| |
Hepatitis C Virus Replicates in the Liver of Patients Who Have a Sustained Response to Antiviral Treatment
|
| |
| |
Clinical Infectious Diseases Nov 15, 2006;43:1277-1283
Inmaculada Castillo, Elena Rodriguez-Inigo, Juan Manuel Lopez-Alcorocho, Margarita Pardo, Javier Bartolome, and Vicente Carreno
Foundation for the Study of Viral Hepatitis, Madrid, Spain
"....In summary, HCV persists and replicates in the livers and PBMCs of a high percentage of patients who received antiviral treatment for years after normalization of liver enzyme levels and clearance of serum HCV RNA. Although it may be suspected that the risk of HCV reactivation is smaller than the risk of hepatitis B virus reactivation [13], persistence of HCV in these patients should be taken into account under special circumstances (e.g., immunosuppression or chemotherapy). For example, there is a report of the case of a single patient with chronic hepatitis C who experienced serum HCV RNA clearance with normalization of aminotransferase levels. After 8.5 years of test results that were negative for HCV RNA, HCV infection reemerged following prednisone therapy [14]."
ABSTRACT
Background. Positive-strand hepatitis C virus (HCV) RNA has been detected in the livers of patients who have achieved a sustained biochemical and virological response to antiviral therapy (hereafter, referred to as sustained responders), but negative-strand HCV RNA was undetectable in the hepatic tissue of these patients. We studied the presence of both positive- and negative-strand HCV RNA in the livers of 20 sustained responders with chronic hepatitis C whose response persisted for a mean (± standard deviation [SD]) of 47.4 ± 32.8 months after treatment.
Methods. HCV RNA was tested by strand-specific, real-time reverse-transcriptase polymerase chain reaction and by in situ hybridization in posttreatment liver biopsy samples (obtained a mean [± SD] 35.4 ± 35.0 months after therapy) and in patients' peripheral blood mononuclear cells.
Results.
Positive-strand HCV RNA was found in 19 (95%) of 20 liver biopsy specimens, and negative-strand HCV RNA was found in 15 (79%) of the 19 samples that had positive-strand HCV RNA. These results were confirmed by in situ hybridization.
Regarding peripheral blood mononuclear cells, 13 (65%) of 20 samples had positive-strand HCV RNA, and negative-strand HCV RNA was detected in 12 (92%) of the 13 samples with positive-strand HCV RNA.
Liver necroinflammation was still present in the posttreatment liver biopsy specimens of 15 patients, and fibrosis was present in 7, although liver damage improved in all but 2 patients.
Conclusions. HCV persisted and replicated in the livers and peripheral blood mononuclear cells of most sustained responders. Thus, these patients did not experience HCV infection clearance, despite apparent clinical disease resolution.
INTRODUCTION
Hepatitis C virus (HCV) infection represents a major health problem: about 170 million people worldwide have chronic HCV infection. It is thought that normalization of liver enzyme levels and loss of serum HCV RNA, either after self-limited acute hepatitis C or after successful antiviral therapy in chronically infected patients, indicates clearance of the virus from the liver and, therefore, resolution of infection. However, recent data suggest that HCV infection may exist in the absence of viremia. We previously reported that HCV can infect and replicate in the livers and PBMCs of patients who have abnormal liver function values of unknown etiology in the absence of anti-HCV and serum HCV RNA (called occult HCV infection) [1, 2]. Similarly, Radkowski et al. [3] reported that HCV RNA persisted in the livers of 3 of 11 patients with chronic hepatitis C who had achieved a sustained biochemical and virological response after antiviral therapy. However, negative-strand HCV RNA (the viral replicative intermediate) was undetectable in the livers of these patients.
Because of the important clinical implications of this finding, we studied the presence of HCV RNA in the liver biopsy specimens of 20 patients with a sustained biochemical and virological response (hereafter, referred to as sustained responders) in an attempt to confirm and expand on those results. Special emphasis has been placed on the study of the possible presence of negative-strand HCV RNA and, thus, on HCV replication.
PATIENTS AND METHODS
Twenty patients with chronic hepatitis C who responded to antiviral therapy were retrospectively included in this study. Patients had been treated thrice weekly with 3 MU of recombinant a-IFN alone (6 patients) or in combination with 1000-1200 mg of ribavirin (4 patients) for 48 weeks, or they were treated once weekly with 1.5 ug per kg of body weight with pegylated IFN alone (2 patients) or in combination with 1000-1200 mg of ribavirin (8 patients) for 24 or 48 weeks. Patients were selected on the basis of the availability of a stored posttreatment liver biopsy specimen and a PBMC sample obtained the same day as the liver biopsy. They were also selected on the basis of persistently normal liver function test results and negative serum HCV RNA test results that remained for at least 16 months after treatment. These 20 patients were contacted and gave their consent for specific testing for HCV RNA in their stored posttreatment liver biopsy and PBMC samples. The study was performed in accordance with the principles of the Declaration of Helsinki.
All patients had histologically proven chronic hepatitis C before treatment, all were anti-HCV and serum HCV RNA positive, and all had abnormal liver enzyme values. Other causes of liver disease were excluded; all patients were HIV-uninfected. The mean age (± SD) at treatment was 42.7 ± 12.6 years, 14 patients (70%) were male, 17 (85%) were infected with HCV genotype 1b, and 3 (15%) were infected with genotypes 2, 3, or 4. During the posttreatment follow-up (mean [± SD] duration, 47.4 ± 32.8 months), liver function tests were performed and serum HCV RNA was checked (Amplicor HCV v2.0; Roche Diagnostics) every 6 months.
For all 20 patients, the posttreatment liver biopsy specimen was obtained 35.4 ± 35.0 months after completion of therapy (range, 8-117 months). The sample obtained during the liver biopsy was cut into 2 portions. One portion was used for histological examination and the second was submerged (no later than 30 s after obtaining the liver sample) in RNALater (Ambion) and stored at -20C until use. PBMC samples, obtained the same day as the liver biopsy and stored at -20C in RNALater (Ambion), were available for all patients. The pre- and posttreatment liver biopsy samples were evaluated simultaneously by a single pathologist and were assigned a Scheuer score [4] to compare histological findings before and after therapy.
To assure the specificity of results, HCV RNA detection was performed in a blinded fashion on different days by 2 different operators. PBMC samples from 10 healthy, anti-HCV-negative subjects, total RNA isolated from HepG2 cells, and blanks were used as negative controls. Finally, all procedures were performed in accordance with the recommendations of Kwok and Higuchi [5].
Quantitative, strand-specific, real-time RT-PCR. Total RNA was isolated from liver and PBMC specimens using the SV Total RNA Isolation kit (Promega), and the concentration was determined by using spectrophotometric analysis.
Quantification of the 5' noncoding (5'NC) region of both HCV RNA strands was performed as described elsewhere [6], using strand-specific, real-time RT-PCR. The thermostable enzyme Tth was used at a high temperature for the synthesis of cDNA. Briefly, for detecting positive-strand HCV RNA, the cDNA was generated at 65C for 20 min in a 20-uL reaction mixture containing 0.5 ug of total RNA, 50 pmol/L of antisense primer UTRLC2 (5'-CAAGCACCCTATCAGGCAGT-3'), 1 mmol/L MnCl2, 200 mol/L of each deoxynucleotide triphosphate, 1X RT buffer (Applied Biosystems) and 5 units of Tth (Applied Biosystems). Thereafter, RT activity was inactivated by Mn2+ chelation with 8 L of 10X chelating buffer (Applied Biosystems), followed by heating at 95C for 30 min. For detecting the negative-strand HCV RNA, the cDNA was synthesized under the same conditions, but with the sense primer UTRLC1 (5'-CTTCACGCAGAAAGCGTCTA-3'). Real-time RT-PCR was performed in a LightCycler (Roche Molecular Biochemicals) with 2 uL of cDNA in a final volume of 20 uL, using the LightCycler FastStart DNA Master SYBR Green I kit (Roche Molecular Biochemicals). The reaction mixture contained 4 mmol/L MgCl2, 0.5 umol/L of primers UTRLC1 and UTRLC2, and 2 L of SYBR Green Master mix (Roche Molecular Biochemicals). Amplification consisted of an initial denaturation and activation of the enzyme for 10 min at 95C, followed by 60 cycles at 95C for 1 s, 60C for 5 s, and 72C for 10 s, and a final step of fluorescence acquisition at 89C for 5 s. Two standard curves with synthetic HCV RNA of positive or negative polarity were constructed for quantification of both HCV RNA strands. Linearity of the assay ranged from 3.2 to 3.2 X 108 copies of positive-strand or negative-strand HCV RNA per reaction. This assay detects 3.2 molecules of the correct strand and nonspecifically detects 107-108 copies of the incorrect strand [6]. Sensitivity and dynamics of each assay were not affected when RNA isolated from HepG2 cells was added to the reaction.
Phylogenetic analysis. To guard against cross-contamination among positive samples, the HCV core region was amplified from total RNA isolated from the postreatment biopsy samples by RT-PCR, as described elsewhere [1]. The 302-base pair core PCR products were cloned into the pCRII-TOPO vector (Invitrogen), and 4 clones from each patient were automatically sequenced. Sequences were aligned with core sequences corresponding to all HCV genotypes retrieved from GenBank using ClustalX, version 1.81 [7]. Phylogenetic and molecular evolutionary analyses were conducted using Mega, version 2.1 [8]. Genetic distances were estimated using Kimura's 2-parameter method, and SEs of distances were calculated using the bootstrap method (1000 replicates). A phylogenetic tree was constructed with the neighbor-joining method, and its statistical significance was tested using the bootstrap method (1000 replicates).
In situ hybridization. Paraffin-embedded liver sections (4 um) were pretreated for in situ hybridization, as described elsewhere [9]. Positive-strand HCV RNA was detected with a complementary RNA probe obtained by in vitro transcription of the pC5' noncoding region plasmid, which contains the complete 5' noncoding region of the HCV genome. Negative-strand HCV RNA was detected with a complementary RNA probe spanning 390 nucleotides of the HCV core coding region, which was obtained by in vitro transcription of the pC core plasmid. Hybridization conditions for detection of HCV RNA of both polarities were described elsewhere [9]. The percentage of infected cells was determined by visual inspection, and at least 2000 cells from each liver section were counted.
Statistical analysis. Statistical analysis was done using SPSS software, version 9.0 (SPSS). Continuous variables were expressed as mean ± SD, except when otherwise indicated. Means were compared using the Mann-Whitney U test, and paired data were compared using the Wilcoxon signed rank test. Correlations were performed using Spearman's test. A 2-sided P value <.05 was considered to be statistically significant.
RESULTS
During the posttreatment follow-up (47.4 ± 32.8 months), liver function tests (measuring aspartate aminotransferase, alanine aminotransferase, and y-glutamyl transpeptidase levels) had persistently normal results, and serum HCV RNA test results were always negative.
Positive-strand HCV RNA was found in 19 (95%) of 20 liver biopsy samples that were obtained 35.4 ± 35.0 months after the end of treatment, and all negative controls were negative. The results of HCV RNA detection performed by different operators on different days were identical in all cases. The mean positive-strand HCV RNA load (± SEM) was 1.9 X 105 ± 5.4 X 104 copies per ug of total RNA. The presence of this intrahepatic positive-strand HCV RNA was confirmed in all cases by in situ hybridization, and the percent of HCV-infected hepatocytes was 4.5% ± 3.7%. The only liver biopsy sample with undetectable viral RNA by strand-specific, real-time RT-PCR also had negative results by in situ hybridization.
Negative-strand HCV RNA was found by strand-specific, real-time RT-PCR in 15 (79%) of the 19 patients with intrahepatic positive-strand HCV RNA. The mean negative-strand HCV RNA load (± SEM) was 7.3 X 104 ± 3.1 X 104 copies per ug of total RNA, which is significantly lower than that of positive-strand HCV RNA (P = .001). The only patient whose biopsy sample did not yield intrahepatic positive-strand HCV RNA also had negative results for negative-strand HCV RNA. The presence of the negative strand was confirmed in all cases by in situ hybridization. The percent of hepatocytes with hybridization signals for negative-strand HCV RNA was 3.0% ± 2.6%, which was significantly lower than the percent of hepatocytes with positive-strand HCV RNA (P = .003). Individual data on HCV RNA detection are shown in table 1.
HCV core sequencing was performed for 9 patients (7 with HCV genotype 1b infection, 1 with genotype 2 infection, and 1 with genotype 3 infection). In the remaining 10 patients with intrahepatic viral RNA, it was not possible to perform core amplification, because RNA samples were exhausted. The phylogenetic analysis demonstrated that the genetic distances among clones from each patient were lower than genetic distances among patients (table 2). The phylogenetic tree shows that the clones were segregated in the 9 patients, indicating that no cross-contamination occurred (figure 1). These results are in agreement with the HCV genotypes that these 9 patients presented in serum before treatment (table 1), because the sequences of the clones from the patients were distributed together with the sequences of their corresponding HCV genotype (1b, 2, or 3).
Positive-strand HCV RNA was detected in 13 (65%) of 20 PBMC samples obtained on the same day that the posttreatment liver biopsy was performed. The patient in whom HCV RNA was not detected in the liver also did not have HCV detected in his PBMC sample. Negative-strand HCV RNA was found in the PBMC samples of 12 (92%) of 13 persons with HCV infection (table 1), and the mean (± SEM) HCV RNA load (1.6 X 105 ± 5.5 X 104 copies per ug of total RNA) was significantly lower than that of the positive-strand HCV RNA (9.5 X 105 ± 6.9 X 105 copies per ug of total RNA) (P = .012). When the amounts of HCV RNA detected in the liver and PBMC samples were compared, we found that the positive-strand HCV RNA load was significantly higher in PBMCs than in liver specimens (mean [± SEM] viral load, 9.5 X 105 ± 6.9 X 105 copies per ug of total RNA vs. 1.9 X 105 ± 5.4 X 104 copies per ug of total RNA; P = .034). Negative-strand HCV RNA load was also significantly higher in PBMCs, compared with livers (mean [± SEM], 1.6 X 105 ± 5.5 X 104 copies per ug of total RNA vs. 7.3 X 104 ± 3.1 X 104 copies per ug of total RNA; P = .037). However, levels of positive- or negative-strand HCV RNA in liver or PBMC samples and time from treatment were not correlated (data not shown).
Regarding histological examination of the liver, there was a significant improvement in necroinflammatory activity (portal plus lobular inflammation), compared with the pretreatment values (score, 3.8 ± 1.6 vs. 1.3 ± 0.9; P < .001). Moreover, the fibrosis score was significantly lower (P = .015) in the post-treatment liver biopsy samples (0.7 ± 1.1) than in the pretreatment ones (1.2 ± 1.0). Nevertheless, necroinflammation was still present in 15 patients, and fibrosis was present in 7, although liver damage improved in all but 2 patients (table 3). Finally, no correlation was found between necroinflammation in the posttreatment liver biopsy samples and levels of positive- or negative-strand HCV RNA (data not shown).
DISCUSSION
We analyzed the presence of positive- and negative-strand HCV RNA in the liver biopsy specimens of 20 sustained responders. Using strand-specific, real-time RT-PCR, positive-strand HCV RNA was detected in 19 (95%) of the 20 liver biopsy specimens that were analyzed. The specificity of the detection of the viral genome was confirmed by in situ hybridization and by the HCV core sequence analysis showing that there was no cross-contamination among liver samples. Negative-strand HCV RNA (the replicative intermediate) was found (both by real-time RT-PCR and by in situ hybridization) in 15 (79%) of the 19 liver biopsy samples that had positive-strand HCV RNA. This indicates that an ongoing viral replication was taking place in the livers of these patients, which explains the persistence of intrahepatic HCV years after a successful antiviral therapy outcome. The differences in the antiviral treatments received by patients, as well as the difference in the dates of the performance of the posttreatment liver biopsies (range, 8-117 months), support the fact that HCV persistence is unrelated to a specific treatment or to time since the end of treatment.
We previously reported the presence and replication of HCV not only in the livers of patients who had abnormal liver function values of unknown etiology [1], but also in the livers of healthy anti-HCV-positive patients who were serum HCV RNA negative and who had persistently normal liver function test results [6]. The patients in the present study comprise a different population, because they were patients with chronic hepatitis C who responded to antiviral treatment. Thus, occult HCV infection may be present in different clinical situations.
Radkowski et al. [3] reported persistence of HCV in sustained responders. Our results differ from those reported by these authors in both the percentage of patients with positive-strand HCV RNA detected in their livers (95% vs. 27%), as well as in the detection of viral replication (79% vs. 0%). There are 2 possible explanations for these discrepancies. First, in our study, liver biopsies were performed a mean (±SD) of 35.3 ± 35.0 months after the end of treatment, and Radkowski et al. [3] included liver biopsies performed 63.6 ± 16.7 months after therapy. Thus, intrahepatic levels of positive- and negative-strand HCV RNA might decrease over time, becoming undetectable in sustained responders. However, there are some data that do not support this hypothesis. We have detected the positive- and negative-strand HCV RNA in liver biopsy samples obtained 50 months after treatment, whereas Radkowski et al. [3] did not find HCV RNA in 4 liver biopsy samples obtained 41, 50, 54, and 55 months after therapy. In addition, 1 of our liver biopsy samples with positive- and negative-strand HCV RNA was obtained 117 months after treatment, which is a longer posttreatment period than that experienced by the patients of Radkowski et al. [3] (maximum length of follow-up, 98 months). Another possibility may lie in the integrity of the intrahepatic HCV RNA. It has been established that the time elapsed between obtaining and freezing the liver biopsy specimen (to inhibit intracellular RNAse) must be no longer than 3 min to prevent RNA degradation and to ensure detection of positive-strand HCV RNA and especially negative-strand HCV RNA [10]. In the present study, liver biopsy specimens were submerged in a chemical agent that inhibits RNAses within 30 s after they were obtaine to preserve the integrity of RNA, but information about measures taken to preserve RNA integrity was not provided by Radkowski et al. [3].
The mean ratio of positive- to negative-strand HCV RNA found in the livers of our sustained responders was 2.6 : 1, differing from the data of a previous study that also used a quantitative, strand-specific, real-time PCR and reported ratios of 100 : 1-1000 : 1 [11]. However, although our patients were serum HCV RNA negative, the study by Pradel-Komurian [11] analyzed viral RNA strands in the livers of chronic hepatitis C patients with HCV RNA in serum. Thus, contamination by circulating virions (that contain positive-strand HCV RNA) cannot be discarded, and this contributes to those high ratios. Moreover, positive strand to negative strand ratios could be affected by HCV RNA degradation if liver biopsy samples are not properly stored, because the decrease in the titer of negative-strand HCV RNA is more pronounced than that of positive-strand HCV RNA [10].
PBMC samples obtained on the same day the liver biopsies were performed were available from all patients, and positive-strand HCV RNA was detected in 13 (65%) of them. Negative-strand HCV RNA was also found in 12 (92%) of the 13 PBMC samples harboring positive-strand HCV RNA. The mean ratio of positive to negative strands in the PBMC samples of our patients was 5.9, which is similar to the mean ratio of 6.6 reported elsewhere in lymphocytes and macrophages of sustained responders [3]. Thus, our findings agree with previous studies, except that we detected HCV replication without exogenous cell activation, and in other reports, negative-strand HCV RNA was detected only after mitogen stimulation of PBMCs [3, 12]. We do not have a clear explanation for this difference. One possible explanation could be that HCV genotypes infecting our patients were different than those found in the other mentioned reports. However, this was not true; most patients in all of these studies were infected with HCV genotype 1. We also found that the loads of positive- and negative-strand HCV RNA in our patients were significantly higher in PBMC samples than in liver biopsy samples, which suggests that the HCV strain detected in our sustained responders replicates more efficiently in PBMCs than in livers.
Finally, regarding histological damage present in posttreatment liver biopsy samples, it should be stated that 15 patients had continued liver necroinflammation, and 1 patient had liver fibrosis; however, there was an overall improvement in histological damage. So, it is difficult to know whether persistence of HCV infection and replication has a clinical relevance until more data become available.
In summary, HCV persists and replicates in the livers and PBMCs of a high percentage of patients who received antiviral treatment for years after normalization of liver enzyme levels and clearance of serum HCV RNA. Although it may be suspected that the risk of HCV reactivation is smaller than the risk of hepatitis B virus reactivation [13], persistence of HCV in these patients should be taken into account under special circumstances (e.g., immunosuppression or chemotherapy). For example, there is a report of the case of a single patient with chronic hepatitis C who experienced serum HCV RNA clearance with normalization of aminotransferase levels. After 8.5 years of test results that were negative for HCV RNA, HCV infection reemerged following prednisone therapy [14].
References
1. Castillo I, Pardo M, Bartolome J, et al. Occult hepatitis C virus infection in patients in whom the etiology of persistently abnormal results of liver-function tests is unknown. J Infect Dis 2004; 189:7-14. First citation in article | Full Text | PubMed
2. Castillo I, Rodriguez-Inigo E, Bartolome J, et al. Hepatitis C virus replicates in peripheral blood mononuclear cells of patients with occult HCV infection. Gut 2005; 54:682-5. First citation in article | PubMed | CrossRef
3. Radkowski M, Gallegos-Orozco JF, Jablonska J, et al. Persistence of hepatitis C virus in patients successfully treated for chronic hepatitis C. Hepatology 2005; 41:106-14. First citation in article | PubMed | CrossRef
4. Scheuer PJ. Classification of chronic viral hepatitis: a need for reassessment. J Hepatol 1991; 13:372-4. First citation in article | PubMed | CrossRef
5. Kwok S, Higuchi R. Avoiding false positive with PCR. Nature 1989; 339:237-8. First citation in article | PubMed | CrossRef
6. Carreno V, Pardo M, Lopez-Alcorocho JM, Rodriguez-Inigo E, Bartolome J, Castillo I. Detection of hepatitis C virus (HCV) RNA in the liver of healthy, anti-HCV antibody-positive, serum HCV-RNA-negative patients with normal alanine aminotransferase levels. J Infect Dis 2006; 194(1):53-60. First citation in article | Full Text | PubMed
7. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 1997; 25(24):4876-82. First citation in article | PubMed | CrossRef
8. Kumar S, Tamura K, Jakobsen IB, Nei M. MEGA2: Molecular Evolutionary Genetics Analysis Software. Bioinformatics 2001; 17:1244-5. First citation in article | PubMed | CrossRef
9. Gosalvez J, Rodriguez-Inigo E, Ramiro-Diaz JL, et al. Relative quantification and mapping of hepatitis C virus by in situ hybridization and digital image analysis. Hepatology 1998; 27:1428-34. First citation in article | PubMed | CrossRef
10. Madejon A, Manzano ML, Arocena C, Castillo I, Carreno V. Effects of delayed freezing of liver biopsies on the detection of hepatitis C virus RNA strands. J Hepatol 2000; 32:1019-25. First citation in article | PubMed | CrossRef
11. Pradel-Komurian F, Perret M, Deinman B, et al. Strand specific quantitative real-time PCR to study replication of hepatitis C virus genome. J Virol Methods 2004; 116:103-6. First citation in article | PubMed | CrossRef
12. Pham TNQ, MacParland SA, Mulrooney PM, Cooksley H, Naumov NV, Michalak TI. Hepatitis C virus persistence after spontaneous or treatment-induced resolution of hepatitis C. J Virol 2004; 78:5867-74. First citation in article | PubMed | CrossRef
13. Lok AS, Liang RH, Chiu EK, Wong KL, Chan TK, Todd D. Reactivation of hepatitis B virus replication in patients receiving cytotoxic therapy. Report of a prospective study. Gastroenterology 1991; 100:182-8. First citation in article | PubMed
14. Lee WM, Polson JE, Carney DS, Sahin B, Gale M. Reemergence of hepatitis C virus after 8.5 years in a patient with hypogammaglobulinemia: Evidence for an occult viral reservoir. J Infect Dis 2005; 192:1088-92. First citation in article | Full Text | PubMed
|
|
| |
| |
|
|
|
