hepatitis c virus and human t cell lymphotropic virus type 1 co infection impact on liver disease virological markers and neurological outcomes

On the other hand, HCV/HTLV-1-coinfected individuals presented with a better prognosis for hepatic involvement when compared with singly HCV-infected subjects.. HTLV-1 causes a persisten

Accepted Manuscript

Title: Hepatitis C virus (HCV) and human T-cell lymphotropic

virus type 1 (HTLV-1) coinfection: Impact on liver disease,

virological markers and neurological outcome

Authors: Ot´avio M Esp´ındola, Alexandre G Vizzoni,

Elisabeth Lampe, Maria Jos´e Andrada-Serpa, Abelardo Q.C

Ara´ujo, Ana Claudia C Leite

liver disease, virological markers and neurological outcome.International Journal of Infectious Diseaseshttp://dx.doi.org/10.1016/j.ijid.2017.01.037

This is a PDF file of an unedited manuscript that has been accepted for publication

As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain

Hepatitis C virus (HCV) and human T-cell lymphotropic virus type 1 (HTLV-1) coinfection: Impact on liver disease, virological markers and neurological outcome

Otávio M Espíndola a,1 , Alexandre G Vizzoni a,1 , Elisabeth Lampe b , Maria José Andrada-Serpa a , Abelardo Q C Araújo a , Ana Claudia C Leite a

a- Laboratory for Clinical Research in Neuroinfections, Evandro Chagas National Institute

for Infectious Diseases (INI) – Oswaldo Cruz Foundation (FIOCRUZ), Rio de Janeiro,

Brazil; b- Laboratory of Viral Hepatitis, Oswaldo Cruz Institute (IOC) – FIOCRUZ, Rio de

Janeiro, Brazil

1- Otávio M Espíndola and Alexandre G Vizzoni contributed equally

Corresponding author: A.C.C Leite LAPCLIN-Neuro – INI/FIOCRUZ, Avenida Brasil,

4365 – Manguinhos – Rio de Janeiro/RJ – Brazil ZIP code: 21040-900 Phone: +55(21)3865-9543 E-mail: ana.leite@ini.fiocruz.br

Highlights

 Frequency of neurological abnormalities is unaltered during HCV/HTLV-1 coinfection

 Hepatic damage is reduced during HCV/HTLV-1 infection, even without HCV clearance

 Coinfection with HCV does not change HTLV-1-directed CD4+ T-cell proliferation

 HCV-related thrombocytopenia is reversed during coinfection with HTLV-1

ABSTRACT

Objectives: 1 infection is associated with neurological abnormalities, such as

HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) and peripheral neuropathy (PN) HCV infection is the leading cause of chronic liver disease worldwide, and causes PN

in approximately 9% of patients Because the interplay between these potentially neuropathogenic viruses in a same individual is still poorly understood we evaluated clinical

and laboratory outcomes in coinfected patients and compared them with controls Methods:

The prevalence of neurological and laboratorial abnormalities was evaluated in 1-coinfected patients (n=50), and in subjects with single HCV (n=46) or HTLV-1 (n=150)

HCV/HTLV-infection Results: A higher frequency of isolated PN was present in HCV-infected patients

and this was not associated with cryoglobulinemia No difference was found in the frequency

of PN or HAM/TSP when compared coinfected with singly infected individuals Hepatic involvement was present in HCV-infected subjects, as shown by increased levels of serum ALT, AST, GGT and bilirubin, in addition to thrombocytopenia On the other hand, HCV/HTLV-1-coinfected individuals presented with a better prognosis for hepatic

involvement when compared with singly HCV-infected subjects Conclusions: Our data

suggest that HCV/HTLV-1 coinfection does not mutualistically alter the outcome of neurological manifestations Nonetheless, changes in the immunological environment induced by HTLV-1 infection could lead to a reduction in hepatic damage, even without significant HCV clearance

Keywords: HCV; HTLV-1; Coinfection; Hepatic involvement; Peripheral neuropathy

1 INTRODUCTION

HTLV-1 infects 5−10 million people worldwide, being endemic in Japan, the Caribbean, Sub-Saharan Africa, the Middle East and South America, particularly in Brazil.1

HTLV-1 causes a persistent infection of CD4+ and CD8+ T-cells that leads to diverse changes

in the immune response,2 and can be associated with the development of adult T-cell leukemia/lymphoma (ATLL) and HTLV-1-associated myelopathy/Tropical spastic paraparesis (HAM/TSP).3 In addition to its pathogenic role in these diseases, HTLV-1 has been also associated with other immune-mediated conditions, such as uveitis, polymyositis, chronic inflammatory arthropathy and increased susceptibility to some opportunistic infections.4

Most of HTLV-1-infected individuals remain asymptomatic carriers (ACs) and only 0.5−5% of them develop disease.5 Several studies have suggested that HTLV-1 is associated with a wider spectrum of neurological manifestations that do not fulfill the diagnostic criteria for HAM/TSP.3 These symptoms and conditions may later progress to classical HAM/TSP or remain as isolated neurological syndromes, such as sensory neuropathy,6 and 7 bladder dysfunction,8 erectile dysfunction,9 amyotrophic lateral sclerosis-like syndrome,10 mild cognitive deficits11 and more rarely, motor neuropathies.12 Events or mechanisms that trigger the neurological involvement in HTLV-1 are still unknown However, a high HTLV-1 proviral load (PVL) in peripheral blood has been associated to the development of neurological abnormalities.13

HCV infection is the leading cause of chronic liver disease worldwide, and it is estimated that about 185 million individuals are or have been infected throughout the world.14The average prevalence of HCV seropositivity in Brazil is around 1.23% among blood donors, and in Rio de Janeiro the estimated prevalence is 2.6%.15 The majority of HCV-infected individuals establish a chronic infection, and 74−86% present detectable viremia in peripheral blood In 25−30% of cases chronic infection can progress to severe complications, such as chronic hepatitis, cirrhosis and hepatocellular carcinoma (HCC).16 It is also estimated that 40−74% of HCV-infected individuals can develop at least one extrahepatic

manifestation, such as mixed cryoglobulinemia, membranoproliferative glomerulonephritis, sicca syndrome, and porphyria cutanea tarda, among others.17 Peripheral neuropathy (PN) can

be found in 9% of patients with chronic HCV infection, regardless the presence of cryoglobulinemia.18 Involvement of the central nervous system (CNS),19 and 20 the peripheral nervous system (PNS)21 and, less frequently, the autonomic nervous system, have also been documented, either as a direct manifestation of the infection or as a paraneoplastic presentation in cases of HCC.22

The HCV/HTLV-1 coinfection may be common, particularly in countries where these viruses are endemic, since they share similar modes of transmission This is especially true in certain population groups, such as in persons who have unprotected sex habits, intravenous drug users and individuals with a history of blood transfusion.23 and 24

The cellular immune response apparently play an important role not only in the viral clearance in both infections25 and 26, but also in the immunopathogenesis of the hepatic lesion

in HCV infection16 and in the neurological involvement in HTLV-1 infection.3 The interplay between those two neuropathogenic viruses in a same individual is still poorly understood It has been shown by some that, when compared to HCV-monoinfected patients, the HCV/HTLV-1 coinfection has a synergistic effect on incident liver disease, as well as on deaths from HCC.23 However, other groups have postulated that HTLV-1 infection can attenuate hepatic damage, as shown by lower serum levels of aminotransferases and lower degree of liver fibrosis. 27 and 28 Similar results were observed in a study with triple HIV, HCV and HTLV-1 coinfection, in which HTLV-1-infected patients presented better outcomes for HCV-associated liver fibrosis.26

Therefore, in order to give new insights about the interactions between HCV and HTLV-1 infections, we evaluated abnormalities on hepatic parameters and the outcome of neurological abnormalities during HCV/HTLV-1 coinfection

2 MATERIAL AND METHODS

2.1 Patients and data collection

This cross-sectional study was conducted from 2008 to 2010 at the Evandro Chagas National Institute for Infectious Diseases (INI/FIOCRUZ), Rio de Janeiro, Brazil The study was approved by the local ethics committee (INI 2009/0028.0.009.000.08), and informed written consent was taken from all patients submitted to the research protocol HTLV-1-infected individuals (n=150) were randomly selected from a cohort of 650 subjects regularly attending the outpatient clinic at INI/FIOCRUZ HCV/HTLV-1-coinfected patients (n=50) from this same cohort were enrolled in the study when inclusion criteria were fulfilled HCV-infected individuals (n=46) were randomly selected from outpatient cohorts followed at the Gaffrée & Guinle University Hospital (HUGG-UNIRIO), Rio de Janeiro, and at the Oswaldo Cruz Institute (IOC/FIOCRUZ) Individuals with concurrent infection with HIV, HTLV-2, syphilis, and HBV, or presenting conditions that could by themselves cause PN (such as vitamin B12 deficiency, chronic use of potentially neurotoxic drugs, diabetes, leprosy, and

renal or genetic diseases) were excluded from this study

2.2 Neurological evaluation

Patients were classified as neurological asymptomatic or as HAM/TSP according to the modified World Health Organization’s criteria, as previously described.29 HTLV-1-infected patients with urinary urgency/incontinency/retention, and with an abnormalurodynamic evaluation, but without PNS or CNS involvement, were classified as presenting

an isolated neurogenic bladder.8

Due to the close relationship between HCV infection and PN in literature,21 a full evaluation of the PNS was prioritized The diagnosis of PN was done when one or more of

the following clinical criteria were fulfilled: numbness, burning sensations, pain in a stocking and glove distribution, distal weakness, and diminished or absent deep reflexes with or without distal muscular atrophy In the electromyography (EMG) and nerve conduction studies (NCS), the following diagnoses were considered according to classical criteria:30polyneuropathy or multiple mononeuropathy Presence of specific histological abnormalities

in sural nerve biopsy was also considered as a hallmark of pathological involvement of the peripheral nerve Isolated PN was identified in patients presenting signs and/or symptoms of

PN and an abnormal NCS/EMG and/or abnormalities in the sural nerve biopsy; or when clinical criteria of PN were present without confirmation by NCS/EMG and/or histological evaluation of the peripheral nerve.12

2.3 Clinical and laboratory data

Sociodemographic data such as gender, age, intravenous drug use (IDU), alcohol consumption, self-reported history of blood transfusion, number of sexual partners and practice of unprotected sex were assessed Peripheral blood was collected in EDTA and used

on the same day for DNA extraction, for subsequent HTLV-1 proviral load quantification, and in determining complete blood counts, and CD4+ and CD8+ T-cells counts by flow cytometry (FACSCalibur, Becton Dickinson) Plasma was obtained from peripheral blood in EDTA by centrifugation at 1,500  g for 10 minutes and used for assessment of cryoglobulins

as previously described.31 Serum samples were used within four hours for quantification of aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP), and total bilirubin levels, or aliquoted and stored at -80C for further analysis Liver fibrosis and cirrhosis were predicted using APRI (AST to platelet count ratio index).32

HTLV-1 diagnosis was based on the detection of specific anti-HTLV-1 antibodies in serum by enzyme immunoassay (EIA) (Anti-HTLV-I ∕ II Sym Solution, Symbiosys, São

Paulo, Brazil) and confirmed by Western blot (HTLV Blot 2.4, Genelabs Technologies, Singapore) The diagnosis of HCV infection was performed by a third-generation EIA (ETI-AB-HCVK-4, Diasorin, Vercelli, Italy), according to the manufacturer’s recommendations, and confirmed by HCV RNA genomic amplification (Amplicor ® HCV Test, v2.0, Roche Diagnostics, France) HCV genotyping was performed using the Versant™ HCV Genotype Assay - LiPA (Bayer Corporation, Tarrytown, NY, USA), which can discriminate HCV genotypes 1 to 6 The quantification of HCV viral load was performed using the Cobas™ Amplicor HCV Monitor Test, v2.0 system (Roche Diagnostics, France), according to manufacturer’s instructions HTLV-1 PVL was determined by quantitative PCR, as previously described,13 and shown as frequency of HTLV-1-infected cells in peripheral blood leukocytes (PBL)

2.4 Statistical Analysis

Statistical analysis was performed using SPSS software v.20 Comparison of categorical variables among groups was performed by Pearson chi-square (2) test, or Fisher’s exact test when cell size was fewer than five Normal distribution of continuous variables was evaluated by Kolmogorov-Smirnov test Comparison of continuous variables was performed

by Student’s t test or Mann-Whitney test Multiple comparisons were carried out by One-way

ANOVA (parametric) or Kruskal-Wallis test (non-parametric) Pairwise comparisons were run in R software v.3.2.5 using Fisher’s exact test or Mann-Whitney test with Bonferroni

correction Results with p<0.05 were considered statistically significant

3 RESULTS

3.1 Sociodemographic characterization of patients

Sociodemographic and risk behavior characteristics of the study population are shown

in Table 1 Mean age ± standard deviation of the patients was 55.04 ± 15.24 years in the HCV/HTLV-1-coinfected group, 53.70 ± 12.99 years for the HCV-infected subjects, and 55.04 ± 15.24 years for the HTLV-1-infected group No significant differences were

observed between groups considering the mean age (ANOVA, p=0.493), or the distribution

of subjects according to age categories (Table 1)

Among HTLV-1-infected individuals, females were statistically predominant, while males were more frequent in the HCV/HTLV-1-coinfected group (Table 1) Considering the most likely routes of infection, significant differences were observed between groups IDU was of high importance among HCV/HTLV-1-coinfected individuals, accounting for 28.2%

of cases (Table 1) HCV infection in our study was strongly associated to blood transfusion, but statistically predominant only in the HCV-infected group In addition, occurrence of blood transfusions prior to 1993, when mandatory serological screening among blood donors was instituted in Brazil, was higher among HCV-infected and HCV/HTLV-1-coinfected individuals On the other side, HTLV-1 single infection was significantly associated to sexual transmission in our study population (Table 1), which was considered when multiple sexual partners and/or practice of unprotected sex were reported and other risk factors were absent Regular alcohol consumption was significantly more frequent among HCV-infected subjects

in comparison to the HTLV-1-infected group, but no differences in alcohol intake were seen between HCV/HTLV-1-coinfected individuals and the other groups (Table 1)

3.2 Neurological evaluation

The plethora of neurological manifestations was more pronounced in HTLV-1 carriers, regardless of mono- or coinfection When comparing the frequency of ACs and patients presenting any neurological condition in each study group, significant differences were observed only for the HCV-infected group (2 test: HTLV-1  HCV, p<0.001;

HCV/HTLV-1  HCV, p<0.001; HCV/HTLV-1  HTLV-1, p=0.093), in which most of the

enrolled individuals were asymptomatic (Table 2) A significant higher frequency of isolated

PN was observed only between HCV-infected and HTLV-1-infected groups, although a trend had been also observed when compared to HCV/HTLV-1 coinfection (Table 2) HCV-related cryoglobulinemia commonly causes disabling complications, including PN.33 However, we observed no association between PN and cryoglobulinemia in our study population (Table 3) Manifestation of PN is also more frequent especially among people over the age of 60 years,34 and can be associated as well to alcohol abuse.35 Hence, we evaluated these variables

as probable confounders for the outcome of PN However, due to the small sample, we were not able to use logistic regression in a multivariate context Thus, individuals were categorized according to age (below or above 60 years-old) or alcohol consumption (yes or no), and analysis was performed by Fisher’s Exact test No association was observed between age above 60 years or alcohol consumption with manifestation of PN, with exception for the HTLV-1-infected group (Supplementary Table 1) In this particular case, 5 of 6 individuals displaying PN had more than 60 years On the other side, all four patients with PN in the HCV/HTLV-1-coinfected group were neither older than 60 years nor alcohol abusers (Supplementary Table 1)

Considering the development of HAM/TSP, which is associated to HTLV-1 infection,

no statistical difference was observed in its frequency during HCV/HTLV-1 coinfection or single HTLV-1 infection (2 test, p=0.141)

3.3 Laboratorial analysis

HCV genotype 1 was found in 82.75% of patients in the HCV/HTLV-1-coinfected group, and in 87.18% of individuals in the HCV-infected group No difference in mean HCV viral loads was observed between the groups with single HCV infection and coinfected with HTLV-1 (HCV: 5.81 ± 0.62 log10 UI/mL; HCV/HTLV-1: 5.98 ± 0.51 log10 UI/mL)

(Student’s t test, p=0.418) (Figure 1A) Moreover, no difference was also observed for

median HTLV-1 PVL levels between individuals with single HTLV-1 infection (5.08 infected cells/100 PBL; IQR25−75: 1.52% − 10.26%) and coinfected with HCV (5.95 infected cells/100 PBL; IQR25−75: 1.06% − 9.32%) (Mann-Whitney, p=0.869) (Figure 1B) In HTLV-

1 infection, high HTLV-1 PVL has been associated to HAM/TSP As expected, median HTLV-1 PVL levels were higher in HAM/TSP patients than in ACs in both HTLV-1-infected group (AC: 2.29 infected cells/100 PBL, IQR25−75: 0.57% − 6.71%; HAM/TSP: 7.32 infected cells/100 PBL, IQR25−75: 3.91% − 12.17%; Mann-Whitney, p<0.001) and HTLV-1/HCV-

coinfected group (AC: 0.56 infected cells/100 PBL, IQR25−75: 0 – 3.15%; HAM/TSP: 7.08 infected cells/100 PBL, IQR25−75: 4.21% − 12.35%; Mann-Whitney, p<0.001) (Figure 1C)

However, no differences were observed between groups

It has been shown that HTLV-1 infection induces proliferation of CD4+ T-cells.2However, no differences between groups were observed for median CD4+ T-cell counts (HCV/HTLV-1: 893 cells/mm3; IQR25−75: 664−1354 cells/mm3; HCV: 719 cells/mm3; IQR25−75: 575−1098 cells/mm3; HTLV-1: 960 cells/mm3; IQR25−75: 725−1196 cells/mm3)

(ANOVA, p=0.129) (Figure 2) Nonetheless, CD4+ T-cell counts positively correlated with

HTLV-1 PVL in the HTLV-1-infected group (Spearman R=0.230, p=0.023), while it negatively correlated with HCV viral load in the HCV-infected group (Spearman R=−0.943, p=0.005) No correlation between these factors was observed in HCV/HTLV-1 coinfection

Moreover, HCV/HTLV-1-coinfected individuals presented higher median CD8+ T-cell counts in comparison to other groups (HCV/HTLV-1: 839 cells/mm3; IQR25−75: 348−1112 cells/mm3; HCV: 458 cells/mm3; IQR25−75: 312−598 cells/mm3; HTLV-1: 499 cells/mm3; IQR25−75: 317−683 cells/mm3) (ANOVA, p=0.004) (Figure 2)

Chronic HCV infection is associated to hepatic damage, which can be easily evaluated

by increased serum concentrations of AST, ALT, GGT and bilirubin.32 However, liver

disease can be also associated to regular alcohol consumption.35 Therefore, we assessed the dependence of abnormal levels of AST, ALT, GGT and total bilirubin to alcohol intake in our study population, but no significant association was observed (Supplementary Table 2) Abnormal levels of serum AST, ALT, GGT and total bilirubin were more frequent among HCV carriers, and their median serum concentrations were higher as well, in comparison to HTLV-1-infected individuals (Table 3) Moreover, HCV/HTLV-1-coinfected patients also presented higher median levels for AST, ALT and GGT, but not for bilirubin, than those of HTLV-1-infected individuals Nonetheless, HCV/HTLV-1-coinfected patients presented better prognosis for hepatic damage than HCV-infected individuals (Table 3), when considering median levels of AST and bilirubin, for instance (Supplementary Table 3) Using APRI, a noninvasive index to predict both significant liver fibrosis and cirrhosis in patients with chronic hepatitis C, we could determine that none of the HTLV-1-infected individuals presented significant hepatic involvement Meanwhile, two HCV/HTLV-1-coinfected patients were manifesting significant liver fibrosis and one of those was more likely to present cirrhosis In turn, HCV-infected subjects had worse scores on APRI than patients from the other groups APRI scores from 11 HCV-infected patients were compatible to cirrhosis, and one exhibited a score indicative of significant liver fibrosis (Table 3)

Changes in platelet counts were also observed The frequency of patients with low platelet counts (<150103 platelets/mm3) was higher among HCV-infected patients, and their median platelet counts was significantly lower as well HCV/HTLV-1-coinfected patients presented lower median platelet counts than individuals with HTLV-1 infection, but still higher than those with HCV monoinfection (Table 3)

4 DISCUSSION

The main goal of the present study was to evaluate whether hepatic or neurological complications are different or more common when comparing patients singly or coinfected

with HCV or HTLV-1 Our study groups had patients with some similar characteristics to previous studies, such as a higher frequency of females among HTLV-1-infected patients,3and a predominance of males among HCV/HTLV-1 coinfected individuals 26, 27 and 28Considering the infection routes, IDU was more frequent in the HCV/HTLV-1-coinfected group, as previously described.36 and 37 It has been shown that IDU is a risk factor for coinfection involving HIV, HCV and specially HTLV-2 in southern Brazil.36 However, as previously suggested by Cardoso et al.27 and now confirmed in our study, IDU seems to be an important risk factor for HCV/HTLV-1coinfection On the other hand, HCV infection was strongly associated to blood transfusion in our study, corroborating previous data from another cohort from São Paulo, Brazil.27 HCV genotype 1 was found in more than 80% of patients with single HCV infection or coinfected with HTLV-1, which corroborates the general prevalence observed in Rio de Janeiro, Brazil.37 and 38

Involvement of CNS and PNS are well-established consequences of HTLV-1 infection, being HAM/TSP the most frequent observed alteration of the CNS.3 However, it has been shown that symptomatic PN can occur in a small proportion of HTLV-1-infected patients.6 Indeed, peripheral manifestations are often masked by symptoms and signs of HAM/TSP However, PN can be found as an isolated manifestation in about 13.4% of HTLV-1-infected individuals without HAM/TSP.12 PN is commonly characterized by alteration of small-fiber functions, with inflammatory infiltrates, axonal degeneration and segmental demyelination, as revealed by sural nerve biopsies, and sometimes is associated with symptomatic polymyositis.39 A sensory-motor polyneuropathy with a predominant axonal involvement is the most common PN observed in these patients.40 The involvement of the nervous system is also observed in HCV-infected individuals Some hypotheses have been proposed to explain this involvement According to some, a "Trojan horse" mechanism could be used by HCV to facilitate its entrance into the CNS.41 HCV-infected peripheral

blood mononuclear cells would enter the CNS and interact with brain cells, especially astrocytes.42

In the present study we showed that PN was more frequent among HCV-infected individuals than in HTLV-1 infected patients On the other hand, HCV/HTLV-1-coinfected individuals presented a marginal tendency to have less PN in comparison to HCV singly infected individuals Also, no significant difference was seen as of the outcome of HAM/TSP among coinfected patients These data suggest that HCV/HTLV-1 coinfection apparently do not mutualistically interfere in the fate of development of neurological manifestations However, due to the nature of our study, it is not possible to draw a definite conclusion about this hypothesis

HCV/HTLV-1-coinfected individuals presented an increased frequency of CD8+ cells but not of CD4+ T-cells, as also shown by Cardoso et al.27 As pointed out here, there was no significant difference in median HTLV-1 PVL levels between HTLV-1-monoinfected and HCV/HTLV-1-coinfected individuals This suggests that HCV apparently does not enhance HTLV-1-directed CD4+ T-cell proliferation in coinfected individuals In addition,

T-we did not find differences in mean HCV viral loads betT-ween those groups The mechanisms

of interaction between CD4+ and CD8+ T-cells in both HTLV-1 and HCV infections are extremely complex and still partially understood, as in determining the efficacy of viral clearance as in the pathogenesis Therefore, it is not possible to define if HTLV-1 interferes with the natural course of HCV replication

The relevance of HCV/HTLV-1 coinfection on liver disease in patients with chronic hepatitis C has been evaluated in prospective and retrospective studies in Japan.23 and 43According to their results, HTLV-1 was found to worsen the clinical course of HCV infection, contributing to the development of HCC in these patients These authors suggest that an impaired T-cell responses resulting from HTLV-1 infection may interfere with HCV