Journal of Medical Virology 78:877–882 (2006)
Multicenter Quality Control of the Detection of
HIV-1 Genome in Semen Before Medically
Christophe Pasquier,1* Deborah Anderson,2Corinne Andreutti-Zaugg,3
Rianne Baume-Berkenbosch,3Florence Damond,4Aviva Devaux,4Yvon Englert,9
Julie Galimand,8Carole Gilling-Smith,6Odile Guist’hau,7Lital Hollander,5
Marianne Leruez-Ville,8Benoit Lesage,9Anne Maillard,7Anne-genevie `ve Marcelin,10
Marie-Paule Schmitt,11Augusto Semprini,5Maria Vourliotis,6Chong Xu,2
Louis Bujan,1and The CREAThE Network
1Laboratoire de Virologie, laboratoire de Spermiologie et CECOS Midi-Pyre ´ne ´es,
Toulouse University Hospital, Toulouse, France
2Brigham and Women’s Hospital, Boston, MA
3La Source Hospital, Lausanne, Switzerland
4Service de Virologie et Service d’Histologie, Ho ˆpital Bichat Claude Bernard, Paris, France
5ESMAN Medical Consulting Srl, Milano, Italy
6Assisted Conception Unit, Chelsea & Westminster Hospital, London, UK
7Laboratoire de virologie, Centre Hospitalier Universitaire, Pontchaillou, Rennes, France
8Laboratoire de Virologie, Ho ˆpital Necker, Paris, France
9Research laboratory on Human Reproduction, ERASME-ULB hospital, Brussels, Belgium
10Laboratoire de Virologie CERVI, Ho ˆpital Pitie ´-salpe ˆtrie `re, Paris, France
11Laboratoire de Virologie, Faculte ´ de Me ´decine-ULP, Strasbourg, France
Couples in whom the man is HIV-1-positive
may use medically assisted procreation in order
to conceive a child without contaminating the
female partner. But, before medically assisted
procreation, the semen has to be processed to
and after processing. The performance was
evaluated of the technical protocols used to
detect and quantify HIV-1 in 11 centers providing
medically assisted procreation for couples with
HIV-1 infected men by testing panels of seminal
plasma and cells containing HIV-1 RNA and/or
DNA. The performance of these tests varied due
for RNA detection in seminal plasma was 500–
1,000 RNA copies/ml, over 500 RNA copies/106
cells in semen cells, and for DNA detection in
of silica-based extraction seemed to increase the
assay performance, whereas the use of internal
controls to detect PCR inhibitor did not. This first
quality control highlights the need for technical
improvements of the assays to detect and
quantify HIV in semen fractions and for regular
evaluation of their performance. J. Med. Virol.
? 2006 Wiley-Liss, Inc.
Advances in antiretroviral therapy have increased the
life expectancy and quality of life of individuals that are
infected with HIV-1 [Al-Khan et al., 2003; Semprini and
Fiore, 2004]. People living with HIV are likely to want to
have children as their clinical condition improves.
Discordant couples must always use condoms during
intercourse since HIV-1 is primarily transmitted sexu-
ally. They are therefore unable to conceive on their own.
Pioneering experiments carried out in 1992 [Semprini,
1993] led to the development of methods of medically
assisted procreation and these are now available to HIV-
when the male partner is infected [Bujan et al., 2004b].
*Correspondence to: Christophe Pasquier, MD, PhD, Labor-
atoire de virologie, variabilite ´ virale, EA2046-IFR30, place du Dr.
Baylac, Ho ˆpital Purpan, TSA 40031, F-31 059 Toulouse Cedex,
France. E-mail: email@example.com
Accepted 18 January 2006
Published online in Wiley InterScience
? 2006 WILEY-LISS, INC.
Fertility centers perform ‘‘sperm washing’’ protocols
before medically assisted procreation, since the excre-
tion of HIV-1 into the semen is not predicted reliably by
the blood virus load and is not necessarily prevented by
potent antiretroviral treatment [Tachet et al., 1999;
Pasquier et al., 2000; Bujan et al., 2004a]. Sperm
washing removes cell-free plasma that can contain free
HIV particles and seminal round cells (mostly white
blood cells) that can be infected with HIV [Quayle et al.,
1997]. It usually involves density gradient centrifuga-
tion to isolate spermatozoa and may be followed by a
centers must ensure that their spermatozoa isolation
method is efficient and check that the processed
spermatozoa are HIV-free before commencing medi-
cally assisted procreation. This is done by testing the
processed spermatozoa or other fractions obtained
during processing for HIV-1 RNA or DNA. The virolo-
tests or modify commercial tests for use on semen since
no commercial assay is available for detection of HIV-1
in seminal plasma or semen cells. These assays have
been validated in each laboratory and are currently in
routine use in several centers, particularly in Europe.
The assays detect and usually quantify HIV-1 RNA in
seminal plasma and detect HIV-1 DNA or both HIV-1
DNA and RNA in semen cells. The spermatozoa can be
used for medically assisted procreation, by techniques
such as intrauterine insemination (IUI), in vitro ferti-
obtained after sperm washing.
This study was undertaken to compare the perfor-
HIV-1in centers involvedin medically assistedprocrea-
tion for couples with HIV-infected men.
MATERIALS AND METHODS
Eleven laboratories took part in the study. Ten were
from European countries, France (6), Switzerland (1),
Belgium (1), United Kingdom (1), and Italy (1) and one
from the United States of America. All the centers,
except the center from the United States of America,
were clinical laboratories. They tested routinely semen
for HIV-1 and were involved in medically assisted
procreation programs for HIV discordant couples. The
center numbers in the results section do not correspond
to the affiliation numbers of the authors.
The sample panels were prepared in the laboratories
of spermiology and virology of the coordinating center
(Toulouse University Hospital, France). On the same
day, nine informed HIV-negative voluntary men pro-
samples were processed in the laboratory, as done
routinely, within 2 hr of ejaculation [Pasquier et al.,
2000; Bujan et al., 2004a]. The semen samples were
pooled toinsureasufficientseminal plasmavolume and
cell count. Seminal plasma and whole sperm cells were
isolated by centrifugation at 11,000g and spermatozoa
were prepared from whole semen by density gradient
centrifugation. The processed seminal plasma (450 ml
aliquots) and harvested spermatozoa (5?106/vial) were
The blood plasma from an HIV-1-infected subject was
quantified (mean of two measurements) by RT-PCR
(COBAS Amplicor HIV-1 Monitor v1.5; Roche Diagnos-
tics, Meylan, France) and stored at ?808C. The quanti-
fied blood plasma samples were diluted in human HIV-
negative plasma to obtain adequate concentrations of
HIV-1 RNA (copies/ml) in 50 ml. The seminal plasma
panel (panel A) was prepared by adding 50 ml aliquots to
450 ml seminal plasma to obtain final concentrations of
HIV RNA of 1,000, 500, 100, and 50 copies/ml. The
semencellRNA panel(panelB) wasprepared byadding
50 ml aliquots to 5?106spermatozoa to obtain final HIV
RNA concentrations of 500, 50, 10, and 5 copies/106
spermatozoa. HIV-negative human plasma (50 ml) was
used to prepare negative samples in both panels.
The semen cell DNA panel (panel C) was prepared by
cell to 5?106semen cells to obtain final concentrations
of HIV DNA of 500, 50, 10, and 5 copies/106semen cells.
All samples were frozen and stored at ?808C. They
were shipped to the participating laboratories in dry ice
and complied with the relevant legislation. They were
received frozenwithin 48hr. Thesesamples weretested
by each laboratory in the sameway as semensamples of
HIV-infected men. No specific techniques or algorithms
Characteristics of Assays Used to Detect and
of their tests to the coordinating center. All 11
laboratories tested the seminal plasma for HIV-1 RNA,
RNA and HIV-1 DNA respectively.
seminal plasma samples, and one performed a qualita-
tive test. Six laboratories used an adapted HIV-1
Cobas1Monitor1assay (centers 1–6), one used an
adapted NASBA1assay (center 7) and 4 used custom-
developed techniques with real-time PCR (centers 8–9)
or conventional PCR with quantification by hybridizing
(centers 10–11) (Table I). The assays differed in the
volume of sample extracted, the extraction technique,
an internal control to detect PCR inhibitors (centers
1–8), the use of reverse transcription before amplifi-
cation or in the amplification tube, the region of the
HIV-1 genome amplified, and the quantification system
The techniques used to detect HIV-1 RNA in semen
cells were all very similar to those used for seminal
plasma. All the assays detected HIV-1 RNA and DNA in
J. Med. Virol. DOI 10.1002/jmv
878Pasquier et al.
the absence of DNAse treatment. Two laboratories used
the HIV-1 Cobas1Monitor1to detect HIV-1 DNA in
semen cells (centers 4–6), one used the HIV-1 Ampli-
cor1assay (center 3), and five used a home-made assay
(two real-time PCR (centers 4–8–9) and two conven-
tional PCR (centers 10–11)) (Table II).
The Chi-square test was used to compare detection
frequencies for the various assays.
HIV-1 RNA Detection in Seminal Plasma
Samples containing no HIV RNA were all negative
except forthree samples
(Table III). They were positive with fewer than 300
RNA copies/ml. The overall specificity was 86%. HIV-1
RNA was detected in 6/11 (55%) samples containing
50 copies/ml, 7/11 (64%) samples containing 100 copies/
ml, 7/9 (78%) samples containing 500 copies/ml, and in
11/11 (100%) samples containing 1,000 copies/ml. The
overall sensitivity (95% detection) was estimated to be
500–1,000 RNA copies/ml. Quantifications of the 1,000
for three that were under-evaluated (88, 134, and
255 copies/ml). Only one laboratory using an internal
control detected a PCR inhibitor in the 500 RNA copies
sample of this panel. There were no significant differ-
ences between the detection frequencies of positive
samples of assays with (22/30, 73%) or without (9/12,
75%) internal control checking for PCR inhibitors. The
Monitor1assays with adapted silica based extraction
gave 20/24 (83%), correct results for positive and
extraction gave 8/12 (67%) correct results, the NASBA1
12 (58%) correct results and the conventional RT-PCR
assays gave 11/11 (100%) correct results. The detection
frequencies of the adapted Monitor1assays using silica
based extraction and those not using it were similar
Detection and Quantification of HIV-1 RNA in
All the tests on samples containing no HIV RNA were
negative except for three positive results for samples
from three laboratories, two gave fewer than 60
RNA copies/106cells, for an overall specificity of 82%
(Table IV). HIV-1 RNA was detected in 5/9 (55%) of the
samples containing 5 HIV-1 RNA copies/106cells, in 6/8
106cells and in 7/8 (88%) of those containing 500 copies/
106cells. The overall sensitivity (95% detection) was
estimated at over 500 RNA copies/106cells. Results of
quantitation of the 500 copies/106semen cells samples
were within the 3-fold range, except for three that were
over-estimated (2,580, 2,060, and 12,000 copies/106
cells). PCR inhibitors were detected in three samples
from this panel; all by the same center (center 6).
There were no significant differences between detection
J. Med. Virol. DOI 10.1002/jmv
TABLE I. Main Characteristics of Assays Used to Quantify HIV-1 RNA in Seminal Plasma and Cells
Volume tested (ml)
Nuclisens Qiagen High pure
Hybridation Hybridation Hybridation Hybridation Hybridation Hybridation Hybrida-
Taqman Hybridation Agarose
Estimated sensitivity (c/ml)
na, not applicable.
*Sensitivity estimated by each laboratory using its own protocol.
Detection of HIV-1 Genome in Semen 879
frequencies for positive samples by assays with (16/22,
73%) and without (8/12, 67%) internal control checking
for PCR inhibitors. The Monitor1assays with adapted
extraction gave correct results for 15/18 (83%) positive
and negative samples, the Monitor1assays without
adapted extraction gave correct results for 8/12 (67%)
samples, the NASBA1for 4/6 (67%) samples, the real-
time PCR for 2/6 (33%) samples, and conventional RT-
PCR for 11/12 (92%) samples.
HIV-1 DNA Detection in Semen Cells
All tests on samples containing no HIV DNA were
negative, except for three (that were positive), giving an
overall specificity of 81%. Tests were positive for 1/8
(12.5%) samples containing 5 HIV-1 DNA copies/106
cells, 1/8 (12.5%) samples containing 10 HIV-1 DNA
copies/106cells, 7/8 (78%) samples containing 50 HIV-1
ing 500 HIV-1 DNA copies/106cells. The overall
sensitivity (95% detection) was estimated at 50–
500 DNA copies/106cells. Quantification of the 500
copies/106semen cells samples were within the 3-fold
detection frequencies of assays with (9/20, 45%) and
without (8/12, 67%) internal control checking for PCR
inhibitors were similar (P¼0.23). The Monitor1assays
gave correct results for 9/18 (50%) positive and negative
samples, the Amplicor1assays gave correct results for
4/6 (67%), the real-time PCR for 9/12 (75%), and the
conventional RT-PCR for 8/12 (67%) of samples
used to check the semen of HIV-infected men before
medically assisted procreation. They are used to assess
safety before processing and/or to validate the sperm
washing process. These tests on semen are required
from HIV infected men in some countries, including
France. We prepared three panels of samples to mimic
observed or plausible situations. Panel A was designed to
B to detect HIV particles that may remain adsorbed onto
spermatozoa and panel C to detect HIV-infected cells
remaining among processed spermatozoa. Panel C could
have been used to detect RNA, but would not have
permitted evaluation of assay sensitivity because the
number of HIV RNA copies/infected cell was unknown.
The technical protocols used to detect and/or quantify
HIV-1 in semen varied considerably and were all
different. This was due mainly to the way in which the
assays were developed from assays used to detect/
quantify HIV-1 in blood, the availability of techniques
and the laboratory expertise. Home-made tests seemed
to be less practical for routine use than commercially
available assay adapted to semen. Previous versions of
commercial kits have already been tested on seminal
plasma [Dyer et al., 1996; Shepard et al., 2000; Dunne
J. Med. Virol. DOI 10.1002/jmv
TABLE II. Main Characteristics of Assays Used to Quantify HIV-1 DNA in Seminal Cells
Volume tested (ml) 1?104cells/ml
Lysis proteinase K
Estimated sensitivity* (c/106cells)
*Sensitivity estimated by each laboratory using its own protocol.
880Pasquier et al.
et al., 2000]; they showed the presence of some false
positive and negative results.
The assays tested in the present study also produced
particularly when the PCR assay used has a very low
et al., 2000]. They may explain, at least partly, detection
observed when testing highly purified motile spermato-
zoa obtained using both density gradient and swim-up
results can increase control tests, the number of semen
samples required, cost and the complexity of medically
assisted procreation management.
By contrast, false negative results can lead to safety
problems. The use of internal controls can help to detect
any PCR inhibitors in seminal plasma. They are not
included systematically in home-made techniques that
rely only on the efficacy of extraction to avoid potential
inhibitors. The results of center 9 for the RNA panels
may thus be due to PCR inhibitors or insufficient
sensitivity. Extraction technique using silica or resin
PCR inhibitors were detected by only one center in this
study. This center was also one of the centers not using
silica for RNA extraction, although this gives better
overall results, as shown for HCV detection in semen
[Bourlet et al., 2003]. All the assays assessed on semen
had similar sensitivities to those obtained with blood
plasma or cells.
in the prepared spermatozoa, this will always be
restricted by the limit of detection of the assay used
Studies on semen spiked with HIV-1 have shown that
HIV cannot be detected in spermatozoa fractions when
the semen contain less than 106HIV RNA copies before
processing, and thus by whatever the sperm washing
technique used [Politch et al., 2004]. This artificial HIV
load is much higher than those usually found in the
semen (5,000–15,000 copies/ml) of untreated men
[Vernazza et al., 1997]. These investigators found no
HIV genome in the semen of HIV-infected men after its
washing by density gradient centrifugation and swim-
high [Kim et al., 1999; Bujan et al., 2002; Bujan et al.,
2004a]. The processing used in this trial thuseffectively
removed HIV. HIV genome was detected rarely in the
prepared spermatozoa, although it is impossible to
guarantee that all HIV particles have been removed
because of the detection threshold. Ultra-sensitive PCR
could be useful [Meseguer et al., 2002], but this
technique is time consuming and needs large volumes
of sample, which is often incompatible with the subse-
in France if the HIV virus load of the seminal plasma is
greaten then 10,000 RNA copies/ml. This threshold was
determined using sperm washing by density gradient
J. Med. Virol. DOI 10.1002/jmv
TABLE III. Results of HIV-1 RNA Quantification in Seminal Plasma Panel (Panel A)
nt, not tested; ini, inibitors detected.
TABLE IV. Results of HIV-1 RNA Quantification in Seminal Cells Panel (Panel B)
nt, not tested; ini, inhibitors detection.
Detection of HIV-1 Genome in Semen 881
centrifugation alone [Leruez-Ville et al., 2002]. Apart
from measures to check that the processed spermatozoa
are HIV-free, basic hygiene and safety rules must be
respected, particularly during semen sampling, proces-
sing and freezing to avoid contaminating other semen
procreation techniques may also reduce the risk by
requiring fewer spermatozoa.
It was found that the performance of the HIV
tion varied greatly. The use of silica-based nucleic acid
inclusion of an internal control to detect PCR inhibitors
does not seem to increase significantly the sensitivity of
the assays tested, but is probably a safe and reassuring
assays. A future study should assess a combination of
For example, semen fraction may be tested using only
one or two assays, on only one semen fraction or more.
The overall safety of the procedure may then be difficult
assays are also important. Lastly, a commercial kit or a
consensus protocol could be a great help in providing
uniform tests performances on semen. This study is the
first quality control evaluation organized for detecting
HIV nucleic acid in semen fractions before medically
assisted procreation. Regular quality control will help
laboratories to improve their practice.
We thank the Association de Recherche sur le Sida
(ARS) for financial support.
Al-Khan A,Colon J,Palta V,Bardeguez A.2003.Assistedreproductive
technology for men and women infected with human immunodefi-
ciency virus type 1. Clin Infect Dis 36:195–200.
Bourlet T, Levy R, Laporte S, Blachier S, Bocket L, Cassuto G,
Chollet L, Leruez-Ville M, Maertens A, Mousnier F, Pasquier C,
Payan C, Pellegrin B, Schvoerer E, Zavadzki P, Chouteau J,
Duverlie G, Izopet J, Lunel-Fabiani F, Pawlotsky JM, Profizi N,
Rouzioux C, Stoll-Keller F, Thibault V, Wattre P, Pozzetto B. 2003.
Multicenter quality control for the detection of hepatitis C virus
RNA in seminal plasma specimens. J Clin Microbiol 41:789–793.
Bujan L, Daudin M, Alvarez M, Massip P, Puel J, Pasquier C. 2002.
Intermittent human immunodeficiency type 1 virus (HIV-1)
shedding in semen and efficiency of sperm processing despite high
seminal HIV-1 RNA levels. Fertil Steril 78:1321–1323.
Berrebi A, Massip P, Pasquier C. 2004a. Factors of intermittent
HIV-1 excretion in semen and efficiency of sperm processing in
obtaining spermatozoa without HIV-1 genomes. AIDS 18:757–766.
Bujan L, Pasquier C, Labeyrie E, Lanusse-Crousse P, Morucci M,
Daudin M. 2004b. Insemination with isolated and virologically
testedspermatozoaisasafeway forhumanimmunodeficiency type
1 virus-serodiscordant couples with an infected male partner to
have a child. Fertil Steril 82:857–862.
Dunne AL, Mitchell F, Allen KM, Baker HW, Garland S, Clarke GN,
Mijch A, Crowe SM. 2003. Analysis of HIV-1 viral load in seminal
plasma samples. J Clin Virol 26:239–245.
Dyer JR, Gilliam BL, Eron JJ, Jr., Grosso L, Cohen MS, Fiscus
SA. 1996. Quantitation of human immunodeficiency virus type
1 RNA in cell free seminal plasma: Comparison of NASBA
with Amplicor reverse transcription-PCR amplification and
correlation with quantitative culture. J Virol Methods 60:161–170.
Fiscus SA, Brambilla D, Coombs RW, Yen-Lieberman B, Bremer J,
Kovacs A, Rasheed S, Vahey M, Schutzbank T, Reichelderfer PS,
Group A. 2000. Multicenter evaluation of methods to quantitate
human immunodeficiency virus type 1 RNA in seminal plasma. J
Clin Microbiol 38:2348–2353.
Hanabusa H, Kuji N, Kato S, Tagami H, Kaneko S, Tanaka H,
eliminating HIV-1. AIDS 14:1611–1616.
Kim LU, Johnson MR, Barton S, Nelson MR, Sontag G, Smith JR,
Gotch FM, Gilmour JW. 1999. Evaluation of sperm washing as a
potential method of reducing HIV transmission in HIV-discordant
couples wishing to have children. AIDS 13:645–651.
Leruez-Ville M, de Almeida M, Tachet A, Dulioust E, Guibert J,
Mandelbrot L, Salmon D, Jouannet P, Rouzioux C. 2002. Assisted
reproduction in HIV-1-serodifferent couples: The need for viral
validation of processed semen. AIDS 16:2267–2273.
Marina S, Marina F, Alcolea R, Exposito R, Huguet J, Nadal J, Verges
A. 1998. Human immunodeficiency virus type 1-serodiscordant
couples can bear healthy children after undergoing intrauterine
insemination. Fertil Steril 70:35–39.
Meseguer M, Garrido N, Gimeno C, Remohi J, Simon C, Pellicer A.
2002. Comparison of polymerase chain reaction-dependent meth-
and hepatitis C virus in washed sperm. Fertil Steril 78:1199–
P,PuelJ,BujanL,IzopetJ.2000.Spermwashing andvirus nucleic
acid detection to reduce HIV and hepatitis C virus transmission in
serodiscordant couples wishing to have children. AIDS 14:2093–
Politch JA, Xu C, Tucker L, Anderson DJ. 2004. Separation of human
immunodeficiency virus type 1 from motile sperm by the double
tube gradient method versus other methods. Fertil Steril 81:440–
Quayle AJ, Xu C, Mayer KH, Anderson DJ. 1997. T lymphocytes and
macrophages, but not motile spermatozoa, are a significant source
of human immunodeficiency virus in semen. J Infect Dis 176:960–
Quayle AJ, Xu C, Tucker L, Anderson DJ. 1998. The case against an
association between HIV-1 and sperm: Molecular evidence.
J Reprod Immunol 41:127–136.
Semprini AE. 1993. Insemination of HIV-negative women with
processed semen of HIV-positive partners. Lancet 341:1343–1344.
Semprini AE, Fiore S. 2004. HIV and reproduction. Curr Opin Obstet
Shepard RN, Schock J, Robertson K, Shugars DC, Dyer J, Vernazza P,
Hall C, Cohen MS, Fiscus SA. 2000. Quantitation of human
ments. J Clin Microbiol 38:1414–1418.
Tachet A, Dulioust E, Salmon D, De Almeida M, Rivalland S,
Finkielsztejn L, Heard I, Jouannet P, Sicard D, Rouzioux C. 1999.
Detection and quantification of HIV-1 in semen: Identification of a
subpopulation of men at high potential risk of viral sexual
transmission. AIDS 13:823–831.
MS. 1997. Quantification of HIV in semen: Correlation with
antiviral treatment and immune status. AIDS 11:987–993.
J. Med. Virol. DOI 10.1002/jmv
TABLE V. Results of HIV-1 DNA Quantification in Seminal
Cells Panel (Panel C)
Target value Center
DNA copies/106cells345689 10 11
nt, not tested.
882 Pasquier et al.