Share
Follow Us
Hepatology, February 1999, p. 585-589, Vol. 29, No. 2
Hepatitis C Virus RNA Profiles in Chronically Infected Individuals: Do They Relate to Disease Activity?
Patrizia Pontisso1,
From the 1Department of Clinical and Experimental Medicine, University of Padova, Italy; 2Division of Gastroenterology, Molinette Hospital, Torino, Italy;3Institute of Internal Medicine, University of Palermo, Italy; 4Department of Internal Medicine, University of Milano, Italy; and 5Roche Molecular Systems, Basel, Switzerland.
ABSTRACT
Fluctuations of Hepatitis C virus (HCV)-RNA serum levels were monitored in a multicenter study in 76 chronic HCV carriers who had been followed longitudinally without receiving antiviral therapy to assess their relation with the course of liver disease activity. Forty-four patients had normal transaminases over more than 2 years, while 32 additional patients had fluctuating levels.Viral load was measured in serial serum samples prospectivelycollected for 10 to 12 months in 54 patients and in sera stored yearly up to 8 years in an additional 22 patients. In patients tested monthly, a lesser extent of fluctuations was detected in cases with constantly normal transaminases as compared with those with fluctuating transaminases. In the former group, the mean difference between maximum and minimum values observed in each individual patient was 0.7 Log, while in the latter group, it was 1.3 Log (P = .0004). Most of these patients experienced, on average, three peaks of viremia over 1 year. The range of variation observed upon yearly testing was between 0.2 and 2.2 Log and did not reach statistical significance between the two groups. Inconclusion, a careful viral replication profile can be achievedonly by monthly testing, because longer time intervals could missviremia fluctuations. HCV-RNA levels are more stable in asymptomaticHCV carriers than in patients with biochemical activity of liverdisease. (HEPATOLOGY 1999;29:585-589.)
INTRODUCTION
Chronic infection with Hepatitis C virus (HCV) is characterized by persistent viremia. HCV RNA is usually detected in serum by sensitive polymerase chain reaction (PCR)-based techniques and has become a useful tool for diagnosis and monitoring. Besides methods for qualitative detection of viremia, a number of procedures to quantify serum HCV RNA have been developed, including end-point dilution PCR,1 competitive PCR,2,3 isothermal nucleic acid amplification,4and signal-amplification branched DNA.5 Routine use of these techniques in a wide clinical setting is hampered by problems of specificity and sensitivity, lack of reproducibility, poor standardization, and high cost.6 Recently, methods forquantitative assessment of HCV-RNA levels have become commerciallyavailable and are being extensively evaluated in clinical studies,particularly in patients treated with interferons and antivirals.7 Little is known about viral kinetics in untreated patients, because most evaluations of the level of viremia are based on single determinations. If wide spontaneous fluctuations occur, they could mimic treatment-inducedeffects or lead to under- or overestimate baseline values. Ina few published studies, different time intervals have been evaluated and discordant conclusions have been reported. Either trivial or consistent differences of viral load over time have been described after observation periods ranging from days to months.8-12 To assess the spontaneous behavior of serum HCV-RNA levels over a reasonable length of time upon close observation, we monitored chronic HCV carriers who had been followed longitudinally without receiving any antiviral therapy and then analyzed their patterns of virus replication in relation to biochemical and histological activity of liverdisease.
PATIENTS AND METHODS
Patients. Seventy-six chronic carriers of HCV were evaluated in a multicenter study (the HEPCCOM Italian Study) in which collection and virological analysis were performed by each participating Center, including the Division of Gastroenterology, Molinette Hospital-Torino, the Department of Internal Medicine-University of Milano, the Department of Clinical and Experimental Medicine-University of Padova, and the Institute of Internal Medicine-University of Palermo. The study population included 44 patients with constantly normal alanine aminotransferase (ALT) values since at least 6 months before enrollment in the study and during a follow-up period lasting for more than 2 years (mean follow-up, 3 years; range,2-8 years), and 32 additional patients with fluctuating ALT. There were 42 males and 34 females, with a median age of 53 years (range, 20-68 years). All were HCV-RNA-positive by qualitative PCR before entering the study. None of the patients had evidence of ongoing Hepatitis B virus or human immunodeficiency virus infection; current alcohol and/or drug abuse or metabolic disorders were also excluded in all patients. None of the patients underwent antiviral treatment during the time of the study. Serial serum samples collected yearly over a period of several years were available for 22 patients (median samples/patient, 4 samples; range, 3-8 samples/patient). In 54 additional cases, monthly serum and plasma samples wereobtained during a prospective follow-up period, lasting 10 ± 2 months (mean ± SD). All samples were collected in sterile conditions, centrifuged, and stored in 500-µL aliquots at -80°C within 3 hoursafter withdrawal. Only unthawed samples were used for HCV-RNAmeasurements. In 49 patients, liver biopsy was obtained just before or during the time of the study and was consistent with minimal liver lesions in 3 patients, mild chronic hepatitis in 25 patients, moderate chronic hepatitis in 20 patients, and severe hepatitis in 1 patient.
Viral Markers. Anti-HCV was determined by commercially available second-generation enzyme-linked immunosorbent assay kits (Ortho Diagnostics, Raritan, NJ) and confirmed by radioimmunoblot assay 2 or 3. Hepatitis B surface antigen and anti-human immunodeficiency virus were detected by enzyme-linked immunosorbent assay (Abbott Diagnostics, North Chicago, IL).
Qualitative HCV-RNA Determination. To assess HCV RNA in serum, either the commercially available qualitative Amplicor HCV assay (Roche Molecular Systems, Branchburg, NJ) or home-made reverse-transcription nested PCR were performed. The former assay showed sensitivity and specificity similar to in-house reverse-transcription PCR in a large confirmatory study13 to a cut-off level of 103 genomes/mL.
HCV Genotype Determination. The HCV genotype was determined by InnoLipa assay (Innogenetics, Gent, Belgium) or by a previously published protocol14 that has been validated by direct sequencing and by InnoLipa assay.15 Briefly, PCR-amplified products were spotted in parallel on nylon filters and hybridized with type-specific probes derived form the 5’noncoding region, labeled with deoxyuridine triphosphate fluoresceine, the reaction being revealed by an enhanced chemiluminescence method (ECL, Amersham International plc, UK) on autoradiographic films.
Quantitation of HCV RNA. Viral load was measured using a recently developed PCR system that includes a quantitation standard that is coamplified with the target HCV RNA (Amplicor HCV Monitor, Roche Molecular Systems).16 RNA was extracted from 100 µL of serum and from the standard containing a known number of RNA copies by a guanidium-thiocyanate method and precipitated by isopropanol. Single-tube reverse transcription and 5′ untranslated region genomic amplification was then performed in a 9600 thermal cycler (Perkin Elmer, Norwalk, CT) using rTthDNA polymerase in which AmpErase enzyme had been introduced toavoid carry-over. Amplified products were immediately denaturedand transferred to probe-coated microwell plates after serialfivefold dilution. Parallel HCV-specific and quantitation standard-specific hybridization was revealed by enzymatic detection and optical density measurement at a wavelength of 450 nm. The concentration of native HCV RNA was automatically calculated and was based on the ratio between the native bound HCV-RNA signal and the bound internal quantitation-standard signal.
Laboratory standardization of the assay included the following evaluations: 1) sera from 4 patients (2 with high viremia and2 with low viremia) run in decuplicate to assess the intra-assay variation; and 2) in 10 patients, HCV-RNA levels were evaluated using both serum and plasma samples obtained monthly, to assess variations of HCV-RNA determination efficiency when using sera.17
Statistical Analysis. The Kruskal-Wallis ANOVA median test, Spearman correlation, <="https://www.hepatitiscentral.com/images/hcv/hepatitis/virus/chi.gif" alt="X" width="9" height="10" />2 and the Fisher’s exact test were used for analysis of the results. Statistical calculations were performed using the program STATISTICA (Statsoft, Inc., Tulsa, OK). HCV-RNA levels were expressed as Log-transformed values to reduce the interference of extreme values.
RESULTS
Serum HCV-RNA Levels
Laboratory Standardization. Using a standard panel of sera at a known HCV-RNA concentration, provided by Roche Molecular Systems, the sensitivity of the assay was estimated between 103 and 104 genomes/mL.
The reproducibility experiments of the HCV quantitation assay showed an intra-assay coefficient of variation ranging from 57% to 23% when sera below or above 104 were used, respectively. Using logarithmic transformation of the results, the standard deviation of HCV-RNA determinations per subject (one serum run in decuplicate) ranged from 0.14 to 0.17. Therefore, in agreement with previously published results,16 changes beyond 0.5 Log were considered as true fluctuations and not as intrinsic test variations.
A good correlation was observed when viral load was assessed using serum or plasma samples, obtained from the same blood withdrawal and run at the same time (Spearman correlation r = .82). These results are in keeping with previous findings,16,18 while they do not confirm other data obtained with a competitive, home-made quantitative test,17 underlying the fact that different technical approaches can influence final results.
Viremia Profiles. Of the 76 HCV-infected patients, 44 had constantly normal ALT during follow-up, while 32 had fluctuating abnormal ALT. Liver biopsies were obtained from 23 patients with a normal transaminase profile and in 26 patients with elevated ALT. table 1 shows thehistological distribution in relation to transaminase and virological profiles. A significantly higher proportion of patients with a normal transaminase profile had minimal liver lesions or mild chronic hepatitis, as compared with patients with fluctuating ALT, who showed more active histology. The rate of fluctuations beyond the variability of the assay was similar in patients with mild or moderate chronic hepatitis. All 3 patients with minimal changes showed stable viremic levels over time.
| table 1. Histological Features Observed in Chronic HCV Patients in Relation to Transaminase Profile and Viremia Levels |
Relation to HCV Genotype. Overall, 38 patients were infected with genotype 1, 14 were infected with genotype 2, and 2 patients were infected with genotype3. HCV carriers infected with genotype 1 showed higher levels of viremia than those infected with genotype 2 or genotype 3 (P < .001), as shown in table 2. The observed differences are likely the result of the underestimation of HCV-RNA load in viral types 2 and 3 by the version of the assay used.6,18
| table 2. Viral Load in Relation to Infecting Genotype |
Viral Fluctuations Over Time in Relation to Clinical Profile. Serial determinations of viral load were performed in 22 patients with yearly collected serum samples (10 with normal ALT and 12 with elevated ALT), while for 54 additional patients (28 with normal ALT and 26 with elevated ALT), monthly collected serum samples were studied. table 3 shows individual changes of viremia, expressed as the maximum Log difference detected during the observationperiod, in relation to transaminase profile, infecting genotype,and histological findings. A significant difference was identifiedin monthly tested samples between patients with a normal ALT profilecompared with patients with elevated ALT. Infecting genotype andoverall histology did not show any influence on viremia profiles,although all the patients with minimal lesions had low fluctuationsof HCV RNA, but their small number did not allow any definitiveconclusion.
| table 3. Maximum Changes of HCV RNA (Log genomes/mL) Observed in Serum During Follow-up in HCV-Infected Patients |
The variability of HCV-RNA values in serial measurements of each patient over time is reported in Fig. 1. In patients withnormal ALT, monthly testing showed changes <1 Log in the majorityof the patients, 23% of them being below the intrinsic variabilityof the assay, while larger changes were observed in patients withfluctuating ALT. These features were not detectable in patientstested yearly, likely because long-term viral kinetics revealedthat in all the patients examined monthly, a maximum of threepeaks of viremia, defined as >50% increase, were experienced over1 year. This occurred in 59% of the patients, while 33% experienced two HCV-RNA peaks, and only 8% had none or one peak. The high likelihood of testing inter-spikes levels of HCV RNA when comparing single points per each year most likely explains this finding.
| Fig. 1. Distribution of the difference between the maximum and minimum value (<="https://www.hepatitiscentral.com/images/hcv/hepatitis/virus/Delta.gif" alt="" width="11" height="10" />) of HCV RNA in samples tested over time in individual patients. The median value for each group is represented by a solid bar. The gray zone indicates the range of intrinsic assay variability. |
Temporal behavior of viremia in relation to ALT profile is described in Fig. 2. HCV-RNA fluctuations above the assay variabilitywere observed in 77% of the patients with normal ALT profile andin 92% of the patients with abnormal ALT. In this latter group,viral profile was synchronous with transaminase behavior in 38%of the patients, while in 62% of them, the two parameters werenot synchronized (Fig. 3C and 3D).
| Fig. 2. Median values of biochemical profiles in relation to HCV-RNA levels over time observed in patients studied monthly with a normal (A) and abnormal (B) ALT profile. |
| Fig. 3. Examples of ALT and HCV-RNA behavior in individual patients with constantly normal ALT (A, B) and in patients with fluctuating ALT (C, D), prospectively studied monthly for 12 months. |
DISCUSSION
Fluctuations over time of HCV replication during the course of chronic HCV infection is a relevant issue that has not been definitely explored. Several clinical relations with replication activity of the virus, including grade of disease and response to antiviral drugs, have been based on single-time-point determinations of viremia.10,19-28 To date, conflicting results have been reported on long-term monitoring of viral replication. Different time intervals, ranging from diurnal to monthly determinations, have been evaluated using different methods of measurement, and discordant conclusions might result from the above-mentioned variables. No fluctuations were reported when daily to monthly analysis over 3 months wasperformed using the branched-DNA assay,11 while wide monthlyfluctuations have been described over 6 months using a competitive reverse-transcription PCR technique.8 In this multicenter study,viremia has been monitored in a large number of untreated patientsboth on a monthly and a yearly basis, and the majority of thesamples has been prospectively collected, following standard criteria of specimen handling and storage conditions. Monthly measurement of HCV RNA revealed that the majority of the patients experienced, on average, three peaks of viremia over 1 year, with a median change of absolute value of about 1 Log, while the remaining patients had a lower frequency of HCV-RNA flares. These findings indicate that a careful viral replication profile can be obtained only by monthly testing, because longer time intervals could miss viremia fluctuations. Monthly variations of HCV RNA were significantly lower in patients with a normal transaminase profile, as compared with those with abnormal ALT, suggesting a relationship between the two parameters. Comparison of ALT and HCV-RNA patterns in patients with fluctuating ALT revealed that in the majority of the patients, fluctuations were asynchronous (i.e., two ALT flares and three HCV-RNA peaks per year). While no impairment of HCV replication seems to occur in patients with normal transaminases, hepatocyte lysis observed in patients with fluctuating ALT could hamper viral replicative activity. As a consequence, a single determination of HCV-RNA level in patients with abnormal ALT does not seem adequate to classify individual patients as “high or low” viremic. It has been repeatedly reported that low viral loads are associated with better response to interferon, either alone or in combination with ribavirin.25-27,29 As a result of the observed viral fluctuations, the suggestion of waiting to start treatment until the patients have lower viremia to increase the response rates could arise. However, prospective trials are needed to specifically address this point. Liver histology did not show any significant correlation with replication activity of the virus, but in the study population, extreme histological features, including minimal changes and active cirrhosis, were not adequately represented. Further studies are required to draw conclusions on the relationship between viral replication profile and liver disease activity at the histological level.
When patients were divided according to the infecting genotype, similar fluctuations were detected in all groups, although absolute values of viremia observed in cases infected with genotype 1 were higher than those detected in patients infected with genotype2 and with genotype 3. This is likely a result of a better efficiency of detection of genotype 1 by the current version of the assay.6,18
In conclusion, long-term monitoring of viral replication showed that HCV RNA is relatively stable over time in asymptomatic HCV carriers, while a low number of viremic flares can occur over a year in patients with biochemical activity of liver disease.
Acknowledgment
The authors are deeply greateful to Professors A. Alberti, F. Bonino, M. Colombo, and A. Craxì for their substantial contribution in the design of the study and in the discussion of the results.
Abbreviations
Abbreviations: HCV, Hepatitis C virus; PCR, polymerase chain reaction; ALT, alanine aminotransferase.
Footnotes
Received June 12, 1998; accepted October 2, 1998.
Address reprint requests to: Patrizia Pontisso, M.D., Dipartimento di Medicina Clinica e Sperimentale-Clinica Medica 5°, Università di Padova, Via Giustiniani, 2 35128 Padova, Italy. E-mail: patrizia@ux1.unipd.it; fax: 49-8754179.
REFERENCES
Copyright © 1999 by the American Association for the Study of Liver Diseases.
