Exploring the Role of Thy‐1 and Its Soluble Form Regarding Fibrosis Severity in Primary Biliary Cholangitis and Metabolic Dysfunction‐Associated Fatty Liver Disease

ABSTRACT

Background and Aim

Portal myofibroblasts (PMF) are known to be critical in bile duct injury, but their role in liver fibrogenesis remains underexplored. Thy‐1, an adhesion molecule detected as soluble Thy‐1 (sThy‐1) in serum, is primarily expressed by portal myofibroblasts (PMF) and, to a lesser extent, hepatic stellate cells (HSC), serving as a marker for myofibroblast activity in liver pathology. This study aimed to elucidate the correlation between Thy‐1 expression in liver histology and sThy‐1 levels in serum with the degree of liver fibrosis in patients with primary biliary cholangitis (PBC) and metabolic dysfunction‐associated steatotic liver disease (MASLD).

Methods

Liver histology samples were stained for Thy‐1, and sThy‐1 was measured using an enzyme‐linked immunosorbent assay (ELISA).

Results

In patients with PBC and MASLD, an association between portal Thy‐1 expression and the extent of fibrosis was observed. Notably, while sinusoidal Thy‐1 expression aligned with fibrosis severity in PBC (p < 0.001), this association was not statistically significant in MASLD (p = 0.059). Moreover, variations in soluble Thy‐1 levels paralleled the progression from mild to advanced fibrosis stages in PBC.

Conclusions

Thy‐1 expression levels were associated with the severity of fibrosis, which could support its role in monitoring the presence and activity of myofibroblasts in liver diseases.

Abbreviations

CI

95% confidence interval

ELISA

enzyme‐linked immunosorbent assay

HC

healthy controls

IgM

immunoglobulin M

INR

International normalized ratio

MASLD

metabolic dysfunction‐associated steatotic liver disease

ns

not significant

OR

odds ratio

PBC

primary biliary cholangitis

α‐SMA

smooth muscle actin antibody

sThy‐1

soluble Thy1

1. Introduction

Liver cirrhosis is the common endpoint of various chronic liver injuries and critically determines the prognosis of chronic liver diseases [1]. Myofibroblasts are central to the fibrotic process, functioning as matrix‐producing cells with contractile properties [2, 3]. Hepatic stellate cells (HSC) and portal myofibroblasts (PMF) are the two primary myofibroblast sources in the liver [4, 5, 6]. While the role of HSC has been studied more extensively [7, 8], the importance of PMF has only more recently come to attention [9]. PMFs are crucial in matrix deposition specifically in the portal areas, whereas HSCs play a key role in the overall initiation, progression, and regression of liver fibrosis [10]. However, the role of PMFs remains poorly defined because of a lack of markers clearly distinguishing this cell population from HSCs at the stage of myofibroblasts [2, 9, 11, 12]. Thymocyte differentiation antigen 1 (Thy‐1, or CD90) is recognized as a surface marker for liver portal myofibroblasts (PMF), aiding in the differentiation of cell sources involved in liver pathology without directly implying a causative role in fibrogenesis [2, 13, 14, 15]. Furthermore, Thy‐1 overexpression correlates with markers of myofibroblast presence in areas affected by fibrosis, serving as a potential indicator of disease progression [16]. The effects of Thy‐1 still need to be explored, as they vary between cell types, tissues, and organs, and between similar tissues in different species [17], with pro‐fibrogenic functions of Thy1+ fibroblasts observed in some studies [18] and anti‐fibrogenic effects in others [19, 20, 21].

The shedding of Thy‐1 from fibroblast surfaces, correlating with a shift toward a profibrogenic myofibroblast phenotype, has been a focus of recent studies [22, 23]. An enzyme‐linked immunosorbent assay (ELISA) for detecting soluble Thy‐1 (sThy‐1) has revealed elevated levels in systemic fibrotic diseases, correlating with disease severity [24]. However, the relevance of sThy‐1 in liver diseases remains unexplored.

Primary biliary cholangitis (PBC) is a chronic autoimmune liver disease characterized by the progressive destruction of small bile ducts, leading to cholestasis and a biliary type of fibrosis.

Metabolic dysfunction‐associated liver disease (MASLD) encompasses a spectrum of liver conditions associated with the metabolic syndrome. In addition to liver inflammation, which is linked to increased mortality [25], MASLD is associated with a higher risk of developing extrahepatic cancers [25, 26]. The disease presents with a hepatocellular pattern of liver injury and perisinusoidal fibrosis that progressively extends to portal fields [27]. To assess liver fibrosis in MASLD and PBC, non‐invasive tools such as the Fibrosis‐4 Index (FIB‐4) [28] and, more recently, ALT levels [29] are used.

In the context of PBC and MASLD, two diseases with distinct fibrosis patterns, this study aims to investigate whether the presence of Thy‐1 in liver histology and its soluble form in serum across different fibrosis stages are different in the two diseases.

2. Patients and Methods

2.1. Patients and Samples

This retrospective monocentric study was conducted at the University Hospital of Bern, a tertiary care center in Switzerland. We included adult patients who received medical care between 2011 and 2022 with a diagnosis of PBC. Eligibility criteria included written informed consent and no concurrent chronic liver diseases. For the liver disease comparator group, we selected 49 patients with MASLD, ensuring a similar fibrosis grading distribution to the PBC patients. Patient samples were sourced from the Hepatology Biobank at the University Hospital Bern, Switzerland. All patients underwent liver biopsy concurrent with or within 2  years of their blood sample collection. The time point of liver biopsy was according to clinical indication, being at diagnosis or during follow‐up. Diagnosis criteria for PBC followed the guidelines of the European Association for the Study of the Liver [30]. Similarly, MASLD diagnosis conformed to the guidelines of the European Association for the Study of the Liver [27]. Definition of portal hypertension was either hepatic venous pressure gradient ≥ 5 mmHg, dilatation of the portal venous system on sonography, splenomegaly, platelets < 150 G/L, or esophageal varices. Control samples from healthy individuals matched the age and sex distribution of the PBC patients and were obtained from the blood donation center.

2.2. Laboratory Methods

For laboratory analysis, sThy‐1 levels were measured using an enzyme‐linked immunosorbent assay (ELISA) as described previously [31]. Liver samples underwent staining using a Thy‐1 antibody (CD90 [5E10] from BD Bioscience #550402) on a BOND Research RX staining platform from Leica Biosystems. The antibody dilution was 1:100, with a 30‐min incubation at room temperature and a 30‐min heat‐induced epitope retrieval in Tris EDTA buffer at 95°C. Additionally, an anti‐alpha‐smooth muscle actin antibody (α‐SMA) was used as a marker for myofibroblasts on liver histology samples. The α‐SMA antibody [1A4] from Sigma #A2547 was used on a BOND Max III in KPATH Diagnostics, with a dilution of 1:8000 and a 15‐min room temperature incubation without heat‐induced epitope retrieval. For fibrosis assessment, Masson's trichome staining was employed. All tissue samples were evaluated by an experienced liver pathologist using a semi‐quantitative approach based on four staining intensity categories: negative, weak, moderate, and strong (Figure 1). Furthermore, the pathologist described the distribution as being sinusoidal or periportal for Thy‐1 and α‐SMA, in order to evaluate the markers regarding co‐localization. The degree of fibrosis in PBC was categorized according to METAVIR. The degree of fibrosis in MASLD was categorized using the Brunt and EPOS classification systems [32, 33]. Patients were categorized into three disease groups based on fibrosis stages: Mild fibrosis (METAVIR F0/1 and Brunt 1/EPOS 1), moderate fibrosis (METAVIR F2 and Brunt 2/EPOS 2 and Brunt 3 if EPOS 3) and advanced fibrosis (METAVIR F3/4 and Brunt 3 if EPOS 4, Brunt 4/EPOS 5–6). FIB‐4 Index was categorized in low risk < 1.3, medium risk between 1.3 and 3.24, and high risk > 3.24 [34].

FIGURE 1.

Immunostaining of Thy‐1 and α‐SMA in liver sections. The images depict separate stainings for Thy‐1 and α‐SMA, highlighting their respective sinusoidal and periportal localizations. This approach allows for a detailed evaluation of marker co‐localization. The images demonstrate the presence of both markers in areas of active fibrogenesis, providing valuable insights into the fibrotic remodeling processes in the liver. The analysis was semi‐quantitative, categorized into four staining intensity levels: negative, weak, moderate, and strong. As no completely α‐SMA negative histology was observed, only the three α‐SMA positive intensities are shown.

2.3. Statistical Analysis

Categorical data are presented as numbers and percentages, while continuous data are expressed as median and interquartile range (IQR). We used the Chi‐square test or Fisher's exact test for comparing categorical data, depending on the data distribution. For quantitative data, the Wilcoxon rank sum test or the Kruskal–Wallis test was applied. We adjusted the alpha level for paired comparisons using the Bonferroni correction. Associations were assessed with ordinal logistic regression, reported as odds ratio (OR) and 95% confidence intervals (CI). Brant's test verified the proportionality of odds in our model. A logarithmic transformation of sThy‐1 measurements was applied to approximate a normal distribution, minimize the impact of outliers, and enhance interpretability. A p‐value < 0.05 (two‐tailed) was considered statistically significant. All analyzes were performed using R software, Version 4.1.2.

Clinical data were missing at random, as most, but not all parameters were available for all patients at the time point of sampling. Missing data is indicated separately in the results section.

2.4. Ethical Approval

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Cantonal Ethics Commission of Bern, Switzerland (approval number 2021–02237). All participants provided informed consent.

3. Results

3.1. Patient Characteristics

3.1.1. Study Inclusion

Among the 2809 patients in the Hepatology Biobank, 69 patients with PBC met the inclusion criteria. Forty nine patients with MASLD were identified and served as a comparative group, ensuring a similar fibrosis grading distribution to the PBC patients. The median time between liver biopsy and blood sample collection was 0 days (0;0) for PBC and MASLD patients.

3.1.2. Baseline Characteristics

Table 1 presents the baseline characteristics of the study population at the time of sample collection. There were no significant differences in bilirubin, liver stiffness measurements, histological fibrosis classification, or the presence of portal hypertension between the two groups (Table 1).

TABLE 1.

Baseline characteristics from patients with primary biliary cholangitis, metabolic dysfunction‐associated steatotic liver disease and healthy controls at the time point of sample acquisition.

	PBC n = 69	MASLD n = 49	Healthy controls n = 20	Missing values PBC/MASLD/HC	p
Demographics
Female sex, n (%)	57 (82.6)	19 (38.8)	18 (90)	0 (0)/0 (0)/0 (0)	< 0.001*
Age at baseline, years	59 (46–66)	55.0 (44.0–60.0)	53.5 (49.5–56.3)	0 (0)/0 (0)/0 (0)	0.091
Body mass index, kg/m2	26.3 (23–30)	30.7 (27.8–35.0)		0 (0)/0 (0)/20 (100)	< 0.001*
Liver assessment
ALT, ×ULN	1.7 (1–2.5)	1.3 (0.9–2.0)		0 (0)/0 (0)/20 (100)	0.203
Alkaline phosphatase, ×ULN	1.7 (1.1–2.8)	0.7 (0.5–1.0)		0 (0)/0 (0)/20 (100)	< 0.001*
gGT, ×ULN	5.5 (3.2–9.5)	1.5 (0.9–3.5)		0 (0)/0 (0)/20 (100)	< 0.001*
Bilirubin total, μmol/L (< 17 μmol/L)	11 (8–16)	11 (7.75–14)		0 (0)/1 (2)/20 (100)	0.541
INR (< 1.2)	1 (1–1.04)	1.03 (1–1.1)		0 (0)/0 (0)/20 (100)	0.008*
Immunoglobulin M g/L (0.4–2.3 g/L)	3.18 (2–4.87)	1.42 (0.76–1.75)		2 (2.9)/23 (47)/20 (100)	< 0.001*
Ferritin μg/L (10–120 μg/L)	109 (65.5–202)	286 (130–547)		26 (37.7)/4 (8.2)/20 (100)	0.002*
M30‐Apoptosens, U/L (< 200)	287 (183–423)	280.5 (172–567.8)		48 (69.6)/17 (34.7)/20 (100)	0.848
Fibroscan at baseline, kPa	7.95 (4.95–14.8)	8.8 (6.4–15.38)		3 (4.4)/5 (10.2)/20 (100)	0.273
Signs of portal hypertension, n (%)	15 (21.7)	11 (22.4)		0 (0)/0 (0)/20 (100)	1.000
Histological fibrosis grading
Mild fibrosis, n (%)	21 (30.4)	12 (24.5)		0 (0)/0 (0)/20 (100)	0.616
Moderate fibrosis, n (%)	22 (31.9)	22 (44.9)		0 (0)/0 (0)/20 (100)	0.212
Advanced fibrosis, n (%)	26 (37.6)	15 (30.6)		0 (0)/0 (0)/20 (100)	0.550

Note: Definition of portal hypertension: hepatic venous pressure gradient ≥ 5 mmHg, dilatation of the portal venous system on sonography, splenomegaly, platelets < 150 G/L or esophageal varices. Mild fibrosis = METAVIR F0/1 or Brunt 1/EPOS 1; moderate fibrosis = METAVIR F2 or Brunt 2/EPOS 2 and Brunt 3 if EPOS 3; advanced fibrosis = METAVIR F3/4 or Brunt 3 if EPOS 4, Brunt 4/EPOS 5–6. Categorical variables presented as n (%). Continuous data presented as median and interquartile range (IQR25‐IQR75). * indicates a significant p‐value (< 0.05).

Abbreviations: HC, healthy controls; MASLD, metabolic dysfunction‐associated steatotic liver disease; PBC, primary biliary cholangitis.

3.2. Liver Histology

Figure 1 illustrates the semi‐quantitative immunostaining intensities used for Thy‐1 and α‐SMA on liver histologies.

3.2.1. Thy‐1 Expression in PBC and MASLD Patients

Figure 2 illustrates the correlation between liver fibrosis and Thy‐1 expression intensity for PBC (Figure 2A) and MASLD (Figure 2B) patients. In PBC, a significant correlation was found (OR 10.37, 95% CI 4.30–30.36, p < 0.001) indicating a higher likelihood of advanced fibrosis, while in MASLD, the correlation between Thy‐1 expression and fibrosis grading was significant but with a lower effect size (OR 3.8, 95% CI 1.63–10.02, p = 0.004). Adjusting for sex and age at baseline did not alter these findings (Table 2).

FIGURE 2.

Intensity of Thy‐1 expression and fibrosis grading on liver histology in patients with PBC and MASLD. Significant correlation between the Thy‐1 intensity and the fibrosis grading in (A) patients with PBC (OR 10.37, 95% CI 4.30‐30.36, p < 0.0001) and (B) patients with MASLD (OR 3.8, 95% CI 1.63‐10.02, p = 0.004). MASLD, metabolic dysfunction‐associated steatotic liver disease; PBC, primary biliary cholangitis.

TABLE 2.

Assessment of the correlation between fibrosis grading and Thy‐1 intensity using ordinal logistic regression.

	Odds ratio	p	CI
Primary biliary cholangitis
Association between fibrosis grading and the following parameters
Thy‐1 expression overall	10.37	< 0.001*	4.30–30.36
Thy‐1 expression overall adjusted	10.78	< 0.001*	4.42–32.04
Thy‐1 expression portal	8.36	0.068	1.01–171.78
Thy‐1 expression portal adjusted	8.42	0.071	1.07–175.40
Thy‐1 expression sinusoidal	6.05	< 0.001*	2.30–17.11
Thy‐1 expression sinusoidal adjusted	6.48	< 0.001*	2.42–18.67
Metabolic dysfunction‐associated steatotic liver disease
Association between fibrosis grading and the following parameters
Thy‐1 expression overall	3.8	0.004*	1.63–10.02
Thy‐1 expression overall adjusted	3.46	0.008*	1.45–9.32
Thy‐1 expression portal	15.78	0.002*	3.10–103.73
Thy‐1 expression portal adjusted	16.84	0.002*	3.12–118.61
Thy‐1 expression sinusoidal	3.73	0.059	0.98–15.55
Thy‐1 expression sinusoidal adjusted	3.06	0.148	0.69–14.86

Note: There was no change in significance after adjusting for sex and age at baseline. * indicates a significant p‐value (< 0.05).

Abbreviations: CI, 95% confidence interval; α‐SMA, smooth muscle actin antibody.

3.2.2. Localization of Thy‐1 According to Fibrosis Stages

Figure 3 shows the portal and sinusoidal localization of the Thy‐1 expression in relation to fibrosis grading. In PBC, patients with mild fibrosis exhibited portal Thy‐1 expression in 85.7% (n = 18), increasing to 95.5% (n = 21) in the moderate fibrosis group, and reaching 100% (n = 24) in the advanced fibrosis group (Figure 3A). As the Thy‐1 expression was already high in the mild fibrosis group, the correlation with the increase in fibrosis grading was not significant (OR 8.36, 95% CI 1.01–171.78, p = 0.068). Notably, the four patients lacking portal Thy‐1 expression only demonstrated weak overall Thy‐1 intensity. Conversely, sinusoidal Thy‐1 expression in PBC was observed in 19% (n = 4) with mild fibrosis, 31.8% (n = 7) with moderate fibrosis, and 70.83% (n = 17) with advanced fibrosis (Figure 3C). This association was statistically significant (OR 6.05, 95% CI 2.3–17.11, p < 0.001), indicating an increase in sinusoidal Thy‐1 expression with advancing fibrosis.

FIGURE 3.

Portal and sinusoidal Thy‐1 expression on liver histology. Portal Thy‐1 expression was present in most patients with PBC (OR 8.36, CI 1.01‐171.78, p = 0.068) (A) and MASLD (OR 15.78, CI 3.10‐103.73, p = 0.002) (B) even in patients with mild fibrosis. Sinusoidal expression of Thy‐1 was significantly more present in patients with advanced fibrosis in patients with PBC (OR 6.05, CI 2.30‐17.11, p < 0.001) (C), but not significantly correlated in patients with MASLD (OR 3.73, CI 0.98–15.55, p = 0.059) (D). MASLD, metabolic dysfunction‐associated steatotic liver disease; PBC, primary biliary cholangitis; α‐SMA, smooth muscle actin antibody.

In MASLD, the detection of Thy‐1 in the portal area was present in 44.4% (n = 4) of patients with mild fibrosis, increasing to 86.4% (n = 19) in moderate fibrosis, and 100% (n = 14) in advanced fibrosis, as shown in Figure 3B. It is important to note that the eight patients who did not exhibit portal Thy‐1 expression demonstrated only weak Thy‐1 intensity. The association between portal Thy‐1 expression and fibrosis grading was significant (OR 15.78, 95% CI 3.10–103.73, p = 0.002). Sinusoidal Thy‐1 expression was present in 11.1% (n = 1) with mild fibrosis, 18.2% (n = 4) with moderate fibrosis, and 42.8% (n = 6) with advanced fibrosis (Figure 3D). This relationship between fibrosis grading and sinusoidal Thy‐1 expression in MASLD was not statistically significant (OR 3.73, 95% CI 0.98–15.55, p = 0.059). The associations did not change relevantly if adjusted for sex and age (Table 2).

One patient had only portal α‐SMA positivity. All other patients had α‐SMA expression portal and sinusoidal. Therefore, the Thy‐1 expression portal and sinusoidal is also reflecting the co‐localization of Thy‐1 with α‐SMA.

3.3. Soluble Thy‐1

A logarithmic transformation of sThy‐1 measurements was applied to approximate a normal distribution, minimize the impact of outliers, and enhance interpretability. The original measurements are available in Table S1.

3.3.1. Soluble Thy‐1 in Serum in Patients With PBC and MASLD According to Fibrosis Groups

In PBC patients, the median sThy‐1 levels varied across fibrosis groups, with the highest levels observed in the advanced fibrosis group, as shown in Figure 4. This group showed significantly higher sThy‐1 levels compared to the group with mild fibrosis (p = 0.043). No significant differences were noted between the other fibrosis groups, as detailed in Figure 4 and Table S2. The overall association between sThy‐1 and fibrosis grading was significant (OR 3.762, 95% CI 1.49–10.93, p = 0.008), persisting after adjusting for sex and age (OR 3.613, 95% CI 1.403–10.696, p = 0.012).

FIGURE 4.

Transformed soluble Thy‐1 measurements in relation to fibrosis groups in patients with PBC and MASLD. The median (IQR25‐IQR75) sThy‐1 measurements according to fibrosis groups in patients with PBC (A) and MASLD (B). Significant differences between groups are indicated. ns indicates a p‐value of ≥ 0.05, * indicates a p‐value of < 0.05, ** indicates a p‐value < 0.01, *** indicates a p‐value < 0.001. HC, healthy controls; MASLD, metabolic dysfunction‐associated steatotic liver disease; ns, not significant; PBC, primary biliary cholangitis.

In MASLD patients, the median sThy‐1 levels were more significantly different between the groups, detailed in Figure 4 and Table S2, with the highest median sThy‐1 levels in the advanced fibrosis group (8.455 pg/mL [8.255; 8.618]). The overall association between sThy‐1 and fibrosis grading was significant (OR 10.661, 95% CI 2.46–64.75, p = 0.005), persisting after adjusting for sex and age (OR 9.277, 95% CI 1.86–64.75, p = 0.013).

No significant differences were observed when directly comparing the sThy‐1 levels of PBC and MASLD patients across comparable fibrosis groups (data not shown). The FIB‐4 score showed a significant association with fibrosis groups and sThy‐1 levels in both PBC and MASLD (p < 0.001). In contrast, no association was found between fibrosis groups and ALT > 30 U/L (p = 0.513). However, ALT > 30 U/L was significantly associated with higher sThy‐1 measurements in PBC and MASLD (p < 0.001).

3.3.2. Soluble Thy‐1 in Serum in Patients With PBC and MASLD According to Thy‐1 Intensity on Liver Histology

In PBC, the median sThy‐1 levels were not significantly different across the different Thy‐1 intensity groups on liver histology, as shown in Figure 5 and Table S2. The overall association between sThy‐1 levels and Thy‐1 staining intensity, however, was significant (OR 2.833, 95% CI 1.11–7.77, p = 0.034), persisting after adjusting for sex and age (OR 2.799, 95% CI 1.08–7.79, p = 0.040). The correlation with the portal Thy‐1 expression alone remained insignificant (data not shown).

FIGURE 5.

Transformed soluble Thy‐1 measurements in relation to Thy‐1 intensities in patients with PBC and MASLD. The median (IQR25‐IQR75) sThy‐1 measurements according to Thy‐1 intensities in patients with PBC (A) and MASLD (B). Significant differences between groups are indicated. ns indicates a p‐value of ≥ 0.05, * indicates a p‐value of < 0.05, ** indicates a p‐value of < 0.01, *** indicates a p‐value of < 0.001. HC, healthy controls; MASLD, metabolic dysfunction‐associated steatotic liver disease; ns, not significant; PBC, primary biliary cholangitis.

In MASLD patients, the median sThy‐1 levels were not significantly different across the different Thy‐1 intensity groups on liver histology, as shown in Figure 5 and Table S2. For patients with MASLD, the overall association between sThy‐1 and Thy‐1 intensity was not significant (OR 2.688, 95% CI 0.71–12.47, p = 0.167), persisting after adjusting for sex and age (OR 2.202, 95% CI 0.47–12.57, p = 0.338). The correlation with the portal Thy‐1 expression alone remained insignificant (data not shown).

4. Discussion

Our study demonstrates a correlation between increased Thy‐1 expression and the degree of fibrosis, suggesting its potential as a marker for fibroblast activity in liver diseases. This correlation does not imply a causal role of Thy‐1 in the fibrogenesis itself. Notably, while PBC primarily involves biliary injury and MASLD hepatocellular injury, the observed associations of Thy‐1 expression with fibrosis stages were consistent across these differing pathologies, with more pronounced findings in PBC.

Our findings support those of Dezsὅ et al. [34], who identified Thy‐1 decorated myofibroblasts in models of both cholangiofibrosis and hepatotoxic cirrhosis, and extended these observations by demonstrating the significant involvement of Thy‐1‐positive cells even in the early stages of liver fibrogenesis [35]. The significant correlation between Thy‐1 expression and fibrosis severity in both PBC and MASLD underscores its potential role as an indicator of disease progression.

In patients with PBC, the Thy‐1 expression was portal in almost all patients, regardless of fibrosis grading. In patients with MASLD, however, the portal expression of Thy‐1 significantly increased with fibrosis progression. This might indicate that portal Thy‐1 positive cells are activated earlier in the context of a biliary type of injury, even in situations where fibrosis is not yet pronounced.

The presence of Thy‐1 positive cells in sinusoidal regions correlated with the severity of fibrosis in PBC patients, reflecting the increased presence of these cells in fibrotic livers. This observation should be interpreted as an association, highlighting potential areas of increased myofibroblast activity, rather than a direct mechanistic insight into fibrogenesis. This is in line with the data from rat models by Dudas et al., which suggest that Thy‐1 may be involved in the process of myofibroblast migration along the liver sinusoids [13]. In MASLD patients, the sinusoidal presence of Thy‐1 expressing cells increased with fibrosis progression, but was observed in a smaller proportion of patients and did not reach statistical significance. Dudas et al. had similar findings in a CCL₄‐induced liver injury model in rats, where Thy‐1 reaction was seen primarily in the periportal area and not significantly sinusoidal [13]. These findings are furthermore in line with the observation from Iwaisako et al., who found that activated PMFs were shown to comprise 70% of myofibroblasts at the onset of the bile duct ligation mouse model of cholestatic fibrosis, while HSCs were increasingly activated with fibrosis progression [12]. Our findings might therefore indicate that also in humans the Thy‐1 positive fibroblasts are more specifically activated and involved in the process of fibrogenesis in patients with a biliary type of liver injury.

Although Thy‐1 expression correlates with fibrosis severity, this relationship primarily supports its role in marking fibroblast presence. This observation could guide further studies to explore whether modulating Thy‐1 signaling pathways might be beneficial in therapeutic approaches to fibrosis, acknowledging that direct evidence of causality in fibrogenesis is not established.

α‐SMA was identified in both portal and sinusoidal areas across all stages of fibrosis in all patients, except for one patient, where it was exclusively portal. Similarly, Thy‐1 distribution varied, being detected in both portal and sinusoidal areas depending on the stage of fibrosis and disease context. This partial overlap with α‐SMA suggests that cells expressing Thy‐1 represent a distinct subset of the myofibroblast population in the liver [9]. These results are in line with previous findings by Dezsὅ [35] and Dudas and colleagues [13], who described α‐SMA staining exceeding Thy‐1 staining.

Significant differences in soluble Thy‐1 levels were observed between early and advanced stages of fibrosis, particularly noted in MASLD. The shedding mechanism of Thy‐1, while not fully understood, may be influenced by varied inflammatory activities within the liver, potentially more pronounced in MASLD [17]. Of note, despite the observed association between sThy‐1 levels and fibrosis grading, the variability of sThy‐1 levels suggests limitations in its utility as a non‐invasive surrogate marker for liver fibrosis. Furthermore, Thy‐1 is expressed not only in the liver and therefore, sThy‐1 measurements are not only representing liver origin. Several factors could contribute to this variability, including the heterogeneous nature of liver diseases and the complex biology of Thy‐1, which may have differential expression patterns and shedding mechanisms under varying pathological conditions [17]. Unexpectedly, the sThy‐1 measurements of healthy controls were significantly higher than those of MASLD patients with mild fibrosis. Initially, we considered whether this difference might be influenced by the higher proportion of women in the control group; however, adjusting for sex did not alter the findings. While the underlying cause remains unclear, it is important to note that the observed values fall within the expected assay‐specific variability, suggesting inter‐individual differences rather than measurement inconsistencies. Additionally, these findings may reflect baseline variability in sThy‐1 levels or contributions from non‐hepatic sources. This underscores the need for further validation of sThy‐1 as a clinical marker in larger, diverse cohorts to fully understand its physiological and pathological relevance. Our observations suggest an association of Thy‐1 positive cells with increasing fibrosis, notable in conditions with biliary injury but also observable in hepatocellular injury. This expansion, however, should not be interpreted as a direct causative role in fibrogenesis. This highlights the complexity of Thy‐1's involvement and the importance of further investigations to clarify its potential roles in liver diseases.

4.1. Strengths and Limitations of the Study

Our study contributes to understanding the associations of Thy‐1 expression with liver fibrosis by analyzing Thy‐1 in liver histologies and measuring soluble Thy‐1 across various fibrosis stages. While it represents the largest cohort of PBC and MASLD patients analyzed for these markers, it is crucial to note the observational nature of these findings, which do not imply direct causative roles. To our knowledge, this represents the largest study of patients with PBC and MASLD analyzed for Thy‐1 expression in the liver and the first study where histological Thy‐1 expression and soluble Thy‐1 measurements were conducted simultaneously in humans.

The significance of our study is underscored by the rarity of PBC as a liver disease and the challenges inherent in measuring sThy‐1 in experimental models [35].

A limitation is the lack of comprehensive characterization of Thy‐1 positive cells using additional markers, which could provide deeper insights into their biological roles beyond their association with fibrosis stages. Furthermore, the small sample sizes of the individual groups represent a limitation to our study. Additionally, it was not possible to retrospectively measure ELF parameters such as tissue inhibitor of metalloproteinases 1 and amino‐terminal propeptide of type III procollagen due to sample limitations, making prospective studies essential.

5. Conclusions

Our study highlights the association of Thy‐1 positive cells with stages of liver fibrosis, particularly in biliary injuries. The findings suggest that while sThy‐1 shows potential, its role as a non‐invasive biomarker requires further exploration to fully understand its utility in clinical diagnostics.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1. Soluble Thy‐1 measurements in relation to fibrosis groups and Thy‐1 intensities in patients with PBC and MASLD.

HC, healthy controls; MASLD, metabolic dysfunction‐associated steatotic liver disease; PBC, primary biliary cholangitis.

Table S2. Significance levels comparing median soluble Thy‐1 levels in the different fibrosis groups and according to Thy‐1 intensities in patients with PBC and MASLD.

Calculations were done with the Wilcoxon rank sum test. The alpha level for the paired comparisons was adjusted applying the Bonferroni correction. * indicates a significant p‐value (< 0.05).