Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Mar 24.
Published in final edited form as: Dig Dis Sci. 2019 Aug 12;65(2):524–533. doi: 10.1007/s10620-019-05760-x

Exploring the link between platelet numbers and vascular homeostasis across early and late stages of fibrosis in hepatitis C

Rabab O Ali 1,*, Mi Sun Moon 1,*, Elizabeth C Townsend 1, Kareen Hill 1, Grace Y Zhang 1, Alyson Bradshaw 1, Hannah Guan 1, Destanee Hamilton 1, David E Kleiner 2, Sungyoung Auh 3, Christopher Koh 1, Theo Heller 1
PMCID: PMC7988415  NIHMSID: NIHMS1537022  PMID: 31407130

Abstract

Background:

Thrombocytopenia is a hallmark of advanced liver disease. Platelets, growth factors (GFs), and vascular integrity, are closely linked factors in disease pathogenesis and their relationship, particularly in early disease stages is not entirely understood. The aim was to compare circulating platelets, growth factors, and vascular injury markers (VIMs) in Hepatitis C infected (HCV) patients with early fibrosis and cirrhosis.

Methods:

Reerospective evaluation of serum GFs and VIMs by ELISA from twenty-six HCV patients. Analytes from an earlier time-point were correlated with MELD at a later time-point.

Results:

Platelets and GFs decreased, and VIMs increased with fibrosis. Platelets correlated positively with PDGF-AA, PDGF-BB, TGFB1, EGF, P-selectin, and negatively with ICAM-3 and VCAM-1. P-selectin showed no correlations with VIMs but positively correlated with PDGF-AA, PDGF-BB, TGFB1, and EGF. Soluble VCAM-1 and ICAM-3 were linked to increasing fibrosis, liver enzymes, and synthetic dysfunction. Higher VCAM-1, and ICAM-3, and lower P-selectin at an earlier time-point were linked to higher MELD score at a later time-point.

Conclusion:

In chronic HCV, progressive decline in platelets and growth factors with fibrosis and their associations suggest that platelets are an important source of circulating GFs and influence GF decline with fibrosis. Enhanced markers of vascular injury in patients with early fibrosis suggests an earlier onset of endothelial dysfunction preceding cirrhosis. Associations of VIMs with platelets suggests a critical link between platelets and vascular homeostasis. Circulating markers of vascular injury may not only have prognostic importance but emphasize the role of vascular dysfunction in liver disease pathogenesis (NCT00001971).

Keywords: Hepatitis C Virus (HCV), Platelets, Growth factors, Vascular Injury

Introduction

Chronic Hepatitis C infection is a leading cause of cirrhosis worldwide [1]. One of the pathological hallmarks of chronic liver disease (CLD) is the sharp decline in platelet numbers with disease progression [2]. This thrombocytopenia has long been known and conceptualized as one of the secondary phenomena of CLD and a complication of portal hypertension. Most well-established factors known to play a role include excessive splenic sequestration and decreased hepatic synthesis of thrombopoietin, a glycoprotein hormone responsible for platelet synthesis [3] [4]. However, as the complexity and breadth of platelet function has become better appreciated, our understanding of the pathogenesis and implications of thrombocytopenia in CLD remains incomplete [5].

Platelets have emerged as mediators of a wide array of host responses extending far beyond coagulation. They play a critical role in inflammation, wound healing, tissue repair, angiogenesis, and maintenance of endothelial/vascular integrity [6]. Platelets perform these functions by regulated release of various cytokines and growth factors into the systemic circulation [7] [8] [9]. Recently, platelets have been implicated as possible mediators of liver inflammation and fibrosis [10] [11] [12]. Expression of cell adhesion molecules, such as P-selectin enables platelets to influence immune responses during tissue injury. P-selectin drives lymphocyte recruitment into damaged tissues [13]. TGFβ1, a growth factor released from platelets, is known to play a role in development of fibrosis [14]. However, the mechanisms by which platelets serve to promote fibrosis progression, resolution, or both processes, is a possibility that has only recently begun to be appreciated [12] [15].

Chronic liver diseases, including HCV, are associated with alterations in both intra- and extra-hepatic vasculature [16]. Chronic inflammation and fibrosis can induce intrahepatic resistance to blood flow and damage liver parenchyma and sinusoidal endothelium. In response to this injury, endothelial cells release soluble vascular injury markers (VIMs) [16] [17]. VIMs initiate mechanisms of host defense, immune cells trafficking, and vascular repair [18] [16] [19] [17]. Given the function of platelets as a source of growth factors, depletion of these growth factors secondary to thrombocytopenia might compound the endothelial dysfunction in patients with cirrhosis and portal hypertension.

We hypothesize that platelet abundance, platelet function by release of growth factors, and vascular integrity, are closely linked factors and together play a crucial role in liver disease progression. To explore this hypothesis, we evaluated the relationship between platelet counts and levels of growth factors and vascular injury markers in the systemic circulation of HCV patients with early fibrosis and cirrhosis. Our findings have the potential to provide insights into the underlying mechanisms involved in liver disease progression.

Materials and Methods

Patients

Serum samples were obtained retrospectively from twenty-six chronic Hepatitis C (HCV) infected patients during participation in a natural history study of patients with liver diseases at the National Institutes of Health Clinical Center (NCT00001971). Patients were evaluated in the outpatient clinic with routine laboratory assessment for HCV associated liver disease and to rule out other forms of chronic liver disease. All patients signed consent for collection, storage, and use of their serum for future research. Samples were collected from HCV patients at two time-points: the earliest available sample from each patient was designated as time-point-A and the most recent sample prior to any therapy as time-point-B. Mean duration between earliest to latest time-point for all patients was approx. 6.8 years. Serum samples from anonymized healthy donors were obtained from NIH Clinical Center blood bank. Blood from an antecubital vein was drawn into a 3.5 ml Z Serum Sep. Clot Activator (Ref 454067P, Greiner Bio-One GmbH, Kremsmunster, Austria). All control and HCV serum samples were processed by centrifugation at 1000 rpm for 10 minutes within 4 hours of being drawn and subsequently stored at −80°C until analysis.

Liver histological and laboratory assessment

The most recent liver biopsy specimen was staged by a single liver pathologist (DEK) to obtain Ishak Fibrosis (IshF) and Hepatic Activity Index (HAI) scores for fibrosis and inflammation [20] [21]. Biochemical assays were performed to measure alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), albumin, total and direct bilirubin. In addition, prothrombin time-international normalized ratio (PT-INR) and platelet counts were measured. All blood was drawn from the anti-cubital vein. Complete blood counts were drawn in 3 ml K2 EDTA tubes (Ref 367856, Becton, Dickinson and Company, Franklin lakes, NJ, USA) and measured on Sysmex system. Serum was collected in a 3.5 ml Z Serum Sep. Clot Activator (Ref 454067P, Greiner Bio-One GmbH, Kremsmunster, Austria) and biochemical parameters were measured on a Cobas C 501 system. PT-INR was drawn in a 3 ml 9NC Coagulation sodium citrate 3.2% (Ref 454334, Greiner Bio-One GmbH, Kremsmunster, Austria). The Model for End Stage Liver Disease (MELD) score was calculated according to the established formula using serum bilirubin (mg/dL), PT-INR, and serum creatinine (mg/dL). The patients were divided into two subgroups according to Ishak Fibrosis (IshF): early fibrosis included IshF of 0, 1, or 2 and cirrhosis included IshF 5 or 6. Of note, all subjects in the cirrhosis category had compensated cirrhosis, Child Pugh Class A.

Measurement of growth factors and vascular injury markers

Growth factors (GFs), and vascular injury markers (VIMs) were measured in the peripheral serum of both healthy controls and HCV infected individuals. TGFβ1 was measured using the milliplex single-plex (48–614MAG, serum diluted to manufacturer recommended starting concentration 40 to 1,000 μg protein/mL); PDGF-AA, and PDGF-BB, within a milliplex three-plex panel (HCYTMAG-60K-PX41, serum diluted to manufacturer recommended starting concentration 1:100 dilution); and EGF within a milliplex eighteen-plex panel (HCYTMAG-60K-PX41, 25 μL) (EMD Millipore Corp., Billerica, MA). E-selectin, ICAM-3, P-selectin, and thrombomodulin (TM) were assessed with V-Plex Human Vascular Injury Panel 1 (K15135C-1, 10 μL); VCAM-1 and ICAM-1 within V-Plex Human Vascular Injury Panel 2 (K15198D, serum diluted to manufacturer recommended starting concentration 1:100 dilution); and VEGF-A within the V-Plex Cytokine Panel 1(K15047D, serum diluted to manufacturer recommended starting concentration 1:4 dilution); (Meso Scale Diagnostics, Rockville, MD). All assays followed manufacturer’s instructions and were performed in duplicate.

Statistical analysis

Subject characteristics were summarized using the mean and standard deviation (SD). HCV patients were compared with healthy controls and between disease categories (early fibrosis and cirrhosis) using Mann-Whitney tests. For all boxplot figures, the central horizontal line denotes the median and length of the box indicates the interquartile range (IQR). Correlations between two parameters of interest at a given time-point (A or B) were performed using non-parametric Spearman’s correlation coefficient. Analyses were performed using GraphPad Prism (Version 6.0; GraphPad Software Inc., La Jolla, CA) and SAS (SAS institute, Cary, NC).

RESULTS

Patient characteristics

Twenty-six HCV infected patients and seventeen healthy controls were included in this study. Basic demographic, clinical, and biochemical parameters of HCV patients at each time-point are shown in Table 1. As previously noted, time-point-A denotes the earliest and time-point-B denotes the most recent time of sample collection. Mean duration of follow-up from the earliest to latest time-point for all HCV subjects was approximately 6.8 years. All patients staged as cirrhotic (F5/6) were within the compensated range and classified as Child Pugh Class A. Approximately 50% of the HCV infected subjects were male and at the latest time-point the mean age was 53.8 years. For the healthy control cohort, about 65% were male and mean age was 43.8 years. Total and Periportal (PP) hepatic inflammation was measured by Hepatic Activity Index (HAI) were higher in cirrhotics compared to patients with early fibrosis. At time-point-A, ALT, AST, and ALP were higher in cirrhotics compared to those with early fibrosis. At time-point-B, only ALP was elevated in the cirrhotic compared to non-cirrhotic patients. Consistent within both time-points, serum albumin and platelet counts were lower in cirrhotics than those with early fibrosis. At time-point-B, cirrhotic patients had higher PT-INR and MELD scores than those with early fibrosis.

Table 1.

Characteristics of Patients with Chronic Hepatitis C (HCV) Infection

Early Fibrosis Cirrhosis
Liver Biopsy
Gender (Male %) 54.5 46.6
Number of subjects 11 15
Total Inflammation Score (mean ± SD) 6.4 ± 1.4 8.3 ± 2.4*
PP Inflammation Score (mean ± SD) 1.5 ± 0.9 3.4 ± 1.6*
Treatment Status
Duration of follow-up (yrs) (mean ± SD) 8.3 ± 4.9 5.05 ± 5.6
Time-Point A
Age (yrs) (mean ± SD) 44 ± 8.8 49.2 ± 7.4
ALT (U/L) (mean ± SD) 46.8 ± 20.55 95.5 ± 51.6*
AST (U/L) (mean ± SD) 35.1 ± 15.1 78.7 ± 40.9*
ALP (U/L) (mean ± SD) 75.8 ± 21.2 107.8 ± 30.5*
Total Bilirubin (mg/dL) (mean ± SD) 0.7 ± 0.25 0.8 ± 0.6
Direct Bilirubin (mg/dL) (mean ± SD) 0.16 ± 0.08 0.3 ± 0.3
Albumin (g/dL) (mean ± SD) 4.4 ± 0.47 3.7 ± 0.4*
Platelet (x1000/mm3) (mean ± SD) 254.3 ± 49.3 161.9 ± 69.3*
Creatinine (mg/dL) (mean ± SD) 0.8 ± 0.1 1.1 ± 1.4
Time-Point B
Age (yrs) (mean ± SD) 53 ± 7 54.1 ± 8.25
ALT (U/L) (mean ± SD) 72 ± 38.5 60.4 ± 22.5
AST (U/L) (mean ± SD) 49.3 ± 21.4 54.3 ± 23.1
ALP (U/L) (mean ± SD) 71.6 ± 28.9 101.9 ± 37.9*
Total Bilirubin (mg/dL) (mean ± SD) 0.7 ± 0.4 0.7 ± 0.62
Direct Bilirubin (mg/dL) (mean ± SD) 0.2 ± 0.1 0.26 ± 0.26
Albumin (g/dL) (mean ± SD) 4.0 ± 0.2 3.59 ± 0.50*
Platelet (x1000/mm3) (mean ± SD) 222 ± 67.4 136.5 ± 60.9*
Creatinine (mg/dL) (mean ± SD) 0.8 ± 0.1 1.1 ± 1.37
PT-INR (mean ± SD) 0.95 ± 0.09 1.16 ± 0.14*
MELD (mean ± SD) 6.8 ± 0.9 9.5 ± 4.9*
Open in a new tab

PP, peri-portal; ALT, alanine transaminase; AST, aspartate transaminase; ALP, alkaline phosphatase; PT-INR, prothrombin time international normalized ratio; MELD, Model for End stage Liver Disease.

*

indicates statistical significance (p<0.05) when compared cirrhotic patients with early fibrotic patients, using unpaired Mann-Whitney test.

Growth factors decrease as platelet numbers decrease with progression of liver fibrosis

Platelet count, and growth factors (GFs) including PDGF-AA, PDGF-BB, TGFβ1, EGF, and VEGF, were measured in controls and HCV patients using serum from most recent time-point-B (Figure 1, all p<0.05). Compared to controls, platelet count was lower in patients with early fibrosis and patients with cirrhosis. Within HCV patients, platelet counts were lower in cirrhosis compared to early fibrosis. PDGF-AA, PDGF-BB, TGFβ1, and EGF were lower in cirrhotics when compared to controls. There was no difference in GFs between controls and patients with early fibrosis. With the exception of VEGF, all measured growth factors were lower in patients with advanced cirrhosis compared to early fibrosis. VEGF showed no difference in HCV patients compared to controls or between disease subcategories. Collectively, these findings illustrate a progressive decline in both platelet numbers and various growth factors with HCV associated disease progression.

Figure 1: Changes in Circulating Platelet Counts and Growth factors (GFs) in HCV Patients.

Figure 1:

Open in a new tab

Growth factors mirror platelet counts in their circulating levels with disease progression to cirrhosis. At time-point B, platelet numbers (A) were lower in HCV patients compared to controls, and lower in the cirrhotic group compared to early fibrosis group. Among growth factors, PDGF-AA, PDGF-BB, TGFβ1, EGF, (B-E) were also lower in HCV patients compared to controls, and lower in the cirrhotics compared to early fibrosis. VEGF (F) showed no difference between HCV patients and controls or between disease stages. For all boxplot figures, the central horizontal line denotes the median and the length of the box indicates the interquartile range (IQR) from 75th to 25th range. P value was calculated with Mann-Whitney test and significant <0.05.

Soluble vascular injury markers increase as HCV disease progresses

Soluble vascular injury markers (VIMs); E-selectin, P-selectin, ICAM-1, ICAM-3, VCAM-1, and thrombomodulin (TM) were measured in serum of controls and HCV patients at the most recent time-point-B (Figure 2, all p<0.05). Compared to controls, HCV patients, in both early fibrosis and cirrhosis categories, displayed higher E-selectin, ICAM-1, ICAM-3, VCAM-1, and TM. Within HCV patients, ICAM-3 and VCAM-1 were higher in patients with advanced cirrhosis (F6) compared to early fibrosis. On the contrary, P-selectin, showed no difference compared to controls or between disease subcategories. Higher VIMs in HCV patients with early fibrosis suggests an early onset of endothelial injury preceding development of cirrhosis. An increase in VIMs from early fibrosis to cirrhosis suggests an additional increase in endothelial damage with disease progression. An alternative explanation is that more of the VIMs are secreted by leukocytes and that this too increases with disease progression.

Figure 2. Enhanced Levels of Circulating Vascular Injury Markers (VIMs) with Fibrosis in HCV Patients.

Figure 2.

Open in a new tab

HCV patients displayed enhanced vascular injury in both early stages of fibrosis and then a progressive increase with cirrhosis. E-selectin, ICAM-1, ICAM-3 and VCAM-1 and TM showed higher levels in both early fibrosis and cirrhosis when compared to controls. (A-E). ICAM-3 and VCAM-1 also showed higher levels with worsening stage of fibrosis. In contrast, P-selectin showed no difference between HCV patients and controls or between disease stages (F). For all boxplot figures, the central horizontal line denotes the median and the length of the box indicates the interquartile range (IQR) from 75th to 25th range. P value was calculated with Mann-Whitney test and significant <0.05. TM, Thrombomodulin.

HCV liver disease is associated with lower platelet counts and growth factors and higher vascular injury

Given the changes in GFs and VIMs with fibrosis, we evaluated associations of GFs and VIMs with clinical parameters (Table 2) (all p<0.05). With the exception of P-selectin and TM, VIMs were linked to higher fibrosis score, ALP, bilirubin, PT-INR, MELD score, and negatively to platelet numbers and albumin. P-selectin was linked to reduced ALT and PT-INR, and showed a positive association with platelet counts. In contrast to VIMs, GFs were associated with lower fibrosis, PT-INR, MELD score, and positively with platelet counts and albumin. Collectively, these findings suggest a distinct separation of GFs and VIMs based on their associations with disease severity. They suggest a possible link of vascular injury and reduced growth factors with the decline in platelet numbers. The strong association between platelet counts and GFs suggests platelets as an importance source of these growth factors in chronic HCV. Regarding P-selectin, the positive association with platelet counts and an inverse correlation with hepatic injury and synthetic function suggests a predominant role of P-selectin as a marker of platelet activation rather than endothelial injury in HCV associated liver disease.

Table 2.

Univariate Associations of Growth Factors and Vascular Injury Markers with Clinical and Biochemical Parameters in HCV Patients

graphic file with name nihms-1537022-t0001.jpg
Open in a new tab

Inf: Inflammation, PP: Periportal, ALT, alanine transaminase; AST, aspartate transaminase; ALP, alkaline phosphatase; Bili: Bilirubin, PT-INR, prothrombin time international normalized ratio; MELD, Model for End stage Liver Disease, Sel: Selectin, TM: Thrombomodulin.

HCV patients show a reciprocal link between growth factors and vascular injury

Spearman’s rank correlation was used to examine the relationship of circulating GFs and VIMs with each other (Table 3) (all p<0.05). Each VIM showed a strong correlation with other VIMs except for P-selectin that did not correlate with other VIMs. Among GFs, PDGF-AA, PDGF-BB, TGFβ1, and EGF showed positive associations with each other. When correlating VIMs and GFs with each other, decreasing levels of many GFs including PDGF-AA, PDGF-BB, TGFβ1, and EGF was linked to a rise in markers of vascular injury. Taken together, these associations may represent a global and reciprocal connection between circulating GFs and VIMs in chronic HCV patients. Again, the lack of an association of P-selectin with other VIMs and a link to GFs, further supports the possible predominance of P-selectin as a marker of activated platelets rather than endothelial activation in HCV.

Table 3.

Univariate Associations between Growth Factors and Vascular Injury Markers in HCV Patients

graphic file with name nihms-1537022-t0002.jpg
Open in a new tab

Soluble VCAM-1, ICAM-3, and P-Selectin from an earlier time-point correlate with MELD score at a later time-point

To elucidate a prognostic value, we evaluated longitudinal associations of GFs and VIMs with MELD score from a later time-point. Spearman’s rank correlation was utilized to assess variables from time-point-A with MELD score from time-point-B. Five variables displayed significant associations with MELD (Supplementary Table 1 p<0.05). These included fibrosis, PT-INR, P-selectin, VCAM-1, and ICAM-3. All associations were positive except P-selectin that was inversely linked to MELD score.

DISCUSSION

In this study an evaluation of circulating platelets, growth factors and vascular injury markers was performed to elucidate a link between thrombocytopenia and disruption of vascular homeostasis in HCV liver disease. By evaluating patients with early and advanced fibrosis, our findings were able to capture associations that may provide insight into the pathogenesis of disease progression. To our knowledge, this is one of the few studies to investigate and compare the relationship between platelet counts and vascular homeostasis in early and advanced stages of fibrosis with longitudinal follow-up of HCV patients.

Our findings suggest that in HCV patients: 1) similar to platelet numbers, circulating growth factors show a progressive decline from minimal to advanced fibrosis; 2) vascular injury markers are elevated even in early stages of fibrosis; 3) vascular injury markers, particularly sVCAM-1 and sICAM-3, are linked to worsening disease and platelet count decline; 4) P-selectin may have a more predominant role in platelet activation than endothelial injury in HCV; and 5) Soluble VCAM-1, ICAM-3 and P-selectin from an earlier time-point correlate with MELD score at a later time-point.

Thrombocytopenia was initially considered a phenomenon secondary to cirrhosis, and a primary role of platelets in fibrosis has recently begun to be appreciated [22] [12] [15]. Animal models studying liver disease associated thrombocytopenia have shown a conflicting role of platelets in fibrosis and have failed to recapitulate the gradual progression of cirrhosis as occurs in humans [23] [11] [10] [24] [25] [26]. Furthermore, most studies in humans have assessed platelets in decompensated cirrhosis. In contrast to these studies, we performed our assessment of platelets within a cohort of HCV patients, ranging from early fibrosis to Child-Pugh class A compensated cirrhosis. As expected, HCV patients with cirrhosis had lower circulating platelet counts compared to controls and displayed a progressive decline from minimal to advanced fibrosis (Fig. 1). A key finding is the decline in platelet counts in patients with early fibrosis when compared to controls.

Platelets release various growth factors (GFs) into the systemic circulation [27]. Although there are many sources of GFs, all GFs measured in our cohort showed a strong correlation with platelet counts (Table 2) suggesting that platelets contribute to the GFs in the circulation. A decline in the circulating levels of GFs noted in cirrhotics could be a consequence of platelet depletion. An alternative explanation could be the intrahepatic accumulation of platelets and GFs with vascular injury, inflammation and fibrosis. Many studies have reported elevated GFs and platelets within the liver supporting the latter explanation [11] [24] [28] [29]. Many GFs e.g. PDGF-AA, and TGFβ-1 are profibrogenic and can activate hepatic stellate cells to differentiate into myofibroblasts [30] [31]. PDGF-BB, in mouse models of liver fibrosis, showed an increase in the liver as more platelets were recruited to the sinusoidal endothelium [24] [12]. However, most studies have only assessed specific growth factors in isolation. Our study has shown changes across multiple GFs including PDGF-AA, PDGF-BB, TGFB1, and EGF in the systemic circulation. Further studies are thus needed to clarify the reason for a decline in circulating platelet numbers and multiple types of GFs by simultaneously evaluating GFs and platelets in liver biopsies and the circulation.

Another important step in pathogenesis of portal hypertension and cirrhosis is vascular or endothelial injury. Studies have shown higher tissue and serum VIMs in both viral and non-viral liver diseases [18] [32]. Here, we showed similar elevation of serum VIMs in HCV patients (Fig. 2). By stratifying patients across early and late stages of fibrosis, we have further shown that many vascular injury markers e.g. sICAM-1, sICAM-3, and sVCAM-1 are not only higher in cirrhotics but also show elevated levels in earlier stages of fibrosis when compared to controls. This may indicate endothelial dysfunction and vascular injury before the onset of cirrhosis suggests a role of distinct VIMs in disease progression.

Of the six VIMs measured, soluble VCAM-1 showed the strongest correlations with disease parameters (Table 2). It is exclusively released from endothelial cells and thus specifically reflects endothelial activation [33]. For this reason, the association of VCAM-1 with fibrosis suggests chronic sinusoidal endothelial activation in HCV associated cirrhosis. This is supported by studies showing higher VCAM-1 and circulating endothelial cells with cirrhosis [34] [35]. In our study, elevated VCAM-1 (and other VIMs) was also associated with a decline in platelet numbers and GFs in the systemic circulation (Table 2 and 3). This suggests that endothelial release of VCAM-1 may be linked to a loss of the ability of platelets to maintain endothelial integrity in HCV induced liver injury. In light of these findings, we propose a complex interplay between platelets and VIMs in chronic liver disease pathogenesis [36] [37].

P-selectin is a well-established marker of activated endothelium or activated platelets in response to injury [38] [39]. In our study, P-selectin, showed no correlations with other markers of endothelial injury (Table 3). Instead, P-selectin was associated with circulating platelet counts suggesting a more significant role for P-selectin as a marker of activated platelets than a marker of endothelial injury in chronic HCV.

One of the advantages of this study was the ability to measure all parameters at an early and a later time-point for each patient. Only VIMs from an early time-point showed associations with MELD at the later time-point (Supplementary Table 1). This not only emphasizes the need to monitor platelet numbers but also further study VIMs when deciphering factors contributing to liver disease progression [40]. The positive association of VCAM-1 and ICAM-3 with patient’s MELD score highlights the potential biological importance of vascular injury and endothelial dysfunction not only in end-stage liver disease but, perhaps more importantly in earlier stages as well. In addition, the inverse association of P-selectin with MELD reiterates a less significant role of P-selectin as a marker of endothelial injury in chronic HCV. However, for all three variables, a small sample size precluded further analysis such as stepwise linear regression. Our findings urge future studies in larger cohorts to elucidate the possible role of VIMs in predicting liver disease progression.

One of the limitations of this retrospective study is the small sample size. Another weakness of our analysis is that platelets are not the only source of investigated GFs. However, strong correlations between platelet counts and GFs supports the notion that platelets are at least a significant source for these GFs in HCV patients. All analytes were measured in circulation and a more complete understanding would require deeper exploration such as from biopsy specimens. Lastly, due to the correlative nature of the study, causality cannot be assumed. Now with availability of Direct Acting Antiviral Therapy (DAA), our understanding of fibrosis and a role of GFs and VIMs in disease pathogenesis and reversal could be validated after HCV clearance.

In summary, we explored the relationship of circulating platelet counts with various growth factors and vascular injury markers in patients with early and late stages of chronic hepatitis C liver disease. The results presented in this study are intended to support and elaborate the role of platelets and vascular injury in liver disease pathogenesis. We propose that despite multiple sources of growth factors in health, platelet count decline in cirrhosis may have an influence on the subsequent decline in circulating levels of growth factors. The associations of markers of vascular injury with decreasing platelet counts and fibrosis reiterates a possible link between platelets and endothelial dysfunction in hepatic fibrogenesis. Furthermore, increased endothelial activation in early stages of fibrosis and the link of VIMs with fibrosis reiterates the role of endothelial injury in pathogenesis of hepatic fibrosis. By assessing a broad panel of growth factors and vascular markers and doing so at an early and a later time-point for each patient, we were able to identify candidate VIMs that may have prognostic value in liver disease progression. These findings emphasize the importance of collectively evaluating pathways involved in vascular homeostasis and platelet function to provide mechanistic insight into liver disease pathogenesis.

Table 4.

Correlations of most recent MELD score with Clinical Parameters, Growth Factors and Vascular Injury Markers measured at an earlier time-point

Parameters (Time-Point A) MELD Score (Time-Point B)
r p n
Ishak fibrosis 0.55 0.0044* 25
PT-INR 0.63 0.0063* 17
VCAM1 0.45 0.0242* 25
P-Selectin −0.42 0.0351* 25
ICAM-3 0.41 0.0412* 25
TGFβ1 −0.39 0.0535 25
Platelets −0.40 0.0545 24
PDGFAA −0.38 0.058 25
PDGF-BB −0.38 0.0583 25
Albumin −0.36 0.0873 24
Direct bilirubin 0.35 0.0955 24
TM 0.32 0.1179 25
PP Inflammation 0.33 0.1202 23
AST 0.31 0.1472 24
Total bilirubin 0.22 0.3067 24
Total Inflammation 0.20 0.371 23
E-Selectin 0.16 0.4457 25
ALP 0.15 0.4883 24
ALT 0.14 0.5048 24
VEGF −0.13 0.5223 25
ICAM-1 0.13 0.5235 25
Open in a new tab

Correlations were calculated by Spearman’s rank correlation between MELD score from the most recent time-point (B) and biopsy scores, GFs and VIMs, and clinical parameters, measured from an earlier time-point (A) in the same patients. Spearman’s rank correlation coefficient, r (rho), p value, and n, number of data points for each variable, are shown above.

*

indicates statistical significance using p-values at a cut-off of <0.05.

PP: Periportal, ALT, alanine transaminase; AST, aspartate transaminase; ALP, alkaline phosphatase; PT-INR, prothrombin time international normalized ratio; MELD, Model for End stage Liver Disease, TM: Thrombomodulin.

Acknowledgements

We thank the patients, fellows, nurses, all staff, and Institutional Review Board that assisted in the study.

Financial Support

This work was supported by the Intramural Research Program, National Institute of Diabetes and Digestive and Kidney Diseases.

Abbreviations

ALP

alkaline phosphatase

ALT

alanine aminotransferase

AST

aspartate aminotransferase

DEK

David E Kleiner

HCV

Hepatitis C Virus/HCV infected subjects

IshF

Ishak Fibrosis

HAI

Hepatic Activity Index

PT-INR

prothrombin time-international normalized ratio

TB

Total Bilirubin

PP

Periportal

TM

Thrombomodulin

MELD

Model for End stage Liver Disease

PDGF-AA

Platelet Derived Growth Factor A

PDGF-BB

Platelet Derived Growth Factor B

TGFB1

Transforming Growth Factor beta isoform 1

EGF

Epidermal Growth Factor

VEGF

Vascular Endothelial Growth Factor

ICAM

Intercellular Adhesion Molecule

VCAM

Vascular Cell Adhesion Molecule

TM

Thrombomodulin

GF

Growth Factor

VIM

Vascular Injury Marker

CLD

chronic liver disease

DEK

David E. Kleiner

IQR

interquartile range

SD

Standard Deviation

DAA

Direct Acting Antiviral Therapy

Footnotes

Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.

Conflicts of Interest

None of the authors has financial interests or conflicts of interest related to this research.

References

  • 1.Gower E, et al. , Global epidemiology and genotype distribution of the hepatitis C virus infection. J Hepatol, 2014. 61(1 Suppl): p. S45–57. [DOI] [PubMed] [Google Scholar]
  • 2.Giannini EG, Review article: thrombocytopenia in chronic liver disease and pharmacologic treatment options. Aliment Pharmacol Ther, 2006. 23(8): p. 1055–65. [DOI] [PubMed] [Google Scholar]
  • 3.Hernandez-Gea V and Friedman SL, Platelets arrive at the scene of fibrosis……studies. J Hepatol, 2011. 54(5): p. 1063–5. [DOI] [PubMed] [Google Scholar]
  • 4.Witters P, et al. , Review article: blood platelet number and function in chronic liver disease and cirrhosis. Aliment Pharmacol Ther, 2008. 27(11): p. 1017–29. [DOI] [PubMed] [Google Scholar]
  • 5.Tana MM, et al. , Factors associated with the platelet count in patients with chronic hepatitis C. Thromb Res, 2015. 135(5): p. 823–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Ho-Tin-Noe B, et al. , Platelet granule secretion continuously prevents intratumor hemorrhage. Cancer Res, 2008. 68(16): p. 6851–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Italiano JE Jr., et al. , Angiogenesis is regulated by a novel mechanism: pro- and antiangiogenic proteins are organized into separate platelet alpha granules and differentially released. Blood, 2008. 111(3): p. 1227–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Nurden AT, et al. , Platelets and wound healing. Front Biosci, 2008. 13: p. 3532–48. [DOI] [PubMed] [Google Scholar]
  • 9.Danielli JF, Capillary permeability and oedema in the perfused frog. J Physiol, 1940. 98(1): p. 109–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Watanabe M, et al. , Platelets contribute to the reduction of liver fibrosis in mice. J Gastroenterol Hepatol, 2009. 24(1): p. 78–89. [DOI] [PubMed] [Google Scholar]
  • 11.Lang PA, et al. , Aggravation of viral hepatitis by platelet-derived serotonin. Nat Med, 2008. 14(7): p. 756–61. [DOI] [PubMed] [Google Scholar]
  • 12.Kondo R, et al. , Accumulation of platelets in the liver may be an important contributory factor to thrombocytopenia and liver fibrosis in chronic hepatitis C. J Gastroenterol, 2013. 48(4): p. 526–34. [DOI] [PubMed] [Google Scholar]
  • 13.Wynn TA, et al. , P-selectin suppresses hepatic inflammation and fibrosis in mice by regulating interferon gamma and the IL-13 decoy receptor. Hepatology, 2004. 39(3): p. 676–87. [DOI] [PubMed] [Google Scholar]
  • 14.Brunner G and Blakytny R, Extracellular regulation of TGF-beta activity in wound repair: growth factor latency as a sensor mechanism for injury. Thromb Haemost, 2004. 92(2): p. 253–61. [DOI] [PubMed] [Google Scholar]
  • 15.Bitto N, Liguori E, and Mura V, Coagulation, Microenvironment and Liver Fibrosis. Cells, 2018. 7(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Iwakiri Y, Shah V, and Rockey DC, Vascular pathobiology in chronic liver disease and cirrhosis - current status and future directions. J Hepatol, 2014. 61(4): p. 912–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Aldamiz-Echevarria T, et al. , Soluble Adhesion Molecules in Patients Coinfected with HIV and HCV: A Predictor of Outcome. PLoS One, 2016. 11(2): p. e0148537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Butcher EC, Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell, 1991. 67(6): p. 1033–6. [DOI] [PubMed] [Google Scholar]
  • 19.Kaplanski G, et al. , Vascular cell adhesion molecule-1 (VCAM-1) plays a central role in the pathogenesis of severe forms of vasculitis due to hepatitis C-associated mixed cryoglobulinemia. J Hepatol, 2005. 42(3): p. 334–40. [DOI] [PubMed] [Google Scholar]
  • 20.Knodell RG, et al. , Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology, 1981. 1(5): p. 431–5. [DOI] [PubMed] [Google Scholar]
  • 21.Ishak K, et al. , Histological grading and staging of chronic hepatitis. J Hepatol, 1995. 22(6): p. 696–9. [DOI] [PubMed] [Google Scholar]
  • 22.Toghill PJ, Green S, and Ferguson F, Platelet dynamics in chronic liver disease with special reference to the role of the spleen. J Clin Pathol, 1977. 30(4): p. 367–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Kodama T, et al. , Thrombocytopenia exacerbates cholestasis-induced liver fibrosis in mice. Gastroenterology, 2010. 138(7): p. 2487–98, 2498 e1–7. [DOI] [PubMed] [Google Scholar]
  • 24.Yoshida S, et al. , Extrahepatic platelet-derived growth factor-beta, delivered by platelets, promotes activation of hepatic stellate cells and biliary fibrosis in mice. Gastroenterology, 2014. 147(6): p. 1378–92. [DOI] [PubMed] [Google Scholar]
  • 25.Leon C, et al. , The contribution of mouse models to the understanding of constitutional thrombocytopenia. Haematologica, 2016. 101(8): p. 896–908. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Liu Y, et al. , Animal models of chronic liver diseases. Am J Physiol Gastrointest Liver Physiol, 2013. 304(5): p. G449–68. [DOI] [PubMed] [Google Scholar]
  • 27.Smyth SS, et al. , Platelet functions beyond hemostasis. J Thromb Haemost, 2009. 7(11): p. 1759–66. [DOI] [PubMed] [Google Scholar]
  • 28.Iannacone M, Platelet-mediated modulation of adaptive immunity. Semin Immunol, 2016. 28(6): p. 555–560. [DOI] [PubMed] [Google Scholar]
  • 29.Murata S, et al. , Platelets promote liver regeneration in early period after hepatectomy in mice. World J Surg, 2007. 31(4): p. 808–16. [DOI] [PubMed] [Google Scholar]
  • 30.Geremias AT, et al. , TGF beta1 and PDGF AA override collagen type I inhibition of proliferation in human liver connective tissue cells. BMC Gastroenterol, 2004. 4: p. 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hayashi H and Sakai T, Biological Significance of Local TGF-beta Activation in Liver Diseases. Front Physiol, 2012. 3: p. 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Adams DH, et al. , Endothelial activation and circulating vascular adhesion molecules in alcoholic liver disease. Hepatology, 1994. 19(3): p. 588–94. [DOI] [PubMed] [Google Scholar]
  • 33.Kaplanski G, et al. , Increased levels of soluble adhesion molecules in the serum of patients with hepatitis C. Correlation with cytokine concentrations and liver inflammation and fibrosis. Dig Dis Sci, 1997. 42(11): p. 2277–84. [DOI] [PubMed] [Google Scholar]
  • 34.Buck M, et al. , Novel inflammatory biomarkers of portal pressure in compensated cirrhosis patients. Hepatology, 2014. 59(3): p. 1052–9. [DOI] [PubMed] [Google Scholar]
  • 35.Abdelmoneim SS, et al. , A prospective pilot study of circulating endothelial cells as a potential new biomarker in portal hypertension. Liver Int, 2010. 30(2): p. 191–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Chauhan A, et al. , Platelets: No longer bystanders in liver disease. Hepatology, 2016. 64(5): p. 1774–1784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Lalor PF, et al. , Hepatic sinusoidal endothelium avidly binds platelets in an integrin-dependent manner, leading to platelet and endothelial activation and leukocyte recruitment. Am J Physiol Gastrointest Liver Physiol, 2013. 304(5): p. G469–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.McEver RP, Role of selectins in leukocyte adhesion to platelets and endothelium. Ann N Y Acad Sci, 1994. 714: p. 185–9. [DOI] [PubMed] [Google Scholar]
  • 39.Kamel MM, Fouad SA, and Basyoni MM, P selectins and immunological profiles in HCV and Schistosoma mansoni induced chronic liver disease. BMC Gastroenterol, 2014. 14: p. 132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Lo Iacono O, et al. , Serum levels of soluble vascular cell adhesion molecule are related to hyperdynamic circulation in patients with liver cirrhosis. Liver Int, 2008. 28(8): p. 1129–35. [DOI] [PubMed] [Google Scholar]

RESOURCES