Hepatocellular Carcinoma in Patients with HIV

Abstract

PURPOSE OF REVIEW

Hepatocellular carcinoma (HCC) is becoming an important cause of mortality in patients with HIV, attributed to coinfection with hepatitis C virus, hepatitis B virus, and the longer survival advantage these patients are achieving after introducing the highly active antiretroviral therapy (HAART) regimens.

RECENT FINDINGS

In addition to hepatitis infection, immunosuppression secondary to HIV infection, direct impact of the virus on liver parenchyma, and the use of hepatotoxic antiretroviral drugs, all contribute to HCC pathogenesis. Screening is very important in this particular population; data on population-specific guidelines are still controversial and scarce. Liver transplantation remains the treatment of choice in eligible patients. Trials on sorafenib have not included patients with HIV; yet, we know from small retrospective series that it might be safe and effective.

SUMMARY

In the HAART era, HCC is arising as a common non-AIDS defining cancer with high impact on morbidity and mortality of HIV-infected patients. Candidates for liver transplantation should be offered this option regardless of HIV infection. Safety and efficacy of sorafenib and other treatment modalities should be further studied and offered as deemed applicable to HIV patients diagnosed with HCC.

INTRODUCTION

HIV-related malignancies have been very well described, since the beginning of the AIDS epidemic. Cancer types have shifted from the common AIDS-defining malignancies, linked to the severe immunosuppression (e.g., Kaposi’s sarcoma, non-Hodgkin’s lymphoma, and cervical cancer), to non-AIDS-defining cancers in the highly active antiretroviral therapy (HAART) era. Twenty years after its introduction since 1996, patients are living longer, even approaching normal life expectancy [1]. Hepatocellular carcinoma (HCC) is an uncommon cancer in the general population, occurring primarily in persons with underlying liver disease. But current evidence suggests that HCC is rising as an important cause of morbidity and mortality in HIV-infected patients. Sahasrabuddhe et al. [2] reported a four-fold higher risk of HCC in HIV-infected patients than in the general population. In 2009, it was estimated at 30 per 100 000 individuals in the United States. Pooled HAART-era incidence data suggested an approximately five-fold to six-fold risk increase in HCC incidence in HIV-infected persons, compared with the general population [3]. This is mainly thought to be due to higher hepatitis C virus (HCV) and/or hepatitis B virus (HBV) coinfection. The possible role for immunosuppression, and higher prevalence of other risk factors such as alcohol abuse [4], nonalcoholic steatohepatitis (NASH) [5], and diabetes, should also not be neglected.

CLINICAL PRESENTATION AND PROGNOSIS

HIV-infected patients with HCC are younger at presentation, with a more symptomatic disease and advanced stage [6]. Median survival revolves around 7 months [7]. Looking into the HIV-infected patients from the Brescia HCC Study Group, liver disease stage was significantly more advanced, and liver tumors had a significantly higher prevalence of being multifocal and infiltrative in nature. A shorter interval was noted between the estimated date of first HCV exposure and initial diagnosis of HCC, in this population [8]. A dramatic reduction in mortality of HIV-infected patients was observed between 1996 and 2009 in a Veterans Affairs (VA) hospitals study. This reduction was less pronounced for patients with HCV coinfection, coinciding with the higher prevalence of cirrhosis (3.7-fold increase), decompensated cirrhosis (three-fold increase), and HCC in those patients (23-fold increase). Risk factors included HCV infection, HBV infection, age, Hispanic ethnicity, white race, diabetes, and alcohol abuse [9]. Two prospective cohorts, one for HIV–HCV patients and one for HCV-infected patients alone, were compared. Patients with cirrhosis, coinfected with HIV and HCV, frequently present with infiltrative-type HCC and portal-obstructing tumors on diagnostic imaging, which results in dramatically shorter survival [10■]. A multicohort study was conducted, including 7229 HIV/HCV coinfected individuals, using four prospective cohorts of HIV-positive individuals: The EuroSIDA study, the Southern Alberta Clinic Cohort, the Canadian Co-infection Cohort study, and the Swiss HIV Cohort Study. Higher age, HBV coinfection, lower current CD4 cell count, and cirrhosis were all associated with a higher incidence of HCC in a logistic multivariate analysis. Alcohol abuse, diabetes mellitus, or detectable HIV RNA had no impact on overall incidence of HCC [11■].

SCREENING

Cancer screening has become more relevant in an aging HIV-infected population; yet, available guidelines are still limited to the general population. HCC, diagnosed at screening, had a better survival in many retrospective studies. The American Association for the Study of Liver Diseases advocates liver ultrasonography every 6 months, in population at high risk for HCC [12]. The alpha-fetoprotein test lacks sensitivity or specificity to be useful as a screening test and is not included in recent screening protocols. Lim et al. [13] reported that poor compliance with screening schedule was associated with impaired liver function, a more advanced tumor stage, and, more important, poorer overall survival. Trinchet et al. [14] evaluated the influence of reducing the screening interval in non-HIV cirrhotic patients in a large randomized prospective trial, but found no impact on survival. A similar study is warranted in HIV-infected patients.

HIV AND HEPATITIS COINFECTION

HIV accelerates the progression of chronic viral hepatitis, alcoholic hepatitis, and NASH to cirrhosis [15]. Rapid progression to cirrhosis has recently been related to a direct effect of HIV on stellate cells, which plays an important role in the progression of liver fibrosis. A meta-analysis by Graham et al. [16] demonstrated an increased relative risk of 2.92 for the development of liver cirrhosis and hepatic decompensation in coinfected patients. HIV worsens the outcome of both HCV and HBV infections, mainly by increasing viral load; thus, increasing the likelihood of progression to chronic infection and cirrhosis, especially in patients with low CD4⁺ cell count. It has also been shown that HIV replication can induce miR-122 synthesis, which is essential for HCV replication. The effect of HCV on HIV disease is less elucidated and seems to be related to an additional chronic inflammatory focus, which can lead to progression of HIV-associated diseases [17]. In a recent Dutch study, a significant decrease in CD4⁺ cells was reported in the first months after acute HCV infection, in patients maintained on HAART [18].

IMMUNOSUPPRESSION AND CD4 CELL COUNT

The quantitative decline of T cells leads to increase in viral replication, indirectly increasing the prooncogenic effects of HCV proteins, thus accelerating the course of HCV. A lower CD4⁺ cell count also alters the HCV-specific interferon-g, thus affecting inhibition of HCV replication leading to fibrosis. HIV also affects the function of dendritic and natural killer cells, contributing to disruption in the HCV-specific immune response [19]. Once HCC has developed, HIV coinfected patients express a weaker antitumor response because of chronic immunosuppression.

A Spanish study had assessed prospectively the incidence of HCC in a cohort of HIV-infected patients with liver cirrhosis. 2.7% of patients developed HCC after 60 months of follow-up. At the clinic visit immediately before HCC diagnosis, all patients had CD4⁺ cell counts more than 100 cells/ml, all were receiving antiretroviral therapy (ART), and 90% had undetectable HIV viral load [20]. A retrospective case–control study from the Swiss HIV Cohort Study investigated the role of HIV-induced immunodeficiency in the development of HCC. HAART use was not significantly associated with HCC development. Lower CD4⁺ cell counts increased the risk for HCC among persons infected with HIV [21]. In the ANRS-CO3 Aquitaine Cohort, which studied the effect of immunodeficiency on the risk of HCC development in HIV patients, CD4⁺ less than 350 cells/ml or less than 500 cells/ml was independently associated with a higher risk for HCC [22]. Sixty-six thousand nine hundred ninety-one veterans with HIV were identified from the VA HIV Clinical Case Registry, to study the impact of HIV-related immune suppression on the risk of HCC. Patients who had a recent CD4⁺ cell count less than 200 had a 71% increased risk of HCC [23■].

HIV AND DIRECT LIVER TOXICITY

Patients with HIV have predominant CD8⁺ cell response, mainly mediated by cytokines [interleukin 4 (IL-4), IL-5 and tumor growth factor beta], which increase inflammation and fibrosis. HIV-envelope protein gp120 induces the apoptosis of hepatocytes and release of inflammatory chemokines, thus promoting fibrosis. The well-described disruption of gut integrity in HIV leads to overexpression of lipopolysaccharide. The latter emits signals via Toll-like receptor 4 that upregulates proinflammatory and profibrotic cytokines, thus contributing to liver inflammation [24]. Some indirect liver toxicity may also occur during the immune reconstitution syndrome after initiating ART.

HEPATOTOXIC DRUGS

Most antiretroviral drugs used in HIV have the intrinsic potential to cause hepatotoxicity. Associations between end stage liver disease (ESLD)/HCC, and the cumulative use of HAART, were investigated in a prospective cohort study. Cumulative use of stavudine (d4T), didanosine (ddI), tenofovir (TDF) and amprenavir (APV) was independently associated with increased ESLD/HCC rates, whereas the use of emtricitabine (FTC) and nevirapine (NVP) was associated with decreased rates, instead. Intensified monitoring of liver function should be considered in all patients [25■]. HIV patients are prone to develop insulin resistance, which plays an important role in the development of nonalcoholic fatty liver disease, due to exposure to nucleotide reverse transcriptase inhibitors such as didanosine and stavudine [26].

TREATMENT

Preliminary data from a retrospective analysis of 104 Italian patients with HIV with a diagnosis of HCC showed that anti-HIV treatment along with an early diagnosis and aggressive treatment strategy is associated with increased survival [27].

ANTIVIRAL THERAPY

Anderson et al. [28] demonstrated that ART significantly reduces the rate of hepatic decompensation events by 28–41%, on average, compared with untreated HIV-positive patients. HIV-positive patients who require ART and have chronic HBV infection can be treated with lamivudine and tenofovir as part of triple ART. Direct-acting antiviral-based regimens for HIV–HCV coinfection have shown excellent efficacy, with cure rates similar to HCV monoinfection. Either in combination with interferon (IFN) and ribavirin (RBV), or with IFN-free regimens alone, cure rates of over 90% are the goal for all HIV–HCV-infected individuals. There are several available options for patients with genotype 1 infection. PHOTON and ALLY-2 data support the use of sofobuvir/RBV for 12 weeks in genotype 2 and 24 weeks in genotypes 3 and 4, with small number of patients. Paritaprevir/rombitasvir, dasabuvir, and RBV [29] showed sustained viral response (SVR) rates above 90% in a trial of HIV/HCV coinfected patients irrespective of treatment duration. Combination of grazoprevir and elbasvir [30] showed comparable results between monoinfected and coinfected patients. In genotypes 5 and 6, data are limited to extrapolation from HCV monoinfection [31■]. Protease inhibitors may protect from HCV-related fibrosis.

In a retrospective analysis of 167HIV/HCV coinfected patients treated before December 2012 [32], HCC was diagnosed after achieving an SVR in 7.8% of the overall cases. The median time from SVR to diagnosis of HCC was 28 months. Several studies have shown a decrease in HCC incidence after SVR in HCV monoinfected patients [33,34]. Widespread use of the new HCV therapies in HIV patients might reduce the risk of HCC. Studies are needed to demonstrate this effect on coinfected patients in the era of new antiviral therapy.

Antiretroviral drug regimens should be carefully considered in individuals with HIV and ESLD, and some drugs are contraindicated in patients with cirrhosis.

LIVER TRANSPLANTATION

In recent years, liver transplantation has been performed in patients with HIV infection and HCC. However, data are still scant and controversial regarding outcome. In a prospective Spanish nationwide cohort of HIV-infected patients undergoing liver transplantation for HCC, between 2002 and 2014, HIV infection had neither impact on the recurrence of HCC nor on survival, after liver transplantation. Microscopic vascular invasion was the only factor independently associated with HCC recurrence [35■]. The current criteria for liver transplantation in HIV-infected patients do not differ from those indicated for non-HIV-infected patients. However, an undetectable HIV viral load and a CD4⁺ cell count higher than 150 cells/ml are required for liver transplantation [36]. Overall, short-term survival rates of HIV-infected patients who undergo liver transplantation have been reported to be similar to those of HIV-negative patients without HCV coinfection. No increased risk of postoperative complications or higher incidence of opportunistic infections was noted. Mid-term and long-term survival is mainly affected by recurrent hepatitis C infection, regardless of the HIV status [37].

TARGETED THERAPY

In the phase III SHARP trial, sorafenib has shown a significant survival benefit (10.7 vs. 7.9 months with placebo) and good tolerability in patients with HCC [38]. Unfortunately, no patients with HIV infection were included in this trial. Thus, little information is known about its safety and efficacy with concomitant HAART in HIV-infected patients. A retrospective study observed the efficacy and safety of sorafenib in 27 HIV-infected patients with unresectable HCC. The trend in CD4 cell count and HIV-viral load during the concomitant treatment did not show a significant decrease. The combination was reasonably safe and effective [39]. A drug interaction was predicted between ritonavir, an inducer of CYP3A2, and sorafenib. MicroRNAs play a key role in immune response regulation and carcinogenesis (through phosphoinositide 3-kinase/AKT, extracellular signal-regulated kinase/mitogen activated protein kinase, and Janus activating kinase/Signal Transducer and Activator of Transcription pathways). Many deregulated miRNAs have been involved in HCC, and antitumor effect of miRNA-122 has been studied [40,41]. Hou et al. [42] demonstrated that miR-199a/b-3p expression is decreased in a large part of HCC tissue samples, with an antitumor activity exhibited through the inhibition of serine/threonine-protein kinase and downstream ERK activation. MiR-199a-3p was reported to inhibit HCC cell cycle progression in vitro by targeting c-Met and mammalian Target Of Rapamycin expression. Interaction between cellular miRNAs and HIV has been well described, but needs to be further clarified [43]. Modulation of miRNA activity could be a novel approach to treating HCC in this population [44]. Maraviroc, a C-C chemokine receptor type 5 inhibitor, was shown to decrease HCC progression in a mouse model [45]. Dianne et al. [46] demonstrated that gut sterilization reduced HCC in mouse at the late stage of hepatocarcinogenesis, hence, the intestinal microbia, and TLR4 which seem also to be a potential target in HCC. Amprenavir, a protease inhibitor, has been studied in combination with doxorubicin in xenografts, and was shown to cause tumor regression [47].

One area of clear therapeutic interest is the development of immune checkpoint inhibitors for advanced HCC patients [48,49]. Nivolumab, a monoclonal antibody to programmed death-1, is tolerable with an overall response rate of 15% and favorable survival data in a small cohort of non-HIV-infected HCC patients. Anecdotally, checkpoint blockade in a patient with HIV–HCV coinfection suggests safety in patients with advanced melanoma [50]. It is intriguing to speculate that tumors that develop in a state of relative immune suppression (i.e., HIV infection) may undergo less immune editing, and thus may be more response to immune-based therapies. Currently, several immune checkpoint molecules are being assessed in solid tumor patients, including HCC, with known HIV to assess the safety and efficacy of this approach.

SPECIAL CONSIDERATIONS

Most of the HIV-infected and coinfected people live in resource-constrained settings. Ten percent of the HIV-infected patients in Africa are coinfected with HBV [51]. The burden of HIV remains higher in sub-Saharan Africa, with 1.6 million new infections every year [52]. This population has limited access to HCC screening and data, leading to underestimation of HCC incidence [53]. The highest incidence of HCC has been recorded in The Gambia, Guinea, Mozambique, South Africa, and Zimbabwe [54].

CONCLUSION

Patients with HIV/HCV coinfection have higher viral loads, and are more prone to develop liver decompensation, cirrhosis, and HCC. Despite the use of ART, HIV-infected patients who develop HCC have a more aggressive disease and present at a later stage.

Population-specific screening modalities, such as shorter interval ultrasound, and validated diagnostic nomograms for suspicious lesions should be considered in those patients. Better understanding of the role of HIV in liver injury would guide future studies in identifying higher risk patients and possible specific biomarkers.

Unfortunately, HIV coinfection is contraindicated in most, if not all, studies on targeted agents in HCC. This should be addressed by including this population in larger prospective randomized trials, or by adopting parallel studies with specific HIV cohorts. Our group and many others have already embarked on a global effort that will work collaboratively with colleagues from sub-Saharan African countries on clinical trials aimed at providing therapeutic opportunities for patients with advanced HCC and HIV, embedded within correlative batteries of studies that may help with a better understanding of HCC within the specific context of HIV.

Collaborative effort among HCC and HIV specialists is crucial in managing those patients. Hepatocarcinogenesis seems to become a prompt process in those patients; with HAART, increased awareness and access to better care, people are living longer, and hence, we expect an increasing number of cases within the following years.

KEY POINTS.

• HIV infection has a role in progression to liver cirrhosis and HCC, through immunosuppression, direct effect on the liver, and the use of hepatotoxic drugs.

• Screening is crucial for early detection and treatment of HCC, further efforts are needed to study tailored screening guidelines.

• Liver transplantation is safe and efficient in patients with HIV, hence this modality should be offered for eligible candidates.

• Many new targets and therapies are being identified and studied. Patients with HIV should be included in investigational clinical trials for specific safety and drug–drug interaction issues.