844
Determinants of the Quantity of Hepatitis C Virus RNA
David L. Thomas,1,2 Jacquie Astemborski,2
David Vlahov,2,3 Steffanie A. Strathdee,2 Stuart C. Ray,1
Kenrad E. Nelson,2 Noya Galai,2 Karen R. Nolt,1
Oliver Laeyendecker,1 and John A. Todd4,a
1
Division of Infectious Diseases, Johns Hopkins School of Medicine,
and 2Department of Epidemiology, Johns Hopkins School of Hygiene
and Public Health, Baltimore, Maryland; 3Center for Urban
Epidemiologic Research, New York Academy of Medicine, New York,
New York; 4Bayer Diagnostics, Emeryville, California
To test the hypothesis that person-to-person variability in blood levels of hepatitis C virus
(HCV) RNA can be explained, the quantity of HCV RNA was assessed in 969 persons who
acquired HCV infection in the context of injection drug use. Serum HCV RNA levels ranged
from 200,000 to 1120 million equivalents/mL (the linear range of the assay). The median log10
HCV RNA level was 0.46 higher in 468 human immunodeficiency virus (HIV)–positive persons
than in 501 HIV-negative persons (P ! .001 ). In addition, among HIV-negative persons, lower
HCV RNA levels were independently associated with younger age (P ! .001 ), ongoing hepatitis
B infection (P = .005), and the absence of needle sharing (P = .02). However, 190% of the
person-to-person HCV RNA level variability was not explained by these sociodemographic,
environmental, and virologic factors. Additional research is necessary to ascertain what determines the level of HCV RNA in blood.
It is estimated that persons infected with human immunodeficiency virus (HIV) make 1011 virus particles a day [1]. The
serum HIV level, which represents the equilibrium of virus production and clearance, is the strongest predictor of clinical outcome. Persons with low serum HIV RNA levels are more likely
to clear viremia with treatment and less likely to acquire opportunistic infections [2–5]. Determinants of HIV levels have
been investigated, and host genetic polymorphisms and sex differences have been described elsewhere [6, 7]. For example,
Farzadegan et al. [6] found that HIV RNA levels were lower
among women.
It is also estimated that persons infected with hepatitis C
virus (HCV) make 1012 HCV particles a day [8]. As with HIV,
a low serum HCV RNA level has been associated with improved treatment response [2, 9]. However, the association between serum RNA level and disease progression is less evident
with HCV than with HIV and has not been found in all investigations [10–13].
The variability in research findings may be explained by difReceived 24 June 1999; revised 25 October 1999; electronically published
9 March 2000.
Presented in part: 49th meeting of the American Association for the Study
of Liver Diseases, Chicago, November 1998.
Informed consent was obtained from all participants in accordance with
the guidelines of the Johns Hopkins Committee on Human Research.
Grant support: NIH (DA-10627, U19 A140035, DA-08004, and DA04334).
a
J.A.T. is an employee of Bayer Diagnostics, which makes the assay used
in this investigation.
Reprints or correspondence: Dr. David L. Thomas, Division of Infectious
Diseases, Johns Hopkins University, 1147 Ross Research Bldg., 720 Rutland
Ave., Baltimore, MD 21205.
The Journal of Infectious Diseases 2000; 181:844–51
q 2000 by the Infectious Diseases Society of America. All rights reserved.
0022-1899/2000/18103-0005$02.00
ferences in factors that determine the level of serum HCV RNA,
which may vary by up to 6 orders of magnitude among untreated persons. However, although the HCV RNA level is an
important indicator of treatment response and a possible indicator of the natural history of infection, there have been few
studies to ascertain the regulating factors. To test the hypothesis
that the level of HCV RNA differs according to virologic, environmental, or sociodemographic factors, serum HCV RNA
quantities were assessed in a large cohort of HCV-infected persons who were at risk for HIV and hepatitis B virus (HBV)
infections.
Research Methods
Participants. Since 1988–1989, members of a cohort of former
and current injection drug users in the AIDS Linked to the Intravenous Experience (ALIVE) study have been followed semiannually at Johns Hopkins University (Baltimore) [14]. At enrollment,
the cohort was 81% male, 90% black, and 25% HIV positive; median age was 34 years and median duration of injection drug use
was 13 years. A standardized questionnaire, administered at enrollment and readministered in modified form at each visit, elicited
the following patient characteristics: sex, age, race, illicit drug use
(including frequency of use, years of use, and specific information
regarding method of heroin and cocaine use), methadone use, alcohol use (including information regarding frequency and quantity), and tobacco use. Members of the original 1988–1989 cohort
with >1 visit between January 1995 and March 1996, when serum
samples were obtained for HCV RNA studies, were included in
this analysis.
Laboratory studies. Blood was obtained from study participants at study visits and sent to a regional commercial laboratory
(Quest Diagnostics, Baltimore) for liver enzyme testing. Liver enzymes, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), and g-glutamyl transferase (GGT), were eval-
JID 2000;181 (March)
Evaluation of Hepatitis C RNA
uated within 24 h of serum procurement by a multichannel
chemistry analyzer (model 5200; Olympus, Lake Success, NY). At
least 3 serum aliquots were also stored within 2 h of procurement
at 2207C for !1 week and then transferred without thawing to a
permanent repository and stored at 2807C. HCV antibody was
tested by use of EIA (Ortho Diagnostic Systems, Raritan, NJ) on
samples taken at enrollment and on subsequent visits for seronegative participants as described elsewhere [15]. HIV testing was also
performed at each study visit for HIV-negative participants, and
CD4 lymphocyte counts were measured in HIV-positive participants as described elsewhere [14]. Sera were also tested at enrollment for hepatitis B surface antigen (HBsAg), antibody to hepatitis
B core, and antibody to HBsAg by commercial kits (Abbott Laboratories, Abbott Park, IL), as described elsewhere [16]. Serum
samples from the same visit that were tested for HCV RNA level
were also tested for HBsAg for those who were capable of being
chronic carriers, including those who were HBsAg positive or negative for all 3 markers at enrollment.
HCV RNA levels were determined on previously unthawed serum aliquots collected at consecutive study visits from January
1995 through March 1996 (further described in Results). Baseline
values refer to the first specimen tested per subject. HCV RNA
levels were assayed by 1 technician according to the manufacturer’s
instructions (bDNA Quantiplex HCV RNA 2.0; Bayer Diagnostics, Emeryville, CA) [17]; the assay uses synthetic oligonucleotide
probes to bind HCV RNA molecules to a solid phase. A second
set of probes hybridizes HCV RNA to a branched DNA amplifier,
which generates signal (chemiluminescence) that is proportional to
the quantity of HCV RNA in the sample. The HCV RNA level is
assigned units of equivalents per milliliter (eq/mL); the assay has
a linear range of 200,000 to 120 million eq/mL. All samples were
run in duplicate, and the average of the 2 values was reported as
the HCV RNA level. If the coefficient of variance between the 2
samples exceeded 20%, testing was repeated.
HCV genotype was assessed by restriction length polymorphism
according to the method of Davidson et al. [18]. On a subset of
42 participants, HCV genotype was also ascertained by phylogenetic analysis of core-E1 sequence, as described elsewhere [19].
Core-E1 sequence was also assessed by the same method to examine
whether reinfection occurred in a subset of participants who acknowledged ongoing drug use and needle sharing. HCV sequence
corresponding to the putative NS5A interferon sensitivity–
determining region (ISDR) was amplified, as described with coreE1, with the following modifications—Hotstar Taq (Qiagen,
Valencia, CA) was used with the following primers: 50-CRKCBCCYGAATTYTTCAC-30 (nt 6691–6710) and 50-TCGAARGAGTCCAGRATYAC-30 (nt 7131–7111) and cycling conditions 15 min
957C, then 70 cycles of 947C 15 s, 627–527C for 30 s stepping down
by 27C every four cycles until the final annealing temperature was
reached, 727C for 1 min. Degenerate primers are indicated with
standard codes of the International Union of Pure and Applied
Chemistry. Nucleotide positions are relative to HCV-1 (accession
no. M62321).
Statistical analysis. The overall distribution of HCV RNA
levels was examined in HCV antibody-positive participants who
had HCV RNA detectable by bDNA assay. HCV RNA levels
exceeding the linear range of the assay were assigned 120 million
eq/mL, the highest recorded value, whereas participants with base-
845
line HCV RNA levels not detected by bDNA were excluded. HCV
RNA levels were log10 transformed to approximate a normal distribution, but they remained skewed among HIV-positive participants. Thus, statistical comparisons were made by nonparametric
tests (Kruskal-Wallis and Spearman’s rank). Liver enzymes were
categorized into clinically meaningful groups as normal, 11.5 times
normal, and others, whereas other continuous variables were categorized into quartiles. The categorical format was shown in tables
for consistency with other variables and to optimally demonstrate
the magnitude of differences in distribution. However, for liver
enzymes, the correlation with HCV RNA level was also expressed
in text as Spearman’s correlation coefficient.
The independence of the association of variates that appeared
to be correlated with HCV RNA level on univariate analysis was
further assessed by fitting multivariate linear regression model of
the log10 HCV RNA levels. Because of the skewed distribution of
data, this analysis was repeated by using a nonparametric multivariate analysis of the rank order of HCV RNA levels, and similar
results were obtained. However, because the parameter estimates
provide a more meaningful representation of the magnitude of
effects, the nonparametric linear regression model is shown. The
degree to which the final model explained the overall variance was
indicated by the adjusted r2.
Results
The quantity of HCV RNA was evaluated in 969 anti–HCVpositive persons with HCV RNA detectable by bDNA. Cohort
members were excluded from this analysis if they had HCV
RNA levels that were undetectable by bDNA (n = 135), were
anti-HCV negative (n = 72), or were not tested for anti-HCV
(n = 3). The sociodemographic, drug use, and serologic status
of participants with HCV load investigation are shown in table
1. The age, sex, drug use patterns, and income of study participants were similar to the original cohort at enrollment. However, the subset under study was more likely to be HIV positive
at enrollment. By 1995 when HCV levels were assessed, 48%
were HIV positive and 36% had !200/mL CD4 lymphocytes.
No study members were being treated for HCV infection during
the interval of quantitative HCV RNA testing. Overall, the
median HCV RNA level in the cohort was 9.3 million eq/mL.
Values spanned the linear range of the assay and included 17
participants with HCV RNA levels above the range, who were
assigned the value of the upper limit, 120 million eq/mL.
HIV infection. The median HCV RNA level was higher
(7.19 log10 eq/mL) in 468 HIV-positive participants than in 501
HIV-negative persons (6.73 log10 eq/mL; P ! .001; figure 1).
However, among HIV-positive participants, increased HCV
RNA levels were not detected in those with lower CD4 lymphocyte counts (P 1 .1; table 2). Similarly, among HIV-positive
participants, no significant differences in HCV RNA levels were
detected by age, race, sex, alcohol or drug use, HCV genotype,
or HBsAg status. Statistically significant (P ! .01 ) associations
were observed between HCV RNA level and the serum AST
and GGT but not with serum ALT; however, the magnitude
846
Thomas et al.
of each liver enzyme correlation was small (Spearman’s correlation coefficient, r 2 ! 0.25).
HIV-negative participants. Among 501 HIV-negative participants, HCV RNA levels were higher in older participants,
males, HBsAg-negative persons, and those with high risk drugusing practices (sharing needles; table 3). However, after adjusting for age, males no longer had significantly higher HCV
RNA levels (table 4). Older age and needle sharing were independently associated with higher HCV RNA levels, whereas
serum HBsAg detection was inversely associated. No significant
differences were detected in HCV RNA levels among participants with different alcohol use patterns. As observed among
HIV-positive participants, a statistically significant association
was detected between the level of HCV RNA and serum GGT,
but not AST or ALT. However, the magnitude of each liver
enzyme correlation was small (r 2 ! 0.25).
No differences were detected in the HCV RNA levels by HCV
subtype. To further examine the hypothesis that polymorphisms
in the NS5A ISDR correlate with viral load differences, we
compared the sequences of 14 HIV-negative participants in the
highest viral load quartile with HCV subtype-, age-, sex-,
and HBsAg status–matched control subjects who had HCV
RNA levels in the lowest quartile. No differences were noted
(figure 2).
To estimate the degree to which the factors assessed in this
investigation accounted for the person-to-person variability in
HCV RNA levels, a linear regression model was constructed
of all variates associated with the log10 HCV RNA levels on
univariate analysis. However, in this model that contained HIV
infection, HBsAg status, age, and needle sharing, !10% of the
variance in HCV RNA was accounted for (adjusted r2, 0.085).
Change in HCV RNA over time. In total, 687 participants
had 11 HCV RNA level in the linear range of the bDNA assay:
473 had 2 tests and 214 had 3 tests constituting 901 pairs. The
median (interquartile range [IQR]) time between the 901 visits
was 5.7 months (range, 5.5–6.0 months). From visit to visit,
Table 1.
JID 2000;181 (March)
Figure 1. Distribution of log10-transformed hepatitis C virus (HCV)
RNA levels in 468 human immunodeficiency virus (HIV)–positive and
501 HIV-negative participants. In HIV-positive participants, distribution of log10 HCV RNA levels is skewed toward higher values. % HIVnegative and -positive participants in each HCV RNA stratum is in
parentheses above bars.
the median (IQR) change in the magnitude of the HCV RNA
level was 0.26 log10 (range, 0.12–10 log10). By using the first
pair of visits, participants who reported injection drug use in
the interval between HCV RNA assessments had higher visitto-visit differences than those who did not report injection drug
use (P = .038), but this difference was small (0.27 vs. 0.23 log10,
respectively). No differences were detected in visit-to-visit HCV
RNA variance according to HIV or HBV infections, alcohol
use, or sociodemographic factors (data not shown). During the
relatively short observation period, nearly equal proportions of
participants experienced increases or decreases in HCV RNA
level; the median intervisit slope per semester of follow-up
(IQR) was 20.029 log10 (range, 20.233 to 0.178 log10).
Characteristics of study participants.
Characteristic
Median years of age (IQR)
Male
Black
Years (IQR) since first IDU
Using injection drugs
Sharing needles
Legal income !$5000
HIV seropositive
!200 CD4 cells/mL
Anti-HCV positive
Enrollment cohort
in 1988–1989
(n = 2946)
Viral load subset
in 1988–1989
(n = 969)
Viral load subset
in 1995
(n = 969)
34 (29–38)
81
89
13 (6–19)
89
63
73
24
7
94
34 (30–38)
77
96
14 (7–19)
95
69
75
31
2
100
41 (37–45)
—
—
21 (14–26)
60
15
88
48
36
—
NOTE. Data are % unless otherwise indicated. Viral load subset represents participants
who presented between January 1995 and March 1996 and who were anti-HCV positive with
detectable HCV RNA by bDNA assay. Their enrollment characteristics (col. 3) are alongside
those of original cohort (col. 2) and as they were when first tested by bDNA (col. 4). Data
are at 6 months before visit unless stated otherwise. HIV, human immunodeficiency virus;
anti-HCV, hepatitis C virus antibody; IDU, injection drug use; IQR, interquartile range.
JID 2000;181 (March)
Evaluation of Hepatitis C RNA
Table 2. Correlates of hepatitis C virus (HCV) RNA level among
468 human immunodeficiency virus–positive participants.
Log10 HCV RNA level
a
Factor
Age, years
!37
37–40
41–44
>45
Sex
Male
Female
Race
Black
Nonblack
Years from first injection drug use
!14
14–20
21–25
>26
Needle sharing
No
Yes
Alcohol use, drinks/week
None
1–21
121
Body mass index
!20.02
20.02–21.64
21.65–23.87
>23.88
CD4 lymphocyte count, cells/mL
>500
200–499
!200
Hepatitis B surface antigen
Negative
Positive
Alanine aminotransferase (ALT), IU/L
!40
40–59
>60
Aspartate aminotransferase (AST), IU/L
!35
35–52
>53
g-Glutamyl transferase (GGT), IU/L
!55
55–82
>83
HCV genotype
1a
1b
2
Other
n
Median
IQR
P
156
114
101
97
7.17
7.17
7.32
7.15
6.67–7.61
6.43–7.52
6.75–7.57
6.78–7.56
—
—
—
1.10
354
114
7.22
7.16
6.72–7.57
6.57–7.53
1.10
452
16
7.19
7.36
6.66–7.56
6.64–7.66
1.10
139
128
95
105
7.25
7.17
7.31
7.16
6.54–7.57
6.57–7.55
6.76–7.62
6.75–7.55
—
—
—
1.10
401
65
7.22
7.16
6.64–7.57
6.68–7.51
—
1.10
186
214
68
7.19
7.17
7.27
6.63–7.55
6.57–7.56
6.72–7.65
—
—
1.10
114
113
116
114
7.25
7.17
7.17
7.16
6.75–7.57
6.53–7.62
6.75–7.53
6.57–7.55
—
—
—
1.10
91
201
167
7.16
7.22
7.19
6.47–7.53
6.68–7.63
6.73–7.53
—
—
1.10
428
21
7.22
7.00
6.71–7.56
6.15–7.40
—
.10
287
98
82
7.28
7.11
7.16
6.55–7.57
6.72–7.54
6.75–7.53
—
—
1.10
156
136
175
7.06
7.25
7.23
6.19–7.56
6.74–7.57
6.85–7.56
—
—
.012
197
63
207
7.01
7.28
7.34
6.20–7.52
6.89–7.51
6.88–7.61
—
—
!.001
240
90
10
31
7.30
7.15
7.38
7.33
6.80–7.62
6.57–7.52
6.74–7.84
6.57–7.90
—
—
—
1.10
—
—
NOTE.
Statistical comparisons of log-transformed HCV RNA levels were
made by Kruskal-Wallis test. Continuous variables were assessed with log-transformed HCV RNA levels by Spearman’s correlation (see text). Continuous variates were categorized for consistency as follows: age and years from first injection
drug use by quartile, alcohol use to differentiate >3 daily drinks from other use,
and liver enzymes (ALT, AST, and GGT) as less than upper limit of normal
(ULN), ! 1.5 3 UNL, and 1 1.5 3 UNL. Difference between 468 and sum of
column 2 represents no. with data missing for each variate. IQR, interquartile
range.
a
Refers to practices in 6 months before collection of serum samples tested for
HCV RNA level.
847
To further examine the possibility that HCV reinfection occurred and explained changes in HCV RNA levels, we examined the genetic distances in core-E1 sequence between paired
serum samples from 20 participants who acknowledged ongoing drug use and needle sharing between visits. Ten participants were chosen from HCV genotype 1a and 10 from HCV
genotype 1b; the mean time between visits was 6 months (SD,
3 weeks), and there was a mean of 491 injections (SD, 317
injections). Seven participants had intervisit HCV RNA level
differences >0.5 log10. The intervisit genetic distance was similar
to previously observed genetic distances among individual viral
variants within a specimen and were lower than intersubject
genetic distances (figure 3) [20]. However, the intervisit genetic
distances of these 20 participants were lower than all possible
intersubject comparisons of the 10 HCV 1a participants.
Discussion
In this large-scale investigation of HCV-infected participants,
the largest differences in serum HCV RNA levels were between
HIV-positive and -negative persons. This finding was expected,
since in 29 members of this cohort we previously observed that
HCV RNA levels increased after HIV infection was acquired
[21]. Similarly, others have found higher serum HCV RNA
levels in HIV-positive persons [22, 23].
Among HIV-positive persons in this study, no significant correlates of serum HCV RNA level were detected. A relationship
was expected with CD4 lymphocyte counts, since we previously
found an inverse association (albeit small) in 29 HIV seroconverters, and Eyster et al. [21, 22] demonstrated a strong inverse
correlation between CD4 lymphocyte count and serum HCV
RNA level in hemophiliacs. The lack of association with CD4
lymphocyte count in the present analysis may reflect changes
in the HIV-positive cohort (survival bias). In our former investigation, we only studied persons whose onset of HIV infection was known and a subset with rapid progression of HIV
infection. In contrast, the present study only includes participants who survived to the initiation of this study. Of interest,
others have also failed to detect an association between HCV
serum RNA level and CD4 lymphocyte count in HIV-positive
persons [23–25].
In HIV-negative persons, higher serum HCV RNA levels
were found among older participants. In this cohort, age was
a strong correlate of the length of drug use and with duration
of HCV infection [26], suggesting that the level of serum HCV
RNA may increase over time from HCV infection. Others have
reported findings to support this hypothesis [27]. Although the
median intervisit slope observed in this study was relatively
stable, the observation time was short (!2 years). Our findings
could also be consistent with age-related immunosuppression,
since other biologic differences (e.g., vaccine responses) have
been described within this age range. In either case, these data
emphasize the importance of controlling for age in evaluations
848
Thomas et al.
Table 3. Correlates of hepatitis C virus (HCV) RNA level among
501 human immunodeficiency virus–negative participants.
Log10 HCV RNA level
a
Factor
Age, years
!37
37–40
41–44
>45
Sex
Male
Female
Race
Black
Nonblack
Years from first injection drug use
!14
14–20
21–25
>26
Needle sharing
No
Yes
Alcohol use, drinks/week
None
1–21
121
Hepatitis B surface antigen
Negative
Positive
Alanine aminotransferase (ALT), IU/L
!40
40–59
>60
Aspartate aminotransferase (AST), IU/L
! 35
35–52
>53
g-Glutamyl transferase (GGT), IU/L
!55
55–82
>83
HCV genotype
1a
1b
2
Other
n
Median
IQR
P
116
110
124
151
6.45
6.75
6.78
6.90
5.95–6.89
6.28–7.21
6.12–7.25
6.43–7.37
—
—
—
!.001
396
105
6.75
6.59
6.23–7.23
5.92–7.11
—
.049
476
24
6.70
6.89
6.17–7.21
6.64–7.30
1.10
123
112
123
143
6.48
6.76
6.74
6.90
6.03–7.07
6.29–7.13
6.08–7.31
6.33–7.30
—
—
—
.033
424
76
6.70
6.93
6.19–7.17
6.13–7.38
—
.044
170
231
97
6.74
6.69
6.65
6.22–7.15
6.18–7.16
6.03–7.41
—
—
1.10
478
7
6.74
5.94
6.21–7.23
5.60–6.31
—
.011
267
113
120
6.75
6.69
73
6.06–7.23
6.27–7.10
6.23–7.20
—
—
1.10
210
153
137
6.71
6.69
6.73
5.99–7.18
6.20–7.22
6.39–7.26
—
—
1.10
209
87
204
6.56
6.81
6.83
5.93–7.07
6.27–7.25
6.39–7.31
—
—
!.001
253
124
13
36
6.78
6.74
7.23
6.90
6.28–7.26
6.06–7.14
6.85–7.47
6.48–7.34
—
—
—
.064
—
NOTE.
Statistical comparisons of log-transformed HCV RNA levels were
made by Kruskal-Wallis test. Continuous variables were also assessed with logtransformed HCV RNA levels by Spearman’s correlation (see text). Continuous
variates were categorized for consistency as follows: age and years from first
injection drug use by quartile, alcohol use to differentiate >3 daily drinks from
other use, and liver enzymes (ALT, AST, and GGT) as within upper limit of
normal (UNL), ! 1.5 3 UNL, and 1 1.5 3 UNL. Difference between 501 and sum
of column 2 represents no. with data missing for each variate. IQR, interquartile
range.
a
Refers to practices in 6 months before collection of serum samples tested for
HCV RNA level.
of HCV RNA level and underscore the need for prospective
investigations of the relationship between HCV RNA level and
disease outcome.
The association of serum HBsAg positivity with lower HCV
RNA level has been reported elsewhere [28–32]. Because
HBsAg was inversely associated with age in this cohort, it was
important that the correlation of HBsAg and lower HCV RNA
JID 2000;181 (March)
level was independent of age. Although a virologic interaction
has been proposed, the data in this study do not provide insight
into the biologic basis for this relationship.
In this investigation, serum HCV RNA level was also positively associated with needle sharing. Among injection drug
users, needle sharing could be a marker for HCV reinfection,
which has been demonstrated in humans with thalassemia and
in chimpanzees [33, 34]. Supporting this hypothesis are the lack
of association of HCV load with drug use without needle sharing, the lack of association with drug type (heroin or cocaine),
and the additional association with use of a shooting gallery,
another plausible marker for HCV reinfection (data not
shown). No evidence for HCV reinfection was found in the
majority of core-E1 sequences of 20 participants with ongoing
drug use and needle sharing. Although these data may suggest
other explanations for the association between drug use and
HCV RNA level, it is impossible to exclude reinfection, since
a limited number of participants were assessed and all variants
in the HCV quasispecies could not be examined. Fortunately,
the proportion of participants actively injecting drugs decreased
over the course of follow-up.
Women in this cohort had lower serum HIV RNA levels [6].
However, unlike HIV RNA levels, after adjusting for age, no
sex difference was detected in the level of HCV RNA [6]. It
was also anticipated that alcohol use and HCV genotype might
be associated with higher HCV RNA level, since these associations have been found in some, but not all, prior studies
[35–37]. If the alcohol effect on HCV replication was short lived
and alcohol use was irregular in the cohort, the association
could have been missed, since alcohol use was assessed as it
occurred in the 6 months before RNA testing. Since ∼90% of
the cohort have genotype 1 HCV infection, differences in HCV
RNA level between HCV genotypes would be difficult to detect.
This analysis was designed to assess determinants of HCV
RNA levels, not the relationship between the progression of
HCV infection and HCV RNA level. The correlation of HCV
Table 4. Multivariate analysis of hepatitis C virus (HCV) RNA levels
among 501 human immunodeficiency virus–negative participants.
Factor
Age, years
!37
37–44
>45
Needle sharing
No
Yes
Hepatitis B surface antigen
Negative
Positive
Sex
Male
Female
Parameter estimate
P
—
0.26
0.39
!.001
—
0.19
—
.02
—
20.71
—
—
—
!.001
—
.005
1.05
—
NOTE. Results of multivariate linear regression of log-transformed HCV
RNA levels. Similar results were obtained by nonparametric multivariate model
of rank order of results. Sex was not included in the final model from which
other shown data were derived.
JID 2000;181 (March)
Evaluation of Hepatitis C RNA
Figure 2. Amino acid (aa) sequences of putative NS5A interferon
sensitivity–determining region (ISDR). aa sequences of NS5A ISDR
(positions 2209–2248) are shown for 14 human immunodeficiency virus
(HIV)–negative participants selected from those in the highest hepatitis
C virus (HCV) RNA quartile and from 14 HIV-negative participants
from those in the lowest HCV RNA quartile, all infected with HCV
subtype 1a. Participants did not acknowledge injection drug use in 6
months before testing and were matched by other factors known to
affect HCV RNA concentration, including age, sex, and hepatitis B
surface antigen status.
RNA level with liver enzymes was considered. Although a statistically significant positive correlation was found with GGT
and, with HIV-positive participants, AST, the strength of each
correlation was weak (r 2 ! 0.25). In addition, liver enzymes are
only crude indicators of the stage of liver disease.
It is unlikely that the results of this investigation were substantially altered by technical limitations. Although some specimens were stored at 2807C for 2 years before testing, recent
data suggest that storage does not affect the HCV RNA level
determined by bDNA [38]. Repeated thawing and prolonged
time before centrifugation have been associated in some, but
not all, studies with lower HCV RNA levels [38–40]. However,
in this investigation, testing was on an aliquot not previously
thawed, and serum samples were centrifuged and frozen within
hours of collection.
Although the analytic sensitivity of the bDNA assay is lim-
849
ited, most individuals (88% of study participants) have HCV
RNA levels within the linear range of the assay, making it
optimal for a study of HCV RNA levels. Additional testing
will be necessary for participants with RNA levels less than the
bDNA linear range, to ascertain whether they have cleared
infection or have low RNA levels. Exclusion of participants
with HCV RNA levels below the level of detection of the assay
ensures that this study did not assess both determinants of HCV
RNA level and HCV RNA clearance, 2 distinct processes. On
the other hand, inclusion of participants with HCV RNA levels
above the level of detection should not substantially affect the
results, since median values were compared and the nonparametric statistical tests consider the rank order of levels rather
than their absolute value.
In this investigation, serum HCV RNA levels ranged from
!200,000 to 1120 million eq/mL. Although numerous virologic,
environmental, and sociodemographic variables were considered and significant correlates were identified, !10% of the overall interperson HCV RNA variability was explained. Little visitto-visit variance was noted in the subset of participants with
>2 HCV RNA assessments and, since the samples were all run
in duplicate, only a small proportion of the variance can be
attributed to the assay. Polymorphisms in the putative NS5A
ISDR have been variably associated with differences in HCV
RNA level in prior studies [41–43]. However, these sequences
are usually similar within HCV subtypes, and we found no
subtype differences in HCV RNA levels. In addition, among
carefully matched, HIV-negative participants within a given
subtype, no association was found between the serum level of
Figure 3. Hepatitis C virus (HCV) core-E1 genetic distances measured as Hamming distances (no. of changes/length of sequenced amplicon). Intervisit distances of 10 HCV 1a and 10 HCV 1b participants
who acknowledged ongoing drug use and needle sharing (mean 5
SD, 491 5 317 injections) between 2 visits mean of 6 months apart
(53 weeks). l, Intrasubject distances represent core-E1 sequences
from clones with different hypervariable region–1 sequences from previous study [20]. n, Intersubject distances are pairwise distances between 10 genotype 1a participants in this study.
850
Thomas et al.
HCV RNA and the NS5A amino acid sequence. Collectively,
these findings indicate that serum HCV RNA levels are chiefly
determined by factors not assessed in this investigation.
In summary, HIV infections, HBV infection, age, and highrisk drug use practices need to be considered in future evaluations of HCV RNA levels. Further research is needed to understand the mechanisms through which these factors affect
HCV RNA levels and to ascertain conditions not measured in
this study that chiefly determine HCV RNA level in blood.
17.
18.
19.
20.
References
21.
1. Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD. HIV-1
dynamics in vivo: virion clearance rate, infected cell life-span, and viral
generation time. Science 1996; 271:1582–6.
2. Katzenstein DA, Hammer SM, Hughes MD, et al. The relation of virologic
and immunologic markers to clinical outcomes after nucleoside therapy
in HIV-infected adults with 200 to 500 CD4 cells per cubic millimeter.
AIDS Clinical Trials Group Study 175 Virology Study Team. N Engl J
Med 1996; 335:1091–8.
3. Marschner IC, Collier AC, Coombs RW, et al. Use of changes in plasma
levels of human immunodeficiency virus type 1 RNA to assess the clinical
benefit of antiretroviral therapy. J Infect Dis 1998; 177:40–7.
4. Mellors JW, Rinaldo CRJ, Gupta P, White RM, Todd JA, Kingsley LA.
Prognosis in HIV-1 infection predicted by the quantity of virus in plasma.
Science 1996; 272:1167–70.
5. Vlahov D, Graham N, Hoover D, et al. Prognostic indicators for AIDS and
infectious disease death in HIV-infected injection drug users: plasma viral
load and CD41 cell count. JAMA 1998; 279:35–40.
6. Farzadegan H, Hoover DR, Astemborski J, et al. Sex differences in HIV-1
viral load and progression to AIDS. Lancet 1998; 352:1510-4.
7. Evans JS, Nims T, Cooley J, et al. Serum levels of virus burden in early stage
human immunodeficiency virus type 1 disease in women. J Infect Dis
1997; 175:795–800.
8. Neumann AU, Lam NP, Dahari H, et al. Hepatitis C viral dynamics in vivo
and the antiviral efficacy of interferon-a therapy. Science 1998; 282:103–7.
9. McHutchison JG, Gordon SC, Schiff ER, et al. Interferon alfa-2b alone or
in combination with ribavirin as initial treatment for chronic hepatitis C.
N Engl J Med 1998; 339:1485–92.
10. Gretch D, Corey L, Wilson J, et al. Assessment of hepatitis C virus RNA
levels by quantitative competitive RNA polymerase chain reaction: high
titer viremia correlates with advanced stage of disease. J Infect Dis 1994;
169:1219–25.
11. De Moliner L, Pontisso P, De Salvo GL, Cavalletto L, Chemello L, Alberti
A. Serum and liver HCV RNA levels in patients with chronic hepatitis
C: correlation with clinical and histological features. Gut 1998; 42:856–60.
12. Haydon GH, Jarvis LM, Blair CS, et al. Clinical significance of intrahepatic
hepatitis C virus levels in patients with chronic HCV infection. Gut
1998; 42:570–5.
13. Negro F, Krawczynski K, Quadri R, et al. Detection of genomic- and minusstrand of hepatitis C virus RNA in the liver of chronic hepatitis C patients
by strand-specific semiquantitative reverse transcriptase polymerase chain
reaction. Hepatology 1999; 29:536–42.
14. Vlahov D, Anthony JC, Munoz A, et al. The ALIVE study: a longitudinal
study of HIV-1 infection in intravenous drug users: description of methods
and characteristics of participants. NIDA Res Monogr 1991; 109:75–100.
15. Villano SA, Vlahov D, Nelson KE, Lyles CM, Cohn S, Thomas DL. Incidence and risk factors for hepatitis C among injection drug users in Baltimore, Maryland. J Clin Microbiol 1997; 35:3274–7.
16. Levine OS, Vlahov D, Koehler J, Cohn S, Spronk AM, Nelson KE. Sero-
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
JID 2000;181 (March)
epidemiology of hepatitis B virus in a population of injecting drug
users—association with drug injection patterns. Am J Epidemiol 1995;
142:331–41.
Lau JY, Davis GL, Kniffen J, et al. Significance of serum hepatitis C virus
RNA levels in chronic hepatitis C. Lancet 1993; 341:1501–4.
Davidson F, Simmonds P, Ferguson JC, et al. Survey of major genotypes
and subtypes of hepatitis C virus using RFLP of sequences amplified from
the 50 non-coding region. J Gen Virol 1995; 76:1197–204.
Villano SA, Vlahov D, Nelson KE, Cohn S, Thomas DL. Persistence of
viremia and the importance of long-term follow-up after acute hepatitis
C infection. Hepatology 1999; 29:908–14.
Ray SC, Wang YM, Laeyendecker O, Ticehurst J, Villano SA, Thomas DL.
Acute hepatitis C virus structural gene sequences as predictors of persistent
viremia: hypervariable region 1 as decoy. J Virol 1999; 73:2938–46.
Thomas DL, Shih JW, Alter HJ, et al. Effect of human immunodeficiency
virus on hepatitis C virus infection among injecting drug users. J Infect
Dis 1996; 174:690–5.
Eyster ME, Fried MW, Di Bisceglie AM, Goedert JJ. Increasing hepatitis C
virus RNA levels in hemophiliacs: relationship to human immunodeficiency virus infection and liver disease. Blood 1994; 84:1020–3.
Sherman KE, O’Brien J, Gutierrez G, et al. Quantitative evaluation of hepatitis C virus RNA in patients with concurrent human immunodeficiency
virus infections. J Clin Microbiol 1993; 31:2679–82.
Telfer PT, Brown D, Devereux H, Lee CA, Dusheiko GM. HCV RNA levels
and HIV infection: evidence for a viral interaction in haemophilic patients.
Br J Haematol 1994; 88:397–9.
Cribier B, Rey D, Schmitt C, Lang JM, Kirn A, Stoll-Keller F. High hepatitis
C viraemia and impaired antibody response in patients coinfected with
HIV. AIDS 1995; 9:1131–6.
Thomas DL, Vlahov D, Solomon L, et al. Correlates of hepatitis C virus
infections among injection drug users in Baltimore. Medicine 1995; 74:
212–20.
Hagiwara H, Hayashi N, Mita E, et al. Quantitation of hepatitis C virus
RNA in serum of asymptomatic blood donors and patients with type C
chronic liver disease. Hepatology 1993; 17:545–50.
Koike K, Yasuda K, Yotsuyanagi H, et al. Dominant replication of either
virus in dual infection with hepatitis viruses B and C. J Med Virol 1995;45:
236–9.
Zarski JP, Bohn B, Bastie A, et al. Characteristics of patients with dual
infection by hepatitis B and C viruses. J Hepatol 1998; 28:27–33.
Pontisso P, Gerotto M, Ruvoletto MG, et al. Hepatitis C genotypes in patients
with dual hepatitis B and C virus infection. J Med Virol 1996; 48:157–60.
Chu CM, Yeh CT, Liaw YF. Low-level viremia and intracellular expression
of hepatitis B surface antigen (HBsAg) in HBsAg carriers with concurrent
hepatitis C virus infection. J Clin Microbiol 1998; 36:2084–6.
Pontisso P, Ruvoletto MG, Fattovich G, et al. Clinical and virological profiles
in patients with multiple hepatitis virus infections. Gastroenterology
1993; 105:1529–33.
Jarvis LM, Watson HG, McOmish F, Peutherer JF, Ludlam CA, Simmonds
P. Frequent reinfection and reactivation of hepatitis C virus genotypes in
multitransfused hemophiliacs. J Infect Dis 1994; 170:1018–22.
Farci P, Alter HJ, Govindarajan S, et al. Lack of protective immunity against
reinfection with hepatitis C virus. Science 1992; 258:135–40.
Oshita M, Hayashi N, Kasahara A, et al. Increased serum hepatitis C virus
RNA levels among alcoholic patients with chronic hepatitis C. Hepatology
1994; 20:1115–20.
McGuinness PH, Bishop GA, Painter DM, Chan R, McCaughan GW. Intrahepatic hepatitis C RNA levels do not correlate with degree of liver
injury in patients with chronic hepatitis C. Hepatology 1996; 23:676–87.
Pessione F, Degos F, Marcellin P, et al. Effect of alcohol consumption on
serum hepatitis C virus RNA and histological lesions in chronic hepatitis
C. Hepatology 1998; 27:1717–22.
Gordon SC, Dailey PJ, Silverman AL, Khan BA, Kodali VP, Wilber JC.
JID 2000;181 (March)
Evaluation of Hepatitis C RNA
Sequential serum hepatitis C viral RNA levels longitudinally assessed by
branched DNA signal amplification. Hepatology 1998; 28:1702–6.
39. Davis GL, Lau JYN, Urdea MS, et al. Quantitative detection of hepatitis C
virus RNA with a solid-phase signal amplification method: definition of
optimal conditions for specimen collection and clinical application in interferon-treated patients. Hepatology 1994; 19:1337–41.
40. Miskovsky EP, Carella AV, Gutekunst K, Sun C, Quinn TC, Thomas DL.
Clinical characterization of a competitive PCR assay for quantitative testing of hepatitis C virus. J Clin Microbiol 1996; 34:1975–9.
41. Chayama K, Tsubota A, Kobayashi M, et al. Pretreatment virus load and
851
multiple amino acid substitutions in the interferon sensitivity–determining
region predict the outcome of interferon treatment in patients with chronic
genotype 1b hepatitis C virus infection. Hepatology 1997; 25:745–9.
42. Enomoto N, Sakuma I, Asahina Y, et al. Mutations in the nonstructural
protein 5A gene and response to interferon in patients with chronic hepatitis C virus 1b infection. N Engl J Med 1996; 334:77–81.
43. Hashimoto M, Chayama K, Kobayashi M, et al. Fluctuations of hepatitis
C virus load are not related to amino acid substitutions in hypervariable
region 1 and interferon sensitivity determining region. J Med Virol 1999;
58:247–55.