU/mL) ≥ 40 | 151 | 1.34 | 0.97–1.86 | 0.078 | |||
| Tumor number ≥ 3 | 34 | 3.01 | 1.98–4.57 | < 0.001 | 2.56 | 1.60–4.07 | < 0.001 |
| Tumor diameter (mm) ≥ 50 | 51 | 0.98 | 0.66–1.48 | 0.940 | |||
| Tumor capsule (+) | 189 | 0.83 | 0.57–1.21 | 0.336 | |||
| Histology: poorly differentiation | 27 | 1.10 | 0.66–1.82 | 0.708 | |||
| Vascular invasion: Vp (+) | 38 | 1.18 | 0.75–1.86 | 0.468 | |||
| Vascular invasion: Vv (+) | 14 | 1.02 | 0.49–2.12 | 0.967 | |||
| IM (+) | 18 | 1.12 | 0.59–2.14 | 0.720 | |||
| Open surgery | 223 | 1.70 | 0.94–3.07 | 0.077 | |||
| Anatomical resection or major hepatectomy | 188 | 1.12 | 0.77–1.63 | 0.543 | |||
| Operative time (min) > 315 | 126 | 1.41 | 1.03–1.94 | 0.034 | 1.42 | 0.92–2.19 | 0.111 |
| Intraoperative blood loss (mL) > 360 | 125 | 1.63 | 1.18–2.24 | 0.003 | 1.15 | 0.76–1.74 | 0.517 |
| Intraoperative blood transfusion (+) | 17 | 1.40 | 0.68–2.87 | 0.360 | |||
| Postoperative complications CD ≥ 3 | 33 | 1.46 | 0.88–2.43 | 0.143 |
Note: The variables in bold are statistically significant (p < 0.05).
Abbreviations: AFP, α‐fetoprotein; ALB, albumin; BMI, body mass index; CD, Clavien‐Dindo; CI, confidence interval; DCP, des‐gamma‐carboxy prothrombin; DM, diabetes mellitus; HBV, hepatitis B virus; HCV, hepatitis C virus; HR, hazard ratio; ICG‐R15, indocyanine green dye retention rate at 15 min; IM, intrahepatic metastasis; PLT, platelet; PT, prothrombin time; T‐Bil, total‐bilirubin; UDCA, ursodeoxycholic acid; Vp, portal vein invasion; Vv, hepatic venous invasion.
3.4. Relation Between Serum Bile Acid Levels and the Risk of Intrahepatic Recurrence in Patients With HCC Following Initial Curative Hepatectomy
Figure 5 shows the relation between serum bile acid levels and the risk of intrahepatic recurrence using an RCS curve. Figure 5a shows the results for all patients in this study. There was a significant association between bile acid levels and the risk of intrahepatic recurrence (Poverall < 0.001). Changes in HR and 95% CI for intrahepatic recurrence after hepatectomy according to different bile acid levels are shown in Table S4. The nonlinearity test was also significant (P nonlinearity = 0.006). The relation between serum bile acid levels and intrahepatic recurrence was J‐shaped, with a nearly equivalent HR for intrahepatic recurrence when bile acid levels were ≥ 19.81 μmol/L. Figure 5b,c showed the results of the RCS analysis for patients divided according to preoperative UDCA use. Although there was no significant overall association between bile acid levels and the risk of intrahepatic recurrence in patients receiving preoperative UDCA (P overall = 0.214; Figure 5b), a significant overall association was observed in patients without preoperative UDCA use (Poverall = 0.014; Figure 5c).
FIGURE 5.
Nonlinear association between postoperative serum bile acid levels and the risk of intrahepatic recurrence using a restricted cubic spline curve.
4. Discussion
This retrospective study investigated the effect of serum bile acid levels on the long‐term prognosis of patients who underwent initial curative hepatectomy for HCC. Compared with patients who had low serum bile acid levels, those with high serum bile acid levels included more patients with a HCV‐positive liver background and low liver reserve function. High serum bile acid levels were also associated with severe liver fibrosis and necro‐inflammatory reactions according to the new Inuyama Classification. In addition, high serum bile acid levels were associated with a high intrahepatic recurrence rate in patients with HCC after initial curative hepatectomy, and this result remained consistent even after adjusting for clinical characteristics. However, once bile acid levels exceeded 19.81 μmol/L, the risk of postoperative intrahepatic recurrence plateaued. This finding may be explained by the mixed presence of patients who received preoperative UDCA. Elevated serum bile acid levels indicate bile acids accumulation in the liver, suggesting that bile acids homeostasis via the gut–liver axis is disrupted. Considering the effects of bile acids on hepatocytes in addition to their digestive functions, which is important to interpret the results of this study.
Bile acids are the main components of bile, and elevated serum bile acid levels are caused by defects in canalicular transport or obstruction of bile flow [20]. These pathologies are caused not only by biliary obstruction but also by drugs, viral hepatitis, alcohol, primary biliary cholangitis, and primary sclerosing cholangitis [21, 22]. Serum bile acids elevation in viral and alcoholic hepatitis, which accounted for a major proportion of the underlying liver diseases in this study, has been reported to result from bile acids accumulation in hepatocytes because of impaired bile excretion caused by hepatocyte damage [23, 24]. In fact, in patients with high serum bile acid levels in this study, liver reserve function, including PLT, PT, ALB, and liver enzymes, was significantly decreased without an increase in bilirubin. Furthermore, a significantly high number of patients exhibited pathologically moderate or greater necro‐inflammatory reaction, which has not been previously reported. Serum bile acids may serve as indicators of the severity of liver damage or necro‐inflammation and as predictors of liver reserve function in patients with nonobstructive jaundice.
Although it is known that bile acids accumulate in hepatocytes because of hepatocyte injury in chronic hepatitis, it has also been demonstrated that the accumulation of bile acids themselves causes hepatocyte damage. Several mechanisms have been proposed for the hepatotoxicity of bile acids, which disrupt cell membranes through their detergent action on lipid components [25] and ultimately induce hepatocyte apoptosis [26]. Liver fibrosis represents the wound‐healing response to hepatocyte injury, and the degree of fibrosis affects the liver functional reserve. In viral hepatitis, which accounted for a large proportion of the underlying liver diseases in this study, serum bile acid levels have been reported to correlate with the degree of liver fibrosis [27, 28]. In this study, it was difficult to clarify whether elevated serum bile acid levels directly accelerated liver damage; however, the association between serum bile acid levels and the degree of liver fibrosis was consistent with previous findings. Moreover, the high‐serum bile acids group showed significant elevation in blood‐based markers for liver fibrosis, which are useful for predicting liver fibrosis. Because serum fibrosis markers are calculated from parameters, such as PLT and liver enzymes, which are themselves associated with serum bile acids, this finding is expected. Several studies have linked liver fibrosis and blood‐based marker for liver fibrosis to postoperative recurrence of HCC [6, 29, 30]. Therefore, serum bile acid levels may reflect the degree of liver fibrosis and be associated with an increased risk of postoperative intrahepatic recurrence of HCC. Serum bile acid levels also reflect findings of liver necroinflammation; therefore, as demonstrated by ROC analysis, they are expected to show higher predictive ability for postoperative intrahepatic recurrence than other serum fibrosis markers. While other blood‐based markers for liver fibrosis showed no significant association, serum bile acid levels were significantly associated with increasing intrahepatic recurrence rates at > 2 years after hepatectomy. These results indicate that serum bile acid levels are more accurately associated with late recurrence and are influenced by the carcinogenic environment of the background liver. Serum bile acids may help identify patients at a high risk of recurrence who would benefit from more rigorous surveillance. In addition, bile acids have been reported to contribute not only to liver injury but also to tumor‐promoting effects.
Certain bile acids have been reported to promote the development and proliferation of HCC cell lines [10]. In addition, serum bile acid levels are associated with an increased risk of HCC in patients with cirrhosis [15]. In this study, high serum bile acid levels did not increase the tumor size or number at the time of HCC resection; however, they significantly increased the postoperative intrahepatic recurrence rate in patients with non‐viral and viral liver backgrounds. In patients with an HBV‐positive liver background, no significant association was found between bile acid levels and the postoperative intrahepatic recurrence rate. However, because the limited sample size likely influences this, reanalysis after accumulating more cases is desirable. The SVR achievement rate for HCV infection has changed dramatically with the advent of direct‐acting antivirals (DAAs) treatment [31], and patients who underwent hepatectomy after 2015 showed a significantly higher SVR rate in this study. However, high serum bile acid levels were significantly associated with increased postoperative intrahepatic recurrence rates in HCV‐positive patients who underwent hepatectomy between 2010 and 2014 and 2015 and 2018. The association between serum bile acid levels and increased postoperative intrahepatic recurrence across different timing and characteristics suggests that serum bile acid levels can accurately predict intrahepatic recurrence after hepatectomy. Even after adjusting for clinical characteristics using PSM, the high‐serum bile acids group was associated with a high postoperative intrahepatic recurrence rate. Serum bile acid levels may be important predictors of postoperative intrahepatic recurrence. Although the risk of postoperative intrahepatic recurrence after hepatectomy for HCC remained almost unchanged when serum bile acid levels rose above 19.81 μmol/L, no significant association was observed in patients who received preoperative UDCA, whereas a significant association was present in those who did not. A report has indicated that UDCA administration increases serum bile acid levels [32]. In this study, patients receiving preoperative UDCA had significantly higher serum bile acid levels than those who did not, thereby weakening the predictive value of serum bile acids for intrahepatic recurrence after HCC resection (data not shown). These results suggest that the increase in serum bile acids concentration caused by UDCA administration in patients with a low risk of recurrence may make it difficult to accurately predict postoperative intrahepatic recurrence. Serum bile acid levels may thus serve as more sensitive indicators of postoperative intrahepatic recurrence in patients not receiving UDCA.
This study had several limitations. First, it was a retrospective, single‐center study with a limited sample size, and not all postoperative prognostic factors for HCC identified in previous studies could be included in the analysis. Second, serum bile acid levels are influenced by diet and are generally measured in the fasting state. In this study, measurements were generally performed during fasting; however, because of the retrospective design, it was difficult to confirm whether all patients were fasting at the time of blood examination. Third, many patients with poor liver function were taking UDCA, and their serum bile acid levels may have been further elevated. Therefore, it was difficult to distinguish patients who developed abnormal serum bile acid levels because of UDCA intake. Prospective studies under standardized conditions, including the timing of blood examinations and UDCA use, are required to confirm these results. Nevertheless, the findings of this study, which demonstrated that serum bile acids are reliable indicators of liver function and fibrosis as well as predictors of postoperative intrahepatic recurrence in patients with HCC following initial curative hepatectomy, are clinically important. Adding serum bile acid levels as a preoperative triage test enables the identification of patients at high risk of intrahepatic recurrence after hepatectomy. In particular, it allows planning long‐term imaging follow‐up to anticipate delayed postoperative intrahepatic recurrence.
5. Conclusion
High serum bile acid levels were associated with a high intrahepatic recurrence rate in patients with HCC following initial curative hepatectomy. Serum bile acids are linked to the intrahepatic environment, including fibrosis and necro‐inflammatory reaction, and may contribute to creating conditions conducive to tumor recurrence.
Author Contributions
Tomoaki Bekki: conceptualization, writing – original draft, investigation, formal analysis. Masahiro Ohira: funding acquisition, writing – review and editing, investigation. Yuki Imaoka: writing – review and editing, funding acquisition, methodology, investigation. Seiichi Simizu: writing – review and editing, investigation. Tomoyuki Abe: writing – review and editing, investigation. Shintaro Kuroda: funding acquisition, writing – review and editing, investigation. Hiroyuki Tahara: writing – review and editing, investigation. Kentaro Ide: writing – review and editing, investigation. Tsuyoshi Kobayashi: writing – review and editing, funding acquisition, investigation. Hideki Ohdan: funding acquisition, writing – review and editing, supervision.
Funding
This work was supported in part by JSPS KAKENHI (JP25H01061, JP23H02981, JP24K02522, JP24K11931, and JP22K16535) and AMED (JP25fk0210165). The funders had no role in the study design, data collection and analysis, decision to publish or manuscript preparation.
Ethics Statement
There is no need for consent to participate to be obtained due to retrospective study. Administrative permissions were not required to access and use the medical records described in this study. The opt‐out method to obtain patient consent was utilized. This study was authorized in advance by the institutional review board of the Hiroshima University Hospital (approval number: E2019‐1580).
Conflicts of Interest
The authors declare no conflicts of interest. Hideki Ohdan is Editorial Bord Members of Annals of Gastroenterological Surgery.
Supporting information
Figure S1: Kaplan–Meier curves for intrahepatic recurrence rates at the landmark time 2 years after hepatectomy, stratified by blood‐based markers for liver fibrosis.
Figure S2: (a–d) Kaplan–Meier curves for intrahepatic recurrence rates in patients with different underlying liver diseases, comparing high‐ and low‐serum bile acid groups.
Figure S3: (a, b) Kaplan–Meier curves for intrahepatic recurrence rates at different surgical timings in HCV‐positive patients, comparing high‐ and low‐serum bile acid groups.
Figure S4: Association between serum bile acid levels and the new Inuyama Classification after propensity score‐matching.
Table S1: Comparison of sustained virological response ratios in HCV‐positive patients according to timing of hepatectomy.
Table S2: Comparison of clinical characteristics of patients according to serum bile acid levels after propensity score‐matching.
Table S3: Univariate and multivariate analyses of prognostic factors for overall survival.
Table S4: Changes in HR and 95% CI for intrahepatic recurrence after hepatectomy according to serum bile acid levels.
Acknowledgments
We would like to thank Editage (www.editage.com) for the English language editing.
References
- 1. Bray F., Laversanne M., Sung H., et al., “Global Cancer Statistics 2022: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 74, no. 3 (2024): 229–263. [DOI] [PubMed] [Google Scholar]
- 2. Hirokawa F., Hayashi M., Miyamoto Y., et al., “Appropriate Treatment Strategy for Intrahepatic Recurrence After Curative Hepatectomy for Hepatocellular Carcinoma,” Journal of Gastrointestinal Surgery 15, no. 7 (2011): 1182–1187. [DOI] [PubMed] [Google Scholar]
- 3. Harimoto N., Tsukagoshi M., Seki T., et al., “Predictors for Early Recurrence Beyond Up‐To‐7 or Distant Metastasis After Hepatocellular Carcinoma Resection: Proposal for Borderline Resectable HCC,” International Journal of Clinical Oncology 29, no. 2 (2024): 195–204. [DOI] [PubMed] [Google Scholar]
- 4. Harimoto N., Yoshizumi T., Inokuchi S., et al., “Prognostic Significance of Preoperative Controlling Nutritional Status (CONUT) Score in Patients Undergoing Hepatic Resection for Hepatocellular Carcinoma: A Multi‐Institutional Study,” Annals of Surgical Oncology 25, no. 11 (2018): 3316–3323. [DOI] [PubMed] [Google Scholar]
- 5. Tung‐Ping Poon R., Fan S. T., and Wong J., “Risk Factors, Prevention, and Management of Postoperative Recurrence After Resection of Hepatocellular Carcinoma,” Annals of Surgery 232, no. 1 (2000): 10–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ko S., Kanehiro H., Hisanaga M., Nagao M., Ikeda N., and Nakajima Y., “Liver Fibrosis Increases the Risk of Intrahepatic Recurrence After Hepatectomy for Hepatocellular Carcinoma,” British Journal of Surgery 89, no. 1 (2002): 57–62. [DOI] [PubMed] [Google Scholar]
- 7. Miyata T., Hayashi H., Yamashita Y. I., et al., “The Impact of Histologic Liver Inflammation on Oncology and the Prognosis of Patients Undergoing Hepatectomy for Hepatocellular Carcinoma,” Annals of Surgical Oncology 29, no. 2 (2022): 893–902. [DOI] [PubMed] [Google Scholar]
- 8. Attili A. F., Angelico M., Cantafora A., Alvaro D., and Capocaccia L., “Bile Acid‐Induced Liver Toxicity: Relation to the Hydrophobic‐Hydrophilic Balance of Bile Acids,” Medical Hypotheses 19, no. 1 (1986): 57–69. [DOI] [PubMed] [Google Scholar]
- 9. Feng S., Xie X., Li J., et al., “Bile Acids Induce Liver Fibrosis Through the NLRP3 Inflammasome Pathway and the Mechanism of FXR Inhibition of NLRP3 Activation,” Hepatology International 18, no. 3 (2024): 1040–1052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Xie G., Wang X., Huang F., et al., “Dysregulated Hepatic Bile Acids Collaboratively Promote Liver Carcinogenesis,” International Journal of Cancer 139, no. 8 (2016): 1764–1775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Shen R., Ke L., Li Q., et al., “Abnormal Bile Acid‐Microbiota Crosstalk Promotes the Development of Hepatocellular Carcinoma,” Hepatology International 16, no. 2 (2022): 396–411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Yang R. and Qian L., “Research on Gut Microbiota‐Derived Secondary Bile Acids in Cancer Progression,” Integrative Cancer Therapies 21 (2022): 15347354221114100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Wang H., Shang X., Wan X., et al., “Increased Hepatocellular Carcinoma Risk in Chronic Hepatitis B Patients With Persistently Elevated Serum Total Bile Acid: A Retrospective Cohort Study,” Scientific Reports 6 (2016): 38180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Thomas C. E., Luu H. N., Wang R., et al., “Association Between Pre‐Diagnostic Serum Bile Acids and Hepatocellular Carcinoma: The Singapore Chinese Health Study,” Cancers 13, no. 11 (2021): 2648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. El‐Serag H. B., Thrift A. P., Duong H., et al., “Serum Levels of Total Bile Acids Are Associated With an Increased Risk of HCC in Patients With Cirrhosis,” Hepatology Communications 8, no. 11 (2024): e0545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Ichida F., Tsuji T., Omata M., et al., “New Inuyama Classification; New Criteria for Histological Assessment of Chronic Hepatitis,” International Hepatology Communications 6 (1996): 112–119. [Google Scholar]
- 17. Sterling R. K., Lissen E., Clumeck N., et al., “Development of a Simple Noninvasive Index to Predict Significant Fibrosis in Patients With HIV/HCV Coinfection,” Hepatology 43, no. 6 (2006): 1317–1325. [DOI] [PubMed] [Google Scholar]
- 18. Wai C. T., Greenson J. K., Fontana R. J., et al., “A Simple Noninvasive Index Can Predict Both Significant Fibrosis and Cirrhosis in Patients With Chronic Hepatitis C,” Hepatology 38, no. 2 (2003): 518–526. [DOI] [PubMed] [Google Scholar]
- 19. Sripongpun P., Kim W. R., Mannalithara A., et al., “The Steatosis‐Associated Fibrosis Estimator (SAFE) Score: A Tool to Detect Low‐Risk NAFLD in Primary Care,” Hepatology 77, no. 1 (2023): 256–267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hofmann A. F., “Bile Acids: The Good, the Bad, and the Ugly,” Physiology 14, no. 1 (1999): 24–29. [DOI] [PubMed] [Google Scholar]
- 21. Jüngst C., Berg T., Cheng J., et al., “Intrahepatic Cholestasis in Common Chronic Liver Diseases,” European Journal of Clinical Investigation 43, no. 10 (2013): 1069–1083. [DOI] [PubMed] [Google Scholar]
- 22. Pollheimer M. J., Fickert P., and Stieger B., “Chronic Cholestatic Liver Diseases: Clues From Histopathology for Pathogenesis,” Molecular Aspects of Medicine 37 (2014): 35–56. [DOI] [PubMed] [Google Scholar]
- 23. Trauner M., Meier P. J., and Boyer J. L., “Molecular Pathogenesis of Cholestasis,” New England Journal of Medicine 339, no. 17 (1998): 1217–1227. [DOI] [PubMed] [Google Scholar]
- 24. Eller C., Heydmann L., Colpitts C. C., Verrier E. R., Schuster C., and Baumert T. F., “The Functional Role of Sodium Taurocholate Cotransporting Polypeptide NTCP in the Life Cycle of Hepatitis B, C and D Viruses,” Cellular and Molecular Life Sciences 75, no. 21 (2018): 3895–3905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Billington D., Evans C. E., Godfrey P. P., and Coleman R., “Effects of Bile Salts on the Plasma Membranes of Isolated Rat Hepatocytes,” Biochemical Journal 188, no. 2 (1980): 321–327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Sokol R. J., Straka M. S., Dahl R., et al., “Role of Oxidant Stress in the Permeability Transition Induced in Rat Hepatic Mitochondria by Hydrophobic Bile Acids,” Pediatric Research 49, no. 4 (2001): 519–531. [DOI] [PubMed] [Google Scholar]
- 27. Shlomai A., Halfon P., Goldiner I., et al., “Serum Bile Acid Levels as a Predictor for the Severity of Liver Fibrosis in Patients With Chronic Hepatitis C,” Journal of Viral Hepatitis 20, no. 2 (2013): 95–102. [DOI] [PubMed] [Google Scholar]
- 28. Wang X., Xie G., Zhao A., et al., “Serum Bile Acids Are Associated With Pathological Progression of Hepatitis B‐Induced Cirrhosis,” Journal of Proteome Research 15, no. 4 (2016): 1126–1134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Toyoda H., Kumada T., Tada T., Kaneoka Y., and Maeda A., “A Laboratory Marker, FIB‐4 Index, as a Predictor for Long‐Term Outcomes of Hepatocellular Carcinoma Patients After Curative Hepatic Resection,” Surgery 157, no. 4 (2015): 699–707. [DOI] [PubMed] [Google Scholar]
- 30. Bekki T., Ohira M., Imaoka Y., et al., “The Steatosis‐Associated Fibrosis Estimator Score Is a Useful Indicator of Recurrence and Survival After Initial Curative Hepatectomy for Hepatocellular Carcinoma,” Annals of Gastroenterological Surgery 9, no. 1 (2025): 178–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Carrat F., Fontaine H., Dorival C., et al., “Clinical Outcomes in Patients With Chronic Hepatitis C After Direct‐Acting Antiviral Treatment: A Prospective Cohort Study,” Lancet 393, no. 10179 (2019): 1453–1464. [DOI] [PubMed] [Google Scholar]
- 32. Sinakos E., Marschall H. U., Kowdley K. V., Befeler A., Keach J., and Lindor K., “Bile Acid Changes After High‐Dose Ursodeoxycholic Acid Treatment in Primary Sclerosing Cholangitis: Relation to Disease Progression,” Hepatology 52, no. 1 (2010): 197–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Kaplan–Meier curves for intrahepatic recurrence rates at the landmark time 2 years after hepatectomy, stratified by blood‐based markers for liver fibrosis.
Figure S2: (a–d) Kaplan–Meier curves for intrahepatic recurrence rates in patients with different underlying liver diseases, comparing high‐ and low‐serum bile acid groups.
Figure S3: (a, b) Kaplan–Meier curves for intrahepatic recurrence rates at different surgical timings in HCV‐positive patients, comparing high‐ and low‐serum bile acid groups.
Figure S4: Association between serum bile acid levels and the new Inuyama Classification after propensity score‐matching.
Table S1: Comparison of sustained virological response ratios in HCV‐positive patients according to timing of hepatectomy.
Table S2: Comparison of clinical characteristics of patients according to serum bile acid levels after propensity score‐matching.
Table S3: Univariate and multivariate analyses of prognostic factors for overall survival.
Table S4: Changes in HR and 95% CI for intrahepatic recurrence after hepatectomy according to serum bile acid levels.