Viral Eradication of HCV

[Guest guest]

Viral Eradication of HCV

“Long-term evolution of serum and liver viral markers in patients

treated for chronic hepatitis C and sustained responseâ€

Journal of Viral Hepatitis

Volume 13 Page 28 - January 2006

M. Moreno1, R. Pérez-Alvarez1, L. Rodrigo1, R. Pérez-López1, P.

Suárez-Leiva2

1Gastroenterology and 2Microbiology Services, Hospital Central de

Asturias, Oviedo, Spain

“……Our four SR patients with HCV-RNA (+) in the liver have now

been followed up since the end of treatment over a long period

(between 4 and 9 years) and no clinical relapse has been observed,

but we consider that this special group needs a longer surveillance,

especially in cases of use of concomitant immunosuppressive therapy.

We do not believe it to be necessary to routinely determine HCV-RNA

in the liver in order to check viral eradication, as positive

patients do not show relapse…… Our results indicate that

antibodies anti-E2 and NS4 are the first to disappear, after

achieving a sustained response, a finding not previously described

and whose significance is unknown…..â€

Summary.

Few studies have analysed the evolution of HCV markers in chronic

hepatitis C (CHC)-treated patients. We have evaluated the presence or

absence of serum and liver HCV-RNA, the core antigen (HCV-cAg) and

the loss of specific antibodies (anti-HCV), in long-term sustained

responders (SR). One hundred and seventy-six patients (132 SR and 44

nonresponders (NR) were included in the study. HCV-RNA was determined

in serum and liver by a commercial PCR-kit. HCV-Ag was determined by

ELISA and specific antibodies against HCV by means of a commercial

line immunoblot assay (LIA) technique. Serum HCVcAg was found

positive in three (4.2%) SR and in one (4%) NR (NS). Four SR (3.6%)

and 44 NR (100%) were also HCV-RNA (+) in liver tissue. Two patients

were HCV-cAg (+). A good correlation was found between the serum

levels of HCV-cAg and HCV-RNA (r = 0.847, P < 0.001). Specific

antibodies (anti-HCV) were determined by LIA in 45 patients. A

decrease was found in the number of patients who presented reactivity

to bands E2 and NS4 when we compared SR with a follow-up of more than

5 years with NR and SR with a follow-up <5 years (P < 0.01 and

0.005). A good correlation was found between the HCV-cAg and HCV-RNA

serum levels in CHC-treated patients (P < 0.001). Few SR (3.6%) had

HCV-RNA in the liver, and HCV-cAg (1.8%) in serum. In SR with more

than 5 years of follow-up a clear tendency exists in the trend to

clarify the bands E2 and NS4 of anti-HCV in serum.

Introduction

Infection with the hepatitis C virus (HCV) is a health problem that

affects more than 170 million individuals worldwide.

Three types of viral markers of hepatitis C infection are clinically

used in the management of patients with chronic hepatitis C (CHC):

the antibodies against the HCV (anti-HCV) determined by ELISA, the

viral load or (HCV-RNA) measured by PCR and the genotype assayed by

LINNO-LIPA.

The determination of the core total antigen of HCV is not usual in

daily clinical practice, although several studies exist which try to

support its routine use. Diagnosis of HCV infection is based on the

detection of specific antibodies (anti-HCV) in serum and the

quantification of its viral load, through the use of molecular

techniques or determination of its structural proteins (antigen

detection). Direct and indirect tests of the presence of HCV play a

key role in the diagnosis of the infection, and help the clinician in

making the appropiate therapeutic decision necessary for confirmation

of the virological response to treatment and the long-term follow-up

of the patients.

Few reports exist concerning the long-term evolution of responders to

the treatment, and in particular, about the proportion and the time

when they have cleared the virus from the liver.

We therefore carried out a study on a large series of CHC-treated

patients with SR in order to evaluate the persistence or absence of

the viral genome in serum and the liver as well as other markers of

viral infection, in a long-term follow-up. The results will help in

understanding better the evolution of these patients and provide more

information related to the precise moment of the cure of HCV

infection.

Methods

Patients

A retrospective study was performed in a total of 176 patients with

CHC, treated consecutively with antiviral therapy in the same

hospital from January 1989 to December 2003. Patients were divided

into two groups: 132 (75%) SR and 44 (25%) NR (118 males, 58 females,

2:1). All patients were treated with standard or pegylated alpha-IFN

alone, or in combination with ribavirin and received several cycles

of treatment over the study period, before obtaining an SR or being

classified definitely as NR.

The time elapsed from the infection until the beginning of the

effective treatment and the follow-up time from the beginning of the

effective treatment in SR and from the first treatment (if there was

more than one) in NR, until the end of the study were evaluated. All

patients were followed up for at least 1 year after the end of

treatment.

>From 1989 to 1993, all patients were diagnosed as having CHC by the

detection of anti-HCV(+) in serum and later confirmation by the RIBA

test. Since 1994, viraemia as measured by PCR, was confirmed in all

patients. Of the SR, viraemia was confirmed prior to treatment in

73%. All NR presented HCV-RNA (+) in serum. Sera and liver samples

were obtained and analysed between September 2001 and January 2003.

Methods

Liver biopsy

A liver biopsy was carried out in 104 SR and in 39 NR. Two samples of

liver tissue of 2 mm in length were cut and kept under strict aseptic

conditions in Eppendorf tubes (free of RNA-ase) and frozen to -70 °C

until needed.

Serum studies

Anti-HCV antibodies.

Total anti-HCV was determined by means of an EIA commercial test and

third-generation line immunoblot assay (LIA) in serum samples (Inno-

Lia ACV Ab III; Innogenetics, Ghent, Belgium).

HCV-cAg

This was determined by means of a quantitative EIA Track-C core

antigen test. The lower limit of sensitivity of this technique is 1.5

pg/mL (equivalent to 12 000 IU/mL by PCR) (Ortho-Clinical

Diagnostics, Raritan, NJ, USA).

HCV-RNA in serum

This was determined qualitatively by a commercial RT-PCR test

(Amplicor HCV; Roche Mannheim, Germany) and quantified using an

Amplicor HCV Monitor (version 2.0; Roche); and lower detection limit

was 50 IU/mL.

HCV-genotypes

These were used for analysing the six most commonly known genotypes

with their different subtypes and performed by HCV (LIPA) Versant®

(Bayer, Tarrytown, NY, USA).

HCV-RNA in liver

This was carried out using the same quantitative test employed for

serum samples (Amplicor v. 2.0; Roche). For the extraction of the

nucleic acid from the sample, a small fragment of liver biopsy was

diluted with 560 μL of lysis buffer (guanidine isothiocianate), and

140 μL of saline serum was added, mixed together, left for 10 min

and immediately frozen until required for analysis. In order to

evaluate the technique, a total of 23 liver samples from HCV-RNA

negative patients were studied, liver HCV-RNA was negative in all.

Statistical analysis

A descriptive study was performed. For the comparison of qualitative

parameters and percentages the chi-squared test with Yates correction

for continuity was used if necessary. In order to compare

quantitative data the Student's t-test was used for nonpaired groups,

with the Yates correction when the differences between variances were

significant. For nonparametric variables Mann-Whitney, Wilcoxon and

ANOVA tests were employed. A significant difference was assessed by

an estimated P ≤ 0.05.

Results

Of the 132 SR, 36% were females and 64% were males. Of the 44 NR, 25%

were females and 75% were males (NS). The mean age at the beginning

of the treatment was 37.4 ± 0.7 years for SR and 42.5 ± 1.4 years

for NR (P < 0.0001). The period of follow-up from the beginning of

the effective treatment in SR or from the first treatment in NR was

4.3 ± 0.04 years and 4 ± 0.4 years respectively (range 1-13 years)

(NS). The time lapsed from the end of treatment was 3.5 ± 0.2 years

for SR and 4.2 ± 0.4 years for NR (NS).

Mean viral load was 2.2 × 106 IU/mL in SR and 4.2 × 106 IU/mL in NR

(NS).

Genotype 1 was most prevalent in NR (P < 0.01), especially the

subtype-1b (41% SR and 65% NR) (NS). Genotype 3 was more frequent in

SR (P < 0.0001) (Figs 1 and 2).

All SR were HCV-RNA (-) in serum at the time of the study and all NR

were positive, with a mean viral load of 0.7 × 106 IU/mL.

HCVcAg was positive in three of 71 (4.2%) SR and in 43 of 44 (98%)

NR. Its mean concentration was 72.6 ± 7.9 pg/mL in NR and 7.7 ± 6.8

pg/mL in three SR. One NR with undetectable HCVcAg had a low viral

load of 0.3 × 106 IU/mL.

A good linear correlation was found between the serum HCVcAg

concentration and viral load in NR (r = 0.847, P < 0.001) (Fig. 3).

HCV-RNA concentration in the liver was positive in all NR and in four

of 104 SR (3.8%; P < 0.0001). Of the four SR with positive HCV-RNA in

liver, two had HCV-cAg detectable in serum, at a low concentration of

2.9 pg/mL.

We studied the persistence or the loss of the different bands of

antibodies of six antigens of HCV (regions C1, C2, E2, NS3 and NS4

NS5) in 45 patients. These patients were divided into three groups:

15 NR, 15 SR with <5 years of follow-up and 15 SR with five or more

years of follow-up. The positivity of the bands was measured

qualitatively, ranging from 0 to ++++. The intensity of bands E2 and

NS4 decreased markedly in SR with <5 years of follow-up compared with

NR, and also in SR with five or more years of evolution compared with

SR of <5 years and NR. Differences were also found (NS) for the rest

of the NS bands, with their intensity decreasing in SR, compared with

NR (Fig. 4).

Discussion

Few reported long-term studies of the virological evolution in serum

and liver of patients treated for chronic hepatitis who have

responded to antiviral treatment exist.

As expected, NR had a higher basal viral load and a greater

percentage of genotype 1 than patients with SR. However, no

significant differences were found between either group, this

probably being due to the wide dispersion of the viral load. The

presence of a high basal viral load seems to be an important factor

in the lack of response observed in a higher percentage of cases, but

no evidence exists at present. This finding seems to predict a worse

effect on the progression of the illness [1].

In two large clinical trials performed in naïve patients it was

postulated that the presence of a low basal viral load (<1 × 106

IU/mL) was one of the best predicting factors for SR of CHC patients

[2,3].

In our study the more prevalent genotype was type-1, amounting to 69%

in SR and 89% in NR (NS). Subtype-1b was also the most frequently

found in both groups, this being 41% in SR and 65% in NR (NS). We

also found differences for genotype 3, as its proportion was clearly

higher in SR than in NR. Our results are in accordance with those

found in previous studies, as genotype 1, besides being more

prevalent, is associated with a worse response to treatment, and

genotypes 2 and 3, to a better response. However, older age at the

beginning of treatment is also associated with a worse response to

treatment. Our SR were younger than NR, for which reason, the higher

prevalence of genotype 1 in NR could be due to, at least in part, a

cohort effect related to age [4,5].

All SR patients presented undetectable HCV-RNA in serum, whereas all

NR were positive with a mean viral load of around 0.7 × 106 IU/mL

(approximately three times lower than basal viral load).

HCV-cAg in serum was negative in all patients, except in three SR, at

a very low concentration (7.7 pg/mL), compared with that found in NR.

All of these NR patients, with one exception, presented a positive

antigenaemia, with a mean concentration of 72.6 pg/mL. One NR patient

with serum HCV-RNA positivity had negative HCV-cAg and a very low

viral load.

We found with the HCV-cAg, three possible false HCV-RNA negatives who

were positive for core antigen, and another patient who was HCV-cAg

negative and HCV-RNA positive. In order to discard a false

positivity, we extracted two more aliquots from each patient and

these were analysed in parallel with a control, the same results

being achieved on two separate occasions with the same patient.

The lowest sensitivity of the HCV-RNA of the assay used in this study

was 50 IU/mL. If we consider this technique as the 'gold standard'

the sensitivity and specificity of the HCV-cAg in our series was 96

and 100%, respectively, similar to that observed by others. It is

possible to occasionally find the presence of 'true' negatives by HCV-

RNA, when using commercial techniques. These negative results could

be changed into positives, when using specific primers of the NS5

region [6-8].

The observed correlation between HCV-cAg and RNA-HCV serum levels was

excellent (r = 0.847, P < 0.0001). Therefore, the core-HCV antigen

measurement could also be used as an indirect replication marker, as

it is an easier and cheaper technique [8-10].

It is considered that 1 pg/mL of total core is equivalent to

approximately 8000 IU/mL of HCV-RNA, with slight differences existing

from one patient to another [9]. Nevertheless, the current method

does not detect HCV-cAg, when HCV-RNA levels are below 20 000 IU/mL,

thus limiting its widespread use. This commercially available method

has not yet been approved by the FDA for use in clinical practice

[11].

Although HCV-cAg determination is not commonly employed, several

studies support its routine use [12]. In the 'window' phase of acute

C hepatitis, its clinical sensitivity compared with HCV-RNA has been

100%, as determined in several studies [6,7,11,13]. The results

obtained to date and also corroborated by our findings seem to

confirm that this test is a good alternative to current methods of

genomic amplification in several diagnostic aspects, in the follow-up

of acute and chronic infection by HCV and also in the control of

treated patients [14-16].

In patients with CHC infection, HCV-RNA determination shows a high

sensitivity of 95% and a maximum specificity of 100% in HCV-cAg (+)

and RIBA positive [17]. Moreover, the predictive negative value for

NR was excellent, this reached 93% at week 4 and 100% at week 12.

This allows us to decide whether to continue or stop the combined

treatment, based on its determination at this time [18,19].

In a study performed with a wide sample of infected blood donors with

genotypes 1-4, carried out in Japan, Egypt and Uzbekistan, the

sensitivity of HCV-cAg was 100% for genotypes 2 and 3, and 93 and 95%

for genotypes 1 and 4 respectively [20].

Nevertheless, HCV-RNA measurement continues to be the technique of

choice, and it appears that it cannot be replaced at the present time

by core antigen determination, although the latter could be used in

the basal diagnosis and in some cases for the evolution of treatment

or in laboratories that do not possess sufficient infrastructure for

PCR techniques, based on their greater simplicity and lower cost [21-

25].

HCV-RNA in the liver was positive in all NR and also in four SR. The

mean tissue viral load of NR was 0.7 × 106 IU/mL (similar to the

mean found in serum). Two SR who were liver HCV-RNA (+), were also

serum HCV-cAg (+), at a low concentration (2.9 pg/mL) in both cases.

This technique has not been standardized as yet and it is very

difficult to discuss its usefulness and reliability, especially when

positive results are obtained in SR. However, HCV-RNA determination

in the liver seems to be one of the most sensitive markers for

assessing the cure or the persistence of HCV infection after one

episode of acute hepatitis, or after treatment of CHC, because its

disappearance is rarely associated with posterior relapse [26-28].

A recent study compared serum and hepatic HCV-RNA in CHC patients,

and a higher concentration was found in the liver, approximately 104

times than found in serum. This would explain the fact of detecting

HCV-RNA (+) in the liver in the same patient, in the absence of

positive viraemia. Thus, it would be of interest to use this

technique in some doubtful cases [29].

Our data could explain that certain virologically serum-negative

patients, may be HCV-RNA (+) in the liver, and present a clinical or

biochemical relapse several years after response of treatment. It is

unknown whether this finding will predict a relapse in future. Our

four SR patients with HCV-RNA (+) in the liver have now been followed

up since the end of treatment over a long period (between 4 and 9

years) and no clinical relapse has been observed, but we consider

that this special group needs a longer surveillance, especially in

cases of use of concomitant immunosuppressive therapy. We do not

believe it to be necessary to routinely determine HCV-RNA in the

liver in order to check viral eradication, as positive patients do

not show relapse.

The evolution of antibody bands of HCV was analysed in 15 NR and 30

SR. We measured the intensity of positivity of bands and it was found

that two of them, E2 and NS4, showed marked differences, losing

intensity in SR (with five or more years of follow-up compared with

those of <5 years) and comparing these with NR. The decrease in HCV-

Ab titres after sustained clearance of HCV has been rarely reported

but there are some studies that have already reported the same

phenomenon [30]. We also observed this tendency as regards loss of

specific antibodies, for the rest of the NS bands, although without

any significant differences.

Our results indicate that antibodies anti-E2 and NS4 are the first to

disappear, after achieving a sustained response, a finding not

previously described and whose significance is unknown.