Markers of Natural Killer Cell Exhaustion in HIV/HCV Coinfection and Their Dynamics After HCV Clearance Mediated by Direct-Acting Antivirals

Abstract Background Liver fibrosis is a leading cause of morbimortality in people with HIV/hepatitis C virus (HCV). Natural killer (NK) cells are linked with amelioration of liver fibrosis; however, NK cells from individuals coinfected with HIV/HCV with cirrhosis display impaired functionality and high PD-1 expression. Here, we aimed to study PD-1, TIGIT, and Tim3 as potential exhaustion markers in NK cells from persons coinfected with HIV/HCV with mild and advanced liver fibrosis. We also evaluated the role of PD-1 expression on NK cells after HCV clearance by direct-acting antivirals (DAAs). Methods Peripheral blood mononuclear cells were isolated from individuals coinfected with HIV/HCV (N = 54; METAVIR F0/F1, n = 27; F4, evaluated by transient elastography, n = 27). In 26 participants, samples were collected before, at the end of, and 12 months after successful DAA treatment. The frequency, immunophenotype (PD-1, TIGIT, and Tim3 expression), and degranulation capacity (CD107a assay) of NK cells were determined by flow cytometry. Results Unlike PD-1, Tim3 and TIGIT were comparably expressed between persons with mild and advanced fibrosis. Degranulation capacity was diminished in NK/TIGIT+ cells in both fibrosis stages, while NK/PD-1+ cells showed a lower CD107a expression in cirrhotic cases. Twelve months after DAA treatment, those with advanced fibrosis showed an improved NK cell frequency and reduced NK/PD-1+ cell frequency but no changes in CD107a expression. In individuals with mild fibrosis, neither PD-1 nor NK cell frequency was modified, although the percentage of NK/CD107a+ cells was improved at 12 months posttreatment. Conclusions Although DAA improved exhaustion and frequency of NK cells in cirrhotic cases, functionality was reverted only in mild liver fibrosis, remarking the importance of an early DAA treatment.

Nearly 71 million people worldwide are chronically infected with hepatitis C virus (HCV) [1].Estimates indicate that 25% to 30% of those will develop cirrhosis within 20 years of infection [2].Among other factors, HIV coinfection has been demonstrated to significantly accelerate fibrosis progression and development of end-stage liver disease [3,4], even in the presence of antiretroviral therapy [3].With a global assessment of 2.3 million people coinfected with HIV/HCV [5], monitoring hepatic disease and preventing liver damage constitute essential recommendations for the clinical management of this population [6].
Direct-acting antivirals (DAAs) represent a major advancement in the treatment of HCV infection, since HCV clearance can be achieved in >90% of the cases, with minimal side effects [7].However, the impact of DAA treatment on hepatic fibrosis still needs to be evaluated in larger populations and broader follow-ups [8].Currently, studies show that DAA treatment is associated with the resolution of hepatic inflammation and improvement of fibrosis, especially among persons with mild to moderate levels of liver damage.Nevertheless, a great proportion of those with baseline advanced liver disease stay within cirrhotic scores [8].Unraveling the mechanisms that regulate liver fibrosis is vital since nearly half a million people die annually from decompensated cirrhosis related to chronic HCV infection [1].Moreover, end-stage liver disease constitutes a major cause of death among persons coinfected with HIV/HCV [9].
Natural killer (NK) cells play an important role in inhibiting hepatic fibrosis by killing early activated or senescence-activated hepatic stellate cells [10].It has been shown that peripheral NK cell frequency is significantly decreased in HCV monoinfection and HIV/HCV coinfection [11,12].The functionality of these Markers of NK Cell Exhaustion in HIV/HCV Coinfection • OFID • 1 Open Forum Infectious Diseases M A J O R A R T I C L E cells has also been reported to be severely compromised in both groups [11,13,14].Previously, we demonstrated that persons coinfected with HIV/HCV with advanced fibrosis are characterized by having low peripheral NK cell frequencies as compared with healthy volunteers and those with minimal fibrosis.
Furthermore, NK cell degranulation and cytokine secretion were significantly diminished in samples from patients with higher levels of fibrosis.When PD-1 expression was assessed on the NK cell compartment, PD-1 expression was significantly upregulated in cirrhotic cases, despite presenting similar times of known HIV and HCV infection, time of antiretroviral therapy, HCV viral load, and HCV genotype to those observed in cases of mild hepatic disease [12,15].
The PD-1 molecule has been extensively studied in T cells, B cells, and dendritic cells [16]; however, less is known about PD-1 expression and NK cell functionality.Several studies have shown that PD-1 is highly expressed on peripheral and tumor-infiltrating NK cells in patients experiencing different malignancies and that PD-1 axis blocking significantly enhances the cytokine production, degranulation, and viability of these cells [17].PD-1 expression on NK cells was also linked to chronic HCV infection [18], although the relationship between progression of liver fibrosis and the PD-1 axis in NK cells has not been completely addressed.Additional surface proteins have been evaluated as potential NK cell immune checkpoints, including CD96, LAG3, T-cell immunoglobulin, mucin domain-containing 3 (Tim3), and T-cell immunoreceptor with Ig and ITIM domains (TIGIT) [19][20][21].Of those, Tim-3 and TIGIT have been studied in viral hepatitis, and it has been suggested that expression of these markers is linked to exhausted phenotypic characteristics [22].
The aim of this work was to characterize PD-1, TIGIT, and Tim3 as potential biomarkers for NK cell exhaustion among persons coinfected with HIV/HCV with different degrees of liver fibrosis.In this regard, we evaluated the association between NK cell exhaustion and progression of liver damage.Additionally, we explored how baseline hepatic fibrosis levels affect PD-1 expression and NK cell dynamics following HCV eradication by DAA.By studying peripheral blood mononuclear cells (PBMCs) from individuals coinfected with HIV/HCV with mild and advanced liver fibrosis, we showed that PD-1 is a selective marker of NK cell exhaustion and is significantly linked to advanced fibrosis stages.Last, although NK cell frequency is mildly improved, HCV clearance does not completely restore NK cell functionality in cases of HIV/HCV coinfection and advanced liver fibrosis.

Study Cohort
This study included individuals who were coinfected with HIV/ HCV (n = 54; 27 with METAVIR F0/F1 and 27 with F4), HIV seropositive (HIV + , n = 6), HCV seropositive (HCV + , n = 9), and HIV/HCV seronegative (healthy controls, n = 5).Written informed consent was obtained, and 60 mL of peripheral blood was drawn.The study was conducted in accordance with the Declaration of Helsinki and was approved by the Bioethics Committee of Fundación Huésped.Persons coinfected with HIV/HCV were allocated to 2 groups based on their levels of fibrosis according to transient hepatic elastography (FibroScan and SuperSonic Imagine's Aixplorer).Those with a result ≤7.1 kPa were classified as compatible with a METAVIR score of F0/F1 (absent or minimal fibrosis) and those with ≥12.5 kPa as compatible with F4 (cirrhosis) [23].Participants enrolled in this study were not acutely or chronically HBV infected (determined by serology) and denied current use of recreational drugs and 14 units per week of alcohol intake on a regular basis.From the 54 individuals who were coinfected, 26 blood samples (14 with F0/F1 and 12 with F4) were collected at 3 times: baseline (prior to DAA treatment), end of treatment (EOT), and 12 months posttreatment (12MPT).

Cell Isolation and Culture
PBMCs were obtained from whole blood by Ficoll-Hypaque centrifugation (GE Healthcare) and cryopreserved at −80 °C for up to 4 months.Cells were cultured in complete RPMI-1640 medium (cRPMI) containing 10% fetal bovine serum, 2mM L-glutamine, 100 IU/mL of penicillin, and 100 μg/mL of streptomycin (all reagents, Gibco SRL).Three days before the experiments, the chronic myelogenous leukemia K562 cell line was thawed and grown at 37 °C and 5% CO 2 in cRPMI.

CD107a Assay
For degranulation assays, the K562 cell line was used as a stimulus and a sensible target for NK cells.PBMCs were thawed and cultured for 2 hours in cRPMI, at 37 °C with 5% CO 2 .Next, 1 million viable PBMCs were coincubated for 5 hours with 10 5 K562 cells, in the presence of anti-CD107a-FITC mAb, brefeldin, and monensin (4 μL, 10 μg/mL, and 0.7 μg/mL, respectively; BD Biosciences).To assess basal levels of degranulation, PBMCs were incubated in the absence of K562 cells.

Multicolor Flow Cytometry
Cells were immunophenotyped by flow cytometry on a FACS Canto Flow Cytometer (BD Biosciences).For antibodies and gating strategies, see Supplementary Table 1 and Supplementary Figure 1.Flow cytometry data was analyzed with FlowJo software version 10 (BD Biosciences).Phenotype and functionality assays were performed according to cell availability.

Data Analysis
Statistical analysis was performed with Prism version 8 (GraphPad).Data normality was assessed with the Shapiro-Wilk test and subsequently analyzed by nonparametric methods.All tests were considered significant when P < .05.

RESULTS
To evaluate biomarkers for NK cell exhaustion and its relationship with liver fibrosis, blood samples were obtained from persons coinfected with HIV/HCV with mild and advanced hepatic fibrosis, as assessed by transient elastography and METAVIR staging.Only participants with extreme F0/F1 and F4 levels were included in this study.This strategy was used to minimize METAVIR misclassification by transient hepatic elastography, which is less accurate than liver biopsies to determine liver fibrosis.PBMCs were isolated as described previously.The characteristics of study participants are summarized in Table 1.In accordance with METAVIR stage, individuals with advanced fibrosis presented higher indicators of liver stiffness, AST-platelet ratio index (APRI), AST, and total bilirubin than those with mild fibrosis, as well as a lower platelet count.

PD-1 Expression in NK Cells Is Associated With Cell Activation and Exhaustion in Individuals Coinfected With HIV/HCV
In comparison with individuals coinfected with HIV/HCV with mild fibrosis, NK cells from those with advanced fibrosis displayed a higher median fluorescence intensity of PD-1 (Figure 1A).Also, the frequency of NK/PD-1 + cells peaked in advanced liver fibrosis as previously demonstrated [15].When PD-1 expression was further analyzed in peripheral NK cells from participants coinfected with HIV/HCV, PD-1 mainly subscribed to the CD56 dim NK cell subset, as described by others [24][25][26].Evaluation of the activation markers CD25   and CD69 showed that PD-1 expression was significantly associated with an activated NK cell phenotype in persons with mild and advanced hepatic fibrosis.In addition to this result, a reduction in Nkp46 expression was registered in NK/PD-1 + cells from those with mild fibrosis.No differences were found in NKG2D expression between NK/PD-1 + and NK/PD-1 − cells from cases of mild or advanced fibrosis (Figure 1B).To evaluate whether PD-1 expression was linked to impaired NK cell functionality, degranulation capacity (externalization of CD107a) of NK/PD-1 + and NK/PD-1 − cells was studied as previously described [15].Briefly, following stimulation with K562 cells, externalization of CD107a was monitored.When compared with the PD-1 − NK cell subset, expression of CD107a was significantly affected in PD-1 + NK cells (Figure 1C).Stratified analysis of the F0/F1 and F4 groups suggests that impaired NK cell degranulation is more frequently observed in cases of advanced liver fibrosis.In line with these latter results, when serum biochemical variables were analyzed, the frequency of peripheral NK/PD-1 + cells was negatively associated with albumin levels and prothrombin time and directly correlated with the liver stiffness, APRI score, and AST levels of individuals coinfected with HIV/HCV (Figure 1D).In sum, these results suggest that not only is PD-1 a marker for mature and activated NK cells but it is also linked to an exhausted phenotype and to more altered liver function tests.

PD-1 Is a Selective Marker of NK Cell Exhaustion in Individuals Coinfected With HIV/HCV With Liver Fibrosis
Next, we evaluated if PD-1 + -exhausted NK cells expressed additional markers of dysfunction.Tim3 and TIGIT have been reported as NK cell exhaustion markers in hepatic diseases [22,[27][28][29].To characterize Tim3 and TIGIT expression in NK cells from individuals coinfected with HIV/HCV, resting PBMCs were analyzed via flow cytometry.When frequencies of Tim3 + or TIGIT + NK cells were compared, no differences were found between groups with mild and advanced fibrosis (Figure 2A).Similar results were obtained when median fluorescence intensity was analyzed (results not shown).To additionally identify alterations in cell functionality due to Tim3 or TIGIT expression, a CD107a degranulation assay was performed as described previously.While similar frequencies of CD107a + NK cells were registered by Tim3 expression (Figure 2B), in cases of mild and advanced liver fibrosis, the proportion of degranulating NK cells was significantly reduced when TIGIT was expressed.Finally, we studied the distribution of NK cell populations defined by the expression of PD-1, Tim3, and TIGIT in individuals coinfected with HIV/HCV with mild and advanced liver fibrosis.Although there was a visible expansion of NK cell populations expressing PD-1 and either of the other markers among those with advanced fibrosis, the global distribution between the F0/F1 and F4 groups did not differ statistically (Figure 2C).Nonetheless, the coexpression of PD-1, Tim3, and TIGIT is rarely observed in NK cells in both groups studied.In sum, results show that PD-1 and TIGIT are associated with impaired NK cell functionality in persons coinfected with HIV/HCV, but only PD-1 is differentially expressed throughout the liver fibrosis stages.First, we evaluated PD-1 expression dynamics over time in the F0/F1 and F4 groups.As seen in Figure 3A, baseline levels of PD-1 expression in HIV/HCV are significantly higher than those displayed in HCV or HIV monoinfection or healthy control.Although all achieved a sustained virologic response after DAA treatment, globally, exhaustion of NK cells was not modified.Interestingly, when cases of advanced fibrosis were evaluated, HCV clearance was associated with decreased cell exhaustion in individuals with NK cells showing high PD-1 expression at baseline.On the contrary, for those few who had a low frequency of PD-1 + cells, the expression of this marker increased over time, although not significantly (Figure 3B).NK cell percentage and degranulation capacity were not directly associated in either of the liver fibrosis stages evaluated.While the frequency of NK cells did not change after DAA treatment in participants with mild fibrosis, the degranulation capacity was significantly improved (Figure 3C and 3D, left panel).In cases of advanced fibrosis, although the percentage of those cells was restored, loss of functionality was not.

DISCUSSION
The arrival of DAA treatment in the last decade has represented a great advance in HCV treatment, since viral clearance is possible in >90% of cases, with minimal adverse effects and short treatment schedules (2-3 months) [7].However, the capacity of DAA therapy to improve liver fibrosis, especially in cases of    Markers of NK Cell Exhaustion in HIV/HCV Coinfection • OFID • 7 advanced liver disease, is less clear [8,30].Liver fibrosis is directly correlated with increased morbidity and mortality due to liver failure and hepatocellular carcinoma [31][32][33].As advanced fibrosis persists, HCV elimination is not enough to restore health, and options to decrease the level of liver damage are needed.NK cells play a crucial role in the modulation of hepatic stellate cells, key cells in the generation of liver fibrosis.In the present study, we were able to add evidence regarding NK cells and their association with liver fibrosis in individuals coinfected with HIV/HCV.We have shown that NK cells from cases of HIV/HCV coinfection with advanced hepatic fibrosis express higher levels of PD-1 than those with mild liver disease (Figure 1A).Also, we demonstrated that PD-1 is significantly associated with a lower capacity of NK cells to degranulate (Figure 1C).The possibility of increasing NK cell functionality could benefit individuals and have an impact on liver tissue remodeling.Since the improvement of NK cell degranulation capacity was observed only in samples with lower fibrosis, this underscores the importance of designing strategies that could apply as possible treatments (eg, anti-PD-1 that has already been approved as adjunctive therapy to chemotherapy) to revert exhaustion and modify in vivo the function of NK cells.
Regarding additional markers of NK cell exhaustion that have been reported, we noted that the expression of TIGIT was associated with a decrease in impaired degranulation in cases of mild and advanced fibrosis (Figure 2B).Nevertheless, TIGIT was not differentially expressed on NK cells from individuals with different degrees of liver fibrosis (Figure 2A).Anti-TIGIT therapies are being studied for cancer treatment [34], so it is plausible to consider them as a therapy to improve NK functionality and secondary liver fibrosis.Tim3 was not associated with a lower capacity of degranulation in NK cells.However, a higher percentage of NK/Tim3 + cells was observed in those with mild and advanced fibrosis as compared with the control group (data not shown).Future analysis should aim to completely understand the role that this marker plays in the context of monoinfection and HIV/HCV coinfection.
HCV elimination by DAA treatment may exert a differential effect on different immune parameters, depending on one's level of liver fibrosis.Here, multiple NK cell parameters were longitudinally monitored after HCV eradication with DAA treatment by analyzing 3 sampling times: before DAA treatment (baseline), EOT, and 12MPT.In the case of mild fibrosis, an increase in NK cell degranulation capacity was noted at 12MPT (Figure 3D).Yet, with advanced fibrosis, this improvement in degranulation capacity was not found.Nevertheless, an increase in the frequency of NK cells was seen at 12MPT (Figure 3C), and a decrease in the percentage of NK/PD-1 + cells at EOT and 12MPT was detected (Figure 3B).This incomplete recovery of the NK cell compartment may suggest a persistent exhaustion of NK cells even after HCV has been eliminated by DAA treatment, particularly in those with advanced fibrosis, who are unable to improve their NK cell degranulation capacity.This is consistent with data reported by other groups describing a limited capacity of DAA treatments to revert the METAVIR score in those with F3 and F4 [35][36][37].These data reinforce the possibility to use the aforementioned immune checkpoint inhibitors to improve the functionality of NK cells and, as a consequence, modify their impact on liver tissue fibrosis.
When modulation of PD-1 expression in NK cells was evaluated following DAA therapy, no significant overall changes were documented in samples with mild or advanced fibrosis.Nevertheless, while NK/PD-1 + cell frequency is significantly higher at baseline than in control groups, this significance is lost after DAA treatment, suggesting a trend that could be better elucidated with a larger sample size and at longer followups.Additionally, when analyzing samples with advanced fibrosis, we observed that PD-1 expression at baseline was heterogenous, and we also observed different outcomes over time.In cases of F4 with high baseline expression of PD-1, a decrease of NK/PD-1 + cell frequency was noted at EOT and 12MPT vs baseline (Figure 3B, left panel).This result was as expected, as it has been found in the cohorts with advanced fibrosis [12,15] and could indicate an improvement in NK cell homeostasis; however, this does not seem to be enough to restore its NK cell degranulation capacity since no significant changes in this parameter were detected over time.With respect to individuals with a low baseline percentage of NK/PD-1 + cells, a paradoxical behavior is noted, since this frequency increases after HCV clearance (Figure 3B).In these participants with low PD-1 baseline expression in their NK cells, high cellular death was documented in the viability control in the flow cytometry analysis.Therefore, one possible explanation for this observation is that in those samples, NK cells presented a high level of exhaustion before DAA and died preferentially during the in vitro experimental conditions.If this is the case, the real proportion of PD-1 + NK cells would be underestimated in these samples and could explain the paradoxical outcome seen in this group.As for those with mild fibrosis, although there was no significant decrease in the percentage of NK PD-1 + cells as a function of time (Figure 3A), there was a significant increase in the degranulation capacity of NK cells at 12MPT (Figure 3D).This would suggest that at least in people with mild fibrosis, it is possible to recover the functionality of NK cells once HCV is cleared with DAA.
From the results reported in the present article, it could be inferred that the presence of advanced liver damage exerts a direct effect on NK cells, regardless of the presence of HCV.From our results, PD-1 and TIGIT emerged as markers of exhaustion in NK cells and compromised degranulation capacity.However, only PD-1 was differentially expressed, with a higher percentage of PD-1 + cells in individuals with advanced fibrosis.This observation could reflect more pronounced immune system exhaustion in people with advanced fibrosis.In line with this, it has been reported that those NK cells that coexpress Tim3 and TIGIT are increased in persons with advanced fibrosis [28].
One of the limitations that faced this study included the great heterogeneity of the data obtained from samples of HIV/HCV coinfection with mild and advanced fibrosis.A larger sample size could help to elucidate the role of each of the studied markers more clearly.Another limitation is that the functionality of NK cells was evaluated ex vivo, under conditions that did not allow taking into account the multiple interactions that occur within the organism, at the level of direct cellular interactions and soluble molecules, such as cytokines.

CONCLUSIONS
The results suggest that DAA therapy is not sufficient per se to reverse the state of general exhaustion of the NK cells and their loss of functionality in the short term, particularly in people with advanced fibrosis.Nevertheless, they shed light on possible modifiable immune parameters that could be involved in NK cell alterations.These results highlight the importance of the active search of new therapies (eg, checkpoint inhibitors) that allow reversion of the described exhausted immune phenotype, which may improve the level of liver fibrosis and reduce the risk of mortality associated with it.

Figure 1 .
Figure 1.Functionality of PD-1 + NK cells from individuals coinfected with HIV/HCV with mild and advanced liver fibrosis at baseline, before DAA treatment.A, Expression of PD-1 was evaluated by flow cytometry in resting PBMCs from healthy patients and those coinfected with HIV/HCV with METAVIR F0/F1 or F4 scores.NK cells were identified as CD3 − /CD56 + lymphocytes.The MFI of PD-1 expression on NK cells from the F0/F1 and F4 groups is shown with the frequency of PD-1 + cells in CD56 dim and CD56 bright subsets.B, Expression of CD25, CD69, Nkp46, NKG2D, and PD-1 was evaluated in resting PBMCs by flow cytometry.Frequency of NK cells positive for each activation marker is shown according to PD-1 status in F0/F1 (upper panel) or F4 (bottom) scores.C, CD107a externalization in PBMCs from participants coinfected with HIV/HCV that were incubated with K562 cells.Representative cytometry plots for CD107a expression after stimulation are shown for PD-1 − and PD-1 + NK cells (left panel).Frequency of CD107a + NK cells according to PD-1 expression in the whole population (middle panel) and the F0/F1 and F4 groups (right).D, Correlation between frequency of PD-1 + NK cells and several clinicopathologic characteristics.In each graph, the correlation coefficient and P values are shown.Blue and red points represent mild and advanced fibrosis, respectively.Data in bar graphs are presented as median (line), IQR (box), and range (error bars).Statistical comparisons were performed with the (A-C ) Wilcoxon matched-pairs signed rank test and (D) Spearman correlation.Each set of points represents a different person.DAA, direct-acting antiviral; HCV, hepatitis C virus; MFI, median fluorescence intensity; NK, natural killer; PBMC, peripheral blood mononuclear cell.

Figure 2 .
Figure 2. Coexpression of PD-1, Tim3, and TIGIT in NK cells from individuals coinfected with HIV/HCV and mild or advanced liver fibrosis.A, Expression of Tim3 (top) and TIGIT (bottom) was evaluated by flow cytometry in resting PBMCs from participants with HIV/HCV coinfection and METAVIR F0/F1 or F4 scores.Representative cytometry plots are shown on the left.Individual NK cell frequencies are indicated with the median (line), IQR (box), and range (error bars).B, CD107a externalization in PBMCs from cases of HIV/HCV coinfection that were incubated with K562 cells.Representative cytometry plots for CD107a expression after stimulation are shown for NK cell populations positive and negative for Tim3 and TIGIT (left).Frequency of CD107a + NK cells according to Tim3 or TIGIT expression in F0/F1 and F4 groups (right).Each set of points represents a different person.C, PD-1, Tim3, and TIGIT coexpression analysis in those with METAVIR F0/F1 or F4 scores, according to SPICE 6.0 software.NK cell frequencies are indicated.Statistical comparisons were performed with (A) the Mann-Whitney test, (B) the Wilcoxon matched-pairs signed rank, (C, left) permutation analysis, and (C, right) the Wilcoxon rank sum test.HCV, hepatitis C virus; NK, natural killer; PBMC, peripheral blood mononuclear cell.

Figure 3 .
Figure 3. Evolution of the NK cell compartment after DAA treatment in individuals coinfected with HIV/HCV and minimal or advanced liver fibrosis.A, Frequency of PD-1 + NK cells in those with METAVIR F0/F1 or F4 scores before DAA treatment (baseline), at the end of treatment (EOT), and 12 months posttreatment (12MPT).PD-1 + NK cell percentage is also shown for cases of HIV (HIV + ) and HCV (HCV + ) monoinfection as well as healthy controls.B, Frequency of NK/PD-1 + cells in participants with METAVIR F4 scores with baseline PD-1 expression: high (>5%, left) or low (<5%, right).C, NK cell percentage in those with METAVIR F0/F1 or F4 scores at baseline, EOT, and 12MPT; those who were HIV + or HCV + ; and healthy donors.D, CD107a externalization assessed in NK cells from participants with METAVIR F0/F1 or F4 scores at baseline, EOT, and 12MPT after incubation with K562 cells.Frequency of NK/CD107a + cells is also shown for HIV + or HCV + cases and healthy controls.Data in bar graphs are presented as median (line), IQR (box), and range (error bars).Statistical comparisons were performed with the Wilcoxon and Mann-Whitney test.DAA, direct-acting antiviral; HCV, hepatitis C virus; NK, natural killer.