Insights Into the Pathophysiology of Liver Disease in HCV/HIV: Does it End With HCV Cure?

Abstract

HCV-HIV coinfected patients exhibit rapid progression of liver damage relative to HCV monoinfected patients. The availability of new directly acting antiviral agents has dramatically improved outcomes for coinfected patients as sustained virologic response rates now exceed 95% and fibrosis-related parameters are improved. Nevertheless, coinfected patients still have a higher mortality risk and more severe hepatocellular carcinoma compared to HCV monoinfected patients, implying the existence of pathways unique to people living with HIV that continue to promote accelerated liver disease. In this article, we review the pathobiology of liver disease in HCV-HIV coinfected patients in the directly acting antiviral era and explore the mechanisms through which HIV itself induces liver damage. Since liver disease is one of the leading causes of non-AIDS-related mortality in HIV-positive patients, enhancing our understanding of HIV-associated fibrotic pathways will remain important for new diagnostic and therapeutic strategies to slow or reverse liver disease progression, even after HCV cure.

As of 2015, there were approximately 1.6 million persons in North America living with human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) [1]. The introduction of potent HIV antiretroviral therapy (ART) has substantially prolonged the survival of HIV-infected individuals, but this extension of life expectancy has allowed other comorbid conditions to manifest. In a recent Canadian population-based retrospective cohort study, the prevalence of liver disease was 7 and 10 times greater in HIV-positive men and women, respectively, than in their HIV-negative counterparts, and was in large part attributed to coinfection with the hepatitis C virus (HCV) [2]. Most people who acquire acute hepatitis C are unable to spontaneously clear the virus and develop chronic infection, which considerably increases their risk for cirrhosis [3] and thus hepatocellular carcinoma (HCC). Recent meta-analyses estimate that 20% of HIV-infected persons in North America are coinfected with HCV [1]. Coinfection is a major public health concern owing to similar methods of transmission—parenteral, perinatal, and sexual. Furthermore, HIV enhances HCV transmission and accelerates the course of HCV-related liver disease [4, 5]. While the effect of HCV on the natural history of HIV is unclear, it is evident that HCV coinfection categorically increases the risk of all-cause and liver-related mortality for HIV-positive persons [6].

Available evidence suggests that sustained virologic response (SVR)—undetectable HCV RNA—is associated with reduced incidence of hepatic decompensation and HCC and a lower mortality risk in HCV-infected patients [7]. In the era when pegylated interferon-alfa and ribavirin (PEG-IFN-alfa/RBV) was the standard of care for treating HCV, HCV-HIV coinfected patients did not respond as well as those infected with HCV alone [8, 9]. However, following the development and implementation of directly acting antiviral (DAAs) agents in 2011, HCV-HIV coinfected patients now achieve SVR rates on par with those of HCV monoinfected patients [10] that are also accompanied by improved measures of fibrosis, as assessed by transient elastography [11]. Indeed, the guidelines established by the American Association for the Study of Liver Diseases and the Infectious Diseases Society of America for the management of HCV-HIV coinfected patients recommend the same treatment regimens as those prescribed for patients with chronic hepatitis C [12].

In the pre-DAA era, hepatic fibrosis progressed more rapidly in HCV-HIV coinfected patients than in HCV monoinfected patients [13]. Recent data modeling using the Swiss HIV Cohort Study, however, predicts that this will not be the case if coinfected patients are identified and treated with DAAs at METAVIR F0 or F1 stage fibrosis [14]. Furthermore, a recent prospective study of 2 French national cohorts revealed that, in the context of ART and DAAs, HCV-HIV patients with compensated cirrhosis experience similar risks for further liver events and HCC as HCV monoinfected patients [15]. Therefore, it may seem that research into the pathogenesis of liver disease in HCV-HIV coinfected patients is of little consequence at present. However, coinfected patients still have a greater mortality risk and earlier onset and more aggressive HCC [15], suggesting that DAAs produce incomplete resolution of inflammatory and profibrogenic stimuli. In addition, it is becomingly increasingly evident that HIV infection itself can promote liver damage. For example, hepatic steatosis (fatty liver) is present in up to 36% of HIV-positive patients and portends an increased risk of developing advanced fibrosis [16]. Thus, in the absence of approved antifibrotic drugs, improving our understanding of mechanisms of hepatic fibrogenesis in HCV-HIV coinfection may reveal unique pathways that can be exploited to retard or reverse liver disease progression.

In this article, we will first review the pathogenesis of HCV-HIV related liver disease and elaborate on mechanisms through which HIV infection impacts fibrogenesis. We will then ultimately evaluate whether HCV clearance is sufficient to prevent further liver damage and progression to HCC.

HEPATIC FIBROSIS

Hepatic fibrosis is a reversible wound-healing response to sustained hepatocellular injury that results in increased extracellular matrix (ECM) deposition in the perisinusoidal region (space of Disse) by hepatic myofibroblasts. Progressive ECM accumulation disrupts blood flow through the sinusoids as the normal parenchymal architecture becomes dissected by fibrous septa. While an array of hepatic cells are involved in fibrogenesis, hepatic stellate cells (HSCs) are the primary drivers [17]. Chronic liver injury foments an inflammatory milieu that promotes the transdifferentiation of quiescent HSCs into activated, chemotactic, ECM-producing myofibroblasts. Activation is regulated by a variety of extracellular signals from both immune and parenchymal cells; of particular interest are hepatocytes, macrophages, and natural killer (NK) cells. Free oxygen radicals (reactive oxygen species) and apoptotic bodies released from damaged hepatocytes may directly or indirectly activate HSCs [18]. Macrophages, including both liver-resident Kupffer cells (KCs) and bone marrow monocyte-derived macrophages, release cytokines, such as transforming growth factor-β1 (TGF-β1) and tumor necrosis factor (TNF), that promote HSC activation. Binding of TGF-β to its cell surface receptors initiates phosphorylation of cytoplasmic SMAD proteins, which then shuttle into the nucleus to regulate transcription of type 1 collagen, one of the primary components of the ECM [19, 20]. In mice, TNF secreted by hepatic macrophages does not directly activate HSCs but instead facilitates their survival in a nuclear factor κ-light-chain–enhancer of activated B cells (NF-κB)-dependent manner [21]. In contrast to hepatocytes and profibrotic macrophages, which activate HSCs, NK cells have distinct antifibrotic activities, such as clearing senescent activated HSCs [22] and initiating HSC apoptosis through production of IFN-γ [23]. In the following sections we will review how HCV and HIV directly regulate these processes to promote fibrosis (Figure 1).

Figure 1.

HIV-associated profibrotic pathways. HIV induces ROS production and phosphorylation of NRF2 and SMAD3 in hepatocytes, which in turn activate AREs and TGF-β1 expression, respectively. HIV infection can also lead to hepatocyte apoptosis and ensuing efferocytosis by macrophages and HSCs, which ultimately promotes HSC transdifferentiation into ECM-producing myofibroblasts. HIV can directly activate HSCs and macrophages and initiate the release of cytokines that amplify the immune response. Importantly, macrophages and HSCs also express TLR4 and are sensitive to activation by LPS, circulating levels of which are elevated in HIV-infected patients due to microbial translocation. The intrahepatic accumulation of LPS also increases CXCL16 secretion by myeloid dendritic cells, which functions to attract cytotoxic NK cells to the liver. However, NK cell death increases, thus permitting the survival of activated HSCs. CD4⁺ T cells may also regulate NK cell activity through secretion of IL-2; however, in HIV-infected patients these cells are defective, preventing clearance of activated HSCs. Abbreviations: ARE, antioxidant response element; CCL2, C-C motif chemokine 2; CCR, C-C chemokine receptor; COL1A1, collagen type I α1; CXCL16, C-X-C motif chemokine 16; CXCR, C-X-C chemokine receptor; DR, death receptor; ECM, extracellular matrix; HIV, human immunodeficiency virus; HSC, hepatic stellate cell; IL-2, interleukin-2; LPS, lipopolysaccharide; mDC, myeloid dendritic cell; NK cell, natural killer cell; P-NRF2, phosphorylated nuclear factor erythroid 2-related factor 2; P-RELA, phosphorylated transcription factor p65; P-SMAD3, phosphorylated mothers against decapentaplegic homolog 3; ROS, reactive oxygen species; TGF-β1, transforming growth factor-β1; TIMP1, TIMP metallopeptidase inhibitor 1; TLR4, toll-like receptor 4; TNF, tumor necrosis factor; TRAIL, TNF-related apoptosis-inducing ligand receptor.

DIRECT VIRAL EFFECTS ON HEPATIC CELLS

Profibrogenic Cytokine Release and Oxidative Stress

CD4⁺ T lymphocytes, macrophages, and dendritic cells are well-characterized reservoirs for HIV. By contrast, HIV does not infect hepatocytes, but hepatocytes do express the canonical HIV coreceptors C-C motif chemokine receptor 5 (CCR5) and C-X-C motif chemokine receptor 4 (CXCR4) [24]. In Huh7.5.1 hepatoma cells, HCV JFH1 (Japanese patient with fulminant hepatitis; genotype 2a) and HIV, independently and cooperatively, induced oxidative stress and TGF-β1 signaling, evidenced by increased nuclear factor erythroid 2-related factor 2 (NRF2) and mothers against decapentaplegic homolog 3 (SMAD3) phosphorylation (activation), respectively [25]. When Huh7.5.1 cells were cocultured with LX-2 human HSCs in a transwell system, to facilitate the exchange of secreted soluble factors, HCV infection and HIV exposure similarly promoted reactive oxygen species generation and increased SMAD3 activation, resulting in the upregulation of the ECM-related genes collagen type 1 α 1 (COL1A1) and TIMP metallopeptidase inhibitor 1 (TIMP1) in HSCs [25]. The profibrogenic effects of HCV and HIV have been further buttressed by a 2019 study that modified the coculture system to incorporate HIV-infected primary human macrophages [26]. In this work, HCV (genotype 1a) and HIV induced LX-2 cell transdifferentiation and chemotaxis [26].

Hepatocyte Apoptosis

Phagocytosis of apoptotic hepatocytes by macrophages or HSCs (efferocytosis) may contribute to liver fibrosis [18, 27]. It has previously been shown that in vitro exposure of HCV JFH1 and CCR5- and CXCR4-tropic HIV to Huh7.5.1 cells induces caspase-dependent apoptosis through both the extrinsic and intrinsic (mitochondrial) pathways [28]. As with the synergistic interactions between HCV and HIV in activating profibrogenic pathways, both viruses also cooperatively induce Huh7.5.1 apoptosis [28]. These findings have clinical significance as plasma levels of the TNF superfamily member 10 (TNFSF10 or TRAIL) ligand that binds death receptors 4/5 (also known as TRAIL receptors 1/2), upon which activation of the extrinsic apoptotic pathway is dependent, are increased in HIV-infected individuals [29]. Notably, however, the HIV envelope can also activate caspase-independent apoptotic pathways in Huh7 cells [24].

Direct HSC Activation

In contrast to hepatocytes, in vitro experiments have suggested that HIV can infect HSCs, although through pathways separate from those involving CD4, CCR5, and CXCR4 [30]. HIV exposure to HSCs resulted in the propagation of infectious viral particles that could be transmitted to human MT4 T cells in coculture through physical interactions [30]. Moreover, HIV upregulated expression of type I collagen and monocyte chemoattractant protein-1 (MCP-1, also known as CCL2) in stellate cells [30]. Like hepatocytes, HSCs express low levels of CCR5 and CXCR4 and HIV gp120 can activate HSCs in a CCR5-dependent fashion as well [31]. HIV gp120-induced HSC activation is also, in part, regulated by toll-like receptor 4 (TLR4) as incubation with either a chemical inhibitor (CLI-095) or neutralizing antibody suppressed CCL2 secretion and chemotaxis [32].

Because CCR5 is a key entry receptor for HIV, a number of HIV antiretroviral drugs targeting CCR5 have been developed. One such drug, maraviroc, appears to have antifibrotic potential. A 2014 retrospective study found that maraviroc use over 2 years in HCV-HIV coinfected patients was associated with fibrosis stagnation (in 59.38%) and improvement (in 29.62%) [33]. Furthermore, in vitro treatment of LX-2 cells with maraviroc reduced TGF-β1 signaling and ECM production [34]. Although maraviroc is now not commonly used to treat HIV, the dual CCR2-CCR5 receptor antagonist, cenicriviroc, is currently under clinical trial for both HIV treatment [35] and for patients with nonalcoholic steatohepatitis and liver fibrosis [36]. Interestingly, Sherman and colleagues recently revealed that cenicriviroc treatment was associated with improvement in noninvasive markers of fibrosis in HIV-infected patients [37]. However, the mechanism through which CCR2-CCR5 neutralization ameliorates fibrosis in HIV-infected patients remains to be elucidated.

Macrophage Activation

As in HSCs, HIV infection in macrophages can drive immune activation through stimulation of TLRs. In vitro experiments have demonstrated that suppression of TLR4 signaling in monocyte-derived macrophages reduces HIV gp120-effected CCL2 secretion [32]. TLR4 and CCR5 may constitute a “receptor cluster” that transmits and disseminates HIV gp120 signals because CCR5 antagonism in macrophages mitigated lipopolysaccharide (TLR4 ligand)-induced CCL2 upregulation [32]. In KCs isolated from healthy patients or HBV/HCV-positive patients, in vitro infection of HIV also increased expression of proinflammatory cytokines, such as TNF and interleukin-6 (IL-6), when challenged with lipopolysaccharide (LPS) in a TLR4-dependent fashion [38]. Alternatively, HIV infection in primary human KCs can further amplify the immune response through upregulation of triggering receptor expressed on myeloid cells 1 (TREM1) [39]. Notably, the percentage of CD68⁺ (macrophage marker)/TREM1⁺ peripheral blood mononuclear cells (PBMCs) is greater in HIV-infected patients than in HCV-monoinfected and uninfected individuals [39].

HIV-associated macrophage activation and fibrogenesis have been confirmed in vivo using simian immunodeficiency virus (SIV)-infected rhesus macaques, a well-defined and representative model for studying HIV pathogenesis. In this animal model, SIV infection increased hepatic trafficking of macrophages expressing CCR2, the corresponding receptor for CCL2 [40]. In addition, hepatic macrophage invasion paralleled hepatic CCL3 (CCR5 ligand), TGF-β, and TNF upregulation [40], thus advancing a potential mechanism for HIV/SIV-induced liver fibrosis.

Although hepatic macrophages are important reservoirs for HIV, viral replication is blocked by ART [38, 41]. One plausible basis for liver disease in HIV-diagnosed patients may be elaboration of the macrophage activation marker soluble CD163 (sCD163), which is cleaved from CD163 scavenger receptors expressed on monocyte lineage cells, perhaps by disintegrin and metalloproteinase domain-containing protein 17 (ADAM17) [42, 43]. sCD163 levels are higher in HCV-HIV coinfected patients when compared to either HCV or HIV monoinfected patients and are associated with inflammation and fibrosis [44]. Successful ART reduces sCD163 levels; however, ART produces incomplete reduction of sCD163 levels to those observed in HIV-uninfected persons, suggesting that fibrogenesis may persist even with well controlled HIV [44].

IMMUNE SYSTEM DYSFUNCTION

T-Cell Depletion and Death

Previous studies have attributed accelerated fibrosis progression in HCV-HIV coinfected patients to increased HCV persistence as a result of dysregulated immune responses, notably a reduction in CD4+ T lymphocyte sensitivity to HCV peptides [45]. The chronic phase of HIV infection is characterized by gradual CD4⁺ T-cell depletion and robust activation of CD4⁺ and CD8⁺ T cells, eventually culminating in T-cell death [46, 47]. Consequently, a high CD4 count has been identified as one of a number of factors that can be utilized to predict spontaneous eradication of acute hepatitis C in HIV-diagnosed men [48]. Interestingly, persons who are able to control both HIV and HCV have reduced T-cell exhaustion and activation and their CD4⁺ T cells produce more cytokines than those of patients who can neither control HIV nor spontaneously clear HCV [49]. As previously suggested by Sherman and colleagues, exhaustion of both HCV- and HIV-specific CD8⁺ T cells may be regulated by CD39, which is upregulated in HCV- and HIV-infected patients [50, 51]. CD39 is involved in the hydrolysis of adenosine triphosphate to adenosine [52], which can then inhibit T-cell proliferation through adenosine a2a receptor engagement [53]. HIV can also induce CD8⁺ T-cell death through Fas cell surface death receptor (FAS)-mediated activation of the extrinsic apoptotic pathway [54].

Interleukin Activation

Mechanistically, it has been proposed that CD4⁺ and CD8⁺ T-cell activation drive fibrosis via IL-15, as expression levels parallel HSC activation in HCV-HIV coinfected patients [55]. Interestingly, an IL15 polymorphism (rs10833AA genotype) in coinfected patients is associated with significant liver fibrosis, but patients harboring this genetic variant are also more likely to achieve PEG-IFN-alfa/RBV-induced SVR [56]. The rs12979860 single nucleotide polymorphisms near the IFN-λ3 (IFNL3 or IL28B) gene have also been associated with patient responsiveness to IFN-based HCV treatment; the CC genotype confers a higher SVR rate relative to the TC and TT genotypes [57]. However, in a prospective Canadian cohort of HCV-HIV coinfected patients without fibrosis at baseline, researchers found that the rs12979860CC genotype was associated with increased liver fibrosis [58]. Eslam and colleagues provide congruent support for the profibrogenic effects of rs12979860 genotypes; in a large cohort of chronic hepatitis C patients, nearly 60% possessing the rs12979860CC genotype exhibited mild or advanced steatosis and fibrosis [59]. To reconcile the ostensibly disparate roles of the IFNL genotype in regulating both viral eradication and fibrogenesis, the authors suggest the IFNL genotype mediates activation of inflammatory and apoptotic pathways thereby allowing hepatocytes to clear HCV, yet also permitting sustained inflammation in nonresponders to treatment [59].

CD4⁺ T cells can also influence fibrogenesis by regulating NK cell degranulation (cytotoxicity) via induction of IL-2 [60]. Glässner et al demonstrated that defective CD4⁺ T cells from HCV-HIV coinfected patients were unable to activate NK cell cytotoxicity, observing a diminished number of IL-2–positive CD4⁺ T cells in coinfected patients than in healthy controls or persons with HCV monoinfection [60]. However, a more recent study found no significant difference in NK cell degranulation when cells were cultured in CD4⁺ T-cell conditioned media from HCV-HIV coinfected patients with variegated grades of fibrosis [61]. Instead, NK cell activation may be regulated by dendritic cells. In SIV-infected rhesus macaques, C-X-C chemokine ligand 16 (CXCL16) produced by myeloid dendritic cells increased shuttling of CXCR6-positive NK cells to the liver [62]. There, a cytotoxic subset was preferentially preserved to perhaps mitigate HSC activation, but NK cell death increased, thus maintaining transformed HSCs [62].

Gut Microbiome

Despite suppression of HIV replication with ART, immune activation biomarkers are still upregulated in many HIV-positive patients [63] and intestinal mucosal CD4⁺ T-cell counts do not wholly recoup from nadir [64]. Gastrointestinal tract analyses revealed that T helper 17 cells (T_H17)—a distinct CD4⁺ T-cell effector lineage that produce IL-17 and IL-22—are depleted in HIV infection [65]. Because T_H17 cells play an important role in host antimicrobial defense [66], their loss can foster microbial translocation and gut dysbiosis, as is observed in patients with HIV [67]. As previously discussed, bacterial endotoxin (LPS) can activate HSCs and macrophages through TLRs. Direct LPS treatment in murine HSCs increased their reactivity to activation by KCs and TGF-β [68]. Furthermore, in vivo administration of LPS in mice induced HSC NF-κB transcription factor p65 (RELA) transcriptional activity [68], thus promoting cell survival. A number of studies have also observed that the intestinal microbiota of HIV-diagnosed persons are enriched in Prevotella enterotype species [69], which may be related to sexual practices [70, 71], although the biological implications of this observation are as yet undefined.

STEATOSIS AND STEATOHEPATITIS

Although hepatic fibrosis is the dominant manifestation of liver disease progression in HCV-HIV coinfected patients, hepatic steatosis, the accumulation of triglycerides in hepatocytes, is also a prevalent histologic finding. The reported prevalence ranges from 13% to 60% [16, 72, 73] and is largely dependent on patient characteristics, such as drinking habits or underlying metabolic disorders, and the genotype of HCV infection; for example, HCV genotype 3 has been directly implicated in steatosis development [74].

Insulin resistance is the major metabolic defect underlying steatosis development, as excess carbohydrates increase hepatocyte triglyceride synthesis—through activation of the de novo lipogenesis transcription factors sterol regulatory element-binding protein 1 (SREBP-1c) and carbohydrate-response element binding protein (ChREBP)—and insulin-sensitive adipose triglyceride lipase (ATGL) catalyzes the lipolysis of peripheral adipose tissue and thus increases the hepatic influx of free fatty acids (FFAs). Excessive alcohol consumption can also contribute to steatosis development through SREBP-1c and ChREBP activation [75]. While steatosis itself is generally benign and indolent, in some patients defective β-oxidation and triglyceride synthesis can ultimately lead to the accumulation of hepatotoxic FFA metabolites, including ceramides, diacylglycerols, and lysophosphatidylcholine, that can induce inflammation and hepatocyte death via activation of apoptotic pathways and stress-inducible kinases (eg, c-Jun N-terminal kinases) [76, 77]. These processes contribute to the hallmarks of steatohepatitis (fatty liver with concomitant necroinflammatory activity), which confers a significantly worse prognosis [78, 79]. In the following sections, we will review how HIV and antiretroviral drugs impact steatosis and steatohepatitis development (Figure 2).

Figure 2.

Mechanisms of HIV-associated steatosis and steatohepatitis development. HIV ARVs, including NRTIs, NNRTIs, and PIs, have been associated with metabolic dysfunction and impaired mitochondrial activity, which can contribute to the development of steatosis and steatohepatitis. HIV accessory protein Vpr can directly regulate lipogenesis through transcriptional activation of SREBP-1c and ChREBP. Moreover, Vpr can inhibit mitochondrial β-oxidation, evinced by decreased PPARα expression. Consequently, expression of the PPARα target MTTP is also reduced, which thus impairs VLDL export and permits hepatocyte lipid droplet accumulation (macrovesicular steatosis). Dysfunctional β-oxidation and triglyceride synthesis eventually leads to the production of lipotoxic intermediates that can induce inflammation, oxidative stress, and apoptosis, processes characteristic of steatohepatitis. Abbreviations: ARV, antiretroviral; ChREBP, carbohydrate-responsive element-binding protein; FFA, free fatty acid; HIV, human immunodeficiency virus; HSC, hepatic stellate cell; LXRα, liver X nuclear receptor α; MTTP, microsomal triglyceride transfer protein; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; PPARα, peroxisome proliferator-activated receptor α; SREBP-1c, sterol regulatory element-binding protein 1; TG, triglyceride; VLDL, very low-density lipoprotein.

Drug-Related Liver Injury

Early antiretroviral regimens, particularly those based on nucleoside reverse transcriptase inhibitors (NRTIs) and protease inhibitors (PIs), were associated with mitochondrial toxicity [80] and metabolic derangements, notably insulin resistance and lipodystrophy [81–83]. In human hepatoma cells, it has been shown that efavirenz and nevirapine—nonnucleoside reverse transcriptase inhibitors (NNRTIs)—also disrupt mitochondrial activity and increase intracellular lipid accumulation [84, 85]. These findings appear to have clinical relevance as efavirenz use is also associated with the progression of steatosis [73]. As previously reported by Sherman et al., contemporary antiretroviral regimens may also be associated with steatosis [51]. New integrase strand transfer inhibitors can contribute to the development of fatty liver as dolutegravir has been implicated in weight gain [51, 86, 87].

Lipogenesis and Impaired Mitochondrial β-Oxidation

Hitherto, there have been few studies describing direct links between HIV viral effects and steatosis. While more than half of patients with nonalcoholic fatty liver disease (NAFLD) are obese [88], HIV-positive patients with NAFLD appear to have a lower body mass index and self-report to be more physically active [89], implying a unique role for HIV in the development of steatosis. In a 2017 study, Agarwal et al demonstrated that transgenic mice expressing HIV accessory protein Vpr, under a tetracycline-repressible construct, developed hepatic steatosis through increased de novo lipogenesis, evinced by enhanced SREBP-1c and ChREBP expression [90]. Mechanistically, the authors proposed that Vpr binds to liver X receptor-α (LXRα) to promote LXRα-regulated Srebp1c and Chrebp gene expression [90]. Furthermore, HIV Vpr induced dysfunctional β-oxidation, which consequently reduced very low-density lipoprotein (VLDL) secretion via downregulated microsomal triglyceride transfer protein (MTTP) expression [90]. The HIV matrix protein p17 also activates LXRalpha-mediated transcription and increases intracellular lipid accumulation in HepG2 cells [91]. Studies in pathogenic and nonpathogenic SIV-infected nonhuman primates further suggest a role for HIV in the development of NAFLD. He et al observed that a number of subjects developed NAFLD during late-stage SIV infection and the authors contend that a synergistic relationship between high-fat diet feeding and SIV infection inhibits intestinal CD4⁺ T-cell recovery and promotes CD4⁺ regulatory T-cell depletion [92].

Alcohol Metabolism

Alcohol use is another major determinant of steatosis progression in HCV-HIV coinfected patients [72]. Heavy alcohol use is associated with hastened fibrosis progression while concurrent HIV suppression with ART actually increases HCV RNA [93, 94]. As with HIV-associated immune activation, ethanol metabolites can also deplete gut mucosal integrity thus increasing intestinal permeability and circulating levels of LPS which can, in turn, activate macrophages and HSCs [95]. In vitro experiments utilizing both primary human hepatocytes and Huh7.5 cells demonstrated that enhanced ethanol metabolism, like HCV, augments HIV-induced apoptosis [96]. The effects of ethanol and HIV on hepatocyte death have also been confirmed in mice with humanized livers [96].

EFFECT OF HCV ERADICATION ON MECHANISMS OF LIVER DISEASE PROGRESSION

While HCV clearance has tangible clinical benefits with respect to liver-related complications and death, studies report that inflammatory and fibrotic biomarkers remain elevated in coinfected patients even after achieving SVR (Table 1). As with ART, IFN-based therapies reduce sCD163 levels but do not return them to baseline, implying that latent macrophage activation may continue to promote HSC activation and ECM accumulation [44]. This also appears to be the case when IFN-free regimens are used as levels did not markedly change after DAA-induced viral eradication in chronic hepatitis C patients [97]. In contrast, plasma TGF-β levels do decrease with DAA therapy in coinfected patients, but this difference is not statistically significant and levels are actually higher in patients with hepatic inflammation (METAVIR A1) post-SVR [98]. Likewise, levels of the proinflammatory cytokines TNF and IL-6 remain unaffected by PEG-IFN/RBV treatment in HCV patients with cirrhosis [99]. SVR also has a negligible effect on gut dysbiosis and the elevated serum endotoxin levels observed in HCV patients with cirrhosis [99]. While the aforementioned studies suggest that an aberrant cytokine milieu persists post-SVR, other studies have reported that DAA regimens reduce circulating levels of chemokines such as IFN-γ–induced protein 10 (IP-10), CXCL11, and macrophage inflammatory protein-1β (MIP-1β) in HCV-HIV coinfected patients [100].

Table 1.

Effect of SVR on Mechanisms of Accelerated Liver Disease in HCV-HIV Coinfected Patients

Mechanism	Influence of SVR	Overall Effect
Altered cytokine profile	Persistently elevated serum IL-6 and TNF levels [99] No significant effect on TGF-β levels [98] Decreases IP-10, CXCL11 and MIP-1β [100]	Unclear
Immune system dysfunction	Maintains elevated kynurenine/tryptophan ratio [101]	None
Microbial translocation and gut dysbiosis	No effect on microbiota composition and endotoxin levels [99]	None
Profibrotic macrophage activation	PEG-IFN-alfa/RBV reduces sCD163 levels [44] DAAs reduce sCD163 levels in chronic hepatitis C patients relative to baseline, but difference is not significant [97]	Positive
Steatosis	Improves HOMA-IR and diabetes [102, 103] Increases triglyceride and cholesterol export via VLDL [104]	Positive

Abbreviations: DAA, directly acting antiviral; HCV, hepatitis C virus; HIV, human immunodeficiency virus; HOMA-IR, homeostatic model assessment of insulin resistance; IL-6, interleukin-6; IP-10, interferon-γ–induced protein 10; MIP-1β, macrophage inflammatory protein-1β; PEG-IFN-alfa, pegylated interferon alfa; RBV, ribavirin; SVR, sustained virologic response; TGF-β, transforming growth factor-β; TNF, tumor necrosis factor; VLDL, very low-density lipoprotein.

In addition to its profibrogenic activities, TGF-β has an immunosuppressive effect by restricting HCV-reactive T-cell proliferation. Indeed, this effect continues following SVR as inhibition of TGF-β in PBMCs isolated from patients 4 years after viral clearance increased T-cell proliferation following exposure to HCV antigens [105]. Immunosuppression post-SVR in HCV-HIV coinfected patients may also be ascribed to increased tryptophan catabolism, evinced by the maintenance of elevated kynurenine levels and kynurenine/tryptophan ratios [101].

In regard to steatosis development, it has been shown that SVR improves Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) scores and measures of diabetes in HCV-infected patients [102]. DAAs also increase triglyceride and cholesterol release through VLDL, thus normalizing hepatic lipid homeostasis [104]. Most importantly, magnetic resonance imaging revealed that DAAs reduce steatosis in patients with chronic hepatitis C [103].

Overall, however, there appears to be no unequivocal consensus on how DAA-induced SVR affects these mechanistic derangements. Therefore, until more definitive explanations have been realized, HCV-cured, HIV coinfected patients should continue to be monitored for fibrosis progression.

EFFECT OF DAAS ON HCC DEVELOPMENT

Recent studies have revealed that, in HCV monoinfected patients, there remains a residual risk for HCC occurrence even after DAA therapy, especially among those with a history of HCC [106, 107]. This may be somewhat attributable to an altered cytokine milieu that persists after DAA treatment [108]. Debes et al analyzed 12 biomarkers that were elevated prior to DAA initiation and were associated with HCC development in HCV-infected patients [108]. Notable cytokines identified included IL-6, TNF, and TRAIL [108]. In many patients who developed HCC, TNF levels increased after 12 weeks of DAA treatment [108]. IL-6 also increased but only in patients who developed recurrent HCC [108]. Indeed, genetic polymorphisms in TNF are linked with HCC incidence [109] and in mouse HCC models, IL-6 transsignaling regulates HCC development by activating β-catenin and precluding hepatocyte apoptosis through inhibition of the tumor suppressor p53 [108, 110]. By contrast, TRAIL levels decreased in most patients who developed HCC, thus perhaps inhibiting apoptosis and permitting the survival of malignant hepatocytes [108].

As a further matter, it has recently been shown that HCV infection induces an epigenetic signature associated with HCC development that also remains unaltered by DAA therapy [111]. Interestingly, the authors found that the TGF and WNT signaling pathways were significantly enhanced in both differentially modified and expressed genes post-SVR [111]. However, in vitro treatment of HCV-cured (HCV-infected and DAA-treated) Huh7.5 cells with C646 (histone acetyltransferase inhibitor) or erlotinib (epidermal growth factor receptor inhibitor) could redress these epigenetic differences [111]. Therefore, further examination of this epigenetic signature and how HCV treatment could be made more efficacious to eliminate the risk of future tumor development post-SVR is necessary. Until then, continued surveillance of HCC in HCV-cured, HIV-positive patients, especially in those with advanced fibrosis, is recommended [112].

CONCLUSIONS

Because HCV-HIV coinfected patients now routinely achieve SVR rates comparable to those of HCV-monoinfected patients, it would appear that the contribution of HCV to liver disease in HIV-positive patients will be of limited import. However, despite improved liver function with DAAs, many HCV-infected patients with decompensated cirrhosis still die or require liver transplantation [113] and the incidence of HCC among HCV-HIV coinfected patients continues to rise [114]. As a result, future studies examining the longer-term impacts of DAA therapy on HCV-HIV patients with advanced liver disease should be conducted, especially among those with superimposed conditions that can contribute to further liver damage, such as heavy alcohol use and those related to the metabolic syndrome. In addition, it is apparent that SVR does not ameliorate all immune activation and fibrotic mediators and even suppressed HIV infection itself contributes to hepatic inflammation and fibrosis. Further insight into how HIV precisely influences hepatic pathophysiology will allow us to appropriately manage and improve clinical outcomes for these patients. Until more definitive antifibrotic approaches can be developed, continued clinical follow-up of HCV-cured patients, including assessment of fibrosis progression and HCC surveillance, particularly those with established fibrosis, would appear to be warranted.

Notes

Financial support. This work was supported by the National Institutes of Health (grant number AI136715-01 to R. T. C.).

Supplement sponsorship. This supplement is sponsored by educational grants from Gilead Sciences Inc. and Abbott Laboratories.

Potential conflicts of interest. Both authors: No reported conflicts of interest. Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.