| |
Hepatitis C virus in the semen of men coinfected with HIV-1: prevalence and origin
|
| |
| |
AIDS: Volume 19(16) 4 November 2005
Note from Jules Levin: whether HCV is transmissible sexually is controversial. Some researchers do not believe HCV is transmitted by the exchange of semen, even if HCV is found in semen. Recent observations out of France & the UK report increased incidence of acute HCV among MSM in HIV clinics. The authors of these reports suggest risky sexual behavior that might draw blood (HCV is transmitted by blood-to-blood contact) or the presence of STDS (STD presence may facilitate HCV transmission) may be responsible. However, these recent reports out of Europe ought to spur research here in the US to evaluate this question more closely. The authors of this study suggest that the higher levels of HCV in the blood of coinfected patients compared to HCV monoinfected may be the reason for HCV RNA found more often in semen of coinfected & this might be at least in part responsible for the increases in acute HCV among MSM in their hospital clinics. Here is study published in the March 2005 issue of the American Journal of Public Health: "Lack of Evidence of Sexual Transmission of Hepatitis C Virus in a Prospective Cohort Study of Men Who Have Sex With Men". This study was conducted 5 years ago in Montreal: The Omega Cohort Study is an ongoing, open prospective cohort in which risk factors associated with the transmission of HIV among MSM in Montreal are being evaluated (recruitment began in October 1996). They found a prevalence of only 2.9% & the cases were strongly associated with injection drug use. This finding is in contrast to reports from HIV clinics in the UK & France where acute HCV is being found among MSM & the researchers report an association with unsafe sex & perhaps the presence of STDs. Perhaps circumstances are different. The presence of STDs could be the difference in findings between 2001 in Montreal and current circumstances in the UK & France. Perhaps syphilis & other STDs are more prevalent, perhaps sexual activities are different, and perhaps HCV is more prevalent now. A full report of the AJPH article will follow. The other possibility is that the presence of HIV increases the risk for HCV sexual transmission, particularly if HCV viral loads are higher, which they often are in HIV+ individuals. This in conjunction with a higher prevalence of STDs may be responsible for changing circumstances & increased risk for sexual transmission of HCV among MSM.
"......In this study, HCV RNA was more frequently found in the semen of HIV-1/HCV-coinfected men (37.8%) than in the semen of non-coinfected men (18.4%) (P = 0.033). These interesting data were not reported for previous series containing a smaller number of subjects.....
....HCV/HIV-1-coinfected men with HCV-positive semen had significantly higher HCV blood load than men with HCV-negative semen (P = 0.038). As HIV-1/HCV-coinfected men also had higher HCV RNA blood levels than non-coinfected men (P = 0.017), this may partly explain why the coinfected men had a higher prevalence of HCV in semen than non-coinfected men....
.....should reinforce the current recommendations concerning protected sex in HIV-infected individuals...."
Authors: Briat, Aureliea; Dulioust, Emmanuelb; Galimand, Juliea; Fontaine, Helenec; Chaix, Marie-Laurea; Letur-Konirsch, Helened; Pol, Stanislasc; Jouannet, Pierreb; Rouzioux, Christinea; Leruez-Ville, Mariannea
From the aVirology Laboratory
bLaboratory of Reproductive Biology, Universite Rene Descartes
cHepatology Service
dGenetics Service (CECOS), Hospital Necker-Enfants Malades, Paris, France.
ABSTRACT
Objective: To compare the prevalence of hepatitis C (HCV) RNA in semen from men infected with HCV and those coinfected with HIV-1/HCV and to study the origin of HCV shed in semen.
Design: Two prospective studies (HC EP09 and BINECO) included 120 HCV-positive men, 82 coinfected with HIV-1; all had positive HCV RNA detection in blood.
Methods: Paired blood and semen samples were collected for HCV RNA detection and quantification in seminal plasma and in blood serum; repeated semen samples were obtained for 45 men. HCV RNA was sought in spermatozoa and non-sperm cells. Phylogenetic analysis of the HVR-1 region of HCV compared the quasispecies in blood serum and seminal plasma of two men.
Results:
- HCV RNA was more frequently found in the semen of men coinfected with HIV-1 (37.8%) than in those with only HCV infection (18.4%) (P = 0.033).
- HCV RNA detection in semen was intermittent and was positive in at least one semen sample of 42.8% of HIV-1/HCV-coinfected men who provided repeated samples.
- Men with HCV-positive semen had significantly higher HCV load in blood than men with HCV-negative semen (P = 0.038).
- Phylogenetic comparison of HCV quasispecies in blood and in semen showed no evidence of HCV replication in genital leukocytes; however, a phenetic structure was observed between compartments (P < 0.001).
Conclusions:
HCV particles in semen originate from passive passage from blood, with preferential transfer of some variants.
Nearly half of HIV-1/HCV-coinfected men may intermittently harbour HCV in their semen.
The authors say: Recommendations of protected sex for HIV-infected individuals should be reinforced.
BACKGROUND
Introduction
Sexual transmission of hepatitis C virus (HCV) is a controversial issue. There have been a number of case reports containing strong evidence for HCV sexual transmission [1,2]. However, the importance of sexual transmission in the spread of HCV infection is not known. Many studies indicate that this route of transmission may be minimal in stable heterosexual non-HIV-1-coinfected partners [3]. However, recent data suggest an increasing incidence of acute HCV infections in HIV-1-infected men who have sex with men [4-8]. Unprotected sexual intercourse is the only risk factor associated in all these reported cases, and a concomitant sexually transmitted disease such as syphilis or gonorrhoea was documented in a large number of these men. This supports the hypothesis of sexual transmission as a cause of this recent increase in the incidence of HCV acute infections in HIV-1-infected men and emphasizes the interest in studies focusing on HCV in the genital tract of HIV-1-infected men.
Although the presence of HCV in seminal fluid has been a controversial issue [9-13], it is now generally agreed that HCV RNA is recovered from the semen of 10-30% of HCV-infected men [14-17]. Men with HCV-positive semen have higher blood loads of HCV than men who do not shed virus in their semen [17].
Patients
Two prospective studies between January 2002 and July 2004 comprised 120 HCV-infected viraemic men, 82 (68.3%) of whom were coinfected with HIV-1. The first study (HC EP09) was supported by the Agence Nationale de la Recherche sur le SIDA and designed to assess the frequency and the origin of HCV shedding in the semen of HCV viraemic men whatever their HIV status. The second study (BINECO) was a clinical trial of medically assisted procreation for HIV-1-infected men wishing to have a child with their non-infected partner and was supported by the Delegation ˆ la Recherche Clinique de l'AP-HP. The two studies were conducted at Cochin and Necker Hospitals in Paris, France and the ethics committee of Cochin Hospital approved both studies.
None of the 120 men was treated for HCV infection at inclusion. Among the 82 HIV-1/HCV-coinfected men, seven were not treated for HIV infection and 75 received HAART. Median CD4 cells count was 524 cells/ml (range, 118-1395) and median HIV RNA load was < 50 copies/ml (range, < 50 to 158 000).
RESULTS
Hepatitis C virus RNA in seminal samples
In the 191 seminal plasma samples studied, 51 were positive for HCV RNA (26.7%). Among the 120 men in the study, 38 (31.6%) had at least one seminal plasma sample with positive HCV RNA detection. HCV RNA in seminal plasma was more frequently detected in HIV-1/HCV-coinfected men (31/82, 37.8%) than in non-coinfected men (7/38, 18.4%) (m2 test; P = 0.033).
Among the 82 HIV-1/HCV-coinfected men, there were no differences between men with detectable HCV RNA in semen and those with no detectable RNA for median CD4 cell count [556 cells/ml (range, 232-1031) and 465 cells/ml (range, 118-1395); P = 0.24] or for median blood HIV RNA [< 50 copies/ml (range, < 50 to 158 000] and < 50 (range, < 50 to 76 600); P = 0.91].
Among the 45 men for whom repeated semen samples were collected, HCV RNA was detected in at least one sample for 18 men, with intermittent detection in 13. Men who had repeated samples were more likely to have at least one positive HCV RNA detection (18/45, 40.0%) than men who only gave one semen sample (20/75, 26.6%), but this difference was not significant (m2 test, P = 0.12). Finally, 15 of the 35 (42.8%) HIV-1/HCV-coinfected men who had repeated semen samples had detectable HCV RNA at least once in their seminal plasma.
Among the 51 seminal plasma samples positive for HCV RNA, 42 were available for quantitative PCR and seven samples were insufficient. As the sensitivity of the quantitative PCR was lower (200 IU/ml) than that of the qualitative PCR (50 IU/ml), only 6 of the 42 samples were positive with the quantitative PCR, with a median viral load of 200 IU/ml (range, 200-515).
HCV RNA was not detected in the 51 spermatozoa fractions obtained from the samples that exhibited a positive result for seminal plasma HCV RNA.
Non-spermatic cells were not kept for HIV-positive men included in the BINECO protocol but HCV RNA was assayed in the 39 samples of non-sperm cells available from 32 of the 38 men only infected with HCV. Two non-sperm cell fractions obtained from two men were positive for HCV RNA (6.2%). These two men had a positive HCV RNA result in the seminal plasma of the same semen sample.
Correlation between hepatitis C virus RNA in blood and seminal plasma:
HCV blood PCR was positive for all 120 men in the study. However, actual HCV RNA levels were measured only for the 89 men for whom serum samples were kept in our virology laboratory.
Median HCV blood load was significantly higher in HIV-1/HCV-coinfected men than in non-coinfected ones: 5.93 log10 IU/ml (range, 2.81-7.47) versus 5.68 log10 IU/ml (range, 4.03-6.66) (Mann-Whitney test, P = 0.017).
Median HCV blood load was higher in men who had at least one semen sample positive for HCV RNA than in men who were negative. This difference was significant in the group of HIV-1/HCV-coinfected men, 6.22 log10 IU/ml (range, 5.41-7.47) and 5.95 log10 IU/ml (range, 2.81-6.59), respectively (Fig. 1a; Mann-Whitney test, P = 0.038), but was not significant in non-coinfected men, 6.13 log10 IU/ml (range, 5.36-6.90) and 5.95 log10 IU/ml (range, 4.08-6.66), respectively (Mann-Whitney test, P = 0.14).
No HCV RNA was detected in the seminal plasma of men who had a HCV RNA serum blood load < 230 000 IU/ml (5.36 log10 IU/ml).
Hepatitis C genotype and the presence of hepatitis C RNA in seminal plasma:
HCV genotypes were obtained in 87 men (59 HIV-1/HCV-coinfected men and in 28 non-coinfected men). HCV RNA was detected at least once in the seminal plasma in 15 of 54 (28%) men infected with genotype 1, in 2 of 5 (40%) men infected with genotype 2, in 5 of 15 (33%) men infected with genotype 3 and in 6 of 13 men (46%) infected with genotype 4. The prevalence of HCV RNA in semen was not significantly different among the four genotypes (m2 test, P > 0.05).
Hepatitis C quasispecies in blood and seminal compartments:
Quasispecies were analysed for paired semen and blood samples from two men. Patient A was HIV-1 negative and infected with HCV genotype 1a; patient B was HIV-1/HCV positive and infected with HCV genotype 1b.
A total of 44 and 40 clones obtained by direct cloning were sequenced for patients A and B, respectively; 32 clones obtained after limiting-dilution cloning were sequenced for patient B.
The mean genetic distance within a compartment was higher in the blood than in the semen for patients A and B after direct cloning. However, mean genetic distance was higher in semen than in blood after limiting-dilution cloning for patient B.
Phylogenetic trees constructed from the viral sequences of patient A showed that, even if semen sequences clustered together, blood-derived sequences were found among them. In patient B, after direct cloning there was an obvious phylogenetic grouping by compartment origin, with bootstrap values > 70%. However, after limiting-dilution cloning such significant phylogenetic clustering was not confirmed; semen-derived sequences clustered among blood-derived sequences (bootstrap value < 70%).
However, sequences within a compartment shared significantly more genetic identity than sequences between compartments (Mantel test, P < 0.001 in both cases after direct cloning and P = 0.031 after limiting-dilution cloning).
Discussion
In this study, HCV RNA was more frequently found in the semen of HIV-1/HCV-coinfected men (37.8%) than in the semen of non-coinfected men (18.4%) (P = 0.033). These interesting data were not reported for previous series containing a smaller number of subjects [14,15]. Detection of HCV RNA in semen was intermittent. This may be explained by the low HCV RNA concentrations observed in this fluid, close to the detection threshold [14,17]. However, a true intermittent excretion of HCV in semen is also plausible, since semen composition may exhibit high variation from one ejaculate to another in the same individual [28,29]. Finally, in our study, 42.8% of HIV-1/HCV-coinfected men who provided repeated semen specimens over time had at least one HCV-positive semen sample. Recent reports suggest an unequivocal increase of HCV sexual transmission among HIV-infected men who have sex with men [5-8]. Increased risk of sexual transmission in this group may be a result of exposure to viral particles originating from the seminal fluid itself or blood, if there is traumatic sexual intercourse, or both. The high prevalence of HCV in the semen of coinfected men may, therefore, partly explain the increased risk of sexual transmission in HIV-1-infected men who have sex with men.
HCV/HIV-1-coinfected men with HCV-positive semen had significantly higher HCV blood load than men with HCV-negative semen (P = 0.038). As HIV-1/HCV-coinfected men also had higher HCV RNA blood levels than non-coinfected men (P = 0.017), this may partly explain why the coinfected men had a higher prevalence of HCV in semen than non-coinfected men. There was no correlation between HCV genotypes (1 to 4) and the presence of HCV RNA in seminal plasma in the two groups.
Correlation between blood viral load and the presence of HCV in semen supports a passive transfer of HCV particles from blood serum to genital tract [17]. However, in our study, some men with low blood viral load had detectable HCV RNA in their semen and conversely some men with high blood viral load had no detectable HCV RNA in semen. Such discrepancies could be explained either by local replication in male genital tract leukocytes, as reported in blood lymphocytes and monocytes [21-23,30] or by a selective passage of some HCV variants, which might exhibit a preferential tropism for genital tract.
An examination of the non-sperm cell fraction, which contains leukocytes, in two men detected HCV RNA (6.2%). This HCV RNA could originate from the seminal fluid itself, if the separation gradient and the subsequent non-sperm cell pellet washing failed to eliminate completely all residual HCV RNA molecules, or from local replication in genital leukocytes. To test this latter hypothesis, we compared HCV quasispecies partition between blood and genital compartments for two men. Phylogenetic analysis showed no evidence of HCV local replication in the male genital tract: all quasispecies detected in semen were represented in blood. For patient B, there was a significant phylogenetic grouping of clones according to compartmental origin, as demonstrated by bootstrap values > 70% when the direct cloning method was used, but these results were not confirmed in limiting-dilution PCR cloning. We explain the discrepancy between the two cloning methods by the higher sensitivity of the limiting-dilution technique. Direct cloning followed by sequencing can identify variants when they represent as little as 5% of the virus mixture but limiting-dilution PCR cloning is even more sensitive as it can identify variants in concentrations as low as 0.1% [31]. We concluded that results obtained with the direct cloning method were misleading because this method is not sensitive enough to detect a wide range of variants in samples where the template RNA level is low. This emphasizes the importance of using appropriately sensitive methods to study HCV genetic diversity in samples such as seminal plasma.
Despite the lack of evidence for HCV replication in the male genital tract, the quasispecies sequences within each compartment shared significantly more genetic identity than was seen between compartments in both patients (P < 0.001). This significant difference could be artifactual arising from errors introduced during the PCR reaction, but this is unlikely because the enzyme used had a proofreading capacity that reduced its error rate [32] and a statistical significance was observed whatever cloning method was used. Therefore, the phenetic structure observed between blood and semen compartments raises the possibility that some quasispecies variants may pass preferentially into the male genital tract. However, the fact that no HCV genotype was more frequently associated with the presence of HCV RNA in seminal plasma suggests that this preferential transfer may only exist at the quasispecies level. This hypothesis needs to be confirmed in further phylogenetic studies of HCV quasispecies in semen.
In summary, our data provide evidence that nearly half of HIV/HCV-coinfected men may intermittently harbour HCV in their semen. Furthermore, we report here original observations suggesting that the presence of HCV particles in semen originate via direct transfer from blood plasma to seminal plasma with no evidence of HCV replication in local leukocytes. However, preferential transfer of some quasispecies could be a plausible hypothesis to explain why some men with high plasma viral load do not have detectable HCV RNA in semen and why some with low blood load do have detectable virus.
Our results partially illustrate how the recent increase in sexually transmitted HCV infection in HIV-infected homosexual men [5-8] could arise and should reinforce the current recommendations concerning protected sex in HIV-infected individuals.
|
|
| |
| |
|
|
|
