ABSTRACT
-
Background/Aims
Hepatitis C virus (HCV) infection remains a global health challenge, leading to chronic liver disease, cirrhosis, and hepatocellular carcinoma (HCC). Despite the high efficacy of direct-acting antiviral therapy in achieving sustained virologic response (SVR), concerns persist regarding long-term immune alterations and residual risks, particularly in cirrhotic patients.
-
Methods
This study investigates 75 soluble immune mediator (SIM) profiles in 102 chronic HCV patients, stratified by cirrhosis status, at therapy initiation, end of treatment, and long-term follow-up (median 96 weeks). Findings were compared with 51 matched healthy controls and validated in an independent cohort of 47 cirrhotic patients, 17 of whom developed HCC.
-
Results
We observed significant SIM alterations at baseline, with cirrhotic patients displaying a more profoundly dysregulated inflammatory milieu. Despite an overall decline in inflammatory markers following SVR, persistent alterations were evident, particularly in cirrhotic patients. Notably, those with liver stiffness exceeding 14 kPa exhibited sustained inflammatory dysregulation, correlating with liver elastography values. Key SIM such as interleukin (IL)-6, IL-8, urokinase plasminogen activator, and hepatocellular growth factor remained elevated and were associated with HCC development. Network analysis highlighted their roles in liver fibrosis, regeneration, and carcinogenesis.
-
Conclusions
These findings underscore the importance of early antiviral intervention to prevent cirrhosis-related sequelae. Future studies should explore the mechanistic pathways linking chronic inflammation, fibrosis, and oncogenesis to identify predictive biomarkers and novel therapeutic targets. Addressing persistent immune alterations post-HCV clearance may improve long-term outcomes, particularly in patients with advanced liver disease.
-
Keywords: Hepatitis C virus; Direct-acting antiviral therapy; Cirrhosis; Soluble inflammatory marker; Hepatocellular carcinoma
Study Highlights
• Our study shows that chronic HCV infection leads to alterations in the inflammatory milieu that can persist long after viral elimination. Especially in cirrhotic patients, several soluble immune mediators associated with liver damage and carcinogenesis remain altered. Our results emphasize the importance of viral elimination before extensive liver damage occurs.
Graphical Abstract
INTRODUCTION
Hepatitis C virus (HCV) infection remains a major global health concern, with an estimated 50 million individuals chronically infected as of 2024 [
1]. Despite the availability of highly effective direct-acting antivirals (DAAs), which achieve sustained virologic response (SVR) in over 95% of treated patients, HCV remains a leading cause of liver cirrhosis and hepatocellular carcinoma (HCC), contributing to approximately 290,000 deaths annually [
1,
2]. Beyond liver-related morbidity, chronic HCV infection is associated with systemic inflammation and extrahepatic manifestations, including lymphoma, cryoglobulinemic vasculitis, cardiovascular diseases, type 2 diabetes mellitus, and chronic fatigue [
3].
Although SVR significantly reduces the risk of cirrhosis and HCC, it does not completely eliminate the risk, particularly in patients with advanced fibrosis or cirrhosis [
4-
7]. This residual risk may stem from persistent biological imprints of HCV infection that remain after viral clearance [
8,
9]. Evidence suggests that epigenetic modifications and immune imprinting induced by HCV contribute to ongoing inflammation and a pro-carcinogenic environment in the liver, even after successful treatment [
10-
12].
Imprinting by HCV can also occur on different levels of the immune system [
8]. For example, HCV induces interferon-stimulated genes [
13] and a characteristic inflammatory milieu that may not be fully reversible after viral elimination [
8]. Soluble inflammatory markers have been shown to remain altered for 12–24 weeks after successful DAA therapy in patients with acute [
14] and chronic HCV infection [
15]. However, few long-term data are available on whether the changes may be reversible over the longer observation period [
14,
16]. As most previous studies included a heterogeneous population of patients without further classification based on the degree of liver fibrosis and cirrhosis [
15] or did not include controls without HCV infection [
16], we aimed to decipher the role of cirrhosis on the soluble inflammatory milieu during and after HCV infection.
To better elucidate the role of cirrhosis in shaping the inflammatory milieu post-HCV clearance, we analyzed soluble immune mediator (SIM) profiles in a well-characterized cohort of 102 chronic HCV patients with and without cirrhosis. We measured 92 SIMs, of which 75 were included in the final analysis at therapy initiation, end of treatment, and long-term follow-up (median: 96 weeks) and compared these profiles to those of 51 age- and sex-matched healthy controls. Additionally, we validated our findings in an independent cohort of 47 HCV patients with cirrhosis, 17 of whom developed HCC.
MATERIALS AND METHODS
Study population and design
Out of 799 chronic HCV patients that were treated with DAA therapy at Hannover Medical School between January 2014 and November 2019, we selected a well-characterized cohort of 102 chronic HCV patients achieving SVR for this study. All patients were part of a prospective biobank registry and peripheral blood was collected at different time points, processed and stored after standard operating procedures. All patients gave their written informed consent for this study (
Supplementary Fig. 1).
We divided our patients in two groups based on non-invasive liver elastography [
17]. 46 chronic HCV patients had elastography values of above 14 kPa and were therefore included in the cirrhosis cohort (cohort A), whereas the 56 patients that had elastography values of below 14 kPa were included in the non-cirrhosis cohort (cohort B). 10 of the cirrhotic patients developed HCC. For all patients, plasma samples were available at therapy start, end of treatment and long-term follow-up (median 96 weeks,
Supplementary Table 1). For patients developing HCC, plasma samples were limited, resulting in shorter follow-up times for some of the patients (
Supplementary Table 1). Plasma samples from 51 healthy individuals, 33 of them from the RESIST senior individuals cohort [
18], served as controls (
Table 1). 47 chronic HCV patients with cirrhosis of whom 17 developed HCC were recruited as an independent validation cohort for Cohort A in the course of our analysis (
Table 1). Plasma samples were available at therapy start (n=47), end of treatment (n=45) and follow-up (n=38, median 96 weeks). 17 patients of the validation cohort developed HCC, resulting in shorter follow-up times for some of the patients (
Supplementary Table 1).
All study protocols conformed to the Declarations of Helsinki and Istanbul and were reviewed and approved by the ethics committee of Hannover Medical School (No. 10873_BO_K_2023, No. 9474_BO_K_2020, No. 8615_BO_S_2019).
Sample preparation
After blood draw was performed at the outpatient clinic of Hannover Medical School, all samples were processed according to standard operation procedures. EDTA whole blood was centrifuged and plasma aliquots were stored at a minimum of –20°C.
SIM measurements
For SIM measurement, 50 μl defrosted EDTA plasma were randomized on 96 well plates and stored at –80°C before shipping to Olink Proteomics, Uppsala, Sweden. All analyses were conducted with the Target 96 Inflammation Panel including 92 SIM.
Measurements were performed in four batches. We accounted for a possible batch effect by re-analyzing 15 samples from the first batch again in the other batches (referred to as bridging samples) using the OlinkAnalyze R package [
19]. The proximity extension assay and quantitative PCR readout were performed at Olink Proteomics, Uppsala, Sweden.
All data were delivered in Olink Proteomics’ arbitrary unit on a log2 scale as normalized protein expression (NPX) values. We excluded CDCP1 since the assay failed on one plate and 16 SIM that were below the limit of detection in more than 75% of all samples. A total number of 75 SIM were included in our final analysis.
Statistical analysis
We performed the data analysis using R (version 4.2.2) and R Studio (version 2024.02.2 Build 764 “Chocolate Cosmos”). All data were normalized via bridging normalization using the OlinkAnalyze R package. Welch two sample t-tests and paired Welch two sample t-tests were used when parametrical tests were applicable. If parametrical tests were not applicable, Mann–Whitney U-test and Wilcoxon signed-rank test were used. We conducted all tests using the OlinkAnalyze R package.
All
P-values were adjusted for multiple testing using the Benjamini–Hochberg procedure controlling the false discovery rate at 5%. The estimated difference in mean was determined using the OlinkAnalyze package and correlations were calculated with the rstatix package using Pearson correlation. We used multiple R packages for data visualization (
Supplementary Table 2). The graphical abstract was created using Biorender.com. Network analysis was performed using StringDB (version 12.0).
RESULTS
Cohort characteristics
All chronic HCV patients (n=102) achieved SVR and no HCV-RNA was detectable at long-term follow-up. Cohort A (patients with liver cirrhosis) consisted of a majority of male patients (34/46 males) with a mean age of 58 years. As expected, liver enzymes were increased at baseline (mean AST=119 U/L, mean ALT=124 U/L) and a moderate thrombocytopenia (platelets=121 Tsd/μl) was apparent (
Table 1). 10 patients of cohort A developed HCC after (n=8) or shortly before (n=2) the follow-up sampling timepoint (
Supplementary Table 3).
Cohort B (patients without liver cirrhosis) consisted of 56 patients (27 males) with a mean age of 58 years. Liver enzymes at baseline were increased but overall lower than in cohort A (mean AST=53.9 U/L, mean ALT=73.2 U/L) and platelets were not decreased (mean 209 Tsd/μl) (
Table 1).
51 healthy individuals (39 males) with a mean age of 58 years were matched based on age and sex by 2:1 cardinality matching (
Table 1). The validation cohort includes 47 chronic HCV patients (37 males) with a mean age of 57 years and liver cirrhosis (mean elastography 38.2 kPa). Similar to Cohort A, patients had high liver enzymes at baseline (mean AST=94.7 U/L, mean ALT=92.1 U/L) and low platelets (mean 84.2 Tsd/μl) (
Table 1). 17 patients of the validation cohort developed HCC before end of treatment (n=5), before the follow-up timepoint (n=4) or after the follow-up sampling timepoint (n=8) (
Supplementary Table 3).
At the start of therapy, we detected a total amount of 51 altered SIM in chronic HCV patients, among others the protease CASP-8, the chemokines CXCL10, CXCL11, CCL19, and IL-12B, the growth factors HGF and stem cell factor (SCF) and several surface markers as IL-18R and PD-L1 (
Fig. 1, adjusted [adj.]
P-value <0.05). Most SIM were upregulated (45/51) in comparison to healthy controls and few SIM, like SCF, were downregulated at baseline. CASP-8, CXCL10, CXCL11, CCL19, CCL20, ST1A1, and HGF showed the highest estimated differences.
We compared cirrhotic (cohort A) and non-cirrhotic (cohort B) patients at therapy start to the healthy controls and observed distinct SIM profiles for each cohort. We detected 61 SIM alterations in cohort A and 42 SIM alterations in cohort B (adj. P-value <0.05).
39 SIM were altered in both cohorts at baseline. Whereas 35 of these SIM, among others IL-8, CXCL10, CXCL11, CCL20, HGF, IL-18R, and PD-L1, were increased, only SCF, FGF-21, TWEAK, and IL-17C were decreased compared to the controls (
Fig. 2A). When we compared the expression levels of these 39 SIM in both cohorts, we detected significantly higher NPX values for 23 of the 35 increased SIM in cirrhotic patients in comparison to non-cirrhotic patients, including HGF, CXCL10, CASP-8, IL-8, CCL20, CXCL11, IL-18R, and PD-L1 (
Fig. 2B,
Supplementary Fig. 2). Meanwhile, cirrhotic patients had lower NPX values for SCF compared to non-cirrhotic patients (
Fig. 2B).
In order to further investigate the role of cirrhosis, we compared both SIM profiles and identified 22 out of 61 differential SIM expressions that were restricted to cirrhotic patients, including the serine protease urokinase plasminogen activator (uPA), ligands such as IL-6, IL-10RB, the chemokine CXCL6, the growth factor FGF-23 and soluble surface markers such as CD8A (
Fig. 2A,
2C).
In consequence, cirrhotic patients (cohort A) did not only have a higher frequency of alterations at baseline, but also more severe alterations in comparison to non-cirrhotic patients (cohort B) (
Fig. 3).
To investigate the influence of viral elimination on the inflammatory milieu, we further analyzed the SIM expression of our two cohorts at end of treatment and long-term follow up (median 96 weeks). Even though the levels of most SIM declined, 20 alterations were detectable at long-term follow-up in non-cirrhotic patients and 46 SIM alterations persisted in cirrhotic patients compared to healthy controls (
Fig. 3).
12 of the 22 SIM that differed exclusively in cirrhotic patients, like uPA, IL-6, IL-10RB, CXCL6, FGF-23, CD8A, CCL11, CXCL5, CCL25, and MMP-10 remained altered in cirrhotic patients at follow-up (
Fig. 3).
In non-cirrhotic patients, 17 of the 39 SIM that were altered in both cohorts at baseline remained altered. Among the persistently altered SIM were the protease CASP-8, chemokines like CXCL10, CXCL11, and OSM, the enzymes ST1A1 and STAMBP, the translation initiation factor 4EBP1, the deacetylase SIRT2, the tumor necrosis factor TNSF14 and the growth factor HGF. Meanwhile, 34 of 39 SIM alterations persisted in cirrhotic patients, among others chemokines like CCL19, CCL20, IL-12B, IL-18, and IL-8 and the chemokine receptor IL-18R1 whose levels regenerated in non-cirrhotic patients (
Fig. 4A). In both cohorts, levels of most of the SIM significantly declined over time, but remained higher than in healthy controls, especially in cirrhotic patients (
Fig. 4B,
Supplementary Fig. 3). Similar trends were also visible in the clinical data: Even though the liver enzymes AST and ALT and elastography values decreased, levels remained high in most of our cirrhotic patients (
Supplementary Fig. 4).
To further evaluate the effect of liver damage, we assessed differences in SIM restoration by dividing the cohort of cirrhotic HCV patients into one group of patients that showed signs of liver fibrosis but not cirrhosis at follow-up (elastography <14 kPa, n=15) and one group of patients with persisting signs of liver cirrhosis (elastography ≥14 kPa, n=29) (
Supplementary Fig. 4). Both groups had similar SIM profiles at baseline, but differed at long-term follow up (
Supplementary Fig. 5). Whereas the inflammatory pattern persisted in the patients with high elastography at follow-up, the pattern disappeared in patients below the cutoff at follow-up (
Fig. 5A).
We detected correlations between most of the SIM (30 of 46) that remained altered in all cirrhotic patients at follow-up and the corresponding elastography values. Correlations of increased SIM like uPA, HGF, CD244, IL-6, IL-8, FGF-23, and PD-L1 were positive, while decreased SIM (e.g., SCF) were negatively correlated with elastography (
Fig. 5B,
Supplementary Fig. 6). Only few SIM, like CASP-8, IL-17C, IL-10, and IL-18 did not correlate with elastography (
Fig. 5B).
We aimed to decipher the interactions among the 30 SIM that remained altered in cirrhotic patients and correlated with elastography and therefore performed a network analysis using the STRING database. 26 of 30 SIM were connected on high confidence level (>0.70) with IL-6 in a central position (19 edges to other SIM) (
Fig. 6A).
Based on our network analysis, we performed a functional enrichment analysis focusing on KEGG pathways. Enriched pathways were associated with inflammation and immune responses, including IL-17 signaling, TNF signaling and the Toll-like receptor signaling pathway. Major pathways were linked to NF-kappa B and PI3K-Akt signaling. Interestingly, most of the pathways included IL-6 and IL-8. Moreover, we detected that 8 of the significantly altered SIM were linked to pathways associated with cancer or transcriptional misregulation in cancer, among them IL-6, IL-8, HGF, CD40, SCF, Flt3L, and uPA (
Fig. 6B).
As 10 out of 46 cirrhotic patients developed HCC, we compared the SIM levels of patients developing or not developing HCC over time. Our results showed a trend of persistently higher levels of HGF, IL-6, IL-8, and uPA and lower levels of SCF in patients developing HCC (
Fig. 7A). In order to confirm our results, we compared the cirrhotic patients with a validation cohort of 47 cirrhotic chronic HCV patients at therapy start and follow-up. Even though we saw different SIM levels for 7 SIM at therapy start, the overall signatures were similar and we detected no significant differences between both cohorts at follow-up (
Supplementary Fig. 7). We compared the SIM levels of IL-6, IL-8, HGF, CD40, SCF, Flt3L, and uPA in the patients developing HCC (n=17) and the patients not developing HCC (n=30) over time. We detected a similar trend that patients developing HCC had persistently higher levels of HGF, IL-6, IL-8, and uPA and lower levels of SCF, particularly at follow-up (
Fig. 7B).
When we combined HGF, IL-6, IL-8, uPa, and SCF in a score, cirrhotic patients developing HCC had the highest score levels over all three timepoints, followed by cirrhotic patients not developing HCC, while non-cirrhotic patients had the lowest levels. Comparing the score in the cirrhosis cohort (cohort A) between patients developing HCC and patients not developing HCC, we detected significantly higher levels in patients developing HCC at therapy start (
P=0.04) and follow-up (
P=0.02). The same trend was observed in our validation cohort, even though differences were not statistically significant (
P=0.5 at therapy start,
P=0.12 at follow-up) (
Supplementary Fig. 8). In both cohorts, liver elastography decreased in most patients, but just 3 out of 27 patients reached an elastography value below 14 kPa, whereas this was much more pronounced in the patients that did not develop HCC after SVR (
Supplementary Fig. 9).
DISCUSSION
Chronic HCV infection leads to substantial alterations of the inflammatory milieu, and previous studies raised concerns about persistent changes after viral elimination through DAA treatment [
20]. Our study confirms earlier results suggesting persistent inflammatory changes, but also demonstrates distinctive patterns in cirrhotic and non-cirrhotic patients before and after viral elimination [
14,
15]. These results provide new insights into the lasting effects of HCV infection and underscore the profound impact of liver damage in this context. Although HCV elimination led to an overall reduction in inflammatory alterations, SIM levels did not fully normalize to those of healthy controls. In non-cirrhotic patients, fewer SIM were initially altered, and fewer changes persisted at long-term follow-up compared to cirrhotic patients. Interestingly, among the persisting alterations were the interferon regulatory factor 3 response genes CXCL10 and CXCL11, which were suggested as biomarkers of liver inflammation during chronic HCV infection as well as the hepatocellular growth factor HGF [
21-
23]. While the levels of several chemokines like CCL19, CCL20, IL-12B, IL-18, and IL-8 recovered in non-cirrhotic patients, cirrhotic patients maintained a highly altered inflammatory milieu.
In particular, those with higher liver stiffness measurements exceeding 14 kPa on elastography, indicating significant hepatic scarring, at follow-up exhibited severe inflammatory milieu alterations. Most of the persistently altered SIM in cirrhotic patients correlated with liver elastography, reinforcing the link between liver stiffness and sustained inflammatory dysregulation. Our results suggest that the extent of liver injury is closely associated with the degree of inflammatory milieu changes, a concept previously described on a smaller scale [
24]. For example, Shah et al. [
25] were able to show that IL-6 is associated with the severity of liver fibrosis in patients with chronic HCV infection.
Importantly, elevated levels of IL-6 but also IL-8, uPA, and HGF played a central role in our network analysis and were associated with carcinogenesis according to the KEGG analysis. Previous studies showed increased plasma or serum levels in patients with HCC after DAA therapy, but validation is lacking and no specific biomarker emerged from those studies [
11,
26].
Patients who developed HCC exhibited high levels of IL-6, IL-8, uPA, and HGF, along with low levels of SCF. When these markers were combined into a score, cirrhotic patients who developed HCC had significantly higher levels. A similar trend was observed in the validation cohort, though it did not reach statistical significance.
Beyond their association with carcinogenesis, IL-6, IL-8, uPA, and HGF play pivotal roles in liver regeneration, fibrosis, and cirrhosis [
27]. Recent work by Filliol et al. [
28] has emphasized the critical role of hepatic stellate cells in this process, highlighting opposing effects of HCC promotion and protection among different subpopulations. Several studies have further demonstrated that liver injury induces increased secretion of IL-6 by Kupffer cells, which subsequently stimulates HGF secretion by stellate cells. This process, later activated by uPA, promotes hepatocyte proliferation [
29,
30]. Additionally, liver injury activates hepatic stellate cells and recruits further inflammatory cells via cytokines such as IL-8 and CXCL10, both of which remained altered in cirrhotic patients [
31]. Despite these insights, the precise underlying mechanisms remain elusive, as pro-and anti-carcinogenic effects likely occur simultaneously. The interactions of IL-6, IL-8, CXCL10, uPA, and HGF should be further investigated and considered in future studies as potential biomarker candidates in chronic HCV patients after cure [
30,
32].
The strengths of our study include a sizeable prospective biobank cohort comprising 102 chronic HCV patients, a validation cohort of 47 chronic HCV patients with cirrhosis, comprehensive clinical characterization, a long follow-up period, and the assessment of 75 SIMs. However, our study also has limitations resulting from the monocentric, retrospective design of the analysis and individual disease progression in the patients developing HCC, leading to shorter follow-up times in some of the patients. Our study confirms previous studies, suggesting that cytokine alterations still persist after viral elimination [
33]. Moreover, fibrosis and cirrhosis may influence the regeneration of the inflammatory milieu after HCV elimination. Fittingly, a recently published study has observed different SIM changes in chronic HCV patients after recovery depending on the degree of steatosis [
16]. In line with previous studies, our results show high plasma levels in patients developing HCC. The number of samples, in particular at follow-up, were limited and further studies with larger, more diverse cohorts and mechanistic experiments are required to further investigate the interplay between inflammatory milieu alterations and carcinogenesis [
11,
26].
In summary, the presented data underscore the importance of early DAA therapy before the onset of cirrhosis to prevent severe sequelae, including persistent imprints in the inflammatory milieu. Future studies should focus on the complex immune mechanisms involved in liver regeneration and carcinogenesis to identify novel therapeutic targets for patients who do not recover from cirrhosis after viral elimination, as well as predictive biomarkers for the development of HCC.
FOOTNOTES
-
Authors’ contribution
MC conceived the project and coordinated the analyses. MW, CO, AK, and MC were involved in designing of experiments. MW, CO, AK, and MC drafted the manuscript. BM, KD, and HW were involved in recruitment of patients. LR and TW were involved in acquring samples and clinical data of the RESIST senior individuals cohort. JT was involved in creation of the clinical cohort, JM and HS helped acquiring the data. MW acquired the data. MW, GG, and MB analyzed the data. CO, TK, and MC supervised the analyses. All authors read and approved the manuscript.
-
Acknowledgements
This work was supported by the German Federal Ministry of Education and Research (BMBF) within the framework of the CompLS research and funding concept (grants 031L0294C, 031L0294A). The project was supported by infrastructure of the German Center for Infection Research (DZIF; TTU 05.708_00, TTU-IICH-07-808) and funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy - EXC 2155 - project number 390874280. This project was part of project A5 in the Collaborative Research Center 900 - Microbial Persistence and its Control. This study and MW were supported by the Else Kröner-Fresenius-Stiftung (Promotionsprogramm DigiStrucMed 2020_EKPK.20). CO, JT were supported by the Else Kröner-Fresenius-Stiftung (Promotionsprogramm KlinStrucMed).
We thank Helena Lickei for her assistance with blood sample processing. We thank the study nurses (Neslihan Devici, Carola Mix, Janet Cornberg, Jennifer Witt, Julia Schneider) and the physicians (Christopher Dietz, Kerstin Port, Tammo Tergast) of the Hepatitis Outpatient Clinic of the Department of Gastroenterology, Hepatology, Infectious Diseases and Endocrinology of Hannover Medical School for the care of the patients in the patient registry. We thank all patients for participating in our research study and for donating blood.
Data were presented at the EASL congress 2023 (21.06.–24.06.2023 in Vienna, Austria, abstract n°3043), the DZIF annual meeting 2023 (25.09.2023–26.09.2023 in Hannover, Germany), the 40th Annual Conference German Assocation for the Study of the Liver (GASL, 26.01.–27.01.2024 in Essen, Germany, abstract 100405) and the EASL congress 2024 (05.06.–08.06.2024 in Milano, Italy, abstract n° SAT-363).
The datasets generated and analyzed in the current study are available in the Synapse repository (https://www.synapse.org/#!Synapse:syn52525902).
-
Conflicts of Interest
MW, CO, JT, HS, JM, GG, MB, LR, TW, TK and AK have nothing to disclose.
KD reports research grants and personal fees from AbbVie, Alnylam and Gilead, outside the submitted work.
BM received speaker and/or consulting fees from Abbott Molecular, Astellas, Intercept, Falk, AbbVie, Luvos, Norgine, Gore, Gilead, Fujirebio, Merck (MSD), and Roche. He also received research support from Abbott Molecular, Altona Diagnostics, EWIMED, Fujirebio and Roche, outside the submitted work.
HW reports grants/research support and personal fees from Abbvie, Biotest AG and Gilead. He received personal fees from Aligos Therapeutics, Altimmune, Astra Zeneca, Bristol-Myers-Squibb, BTG Pharmaceuticals, Dicerna Pharmaceuticals, Enanta Pharmaceuticals, Dr. Falk Pharma, Falk Foundation, Intercept Pharmaceuticals, Janssen, Merck KGaA, MSD Sharp & Dohme GmbH, MYR GmbH, Norgine, Novartis, Pfizer Pharma GmbH, Roche and Vir Biotechnology, outside the submitted work.
MC reports honoraria for lectures and consulting from AbbVie Deutschland GmbH & Co. KG, AiCuris AG, Falk Foundation e.V., Gilead Sciences GmbH, GSK Service Unlimited, MSD Sharp & Dohme GmbH, Novartis AG, Roche AG, Spring Bank Pharmaceuticals, Swedish Orphan Biovitrum AB (SOBI), outside the submitted work.
SUPPLEMENTAL MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 1.
Criteria for inclusion and exclusion of HCV patients. DAA, direct-acting antiviral; eGFR, estimated glomerular filtration rate; HCV, hepatitis C virus.
Supplementary Figure 2.
Additional boxplots of SIM altered in both cohorts at therapy start. Significance was determined by using Welch Two-Sample t-test and P-values were adjusted using the Benjamini–Hochberg procedure. Healthy controls (blue), cirrhosis (red), no cirrhosis (turquoise). ***P<0.001, **P<0.01, *P<0.05, NS P>0.05.
Supplementary Figure 3.
Additional longitudinal boxplots of SIM altered at therapy start. Significance was determined by using paired Welch Two-Sample t-test and unpaired Welch Two-Sample t-test and P-values were adjusted using the Benjamini–Hochberg procedure. Healthy Controls (blue), Cirrhosis (red), no Cirrhosis (turquoise). ***P<0.001, **P<0.01, *P<0.05, NS P>0.05.
Supplementary Figure 4.
Selected clinical parameter in cirrhotic HCV patients at therapy start and FU. Shown are clinical parameters for patients with elastography ≥14 kPa at FU (dark red) and <14 kPa at follow up (red). Reference thresholds or areas are marked in grey. ALT, alanine aminotransferase; AST, aspartate aminotransferase; FU, follow-up; HCV, hepatitis C virus; INR, International Normalized Ratio; TS, therapy start.
Supplementary Figure 5.
Heatmap of SIM alterations before start of DAA therapy in patients defined as non-cirrhotic and cirrhotic before DAA therapy (initiation of therapy). Patients with cirrhosis at baseline were categorized into two groups based on their elastography values at FU (n=15 ≥14 kPa; n=29 <14 kPa). Shown are row-wise z-score normalized NPX values. DAA, direct-acting antiviral; FU, followup; NPX, normalized protein expression; SIM, soluble immune mediators.
Supplementary Figure 6.
Additional correlating SIM in all HCV patients at follow-up. Correlations were calculated using two-sided Pearson correlation and P-values were adjusted using the Benjamini–Hochberg procedure. HCV, hepatitis C virus; NPX, normalized protein expression; NS, not significant; SIM, soluble immune mediators. ***P<0.001, **P<0.01, *P<0.05, NS P>0.05.
Supplementary Figure 7.
Differences in SIM expression between cirrhotic patients of cohort A and the validation cohort. (A) Heatmap of the 61 SIM that were altered in cirrhotic patients compared to healthy controls at therapy start in cohort A (n=46) and in the validation cohort (n=47). Shown are significant results compared to healthy controls. (B) Heatmap of the 46 SIM that were altered in cirrhotic patients compared to healthy controls at follow-up in cohort A (n=46) and in the validation cohort (n=47). Shown are significant results compared to healthy controls. (C) Volcano plot of the validation cohort compared to cohort A at therapy start. Significantly increased SIM are labeled in red and significantly decreased SIM in blue (adjusted [adj.] P<0.05). Non-significant SIM are represented as grey dots (adj. P≥0.05). (D) Volcano plot of the validation cohort compared to cohort A at therapy start. Significantly increased SIM are labeled in red and significantly decreased SIM in blue (adj. P<0.05). Non-significant SIM are represented as grey dots (adj. P≥0.05). Estimated differences and P-values were calculated using Welch Two-Sample t-test and P-values were adjusted using the Benjamini–Hochberg procedure. SIM, soluble immune mediators.
Supplementary Figure 8.
Score calculated by adding the SIM levels of HGF, IL-8, uPA, and IL-6 and subtracting the levels of SCF. (A) Score shown for non-cirrhotic patients (cohort B) and cirrhotic patients (cohort A) developing HCC and not developing HCC (yellow) at TS, EOT, and follow-up (FU). Welch Two-Sample t-test was performed comparing the score in cirrhotic patients developing HCC and cirrhotic patients not developing HCC (P=0.04 at TS, P=0.29 at EOT, P=0.02 at FU). (B) Score shown for the validation cohort for patients developing HCC and patients not developing HCC at TS, EOT, and FU. Welch Two-Sample t-test was performed comparing the score in patients developing HCC and patients not developing HCC (P=0.5 at TS, P=0.35 at EOT, P=0.12 at FU). (C) Score in patients developing HCC and patients not developing HCC in the cirrhosis cohort (cohort A) at therapy start. (D) Score in patients developing HCC and patients not developing HCC in the validation cohort at therapy start. (E) Score in patients developing HCC and patients not developing HCC in the cirrhosis cohort (cohort A) at follow-up. (F) Score in patients developing HCC and patients not developing HCC in the validation cohort at follow-up. Significance was determined by using Welch Two-Sample t-test and P-values were adjusted using the Benjamini– Hochberg procedure. EOT, end of treatment; FU, follow-up; HCC, hepatocellular carcinoma; HGF, hepatocellular growth factor; IL-8, interleukin-8; SCF, stem cell factor; SIM, soluble immune mediators; TS, therapy start; uPA, urokinase plasminogen activator.
Supplementary Figure 9.
Liver stiffness in the cirrhosis cohort (cohort A) and the validation cohort. (A) Elastography values of the cirrhosis cohort (cohort A) at therapy start and follow-up. Patients developing HCC are marked in yellow and patients not developing HCC in red. The reference threshold of 14 kPa is colored in grey. (B) Elastography values of the validation cohort at therapy start and followup. Patients developing HCC are marked in yellow and patients not developing HCC in red. The reference threshold of 14 kPa is colored in grey. FU, follow-up; HCC, hepatocellular carcinoma; TS, therapy start.
Figure 1.SIM expression of chronic HCV patients at baseline compared to healthy controls. (A) Significantly increased SIM are labeled in red and significantly decreased SIM in blue (adjusted [adj.] P-value <0.05). Non-significant SIM are represented as grey dots (adj. P-value ≥0.05). Estimated differences and P-values were calculated using Welch Two-Sample t-test and P-values were adjusted using the Benjamini–Hochberg procedure. (B) Adjusted P-values of the 51 SIM changed in chronic HCV patients at baseline compared to healthy controls. P-values were calculated using Welch Two-Sample t-test and adjusted using the Benjamini–Hochberg procedure. HCV, hepatitis C virus; SIM, soluble immune mediators.
Figure 2.The SIM milieu of cirrhotic and non-cirrhotic patients is distinct at baseline. (A) Heatmap of significantly altered SIM in cirrhotic (cohort A) and non-cirrhotic (cohort B) patients compared to healthy controls at baseline. Shown are to controls normalized to the means of NPX values. Significance was determined by using Welch Two-Sample t-test and P-values were adjusted using the Benjamini–Hochberg procedure. The first row shows the significance (adjusted [adj.] P<0.05), SIM changes in both groups are marked in blue, SIM changes in cirrhotic patients only are red, SIM changes in non-cirrhotic patients only are yellow. (B) Boxplots of selected SIM altered in cirrhotic and non-cirrhotic patients at therapy start. Significance was determined by using Welch Two-Sample t-test and P-values were adjusted using false discovery rate. Healthy controls (blue), cirrhosis (red), no cirrhosis (turquoise). ***P<0.001, **P<0.01, *P<0.05, NS P>0.05. (C) Boxplots of selected SIM altered in cirrhotic patients at therapy start. Significance was determined by using Welch Two-Sample t-test and P-values were adjusted using false discovery rate. Healthy controls (blue), cirrhosis (red), no cirrhosis (turquoise). NPX, normalized protein expression; NS, not significant; SIM, soluble immune mediators. ***P<0.001, **P<0.01, *P<0.05, NS P>0.05.
Figure 3.Non-cirrhotic and cirrhotic patients show distinct SIM profiles. Spidergraphs of estimated difference in means compared to healthy controls of all analyzed SIM at therapy start (A), end of treatment (B) and follow-up (C). Chronic HCV patients with cirrhosis (cohort A) are red, chronic HCV patients without cirrhosis (cohort B) are turquoise and healthy controls are blue. Estimated differences were calculated using Welch Two-Sample t-test. HCV, hepatitis C virus; NS, not significant; SIM, soluble immune mediators.
Figure 4.Kinetics of SIM that were altered in non-cirrhotic and cirrhotic patients at therapy start. (A) Heatmap of the 39 SIM that were altered in both cohorts compared to healthy controls at baseline (see Fig. 2A). Shown are significant results for all three time points compared to healthy controls. Significance was determined by using Welch Two-Sample t-test and P-values were adjusted using the Benjamini– Hochberg procedure. (B) NPX values of selected SIM that were significantly altered in both cohorts before therapy with different patterns from cirrhotic (cohort A) and non-cirrhotic (cohort B) chronic HCV patients at follow-up and healthy controls. Significance was determined by using paired Welch Two-Sample t-test and P-values were adjusted using the Benjamini–Hochberg procedure. Healthy controls (blue), cirrhosis (red), no cirrhosis (turquoise). HCV, hepatitis C virus; NPX, normalized protein expression; NS, not significant; SIM, soluble immune mediators. ***P<0.001, **P<0.01, *P<0.05, NS P>0.05.
Figure 5.Persistent SIM alterations are associated with liver stiffness. (A) Heatmap of SIM alterations persistent in cirrhotic patients at follow-up in patients defined as non-cirrhotic and cirrhotic before DAA therapy (initiation of therapy). Patients with cirrhosis at baseline were categorized into two groups based on their elastography values at follow-up (n=15 <14 kPa; n=29 ≥14 kPa). Shown are row-wise z-score normalized NPX values. (B) Correlations between NPX levels and elastography values of selected SIM that persisted in cirrhotic patients at follow-up in all HCV patients at follow-up. Correlations were calculated using Pearson correlation and P-values were adjusted using the Benjamini–Hochberg procedure. DAA, direct-acting antiviral; HCV, hepatitis C virus; NPX, normalized protein expression; NS, not significant; SIM, soluble immune mediators. ***P<0.001, **P<0.01, *P<0.05, NS P>0.05.
Figure 6.Network interaction analysis of persistently altered SIM correlating with elastography. (A) Network was created via StringDB; edges are shown based on confidence (>0.7). The strength describes the enrichment effect based on log10 (observed/expected). (B) KEGG pathway enrichment analysis of altered SIM correlating with elastography. Enrichment significance (–log10 of Benjamini–Hochberg adjusted P-values) is plotted. KEGG, Kyoto Encyclopedia of Genes and Genomes; SIM, soluble immune mediators.
Figure 7.Persistently high plasma levels of HGF, uPA, and IL-8 in patients developing HCC. (A) Line plots with error bars of non-cirrhotic patients (cohort B) and cirrhotic patients (cohort A) developing HCC and not developing HCC over time. (B) Line plots with error bars of the validation cohort of patients developing HCC and not developing HCC over time. HCC, hepatocellular carcinoma; HGF, hepatocellular growth factor; IL-8, interleukin-8; uPA, urokinase plasminogen activator.
Table 1.Clinical characteristics at baseline
Table 1.
|
Characteristics |
Cohort A (cirrhosis) |
Cohort B (no cirrhosis) |
Controls |
Validation (cirrhosis) |
Reference |
|
Number |
46 |
56 |
51 |
47 |
|
|
Age |
58 (±9) |
58 (±12) |
58 (±20) |
57 (±9) |
|
|
Sex |
|
|
|
|
|
|
Female |
12 (26.1) |
29 (51.8) |
21 (41.2) |
10 (21.3) |
|
|
Male |
34 (73.9) |
27 (48.2) |
30 (58.8) |
37 (78.7) |
|
|
BMI (kg/m2) |
27.5 (±3.7) |
26.6 (±5.2) |
- |
27 (±3.9) |
18.5–24.9 |
|
Elastography (kPa) |
34.6 (±17.4) |
7.6 (±2.5) |
- |
38.2 (±17.1) |
2–7 |
|
Leukocytes (Tsd/µl) |
5.5 (±1.9) |
6.9 (±1.8) |
- |
5.1 (±1.5) |
3.9–10.2 |
|
Hemoglobin (g/dL) |
13.9 (±1.6) |
14.8 (±1.4) |
- |
13.8 (±2.2) |
13.5–17.2 (M) |
|
12–15.6 (F) |
|
Platelets (Tsd/µl) |
121 (±54) |
209 (±62) |
- |
84 (±40.8) |
160–370 |
|
Neutrophils (Tsd/µl) |
3.1 (±1.1) |
4.0 (±1.4) |
- |
3 (±1.1) |
1.5–7.7 |
|
INR |
1.3 (±0.3) |
1.0 (±0.2) |
- |
1.3 (±0.3) |
0.90–1.25 |
|
AST (U/L) |
119.2 (±75) |
53.9 (±27.7) |
- |
952 (±39.8) |
0–35 (M) |
|
0–31 (F) |
|
ALT (U/L) |
124 (±93) |
73.2 (±47.3) |
- |
92 (±57.5) |
0–45 (M) |
|
0–34 (F) |
|
HCV-RNA (IU/mL) |
1,318,155 (±1,425,057) |
2,634,406 (±2,972,906) |
- |
1,291,548 (±1,830,400) |
0 |
Abbreviations
aspartate aminotransferase
hepatocellular growth factor
interferon-stimulated gene
normalized protein expression
sustained virological response
urokinase plasminogen activator
REFERENCES
- 1. World Health Organization (WHO). Hepatitis C. WHO web site, <https://www.who.int/news-room/fact-sheets/detail/hepatitisc>. Accessed 22 Oct 2024.
- 2. Huang DQ, Terrault NA, Tacke F, Gluud LL, Arrese M, Bugianesi E, et al. Global epidemiology of cirrhosis - aetiology, trends and predictions. Nat Rev Gastroenterol Hepatol 2023;20:388-398.
- 3. Cacoub P, Saadoun D. Extrahepatic manifestations of chronic HCV infection. N Engl J Med 2021;384:1038-1052.
- 4. Lockart I, Yeo MGH, Hajarizadeh B, Dore GJ, Danta M. HCC incidence after hepatitis C cure among patients with advanced fibrosis or cirrhosis: A meta-analysis. Hepatology 2022;76:139-154.
- 5. Negro F. Residual risk of liver disease after hepatitis C virus eradication. J Hepatol 2021;74:952-963.
- 6. Tahata Y, Hikita H, Mochida S, Enomoto N, Kawada N, Kurosaki M, et al. Liver-related events after direct-acting antiviral therapy in patients with hepatitis C virus-associated cirrhosis. J Gastroenterol 2022;57:120-132.
- 7. Cornberg M, Manns MP. The curing regimens of HCV: A SWOT analysis. Antivir Ther 2022;27:13596535211072672.
- 8. Cornberg M, Mischke J, Kraft AR, Wedemeyer H. Immunological scars after cure of hepatitis C virus infection: Long-HepC? Curr Opin Immunol 2023;82:102324.
- 9. Strunz B, Hengst J, Deterding K, Manns MP, Cornberg M, Ljunggren HG, et al. Chronic hepatitis C virus infection irreversibly impacts human natural killer cell repertoire diversity. Nat Commun 2018;9:2275.
- 10. Hamdane N, Jühling F, Crouchet E, El Saghire H, Thumann C, Oudot MA, et al. HCV-induced epigenetic changes associated with liver cancer risk persist after sustained virologic response. Gastroenterology 2019;156:2313-2329.e2317.
- 11. Owusu Sekyere S, Port K, Deterding K, Cornberg M, Wedemeyer H. Inflammatory patterns in plasma associate with hepatocellular carcinoma development in cured hepatitis C cirrhotic patients. United European Gastroenterol J 2021;9:486-496.
- 12. Du Y, Khera T, Strunz B, Deterding K, Todt D, Woller N, et al. Imprint of unconventional T-cell response in acute hepatitis C persists despite successful early antiviral treatment. Eur J Immunol 2022;52:472-483.
- 13. Heim MH, Thimme R. Innate and adaptive immune responses in HCV infections. J Hepatol 2014;61:S14-25.
- 14. Khera T, Du Y, Todt D, Deterding K, Strunz B, Hardtke S, et al. Long-lasting imprint in the soluble inflammatory milieu despite early treatment of acute symptomatic hepatitis C. J Infect Dis 2022;226:441-452.
- 15. Hengst J, Falk CS, Schlaphoff V, Deterding K, Manns MP, Cornberg M, et al. Direct-acting antiviral-induced hepatitis C virus clearance does not completely restore the altered cytokine and chemokine milieu in patients with chronic hepatitis C. J Infect Dis 2016;214:1965-1974.
- 16. Du Y, Khera T, Liu Z, Tudrujek-Zdunek M, Dworzanska A, Cornberg M, et al. Controlled attenuation parameter is associated with a distinct systemic inflammatory milieu after clearance of HCV infection. Biomedicines 2023;11.
- 17. Coco B, Oliveri F, Maina AM, Ciccorossi P, Sacco R, Colombatto P, et al. Transient elastography: a new surrogate marker of liver fibrosis influenced by major changes of transaminases. J Viral Hepat 2007;14:360-369.
- 18. Roesner LM, Gupta MK, Kopfnagel V, van Unen N, Kemmling Y, Heise JK, et al. The RESIST senior individuals cohort: design, participant characteristics and aims. Geroscience 2025;47:3299-3310.
- 19. Nevola K, Sandin M, Guess J, Forsberg S, Cambronero C, Pucholt P, et al. OlinkAnalyze: Facilitate Analysis of Proteomic Data from Olink. rOpenSci web site, <https://olink-proteomics.r-universe.dev/OlinkAnalyze?utm_source>. Accessed Feb 2023.
- 20. Carlin AF, Aristizabal P, Song Q, Wang H, Paulson MS, Stamm LM, et al. Temporal dynamics of inflammatory cytokines/chemokines during sofosbuvir and ribavirin therapy for genotype 2 and 3 hepatitis C infection. Hepatology 2015;62:1047-1058.
- 21. Helbig KJ, Ruszkiewicz A, Lanford RE, Berzsenyi MD, Harley HA, McColl SR, et al. Differential expression of the CXCR3 ligands in chronic hepatitis C virus (HCV) infection and their modulation by HCV in vitro. J Virol 2009;83:836-846.
- 22. Brownell J, Bruckner J, Wagoner J, Thomas E, Loo YM, Gale M Jr., et al. Direct, interferon-independent activation of the CXCL10 promoter by NF-κB and interferon regulatory factor 3 during hepatitis C virus infection. J Virol 2014;88:1582-1590.
- 23. Chalin A, Lefevre B, Devisme C, Barget N, Amiot L, Samson M. Circulating levels of CXCL11 and CXCL12 are biomarkers of cirrhosis in patients with chronic hepatitis C infection. Cytokine 2019;117:72-78.
- 24. Garcia-Broncano P, Medrano LM, Berenguer J, González-García J, Jiménez-Sousa M, Carrero A, et al. Dysregulation of the immune system in HIV/HCV-coinfected patients according to liver stiffness status. Cells 2018;7.
- 25. Shah S, Ma Y, Scherzer R, Huhn G, French AL, Plankey M, et al. Association of HIV, hepatitis C virus and liver fibrosis severity with interleukin-6 and C-reactive protein levels. Aids 2015;29:1325-1333.
- 26. Debes JD, van Tilborg M, Groothuismink ZMA, Hansen BE, Schulze Zur Wiesch J, von Felden J, et al. Levels of cytokines in serum associate with development of hepatocellular carcinoma in patients with HCV infection treated with direct-acting antivirals. Gastroenterology 2018;154:515-517.e513.
- 27. Michalopoulos GK, Bhushan B. Liver regeneration: biological and pathological mechanisms and implications. Nat Rev Gastroenterol Hepatol 2021;18:40-55.
- 28. Filliol A, Saito Y, Nair A, Dapito DH, Yu LX, Ravichandra A, et al. Opposing roles of hepatic stellate cell subpopulations in hepatocarcinogenesis. Nature 2022;610:356-365.
- 29. Fazel Modares N, Polz R, Haghighi F, Lamertz L, Behnke K, Zhuang Y, et al. IL-6 trans-signaling controls liver regeneration after partial hepatectomy. Hepatology 2019;70:2075-2091.
- 30. Schmidt-Arras D, Rose-John S. IL-6 pathway in the liver: from physiopathology to therapy. J Hepatol 2016;64:1403-1415.
- 31. Wang FD, Zhou J, Chen EQ. Molecular mechanisms and potential new therapeutic drugs for liver fibrosis. Front Pharmacol 2022;13:787748.
- 32. Wang H, Rao B, Lou J, Li J, Liu Z, Li A, et al. The function of the HGF/c-Met axis in hepatocellular carcinoma. Front Cell Dev Biol 2020;8:55.
- 33. Radmanić L, Bodulić K, Šimičić P, Vince A, Lepej S. The effect of treatment-induced viral eradication on cytokine and growth factor expression in chronic hepatitis C. Viruses 2022;14.
