GASTROENTEROLOGY 1999;116:636–642

LIVER, PANCREAS, AND BILIARY TRACT

Real-Time Detection System for Quantification

of Hepatitis C Virus Genome

TOMOKO TAKEUCHI,* ASAO KATSUME,*,‡ TAKESHI TANAKA,§ AKI ABE,\ KAZUAKI INOUE,*
KYOKO TSUKIYAMA–KOHARA,* RYUJI KAWAGUCHI,\ SATOSHI TANAKA,§ and MICHINORI KOHARA*
*Department of Microbiology, The Tokyo Metropolitan Institute of Medical Science, Tokyo; ‡Fuji-Gotemba Research Laboratory,
Chugai Pharmaceutical Co., Ltd., Shizuoka; §Liver Unit, The Tokyo Metropolitan Komagome Hospital, Tokyo;
and \Center for Molecular Biology and Cytogenetics, SRL, Inc., Tokyo, Japan

detection is less sensitive in the early phase of HCV

See editorial on page 763.
infection and cannot differentiate between active infec-
tion or past infection status. The lack of antibody
Background & Aims: For diagnosis of hepatitis C virus production in chronically infected patients has also been
infection and monitoring of viral load in patients, a reported in some immunocompromised individuals such
highly sensitive and accurate hepatitis C virus quantifi- as patients who have undergone bone marrow transplanta-
cation system is essential. Methods: Hepatitis C
tion or hemodialysis. Therefore, direct detection or
virus genome was detected by real/time detection
system using an ABI Prism 7700 sequence detector
quantification of viral load is crucial for diagnosis of HCV
(Perkin Elmer Corp./Applied Biosystems, Foster City, infection or monitoring the efficacy of interferon treat-
CA). Results: As few as 10 copies of the genome ment of patients with hepatitis.
were detected, and the quantification range was be- Currently, HCV RNA is quantified by using nested
tween 101 and 108 copies (r G 0.99). This system reverse-transcription (RT) polymerase chain reaction (PCR;
was 10–100-fold more sensitive than an Amplicor conventional RT-PCR),8–10 branched DNA (bDNA) sig-
monitor (Roche Diagnostic Systems, Branchburg, NJ). nal amplification assay,11–13 or the Amplicor moni-
The coefficient of variation values for both intra-assay tor.14–18 Nested RT-PCR is highly sensitive, but the
precision and interassay reproducibility of identifying quantitatively and reproducibility of this technique are
the genome quantification ranged from 0.37% to insufficient for clinical use. Conversely, the Amplicor
2.00% and 0.88% to 4.66%, respectively. The system monitor (102–103 genome equivalents/mL) and bDNA
could detect the genome in 98% of patients with assay (3.5 ⫻ 105 genome equivalents/mL) lack sensitiv-
chronic hepatitis, 95.8% of patients with liver cirrhosis,
ity. Therefore, a simple, quantitative, and highly sensitive
and 100% of patients with hepatocellular carcinoma
system that obtains reproducible results should be devel-
who had the antibody to hepatitis C virus, but could not
detect the genome in patients without the antibody. oped to overcome these problems.
Conclusions: The establishment of a real-time detec- Recently, real-time detection (RTD)-PCR based on the
tion system enables more accurate diagnosis of infec- TaqMan Chemistry system (Perkin Elmer Corp./Applied
tion and monitoring of viral load in interferon-treated Biosystems, Foster City, CA) was developed. The TaqMan
patients via quantification of viral genome. Chemistry system takes advantage of the reverse-
transcriptase activity and 58-38 nucleolytic activity of Taq
DNA polymerase to digest a probe labeled with a
epatitis C virus (HCV) is the major causative agent
H of posttransfusion-associated hepatitis and more
than half of sporadic non-A, non-B hepatitis. Persistent
fluorescent reporter and quencher dye.19,20 The probe is
added directly to the PCR reaction mixture and is
HCV infections often progress to chronic hepatitis, liver
Abbreviations used in this paper: bDNA, branched DNA; CR-Rel,
cirrhosis, and hepatocellular carcinoma.1–3 Several serodi- complete response and relapsed response; CR-SR, complete re-
agnostic assays of HCV infection have been developed sponse and sustained response; FAM, 6-carboxy-fluorescein; IFN,
using recombinant DNA methods.4–7 interferon; PCR, polymerase chain reaction; RT, reverse-transcrip-
The diagnosis of HCV infection is currently based on tion; RTD, real-time detection; TAMRA, 6-carboxy-tetramethyl-
rhodamine; UTR, untranslated region.
the detection of serum antibodies to HCV antigens using r 1999 by the American Gastroenterological Association
enzyme-linked immunosorbent assay. However, antibody 0016-5085/99/$10.00
March 1999 HEPATITIS C VIRUS GENOME DETECTION 637

based on the detection rates of HCV RNA in HCV-

infected patient or noninfected patient sera and by
monitoring viral loads in the sera of interferon (IFN)-
treated patients.

Materials and Methods

Construction and In Vitro RNA Transcription
of Standard RNA
The plasmids pKI5 (genotype 1b HCV) and pKII5
(genotype 2b HCV)23 were used as sense templates after
digestion with HindIII. RNA was transcribed from the tem-
plates using T7 RNA polymerase and MEGAscript in vitro
transcription kits according to the manufacturer’s protocol
(Ambion, Austin, TX).
Figure 1. Schematic structure of DNA templates used to synthesize To purify standard RNA from template DNA perfectly, the
standard HCV RNA. Vectors were constructed as described previ- synthetic RNA was treated with 50 U of deoxyribonuclease 1
ously.15 PKI5 codes sequences from HCV genotype 1b. PKII5 codes (Boehringer Mannheim GmbH, Mannheim, Germany) at 37°C
sequences from HCV genotype 2b. for 10 minutes, followed by phenol-chloroform extraction.
Next, the extract was precipitated with 4.1 mol/L LiCl,
designed to hybridize to a region between the PCR
followed by purification using the RNeasy system (Qiagen,
primers (Figure 1). The fluorescent reporter dye 6-carboxy- Chatsworth, CA). These steps were important for elimination
fluorescein (FAM) is attached covalently to the 58 end of of template DNA and deoxyribonuclease. The RNA was
the probe. The quencher dye 6-carboxy-tetramethyl- quantified by measuring optical density, and molecular size and
rhodamine (TAMRA) is attached to the 38 end of the purity were confirmed by electrophoresis through a 1%
probe. During PCR, the probe hybridizes to the template agarose/formaldehyde gel in 3-(N-morpholino)propanesulfonic
and is digested as Taq DNA polymerase extends the PCR acid buffer.24 The purified RNA was diluted with 0.01%
primer. Digestion of the probe releases the reporter from diethyl pyrocarbonate–treated water containing 0.2 µg/µL
the activity of the quencher, and successive PCR cycles baker’s yeast transfer RNA (Boehringer Mannheim GmbH)
result in exponential amplification of the PCR product purified by phenol and chloroform extraction (10 times each)
and fluorescence intensity. and 10 mmol/L dithiothreitol and 200 U/mL ribonuclease
The use of an ABI Prism 7700 sequence detector inhibitor. The RNA was divided into small aliquots (10 µL),
placed in siliconized 500-µL tubes, and stored at ⫺80°C to
(Perkin Elmer Corp./Applied Biosystems) allows continu-
avoid repeated freezing and thawing.
ous measurement of the fluorescent spectra of all 96 wells We eliminated the possibility of contamination of the DNA
of a thermal cycler during PCR amplification. The template in the purified RNA by amplification using RTD-
reactions are monitored in real time.21,22 Most PCR PCR without reverse transcription. Synthetic HCV-RNA copy
amplifications reach a plateau phase of reporter fluores- numbers were calculated from the quantity and molecular
cent emission if the reaction is performed to high cycle weight of RNA according to conventional methods.
numbers. The amplification plot is examined early in the
reaction using an ABI Prism 7700 sequence detector and Oligonucleotide Primers and Probe
analyzed at a point that represents the log phase of RTD-PCR was performed using one of two sets of PCR
product accumulation. primers and probe complementary to sequences located in the
We report the development of RTD-PCR using a 58 UTR. Both sets were universally conserved among all known
dye-labeled oligonucleotide probe, primes that bind to sequences. Set 1 consisted of the forward primer KK30
the HCV 58 untranslated region (UTR), recombinant (nucleotides 58–77), 58-CTGTCTTCACGCAGAAAGCG-38;
Thermus thermophilus (rTth) DNA polymerase, and a the reverse primer KM3 (nucleotides 313–294), 58-CACTCG-
CAAGCACCCTATCA-38; and the TaqMan probe R6-84-
sequence detector. This system enables complementary
S26FT (nucleotides 84–109), 58-CATGGCGTTAGTATGAGT-
DNA (cDNA) synthesis and PCR amplification and
GTCGTGCA-38, which were purchased from Takara Syuzo
analysis in a single reaction tube. The sensitivity and (Kusatu, Siga, Japan). Set 2 consisted of the forward primer
specificity of this assay are evaluated using synthetic R6-130-S17 (nucleotides 130–146), 58-CGGGAGAGC-
HCV RNA as the standard, and the correlation between CATAGTGG-38; the reverse primer R6-290-R19 (nucleotides
results obtained using this assay and results obtained 290–272), 58-AGTACCACAAGGCCTTTCG-38; and the Taq-
using the bDNA assay and Amplicor monitor is investi- Man probe R6-148-S21FT (nucleotides 148–168), 58-CTGCG-
gated. The efficacy of the RTD-PCR system is evaluated GAACCGGTGAGTACAC-38, which were also purchased from
638 TAKEUCHI ET AL. GASTROENTEROLOGY Vol. 116, No. 3

Takara Syuzo. The reporter dye FAM was attached covalently to Tokyo, Japan. All patients tested seropositive for HCV antibod-
the 58 end (6-FAM amidite; Perkin Elmer Corp./Applied ies according to a second-generation enzyme-linked immuno-
Biosystems), and the quencher dye TAMRA was joined to the sorbent assay (International Reagents Corp., Kobe, Japan)7 and
38 end of the probe sequence (38-TAMRA CPG 500; Glen for HCV RNA according to bDNA assay and/or Amplicor
Research, Sterling, VA). monitor, and tested seronegative for hepatitis B virus (HBV)
and other autoimmune markers.
Quantification of HCV RNA by RTD-PCR
Samples for Determination
Total RNA was extracted from 250 µL of serum using
the Isogen-LS RNA extraction system (Nippon Gene, Tokyo,
of Prevalence Rate
Japan) or the SepaGene RV-R RNA extraction system (Sanko The positive sensitivity control group consisted of 14
Junyaku Co., Ltd., Tokyo, Japan) containing 5 µL 2-mercapto- patients who were seropositive for the antibody to HCV and
ethanol (ME), 10 µg transfer RNA (see Construction and In had normal alanine aminotransferase (ALT) levels for longer
Vitro RNA Transcription of Standard RNA), and 20 µg than 6 months, 103 patients with chronic hepatitis, 48 patients
glycogen according to the manufacturers’ instructions. The with liver cirrhosis, and 53 patients with hepatocellular
extracted total RNA was used in the PCR reaction. The RNA carcinoma caused by HCV. The negative sensitivity control
was dissolved in diethyl pyrocarbonate–treated water contain- group consisted of 20 healthy donors (13 men and 7 women;
ing 10 mmol/L dithiothreitol and 200 U/mL ribonuclease seronegative for both anti-HCV and HBV surface antigen), 15
inhibitor using a siliconized tube to avoid absorption of the patients with chronic HBV (without HCV antibody), and 15
RNA. Nested competitive RT-PCR,10 bDNA assay, and Ampli- patients with non-B, non-C hepatic failure (seronegative for HBV
cor monitor analysis were performed as described previously. and HCV). Samples were collected according to a protocol that
RTD-PCR was performed using the EZ rTth RNA PCR kit was previously shown to best preserve HCV RNA.10
with the ABI Prism 7700 sequence detector system. The
RTD-PCR reaction mixture was included in the 50 µL PCR Samples for Monitoring of Viral Load
reaction that contained TaqMan EZ buffer (50 mmol/L Bicine, The viral loads of 4 patients with chronic HCV were
115 mmol/L potassium acetate, 0.01 mmol/L EDTA, 60 monitored using RTD-PCR. These patients were treated with
nmol/L 6-carboxy-X-rhodamine, and 8% glycerol, pH 8.2; 6 million units of natural IFN-␣ (Sumitomo Pharmaceuticals,
Perkin Elmer Corp./Applied Biosystems), 200 µmol/L deoxy- Osaka, Japan) daily for 2 weeks and then three times per week
adenosine triphosphate, 200 µmol/L deoxyguanosine triphos- for another 22 weeks. Complete response and sustained
phate, 200 µmol/L deoxycytidine triphosphate, 500 µmol/L response (CR-SR), which was observed in 1 of these 4 patients,
deoxyuridine triphosphate, 200 nmol/L forward primer, 200 was defined as maintaining normal ALT levels throughout the
nmol/L reverse primer, 300 nmol/L TaqMan probe, 3 mmol/L follow-up period (at least 6 months) (Figure 2A). Complete
manganese acetate, 0.5 U AmpErase uracil-N-glycosylase 7.5 response and relapsed response (CR-Rel), which was observed
U rTth DNA polymerase, and template HCV RNA. After in the other 3 patients, was defined as showing normal ALT
initial activation of uracil-N-glycosylase at 50°C for 2 minutes, levels followed by an increase in ALT levels shortly after the
RT was performed at 60°C for 30 minutes, followed by cessation of IFN-␣ treatment (Figure 2B–D).
inactivation of uracil-N-glycosylase at 95°C for 5 minutes.
Subsequent PCR amplification consisted of 53 cycles of
denaturation at 95°C for 20 seconds and annealing and
extension at 60°C for 1 minute (primer and probe set 1) or at
62°C for 1 minute per cycle (primer and probe set 2) in an ABI
7700 sequence detector. After amplification, real-time data
acquisition and analysis were performed. Once the threshold
was chosen, the point at which the amplification plot crossed
the threshold was defined as Ct. The calculated Ct value is pre-
dictive of the quantity of target RNA copies present in the sample.
The standard curve for this assay was calculated using a series of
10-fold dilutions of previously titrated synthetic HCV RNA.
bDNA and Amplicor Monitor Assays
The bDNA11–13 and Amplicor monitor14–17 assays were
performed as described previously.
Samples for Specificity
and Sensitivity Assays Figure 2. Serum ALT levels and HCV-RNA quantity in the 4 patients
treated with IFN-␣. (A ) CR-SR case; and (B–D ) CR-Rel cases. 䊉, ALT
Sera from 15 randomly selected patients with chronic levels; 䊊, RTD-PCR (set 2 primers and probe) titers; 䊐, Amplicor
HCV were collected at the Liver Unit of Komagome Hospital, monitor titers.
March 1999 HEPATITIS C VIRUS GENOME DETECTION 639

Results Table 1. Correlation Between RNA Levels in Sera of Patients

With Different HCV Genotypes Determined
The sensitivity and quantitativity of RTD-PCR by Competitive RT-PCR and RTD-PCR
were examined by using synthetic HCV RNA (Figure 1). Using Set 1 or 2 Primers and Probe
Set 2 primers and probes enabled detection of as few as 10 Competitive RTD-PCR (copies/mL )
copies of genotype 1b and genotype 2b synthetic RNA Sample RT-PCR
no. Genotype (copies/mL ) Set 1 Set 2
(Figure 3). The correlation coefficient between RNA
3.2e ⫹ 06 1.8e ⫹ 06 2.3e ⫹ 06
quantity and threshold cycle (Ct) was ⬎0.98 for samples 1
2
1a
1b 3.2e ⫹ 05 2.8e ⫹ 05 1.4e ⫹ 05
with copy numbers between 101 and 108 (Figure 3B 3 1b 3.2e ⫹ 06 4.4e ⫹ 06 2.1e ⫹ 06
and E). Identical sensitivity and linearity were obtained 4 1b 3.2e ⫹ 04 1.8e ⫹ 04 2.1e ⫹ 04
5 1b 3.2e ⫹ 03 7.9e ⫹ 03 3.0e ⫹ 03
from each primer and probe set. 6 2a 1.0e ⫹ 04 1.2e ⫹ 03 8.6e ⫹ 03
The quantity of viral RNA in the sera of the 15 7 2a 3.2e ⫹ 04 2.1e ⫹ 04 4.4e ⫹ 04
patients infected with different HCV genotypes was 8 2a 3.2e ⫹ 03 1.0e ⫹ 03 1.7e ⫹ 03
9 2a 3.2e ⫹ 05 1.4e ⫹ 05 5.2e ⫹ 05
determined using sets 1 and 2 primers and probes. The 10 2a 1.0e ⫹ 05 1.1e ⫹ 05 2.0e ⫹ 05
viral RNA titers measured by sets 1 and 2 primers and 11 2a 3.2e ⫹ 02 8.2e ⫹ 01 6.6e ⫹ 02
probes were similar, and in the case of all (7 of 7) 12 2a 3.2e ⫹ 03 7.4e ⫹ 02 3.5e ⫹ 03
13 2b 3.2e ⫹ 04 8.2e ⫹ 04 6.0e ⫹ 03
genotype 2a–infected patients, set 2 primers and probe 14 2b 1.0e ⫹ 06 1.5e ⫹ 06 2.0e ⫹ 06
showed higher sensitivity (Table 1). The quantity of HCV 15 2b 3.2e ⫹ 03 9.9e ⫹ 03 2.0e ⫹ 03
RNA (genotype 1b) in the serum of 15 patients with Sets 1 and 2, probe and primers (see Materials and Methods).
chronic hepatitis was detected by the RTD-PCR method
and compared with that detected by bDNA assay and the correlation was found between the results of RTD-PCR
Amplicor monitor (Figure 4A and B). A significant and the two other methods, with a correlation coefficient
(r) of 0.840 (P ⫽ 0.006) vs. the bDNA assay and r ⫽
A D 0.612 (P ⫽ 0.044) vs. the Amplicor monitor. Five of the
11 cases determined to be negative by bDNA assay were
found to be positive by RTD-PCR (Figure 4A). All
samples that tested negative on the Amplicor and
positive on RTD-PCR were confirmed by Southern blot
analysis of the PCR products.
B E In another experiment, RTD-PCR results showed a
very strong correlation with those of the Amplicor
monitor in detection of HCV genotype 1b (r ⫽ 0.991)
(Figure 5A). There was also a significant correlation
between RTD-PCR results and those obtained using the
Amplicor monitor for detection of HCV genotype 2a
(r ⫽ 0.567) and 2b (r ⫽ 0.652) (Figure 5B and C). In
C F addition, the Amplicor monitor could not detect 2.2 ⫻
102 copies of synthetic HCV-RNA of either genotype 1b
or 2b, and the bDNA assay could not detect 2.2 ⫻ 105

Figure 3. Amplification plots of synthetic HCV RNA. Tenfold serial

dilutions of target HCV-RNA were prepared in triplicate, reverse
transcribed and amplified using TaqMan probe, and detected using
ABI Prism 7700 sequence detector. For each dilution, ⌬ normalized
reporter (Rn) is plotted against each cycle number. (A ) Fluorescence
units using set 2 primers, probe, and genotype 1b standard HCV RNA.
(D ) Fluorescence units using set 2 primers, probe, and genotype 2b
standard HCV-RNA. B (genotype 1b) and E (genotype 2b) show Ct
values19,20 plotted against relative copy number for two genotypes of
HCV. C (genotype 1b; set 1) and F (genotype 1b; set 2) show double- Figure 4. Correlation between serum HCV-RNA levels determined
strand DNA products of PCR reaction (53 cycles) with ethidium using RTD-PCR and those determined using (A ) bDNA assay and (B )
bromide. The calculated Ct value is predictive of numbers of copies of Amplicor monitor using set 1 primers and probe. The vertical dotted line
target RNA present in sample. bp, base pairs. indicates the detection limit of (A) bDNA assay and (B) Amplicor monitor.
640 TAKEUCHI ET AL. GASTROENTEROLOGY Vol. 116, No. 3

Table 2. Sensitivity of Amplicor Monitor and bDNA Assay

in Detection of Synthetic HCV RNA
Genotype 1b HCV-RNA Genotype 2b HCV-RNA
Equivalent copies copies detected by copies detected by
of synthetic Amplicor monitor Amplicor monitor
HCV RNA (copies/mL ) a (copies/mL ) a
2.2 ⫻ 108 1.2 ⫻ 107 1.1 ⫻ 107
2.2 ⫻ 106 4.4 ⫻ 105 2.4 ⫻ 105
Figure 5. Correlation between serum HCV-RNA levels determined 2.2 ⫻ 104 1.8 ⫻ 104 8.5 ⫻ 103
using RTD-PCR (set 2 primers and probe) and those determined using 2.2 ⫻ 103 8.6 ⫻ 102 5.3 ⫻ 102
Amplicor monitor. (A ) Genotype 1b; (B ) genotype 2a; and (C ) geno- 2.2 ⫻ 102 ND ND
type 2b. 2.2 ⫻ 101 ND ND
Equivalent copies Genotype 1b HCV-RNA Genotype 2b HCV-RNA
copies of synthetic HCV-RNA of either genotype 1b or of synthetic copies detected by copies detected by
2b (Table 2). Thus, RTD-PCR appears to be 10–100-fold HCV RNA bDNA (Meq/mL ) b bDNA (Meq/mL ) b
more sensitive than the Amplicor monitor for detection 2.2 ⫻ 109 ⬎58 ⬎58
of HCV genotype 1b and 2b RNA. 2.2 ⫻ 107 9.6 1.2
2.2 ⫻ 105 ND ND
Intraassay precision was assessed from four measure- 2.2 ⫻ 104 ND ND
ments of eight specimens of human sera (samples 1–6, 2.2 ⫻ 103 ND ND
positive control sera; and samples 7 and 8, negative 2.2 ⫻ 102 ND ND
control sera). The quantity of viral RNA means ranged ND, Not detected.
from 5.00 ⫻ 103 to 4.50 ⫻ 107 for the positive control aSerial dilutions of synthetic HCV-RNA were measured by Amplicor
monitor.
sera, and the coefficients of variation (CVs) ranged from bSerial dilutions of synthetic HCV-RNA were measured by bDNA assay.
0.37% to 2.00% (Table 3). The reproducibility between
runs (interassay) was assessed from the same specimens.
The mean quantity of viral RNA ranged from 7.80 ⫻ 103 CR-SR patient (Figure 2A), the HCV-RNA quantity
to 4.88 ⫻ 107 for these positive control sera during 4 decreased on day 5 to below the detection limit of each
days of measurements, and the CVs ranged from 0.88% assay. In CR-Rel patients, HCV RNA was detected
to 4.66% (Table 4). during IFN-␣ treatment by RTD-PCR but not by the
HCV genomes could be detected in the positive Amplicor monitor (Figure 2B–D).
control sera from HCV antibody–positive patients.
HCV RNA was detected in 10 (71.4%) of the 14
asymptomatic carrier cases, 101 (98%) of the 103 Discussion
patients with chronic hepatitis, 46 (95.8%) of the 48 We developed a novel HCV-RNA quantification
patients with liver cirrhosis, and 53 (100%) of 53 system using an RTD-PCR assay consisting of a dye-
patients with hepatocellular carcinoma (Figure 6). HCV labeled oligonucleotide probe, primers complementary to
genomes could not be detected in the control sera from the HCV 58 UTR of HCV genomes, rTth DNA polymer-
20 healthy donors, 15 patients with chronic HBV ase, and a sequence detector. This technique enables
infection, or 15 patients with non-B– and non-C–related cDNA synthesis and PCR amplification and analysis in a
liver diseases (Figure 6). single reaction tube. The sensitivity and specificity of this
The quantity of viral RNA in the sera of the CR-SR assay were evaluated using synthetic HCV RNA as the
and CR-Rel IFN-␣–treated patients was monitored using standard and HCV RNA–positive or –negative sera as the
both RTD-PCR and the Amplicor monitor. In the control. We also investigated the correlation between the

Table 3. Intra-assay Precision

Sample no. Experiment 1 Experiment 2 Experiment 3 Experiment 4 Mean SD CV (%)
1 4.0 ⫻ 106 3.0 ⫻ 106 3.2 ⫻ 106 3.6 ⫻ 106 3.45 ⫻ 106 0.055 0.85
2 4.8 ⫻ 107 4.8 ⫻ 107 4.0 ⫻ 107 4.4 ⫻ 107 4.50 ⫻ 107 0.038 0.49
3 3.0 ⫻ 107 3.2 ⫻ 107 2.8 ⫻ 107 3.2 ⫻ 107 3.05 ⫻ 107 0.028 0.37
4 5.6 ⫻ 103 4.2 ⫻ 103 4.6 ⫻ 103 5.6 ⫻ 103 5.00 ⫻ 103 0.063 1.70
5 3.4 ⫻ 104 3.4 ⫻ 104 2.4 ⫻ 104 2.4 ⫻ 104 2.90 ⫻ 104 0.087 1.96
6 4.0 ⫻ 104 2.8 ⫻ 104 2.6 ⫻ 104 2.6 ⫻ 104 3.00 ⫻ 104 0.089 2.00
7 ND ND ND ND
8 ND ND ND ND

ND, Not detected.

March 1999 HEPATITIS C VIRUS GENOME DETECTION 641

Table 4. Reproducibility Between Runs

Sample no. Day 1 Day 2 Day 3 Day 4 Mean SD CV (%)
1 4.0 ⫻ 106 3.5 ⫻ 106 3.6 ⫻ 106 4.7 ⫻ 106 3.95 ⫻ 106 0.058 0.88
2 4.2 ⫻ 107 4.5 ⫻ 107 4.4 ⫻ 107 6.4 ⫻ 107 4.88 ⫻ 107 0.084 1.09
3 4.0 ⫻ 107 3.1 ⫻ 107 5.0 ⫻ 107 4.0 ⫻ 107 4.03 ⫻ 107 0.085 1.12
4 1.3 ⫻ 104 5.0 ⫻ 103 6.0 ⫻ 103 7.2 ⫻ 103 7.80 ⫻ 103 0.180 4.66
5 3.4 ⫻ 104 2.9 ⫻ 104 2.6 ⫻ 104 4.4 ⫻ 104 3.33 ⫻ 104 0.099 2.20
6 4.0 ⫻ 104 3.0 ⫻ 104 3.0 ⫻ 104 3.9 ⫻ 104 3.45 ⫻ 104 0.066 1.46
7 ND ND ND ND
8 ND ND ND ND

ND, Not detected.

results of RTD-PCR and those of bDNA assays and an cor monitor for the detection of HCV-RNA genotypes 2a
Amplicor monitor. and 2b (Figure 5B and C). Furthermore, the HCV
The coefficient of correlation between RNA quantity genome was detected with very high sensitivity in the
and Ct was r ⬎ 0.99 between 101 and 108 copies, which HCV antibody–positive sera from 10 (71.4%) of the
indicates high sensitivity, good linearity, and a wide asymptomatic carriers, 101 (98%) of the patients with
dynamic range (Figure 3B and E). chronic hepatitis, 46 (95.8%) of the patients with liver
Using the RTD-PCR method, HCV RNA was de- cirrhosis, and 53 (100%) of the patients with hepatocellu-
tected in cases determined to be negative by the bDNA lar carcinoma, respectively (Figure 6). In the 4 cases of
assay (Figure 4A and Table 2). Thus, the RTD-PCR asymptomatic carriers in whom HCV-RNA was not
method has much higher sensitivity than the bDNA detected, the level of HCV antibody was low. Thus, in
assay. Furthermore, the RTD-PCR is 10–100-fold more these cases, there may be no HCV particles in the serum,
sensitive than the Amplicor monitor for detection of only an immune memory of HCV infection. In the
HCV genotype 1b and 2b synthetic RNA. negative controls, HCV genomes could not be detected in
Intraassay precision was assessed from four measure- the control sera of 20 healthy donors, 15 patients with
ments of eight specimens of human sera. The CVs ranged chronic HBV infections, and 15 patients with non-B–
from 0.37% to 2.00% (Table 3). The reproducibility and non-C–related liver diseases (Figure 6). These results
between runs was assessed from the same specimens. The indicate that RTD-PCR has a high specificity and
CVs ranged from 0.88% to 4.66% (Table 4). These sensitivity.
results indicate that the RTD-PCR has a high intra-assay RTD-PCR also showed much higher sensitivity in
precision and reproducibility. monitoring the HCV viral loads of IFN-␣–treated pa-
RTD-PCR was 10-fold more sensitive than the Ampli- tients compared with the Amplicor monitor (Fig-
ure 2B–D). These results indicate that using RTD-PCR
to monitor quantities of HCV-RNA could be very useful
in planning IFN treatment.
The enhanced sensitivity, linearity, specificity, and
reproducibility of RTD-PCR may have resulted from
successful selection of the HCV genome region from
which the PCR primers and probes were derived and the
simplicity of performing the whole analysis in a single
tube. RTD-PCR sensitivity also depends on the manner
in which HCV-RNA is prepared.
In conclusion, RTD-PCR has higher simplicity, sensi-
tivity, specificity, accuracy, and reproducibility. Quantifi-
cation can be performed within 3 hours without the
false-positive results that are typically caused by contami-
nation of second-round amplification. Based on our
Figure 6. Prevalence rate of HCV RNA in HCV-positive or -negative
results, RTD-PCR appears to be an extremely useful
control sera. ASC, seropositive for HCV antibody and normal ALT levels
for more than 6 months; CH, patients with chronic HCV; LC, patients HCV quantification system.
with HCV liver cirrhosis; HCC, patients with HCV hepatocellular
carcinoma; HBV carrier, patients with chronic HBV; Non-B, Non-C, HBV References
and HCV marker–negative patients with hepatitis; and Healthy, HBV 1. Alter HJ, Purcell RH, Shih JW, Melpolder JC, Houghton M, Choo QJ,
and HCV marker–negative and healthy donor. Kuo G. Detection of antibody to hepatitis C virus in prospectively
642 TAKEUCHI ET AL. GASTROENTEROLOGY Vol. 116, No. 3

followed transfusion recipients with acute and chronic non-A, human immunodeficiency virus infection and liver disease. Multi-
non-B hepatitis. N Engl J Med 1989;321:1494–1500. center Hemophilia Cohort Study. Blood 84;4:1020–1023.
2. Kiyosawa K, Sodeyama T, Tanaka E, Gibo Y, Yoshizawa K, Nakano 14. Zeuzem S, Ruster B, Roth WK. Clinical evaluation of a new
Y, Furuta S, Akahane Y, Nishioka K, Purcell RH, Alter HJ. polymerase chain reaction assay (Amplicory HCV) for detection
Interrelationship between blood transfusion, non-A, non-B hepati- of hepatitis C virus. Z Gastroenterol 1994;32:342–347.
tis and hepatocellular carcinoma. Hepatology 1990;12:671– 15. Nolte FS, Thurmond C, Fried MW. Preclinical evaluation of
675. AMPLICOR hepatitis C virus test for detection of hepatitis C virus
3. Saito I, Miyamura T, Ohbayashi A, Harada H, Katayama T, Kikuchi RNA. J Clin Microbiol 1995;33:1775–1778.
S, Watanabe Y, Koi S, Onji M, Ohta Y, Choo QL, Houghton M, Kuo 16. Navas S, Carreno V. Semiquantification by Amplicory assay of
G. Hepatitis C virus infection is associated with the development hepatitis C virus genome during therapy. J Hepatol 1995;22:115–
of hepatocellular carcinoma. Proc Natl Acad Sci USA 1990;187: 117.
6547–6549. 17. Miskovsky EP, Carrella AV, Gutekunst K, Sun CA, Quinn TC,
4. Kuo G, Choo QL, Alter HJ, Gitnick GL, Redeker AG, Purcell RH, Thomas DL. Clinical characterization of a competitive PCR assay
Miyamura T, Dienstag JL, Alter MJ, Stevens CE, Tegtmeier GE, for quantitative testing of hepatitis C virus. J Clin Microbiol
Bonino F, Colombo M, Lee WS, Kuo C, Berger K, Shuter JR, Overby 1996;34:1975–1979.
LR, Bradley DW, Houghton M. An assay for circulating antibodies 18. Albadalejo J, Alonso R, Antinozzi R, Bogard M, Bourgault A-M,
to a major etiologic virus of human non-A non-B hepatitis. Science Colucci G, Fenner T, Petersen H, Sala E, Vincelette J, Young C.
1989;244:362–364. Multicenter evaluation of the COBAS AMPLICOR HCV assay, an
5. Choo QL, Richman KH, Han JH, Berger K, Lee C, Dong C, Gallegos integrated PCR system for rapid detection of hepatitis C virus RNA
C, Coit D, Medina-Selby A, Barr PJ, Weiner AJ, Bradley DW, Kuo G, in the diagnostic laboratory. J Clin Microbiol 1998;36:862–865.
Houghton M. Genetic organization and diversity of the hepatitis C 19. Livak KL, Flood SJA, Marmaro J, Giusti W, Deetz K. Oligonucleo-
virus. Proc Natl Acad Sci USA 1991;88:2451–2455. tides with fluorescent dyes at opposite ends provide a quenched
6. Bresters D, Cuypers HTM, Reesink HW, Schaasberg WP, van der probe system useful for detecting PCR product and nucleic acid
Poel CL, Mauserbunschoten EP, Houghton M, Choo QL, Kuo G, hybridization. PCR Methods Applic 1995;4:357–362.
Lesniewski R, Troonen H, Lelie PN. Enhanced sensitivity of a 2nd 20. Morris T, Robertson B, Gallagher M. Rapid reverse transcription-
generation ELISA for antibody to hepatitis C virus. Vox Sang PCR detection of hepatitis C virus in serum by using the TaqMan
1992;62:213–217. fluorogenic detection system. J Clin Microbiol 1996;34:2933–
7. Saito M, Hasegawa A, Kashiwakuma T, Kohara M, Sugi M, Miki K, 2936.
Yamamoto T, Mori H, Ohta Y, Tanaka E, Kiyosawa K, Furuta S, 21. Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative
Wakashima M, Tanaka S, Hattori N. Performance of enzyme- PCR. Genome Res 1996;6:986–994.
linked immunosorbent assay system for antibodies to hepatitis C 22. Gibson UEM, Heid CA, Williams PM. A novel method for real time
virus with two new proteins (c11/c7). Clin Chem 1992;38:2434– quantitative RT-PCR. Genome Res 1996;6:995–1001.
2439. 23. Tsukiyama-Kohara K, Iizuka N, Kohara M, Nomoto A. Internal
8. Kaneko S, Murakami S, Unoura M, Kobayashi K. Quantitation of ribosomal entry site within hepatitis C virus RNA. J Virol 1992;66:
hepatitis C virus RNA by competitive polymerase chain reaction. J 1476–1483.
Med Virol 1992;37:278–282. 24. Sambrook J, Fritsch EF, Maniatis T, eds. Molecular cloning: a
9. Young KKY, Resnick RM, Myers TW. Detection of hepatitis C virus laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring
by a combined reverse transcription–polymerase chain reaction Harbor Press, 1989.
assay. J Clin Microbiol 1993;31:882–886.
10. Kohara M, Tanaka T, Tsukiyama-Kohara K, Tanaka S, Mizokami M,
Lau JYN, Hattori N. Hepatitis C virus genotypes 1 and 2 respond Received November 7, 1997. Accepted December 1, 1998.
to interferon-␣ with different virologic kinetics. J Infect Dis Address requests for reprints to: Michinori Kohara, Ph.D., Depart-
1995;172:934–938. ment of Microbiology, The Tokyo Metropolitan Institute of Medical
11. Urdea MS, Horn T, Fultz TJ, Anderson M, Running JA, Hamren S, Science, 3-18-22, Honkomagome, Bunkyo-ku, Tokyo 113-8613,
Ahle D, Chang CA. Branched DNA amplification multimers for the Japan. e-mail: [email protected]; fax: (81) 3-3828-8945.
sensitive, direct detection of human hepatitis viruses. Nucleic Supported in part by a Grant-in-Aid for Specially Promoted
Acids Symp Ser 1991;24:197–200. Research on Viral Diseases from the Tokyo Metropolitan
12. Lau JNY, Davis GL, Kniffen J, Qian KP, Urdea MS, Chan CS, Government; a grant from the Ministry of Education, Science, and
Mizokami M, Neuwald PD, Wilber JC. Significance of serum Culture of Japan; and a grant from the Ministry of Health and
hepatitis C virus RNA levels in chronic hepatitis C. Lancet Welfare of Japan.
1993;34:1501–1504. The authors thank Dr. Makoto Yoshiba for his thoughtful
13. Eyster ME, Fried MW, Di Bisceglie AM, Goedert JJ. Increasing comments and suggestions and Atsuko Fujisawa for creating the
hepatitis C virus RNA levels in hemophiliacs: relationship to figures.