UvA-DARE (Digital Academic Repository) Detection of blood ...Detection of blood transmissible viral...
Transcript of UvA-DARE (Digital Academic Repository) Detection of blood ...Detection of blood transmissible viral...
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Detection of blood transmissible viral agents: implications for blood safety
Jongerius, J.M.
Link to publication
Citation for published version (APA):Jongerius, J. M. (2002). Detection of blood transmissible viral agents: implications for blood safety.
General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.
Download date: 15 May 2020
Detectionn of blood transmissible viral agents:
implicationss for blood safety
Johnn Jongerius
Detectionn of blood transmissible viral agents:
implicationss for blood safety
Johnn Jongerius
Thiss thesis was prepared at the Sanquin Blood Bank Midden-Nederland, Utrecht, The
Netherlands;; and at the University Medical Center, Utrecht; and at the Viral Diagnostic
Laboratoryy of the Sanquin-CLB division, Amsterdam, The Netherlands.
Cover:: further closing the window (design Frits Fijen)
Copyright:: J.M. Jongerius, IJsselstein, The Netherlands, 2002
Al ll rights reserved. No part of this publication may be reproduced, stored in a retrieval
system,, or transmitted, in any form or by any means, electronic, mechanical,
photocopying,, recording, or otherwise, without the prior written permission of the holder
off the copyright.
ISBN:: 90-9015689-5
Detectionn of blood transmissible viral agents:
implicationss for blood safety
ACADEMISCHH PROEFSCHRIFT
terr verkrijging van de graad van doctor
aann de Universiteit van Amsterdam
opp gezag van de Rector Magnificus
prof.. mr. P.F. van der Heijden
tenn overstaan van een door het college voor promoties ingestelde
commissie,, in het openbaar te verdedigen in de Aula der Universiteit
opp vrijdag 21 juni 2002, te 10.00 uur
door r
Johanness Michiel Jongerius
geborenn te Utrecht
Promotiecommissie e
Promotor: :
Co-promotores: :
Overigee leden:
Faculteit: :
prof.. dr. W.G. van Aken
dr.. E.F. van Leeuwen
dr.. CL. van der Poel
prof.. dr. J. van der Noordaa
prof.. dr. P. Speelman
prof.. dr. GJ. Bonsel
dr.. R.A.F.M. Chamuleau
prof.. dr. A. Brand
Geneeskunde e
Bloodd transfusion is like marriage; it should not be entered upon lightly or wantonly, or
moree often than is absolutely necessary (Beal RW, 1976).
Aan:: Debby, Michael en Danny
Contents s Page e
Chapterr 1 General introduction. 9
Chapterr 2 Human immunodeficiency virus (HIV) antibodies detected by new 27
assayss that are enhanced for HIV-1 subtype O.
(Transfusionn 1997; 37: 841-4)
Chapterr 3 New hepatitis B virus mutant form in a blood donor that is 39
undetectablee in several hepatitis B surface antigen screening assays.
(Transfusionn 1998;38:56-9)
Chapterr 4 A simple strategy to look back on posttransfusion hepatitis B in a 51
multitransfusedd patient.
(Voxx Sang 1998;75:66-9)
Chapterr 5 Evaluation of automated nucleic acid extraction devices for 61
applicationn in HCV NAT.
(Transfusionn 2000;40:871-4)
Chapterr 6 Evaluation of HIV nucleic acid tests in combination with NucliSens 73
Extractorr for application in minipool screening.
(Preliminaryy study for chapter 7)
Chapterr 7 Validation of the NucliSens Extractor combined with the 83
AmpliScreenn HIV-1 vl.5 and HCV v2.0 test for application in NAT
minipooll screening.
(Transfusion,, accepted for publication)
Chapterr 8 GB virus type C viremia and envelope antibodies among population 101
subsetss in The Netherlands.
(Voxx Sang 1999;76:81-4)
Chapterr 9 Post-transfusion hepatitis infections among cardiac surgery patients 113
afterr the introduction of anti-HCV screening of blood donations.
(Submittedd for publication)
Chapterr 10 Summary 127
Chapterr 10 Samenvatting 137
Dankwoordd 145
Overigee publicaties 147
Listt of abbreviations
aa a
AIDS S
anti-E2 2
anti-HBc c
anti-HBe e
anti-HBs s
CO O
CPMP P
DNA A
ECL L
ELISA A
EIA A
FDA A
FFP P
GBV-C C
GBV-CC E2
HBcAg g
HBeAg g
HBsAg g
HBV V
HCV V
HGV V
HIV V
HTLV V
IC IC
MEIA A
NAT T
aminoo acid
acquiredd immunodeficiency syndrome
antibodyy to GB virus typee C E2 antigen
antibodyy to hepatitis core antigen
antibodyy to hepatitis e antigen
antibodyy to hepatitis B surface antigen
cutofff value
Committeee for Proprietary Medicinal Products
deoxyribonucleicc acid
electrochemiluminescence e
enzyme-linkedd immunosorbent assay
enzymee immunoassay
Foodd and Drug Administration
freshfresh frozen plasma
GBB virus type C
GBB virus type C E2 antigen
hepatitiss B core antigen
hepatitiss B e antigen
hepatitiss B surface antigen
hepatitiss B virus
hepatitiss C virus
hepatitiss G virus
humann immunodeficiency virus
humann T-cell lymphotropic virus
internall control
microparticlee enzyme immunoassay
nucleicc acid amplification technology
ntt nucleotides
ODD optical density
OD:COO optical density-to-cutoff ratio
PCRR polymerase chain reaction
PEII Paul Ehrlich Institute
PMSS pooling management software (or system)
PTHH post-transfusion hepatitis
PTH-NANBB post-transfusion hepatitis non-A, non-B
RBCC red blood cells
RIAA radio immunoassay
RIBAA recombinant immunoblot assay
RNAA ribonucleic acid
RT-PCRR reverse-transcription polymerase chain reaction
WBB Western blot
WTT wild-type
Chapterr 1
Generall introduction
9 9
10 0
Introduction n
Thee safety of the blood supply has increased enormously over the last two decades.
Thiss is particularly due to the selection of voluntary nonremunerated blood donors, the
continuouss improvements and new approaches to laboratory testing of donated blood, and
thee use of specific viral inactivation methods for blood products. However, there still
remainss a residual risk of recipient infection from today's blood supply. The predominant
sourcee of this risk is attributable to blood collected from donors who are in the
preseroconversionn phase of infection, the so-called window period. Other sources of
residuall risk are emerging viral mutants, atypical seroconversions, immunosilent
infections,, newly identified blood borne viral agents and last but not least laboratory
testingg errors [1-3].
Inn The Netherlands as well as in many other countries, blood donations are
routinelyy screened for hepatitis B virus (HBV), human immunodeficiency virus type 1 and
22 (HIV-1 and HIV-2), hepatitis C virus (HCV), human T-lymphotropic virus typee I and
III (HTLV-I and HTLV-II ) and Treponema pallidum. Apart from the direct virus-detection
testt for HBV, the screening tests used are based on antibody detection. Despite the
continuouss improvement of such screening tests, the existence of a window period,
remainss a problem. To improve the safety of the blood supply, a new technology, nucleic
acidd amplification technology (NAT), was recently introduced in blood banking. Instead
off screening for viral antibodies or proteins, using NAT, the presence of nucleic acid
sequences,, unique to a particular viral agent, can be determined in blood samples. Thus,
(infectious)) window period blood donations that otherwise escape detection may be
identified.. Routine HCV NAT screening of blood donations was implemented in July
19999 in The Netherlands. The HCV NAT minipool screening program was extended with
HIVV NAT in November 2000. The implementation of NAT for other blood borne viral
agentss is currently under discussion. Efforts to even further improve the safety of the
bloodd supply will continue to be explored. However, the necessity and feasibility to
achievee a "zero-risk" for blood transmissible viral agents is momentarily questioned,
i i i
especiallyy in view of thee high cost benefit of additional measures [4-6].
Thiss thesis focuses on the limitations and new developments of laboratory testing
forr blood transmissible viral agents. In addition, the transmission of the recently identified
GBV-CC virus by blood was studied.
Hepatitiss B virus
Hepatitiss B virus (HBV) is an enveloped virus, 42 nm in diameter, containing a
partiallyy double stranded circular DNA genome. HBV belongs to the family of the
Hepadnaviridae.Hepadnaviridae. At least six HBV genotypes, designated A-F, can be distinguished [7].
HBVV is transmitted by parenteral routes, intimate contact and vertically from mother to
childd at birth. Hepatitis B surface antigen (HBsAg) is the major target for detecting acute
orr chronic infections with HBV. Antibodies to HBsAg confer protection after infection
orr vaccination. Antibodies to the hepatitis B core antigen (anti-HBc) become detectable
3-44 weeks after the appearance of HBsAg and usually persist for many years. Thus, anti-
HBcc antibodies are considered as a reliable marker of (past) HBV infection [8]. Currently,
inn The Netherlands, blood donor screening for HBV infections entirely relies on the
detectionn of HBsAg.
Sincee the early nineteen seventies, diagnostic tests are available for HBV diagnosis.
Att present, HBV testing is performed using third generation immunoassays for the
detectionn of HBsAg. Using these tests in blood donation screening does not completely
eliminatee the risk of HBV transmission by blood products. HBsAg tests may be negative
inn the window period of HBV infection, in the early convalescence phase (core window)
off HBV infection, and during chronic HBV infection when very low levels of HBsAg are
presentt [10-12]. Additionally, emerging HBsAg mutant forms of HBV, expressing the
HBsAgg "a" determinant with a changed amino acid composition, cause a threat to the
safetyy of the blood supply [3,6,9].
Thee major antigenic domain of the HBsAg protein is the "a" determinant which
iss formed by a hydrophilic region located between amino acids 124 to 147. Antibodies to
12 2
thee "a" determinant of HBsAg confer protection to HBV after infection or vaccination.
Variouss HBsAg mutant forms, carrying one or more mutations in the HBV S (surface or
envelope)) gene compared to the wild type virus, have been reported. HBsAg mutant forms
weree initially observed in vaccinees, monoclonal antibody treated patients, as well as in
patientss without antibody to HBsAg. The most common HBsAg mutant form has a R
(Arg)) replacement of G (Gly) at position 145 of the envelope protein [13]. Other
replacementss include E (Glu) at position 141 and N (Asn) at position 126 [14]. Epitope
changess of the 'a' determinant due to mutation(s) of the HBV S gene can result in false-
negativee results in the routinely used HBsAg screening assays [14-16].
Schreiberr et al. estimated the residual risk of transmitting HBV infection by the
transfusionn of screened donor blood in the USA to be 1 in ~ 63,000 [9]. The estimated
lengthh of the HBV window period used for calculating this risk was 59 days [17]. This
riskrisk was estimated by calculating the incidence of infection in repeat blood donors with
thee length of the window period as fraction of a year. As shown in Table 1, the incidence
off HBV infection among repeat donors in The Netherlands during the period of 1995
throughh 2000 was 1.3 per 100,000 donors. Using Schreiber's method, the residual risk of
transfusionn transmitted HBV infection in The Netherlands, for 1995 through 2000, is 1
inn ~ 475,000 screened blood donations.
Too further improve the safety of the blood supply, the implementation of HBV
NATT testing in minipools is discussed. The currently used HBsAg screening tests are
highlyy sensitive and HBV DNA levels are relatively low during the pre-HBsAg-positive
infectiouss window period with an average of less than 1,000 copies per mL (range: 1 to
2,4000 copies per mL; genome doubling time: 4 days) [6], Busch et al. reported that
implementationn of HBV NAT reduces the HBV window period with approximately 6-15
dayss [1]. Based on this window period reduction estimate, the residual risk may decline
inn the Netherlands from 1 in ~ 475.000 to 1 in ~ 530,000 to 635,000 screened blood
donations.. Thus, the protected yield of implementing HBV NAT is small. For this reason,
HBVV is a less feasible target for NAT in minipools compared to HCV and HIV [6].
13 3
TableTable 1. Viral infections per 100,000 donors in The Netherlands [1]
Year r
1995 5
1996 6
1997 7
1998 8
1999 9
2000 0
Mean n
HBVV infection
Neww *
61 1
44 4
54 4
63 3
86 6
44 4
59 9
Repeatt t
2.7 7
1.0 0
1.7 7
0.3 3
0.7 7
1.4 4
1.3 3
HIVV infection
New w
0 0
6.1 1
0 0
0.2 2
2.3 3
3.4 4
2.0 0
Repeat t
0 0
0.5 5
0.7 7
1.9 9
0.5 5
0.2 2
0.6 6
HCV V
New w
31 1
24 4
31 1
38 8
35 5
17 7
29 9
infection n
Repeat t
0.7 7
0.2 2
0.3 3
0.2 2
0.5 5
0 0
0.3 3
77 Sanquin jaarverslag 2000.
** first time blood donors.
tt repeat blood donors.
Note:: incidence among repeat blood donors has been used for calculating residual risks.
Humann immunodeficiency virus
Humann immunodeficiency virus (HIV), the etiologic agent of acquired
immunodeficiencyy syndrome (AIDS), is an enveloped single stranded RNA virus
belongingg to the lentivirus subfamily of the Retroviridae. It was originally named
lymphadenopathy-associatedd virus (LAV) or human T-cell lymphotropic virus type-Ill
(HTLV-III) .. The virus, which was first identified in 1983 is now designated HIV-1
[18,19].. A closely related strain, HIV-2, which is endemic in West Africa, has been
identifiedd in 1986 [20]. HTV-1 can be divided in three genetically distinct groups, group
MM (major), group O (Outlier) and the recently discovered group N [21-24]. In group M,
thee genotypes A through K can be distinguished [23]. HIV is predominantly transmitted
byy sexual intercourse, sharing of contaminated injecting equipment, and vertically to
infantss from HIV infected mothers. In the early stages of the HIV epidemic the receipt of
HTVV contaminated blood components was also an important route of spread. The first
14 4
reportss on transfusion-associated AIDS were published in late 1982 and early 1983 [25-
27].. HIV has been transmitted through receipt of whole blood, cellular components,
plasmaa and clotting factors [28,29].
Afterr the introduction of HTV-antibody testing in 1985, the incidence of transfusion
transmittedd HIV infection declined dramatically. The first commercially available tests
employedd HIV-1 lysate proteins for the detection of HTV-antibodies. Over time, HIV-
antibodyy tests have been improved continuously to increase both sensitivity and
specificity.. At present, HIV antibody testing of blood donations is mainly performed using
thirdd generation "double antigen" immunoassays. These assays employ both recombinant
HIVV antigens and synthetic HIV peptides, enabling simultaneous and earlier IgM and IgG
detectionn of HIV-1 and HIV-2 antibodies. In some countries blood donations are
additionallyy tested for the presence of the HIV p24 antigen, which is detectable
approximatelyy 1 week before the appearance of HTV antibodies [30,31]. Currently, fourth
generationn tests are available which combine detection of HIV p24 antigen and HIV
antibodies. .
Thee main source of residual risk for HIV infection in blood banking is attributable e
too donor blood collected during the window period of HIV infection. The incidence of
suchh events is calculated 1 in - 493,000 in the USA [9]. The incidence of HIV infection
amongg repeat donors in The Netherlands during the period 1995 through 2000 was 0.6 per
100,0000 donors (Table 1). Using Schreiber's method combined with a preseroconversion
windoww period estimate of 22 days, the residual HIV risk for The Netherlands is 1 in ~
2,780,0000 screened blood donations [8,32]. To further diminish the risk of virus
transmissionn via blood donated during the early window period of HIV infection various
nationall blood transfusion organizations presently use, or are in the process of
implementing,, routine HIV NAT testing in so-called minipools varying from 24 to 96
donations.. The Dutch blood supply foundation, Sanquin, has implemented HIV NAT
minipooll screening as of November 2000. HIV NAT testing narrows the HIV window
periodd with approximately 10-15 days [1]. Thus, theoretically, the residual HIV risk
attributedd to window period donations in The Netherlands has declined from 1 in ~
15 5
2,780,0000 to 1 in ~ 5,000,000 to 8,335,000 screened donations following NAT-testing.
Anotherr important residual risk source is the emergence of HIV variants which may
nott be detected by commercially available antibody screening assays [2,3]. The isolation
andd characterization of the Cameroonian variant HIV strains ANT-70 and MVP-5180,
resultedd in modifications of HIV antibody tests to correct deficiencies in sensitivity
[21,22,33].. Modified tests should always be evaluated, even to ensure the sensitivity in
detectingg the more common HIV strains [34]. The sensitivity problems (false-negative
HIVV tests) with a third-generation HIV antibody detection test (IMx HIV-l/HIV- 2 Plus,
Abbott)) which occurred in 1996 in Europe further stress the necessity for continuous
monitoringg of routinely used screening tests [35,36].
Hepatitiss C virus
Hepatitiss C virus (HCV) is an enveloped single-stranded RNA virus [37]. HCV,
thee major causative agent of post-transfusion hepatitis non-A, non-B (PTH-NANB), was
discoveredd in 1989. HCV has been classified within a third independent genus of the
FlaviviridaeFlaviviridae family, which also includes flavivirus and pestivirus genera [38]. Different
stainss of HCV demonstrate a remarkable degree of nucleotide sequence diversity.
Currently,, at least 6 major genotypes can be distinguished [39]. HCV is predominantly
transmittedd by infected blood. The risk of sexual transmission of HCV is absent or very
loww and vertical transmission from an infected mother to child is infrequent [40-42].
Afterr the cloning of a specific part of the HCV genome in 1989, the rapid
developmentt and introduction in 1990 of first generation anti-HCV screening tests for
applicationn in routine donation screening led to the prevention of PTH-NANB in ~ 63%
off cases [43,44], The first commercially available screening tests employed only the CI 00
recombinantt antigen which is derived from the non-structural NS4 region of the HCV
genomee [38]. Second generation anti-HCV tests were introduced during 1991 and
employedd both structural (C22/core) and non-structural (NS3/NS4, CI00 and C200)
derivedd recombinant or synthetic antigens. Second generation HCV screening tests, which
16 6
showedd a largely increased sensitivity in detecting anti-HCV antibodies, reduced the
windoww period by several weeks [45,46]. The sensitivity of third generation anti-HCV
tests,, employing an additional recombinant antigen derived from the NS5 region of the
HCVV polyprotein, which were introduced in 1993, was slightly improved (with regard to
anti-NS33 detection) compared to second generation tests. In contrast, the addition of the
NS55 antigen had adverse effects on the specificity of the third generation tests [47,48].
Despitee the continuous improvement of test sensitivity, the risk of HCV
transmissionn due to window period donations has remained an issue. It may take as long
ass an average of 70 days after HCV infection before anti-HCV antibodies reach detectable
levelss [48]. The residual risk of transfusion-transmitted HCV infection in the USA was
calculatedd to be 1 in ~ 103,000 screened blood donations [9]. The incidence of HCV
infectionn among repeat donors in The Netherlands during 1995 through 2000 was 0.3 per
100,0000 donors (Table 1). Accordingly, for The Netherlands, the residual HCV risk
calculatedd by Schreiber's method combined with a preseroconversion window period
estimatee of 70 days is 1 in ~ 1,725,000 screened blood donations [9,48]. HCV NAT
screeningg in minipools (maximum 48 donations), a method which reduces the HCV
windoww period with approximately 41-60 days [1], was implemented in The Netherlands
inn July 1999 [49]. Due to HCV NAT testing, the residual risk attributed to donations
collectedd during the window period of HCV infection has theoretically declined from 1
inn ~ 1,725,000 (when only screening of anti-HCV antibodies was used) to 1 in ~
4,165,0000 to 12,500,000 screened blood donations. Thus, the application of HCV NAT
testingg virtually eliminates the risk of HCV transmission by blood products. During the
firstfirst two years of routine HCV NAT minipool screening of up to 48 donations, no HCV
RNA-positive,, anti-HCV-negative blood donations were found among -1,800,000 blood
donationss collected in The Netherlands.
Recentt studies have shown that the HCV core antigen is detectable throughout the
viremicc window period of HCV infection [50-52]. It was shown that the core protein
becomess detectable, on average, within 1-2 days after the appearance of viral RNA [50-
52].. Currently, a new diagnostic test is commercially available for the direct detection of
17 7
thee HCV core antigen. This assay appears to be suitable for high-throughput screening of
bloodd donations [53]. HCV core antigen testing may be a useful alternative for HCV
NAT,, especially in countries which lack routine application of NAT.
GBB virus type C
Throughh the identification of HCV and the development of screening tests it
becamee possible to show that HCV was responsible for the vast majority of PTH-NANB
casess [44,45]. However, approximately 10 percent of PTH-NANB are unrelated to HCV
infection,, suggesting the existence of additional causative agents (non-A-E hepatitis) [54].
Inn 1995, Simons et al. identified 3 novel RNA virus, designated GB virus type C (GBV-
C),, in patients with clinical evidence of hepatitis of unknown etiology [55]. In addition
Linnenn et al. independently identified a similar RNA virus, designated hepatitis G virus
(HGV),, from plasma of a patient with chronic hepatitis [56]. Sequence analysis indicated
thatt GBV-C and HGV are different isolates of the same virus. The homology between
GBV-CC and HGV at the nucleotide and amino acid level is approximately 86 and 95%,
respectivelyy [57]. Both viruses are classified in the family of Flaviviridae but are distinct
fromfrom HCV. The homology at the nucleotide and amino acid level between GBV-C/HGV
andd HCV is approximately 25 and 29%, respectively [57,58]. The GBV-C/HGV genome
off 9,392 nucleotides encodes for a single large polyprotein [56-59]. GBV-C/HGV is
transmittedd parenterally, sexually and perinatally [60-63]. In the further part of this section
GBV-C/HGVV will be referred to as GBV-C.
GBV-CC detection predominantly relies on nucleic acid amplification by reverse-
transcriptionn polymerase chain reaction (RT-PCR) using specific primers from the 5' non-
translatedd region, the NS5a or the NS3 region [57,59]. In contrast with HCV, the envelope
E22 region of the genome, encoding for the E2 protein, is not hypervariable [58,64]. This
findingg may be of importance for the effectiveness of the humoral immune response
againstt GBV-C [65]. In 1997, Tacke et al. reported the development of an immunoassay
forr anti-E2 antibody detection based on a recombinant derived E2 protein [65]. It was
18 8
concludedd that a humoral immune response to E2 is associated with a loss of detectable
GBV-CC RNA. Thus, anti-E2 may be a useful marker to confer a past/resolved GBV-C
infection.. However, simultaneous GBV-C RNA-positivity and GBV-C anti-E2-positivity,
representingg early viral clearance, incomplete protection by circulating GBV-C anti-E2
antibodiess or false-positive EIA results, was found in several individuals [65].
Threee out of 79 (4%) patients enrolled in a prospective study on the incidencee of
transfusion-associatedd GBV-C infection were found to be infected with HCV before
transfusionn and the cause of acute hepatitis could not be determined. 63 (80%) had
transfusionn related HCV infections of which 6 (10%) were co-infected with GBV-C.
Threee (4%) recipients were found to be solely GBV-C RNA-positive after transfusion.
Thee remaining 10(13%) recipients had no serological or molecular viral marker for any
off the established hepatitis viruses [59,60]. The 3 recipients infected with GBV-C only
hadd mild hepatitis (no jaundice; mean peak alanine aminotransferase level: 198 U/L) and
thee combined HCV and GBV-C infections were no more severe than HCV infections
alone.. In this study no causal relationship between GBV-C and hepatitis could be
established.. Thus, GBV-C accounts for only a minority of non-A-E hepatitis cases. In
addition,, a study performed by the Sentinel Counties Viral Hepatitis Study Team revealed
thatt GBV-C was not implicated as an etiologic agent of non-A-E hepatitis [66], It was also
foundd that GBV-C infection did not lead to chronic disease and did not affect the clinical
coursee of acute disease among patients infected with hepatitis A, B or C [66]. From these
andd other studies it can be concluded that GBV-C infection is not unequivocally causally
relatedd to acute or chronic hepatitis and does not alter the natural history of other chronic
liverr diseases [59,60,66,67]. Thus, given the apparently low or absent pathogenicity of
GBV-CC the need of blood donation screening procedures to prevent GBV-C infection is
highlyy questionable [68],
19 9
Aimss of the studies described in this thesis
Thee aims of the studies described in this thesis are:
1)) to investigate a number of currently used, improved and or new methods for the
detectionn of blood transmissible viral agents (chapter 2, 3, 5, 6, 7).
2)) to determine the prevalence of GBV-C in various groups of patients and blood donors
(chapterr 8).
3)) to determine the incidence of post-transfusion GBV-C, HCV and HBV infection in
patientss who underwent cardiac surgery (chapter 9).
4)) to investigate the efficacy of a new strategy to retrospectively investigate the
infectivityy of blood products in a case of post-transfusion HBV infection (chapter 4).
20 0
References s
1.. Busch MP, Kleinman SH. Report of the interorganizational task force on nucleic acid
amplificationn testing of blood donors: Nucleic acid amplification testing of blood
donorss for transfusion-transmitted infectious diseases. Transfusion 2000;40:143-59.
2.. Busch MP, Watanabe KK, Smith JW, et al. False-negative testing errors in routine
virall marker screening of blood donors. The Retrovirus Epidemiology Donor Study.
Transfusionn 2000;40:585-9.
3.. Roberts P. Virus safety of plasma products. Med Virol 1996;6:25-38.
4.. Holland PV. Viral infections and the blood supply. New Engl J Med 1996;334:1734-5.
5.. Gallarda JL, Dragon E. Blood screening by nucleic acid amplification technology:
currentt issues, future challenges. Molecular Diagnosis 2000;5:11-22.
6.. Busch MP. HTV, HBV and HCV: new developments related to transfusion safety. Vox
Sangg 2000;78(suppl 2):253-6.
7.. Norder H, Courouce AM, Magnius LO, et al. Complete genomes, phylogenetic
relatedness,, and structural proteins of six strains of the hepatitis B virus, four of which
presentt two new genotypes. Virology 1994;198:489-503.
8.. Laperche S, Guitton C, Smilovici W, et al. Blood donors infected with the hepatitis
BB virus but persistently lacking antibodies to the hepatitis B core antigen. Vox Sang
2001;80:90-4. .
9.. Schreiber GB, Busch MP, Kleinman SH, et al. The risk of transfusion-transmitted viral
infections.. The Retrovirus Epidemiology Donor Study. N Eng J Med 1996;334:1685-
90. .
10.. Jagodzinski L, Kraus F, Garrett P, et al. Detection of hepatitis B viral sequences in
earlyy HBV infection. Transfusion 1994;34 (Suppl):37S.
11.. McMahon BJ, Bender TR, Berquist KR, et al. Delayed development of antibody to
hepatitiss B surface antigen after symptomatic infection with hepatitis B virus. J of Clin
Microbioll 1981;14:130-4.
21 1
12.. Bréchot C, Degos F, Lugassy C, et al. Hepatitis B virus DNA in patients with chronic
liverr disease and negative tests for hepatitis B surface antigen. N Eng J Med
1985;312:270-6. .
13.. Carman WF, Zanetti AR, Karayiannis P, et al. Vaccine-induced escape mutant of
hepatitiss B virus. Lancet 1990;336:325-9.
14.. Carman WF. Vaccine-associated mutants of hepatitis B virus. In: Nishioka K, Suzuki
H,, Mishiro S, Oda T, eds. Viral hepatitis and liver disease: proceedings of the
Internationall Symposium on Viral Hepatitis and Liver Disease. Tokyo: Springer-
Verlag,, 1994:243-7.
15.. Carman WF, Korula J, Wallace L, et al. Fulminant reactivation of hepatitis B due to
envelopee protein mutant that escaped detection by monoclonal HBsAg ELISA. Lancet
1995;345:1406-7. .
16.Zaaijerr HL, Pfeiffer KD, Vrielink H, et al. Early detection of hepatitis B surface
antigenn (HBsAg) and detection of HBsAg-mutants; a comparison of five assays. Vox
Sangg 2001;81:219-21.
17.. Mimms LT, Mosley JW, Hollinger FB, et al. Effect of concurrent acute infection with
hepatitiss C virus on acute hepatitis B virus infection. Br Med J 1993;307:1095-7.
18.. Barré-Sinoussi F, Chermann JC, Rey F, et al. Isolation of a T-lymphotropic retrovirus
fromfrom a patient at risk for acquired immune deficiency syndrome (AIDS). Science
1983;220:868-71. .
19.Galloo RC, Salahuddin SZ, Popovic M, et al. Frequent detection and isolation of
cytopathicc retroviruses (HTLV-III ) from patients with AIDS and at risk for AIDS.
Sciencee 1984;224:500-3.
20.Clavell F, Guétard D, Brun-Vézinet F, et al. Isolation of a new human retrovirus from
Westt African patients with AIDS. Science 1986;233:343-6.
21.. Vanden Haesevelde M, Decourt J-L, De Leys RJ, et al. Genomic cloning and complete
sequencee analysis of a highly divergent African human immunodeficiency virus
isolate.. J Virol 1994;68:1586-96.
22 2
22.. Gürtler LG, Hauser PH, Eberle J, et al. A new subtype of human immunodeficiency
viruss type 1 (MVP-5180) from Cameroon. J Virol 1994;68:1581-5.
23.. Robertson DL, Anderson JP, Bradac J A, et al. In: Kuiken C, McCutchan F, Foley B,
Mellorss JW, Hahn B, Mullins J, Marx P, Wolinsky S, Korber B, eds. Human
Retrovirusess and AIDS: HIV-1 nomenclature proposal; a reference guide to HIV-1
classification.. Los Alamos, NM: Los Alamos national Laboratory, 1999:492-505.
24.. Simon F, Mauclare P, Roques P, et al. Identification of a new human
immunodeficiencyy virus type 1 distinct from group M and group O. Nat Med
1998;4:1032-7. .
25.. Centers for Disease Control. Possible transfusion-associated acquired immune
deficiencyy syndrome (AIDS) - California. MMWR Morb Mortal Wkly Rep
1982;31:652-4. .
26.. Ammann AJ, Cowan MJ, Wara DW, et al. Acquired immunodeficiency in an infant:
possiblee transmission by means of blood products. Lancet 1983;1:956-8.
27.Goodnoughh LT, Brecher ME, Kanter MH, et al. Medical progress, Transfusion
Medicine,, First of two parts, Blood Transfusion (review article). N Engl J Med
1999;340:438-47. .
28.. Curran JW, Lawrence DN, Jaffe H, et al. Acquired immunodeficiency syndrome
(AIDS)) associated with transfusions. N Engl J Med 1984;310:69-75.
29.. Evatt BL, Ramsey RB, Lawrence DN, et al. The acquired immunodeficiency syndrome
inn patients with hemophilia. Ann Intern Med. 1984;100:499-504.
30.. Busch MP, Alter HJ. Will human immunodeficiency virus p24 antigen screening
increasee the safety of the blood supply and, if so, at what cost? Transfusion
1995;35:536-9. .
31.. Le Pont F, Costagliola D, Rouzioux C, et al. How much would the safety of blood
transfusionn be improved by including p24 antigen in the battery of tests? Transfusion
1995;35:542-7. .
32.. Busch MP, Lee LLL, Satten GA, et al. Time course of detection of viral and serologic
markerss preceding human immunodeficiency virus type 1 seroconversion: implications
23 3
forr screening of blood and tissue donors: Transfusion 1995;35:91-7.
33.. De Leys RJ, Vanderborght B, Vanden Haesevelde M, et al. Isolation and partial
characterizationn of an unusual human immunodeficiency retrovirus from two persons
off west-central African origin. J Virol 1990;64:1207-16.
34.. Dondero TJ, Hu DJ, George JR. HTV-1 variants: yet another challenge to public health
(comment).. Lancet 1994;343:1376.
35.Chooo V. False-negative HIV test leads to widespread retesting (news). Lancet
1996;347:1031. .
36.Banatvalaa JE. HIV testing over Easter. Lancet 1996;347:1058-9.
37.. Choo Q-L, Richman KH, Han JH, et al. Genetic organization and diversity of the
hepatitiss C virus. Proc. Natl. Acad. Sci. USA, March 1991;88:2451-5.
38.. Lemon SM, Brown EA. Hepatitis C virus. In: Mandell GL, Bennett JE , Dolin R, eds.
Principless and practice of infectious diseases. New York: Churchill Livingstone,
1995:1474-5. .
39.. Blatt LM, Mutchnick MG, Tong MJ, et al. Assessment of hepatitis C virus RNA and
genotypee from 6807 patients with chronic hepatitis C in the United States. J Viral
Hepatt 2000;7:196-202.
40.Bresterss D, Mauser-Bunschoten EP, Reesink HW, et al. Sexual transmission of
hepatitiss C virus. Lancet 1993;342:210-1.
41.. Murphy EL, Bryzman SM, Glynn SA, et al. Risk factors for hepatitis C virus infection
inn the United States blood donors. Hepatology 2000;31:756-62
42.Polywkaa S, Schröter M, Feucht H-H, et al. Low risk of vertical transmission of
hepatitiss C virus by breast milk. Clin Infect Dis 1999;29:1327-9.
43.. Choo Q-L, Kuo G, Weiner AJ, et al: Isolation of a cDNA clone derived from a blood-
bornee non-A, non-B viral hepatitis genome. Science 1989;244:359-62.
44.. Van der Poel CL, Reesink HW, Schaasberg W, et al. Infectivity of blood seropositive
forr hepatitis C virus antibodies. Lancet 1990;335:558-60.
45.Bresterss D, Cuijpers HTM, Reesink HW, et al. Enhanced sensitivity of a second
generationn ELISA for antibody to hepatitis C virus. Vox Sang 1992;62:213-7.
24 4
46.. Van der Poel CL, Bresters D, Reesink HW, et al. Early anti-hepatitis C virus response
withh second-generation C200/C22 ELISA. Vox sang 1992;62:208-12.
47.. Vrielink H, Zaaijer HL, Reesink HW, et al. Sensitivity and specificity of three third-
generationn anti-hepatitis C virus ELISAs. Vox Sang 1995;69:14-7.
48.. Alter HJ. To C or not to C: these are the questions. Blood 1995;7:1681-95.
49.. Cuijpers HTM, Molijn MHJ, Bos HJ, et al. Validation of the NucliSens Extractor in
combinationn with the hepatitis C virus Cobas Amplicor 2.0 assay in four laboratories
inn The Netherlands utilizing nucleic acid amplification technology for blood screening.
Voxx Sang 2001;81:12-20.
50.. Aoyagi K, Ohue C, Iida K, et al. Development of a simple and sensitive enzyme
immunoassayy for hepatitis C virus core antigen. J. Clin Microbiol 1999;37:1802-8.
51.. Peterson J, Green G, Iida K, et al. Detection of hepatitis C core antigen in the antibody
negativee 'window' phase of hepatitis C infection. Vox sang 2000;78:80-5.
52.Couroucee AM, Le Marrec N, Bouchardeau F, et al. Efficacy of hepatitis C virus core
antigenn detection during the preseroconversion period. Transfusion 2000;40:1198-
1202. .
53.. Lee SR, Peterson J, Niven P, et al. Efficacy of a hepatitis C virus core antigen enzyme-
linkedd immunosorbent assay for the identification of'window-phase' blood donations.
Voxx Sang 2001;80:19-23.
54.. Alter HJ, Jett BW, Polito AJ, et al. Analysis of the role of hepatitis C virus in
transfusion-associatedd hepatitis. In: Hollinger FB, Lemon SM, Margolis H, eds. Viral
hepatitiss and liver disease. Baltimore: Williams & Wilkins, 1991:396-402.
55.. Simons JN, Leary TP, Dawson GJ, et al. Isolation of novel virus-like sequences
associatedd with human hepatitis. Nat Med 1995;1:564-9.
56.. Linnen J, Wages J Jr., Zhang-Keck Z-Y, et al. Molecular cloning and disease
associationn of hepatitis G virus: a transfusion-transmissible agent. Science 1996; 271:
505-8. .
57.. Karayiannis P, Thomas HC. Current status of hepatitis G virus (GBV-C) in
transfusion:: is it relevant? Vox Sang 1997;73:63-9.
25 5
58.. Leary TP, Muerhoff AS, Simons JN, et al. Sequence and genomic organization of
GBV-C:: a novel member of the Flaviviridae associated with human non-A-E hepatitis.
JJ Med Vir 1996;48:60-7.
59.. Schiff ER, de Medina MN. Hepatitis G. In: Schiff ER, Sorrell MF, Maddrey WC, eds.
Schifff s Diseases of the Liver. Philadelphia-New York: Lippincott-Raven, 1999:861-7.
60.. Alter HJ, Nakatsuji Y, Meipolder J, et al. The incidence of transfusion-associated
hepatitiss G virus infection and its relation to liver disease. N Engl J Med 1997;
336:747-54. .
61.. Roth WK, Waschk D, Marx S, et al. Prevalence of hepatitis G virus and its strain
variant,, the GB agent, in blood donations and their transmission to recipients.
Transfusionn 1997;37:651-6.
62.Kaoo JH, Liu CJ, Chen PJ, et al. Interspousal transmission of GB virus-C/hepatitis G
virus:: a comparison with hepatitis C virus. J Med Virol 1997;53:348-53.
63.Zanettii AR, Tanzi E, Romano L, et al. Multicenter trial on mother-to-infant
transmissionn of GBV-C virus. J Med Virol 1998;54:107-12.
64.. Weiner AJ, Brauer MJ, Rosenblatt J. Variable and hypervariable domains are found
inn the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and
thee pestivirus envelope glycoproteins. Virology 1991;180:842-8.
65.. Tacke M, Kiyosawa K, Stark K, et al. Detection of antibodies to a putative hepatitis
GG virus envelope protein. Lancet 1997;349:318-20.
66.. Alter MJ, Gallagher M, Morris TT, et al. The Sentinel Counties Viral Hepatitis Study
Team.. Acute non-A-E hepatitis in the United States and the role of hepatitis G virus
infection.. N Engl J Med 1997;336:741-6.
67.. Lunel F, Frangeul L, Chuteau C, et al. Transfusion-associated or nosocomial hepatitis
GG virus infection in patients undergoing surgery. Transfusion 1998;38;1097-1103.
68.. Sauleda S, Esteban JI, Hernandez JM, et al. Evaluation of RNA and E2 antibodies in
prospectivelyy followed recipients of hepatitis G virus-infected blood. Transfusion
1999;39:633-8. .
26 6
Chapterr 2
Humann immunodeficiency virus (HIV)
antibodiess detected by new assays that are
enhancedd for HIV-1 subtype O
J.M.. Jongerius1, CL. van der Poel', A.M. van Loon2, R. van den Akker3,
W.. Schaasberg4, E.F. van Leeuwen'
11 Red Cross Blood Bank Utrecht 22 Department of Virology, Academic Hospital Utrecht
33 National Institute of Public Health and Environmental Protection, Bilthoven 44 Central Laboratory of The Netherlands Red Cross Blood Transfusion Service,
Amsterdam m
Publishedd in: Transfusion 1997;37:841-4
27 7
28 8
Abstract t
Humann immunodeficiency virus type 1 (HTV-1) subtype O infections are not
reliablyy detected by commonly used anti-HIV-1/2 screening assays. Therefore, anti-HTV-
1/22 assays have been modified to increase their sensitivity in detecting antibodies to HTV-
11 subtype O.
Twoo new anti-HIV-1/2 enzyme-linked immunosorbent assays (ELISAs) (Abbott
Pluss and Ortho Enhanced) were compared with a currently used anti-HIV-1/2 ELISA
(Abbottt Recombinant) in various serum panels: 91 Western blot-confirmed anti-HIV-1
positivee samples, 20 samples from Western blot-confirmed HIV-1-infected patients in log3
seriall dilutions, and 1463 samples from consecutive, volunteer, nonremunerated blood
donors. .
Amongg 91 anti-HIV-1 Western blot-positive samples, 2 (2.2%) were missed by the
Abbottt Recombinant ELISA, but all 91 were detected by the Abbott Plus and Ortho
Enhancedd ELISAs. In contrast, two discrepant samples were found to react in viral lysate-
basedd assays. In serial dilutions, Ortho Enhanced ELISA was significantly less sensitive
thann the Abbott Recombinant and Abbott Plus ELISAs, with the latter two being of
comparablee sensitivity. The specificities of Abbott Recombinant, Abbott Plus, and Ortho
Enhancedd ELISAs in 1463 blood donors were 100, 99.93, and 99.86 percent, respectively.
Routinee testing of 29,102 donations with the enhanced Abbott Plus ELISA revealed a
specificityy of 99.93 percent.
Twoo Western blot-confirmed anti-HIV-1-positive samples were missed by the
Abbottt Recombinant ELISA but detected by the Abbott Plus and Ortho Enhanced
ELISAs.. The analytic sensitivity of the Ortho Enhanced ELISA was inferior to that of
bothh Abbott ELISAs. The specificities of the Abbott Recombinant, Abbott Plus, and Ortho
Enhancedd ELISAs were comparable.
29 9
Introduction n
Thee isolation and characterization of novel strains of human immunodeficiency
viruss type 1 (HTV-1) were reported previously [1,2]. These highly divergent HTV-1 strains,
designatedd ANT-70 and MVP-5180, were isolated from African patients of Cameroonian
originn [1-3]. The genetic structure of these new HIV-1 isolates was found to be largely
differentt from that of the main HIV-1 subtypes. Envelope gene products showed only 50-
percentt homology with HIV-1 reference strains [1]. The ANT-70 and MVP-5180 isolates
weree classified by sequence analysis as HIV-1 subtype O [4].
Thee most commonly used anti-HIV screening tests for blood donors are anti-HIV-
1/22 enzyme-linked immunosorbent assays (ELISAs). In longitudinal studies, assay
sensitivity,, especially in the early phase of seroconversion, was increased by replacing
virall lysate antigens with recombinant proteins and synthetic peptides [5]. The
introductionn of third generation ELISAs based on the double-antigen-sandwich principle
furtherr reduced the interval between HTV infection and antibody detection, referred to as
thee window period [6]. However, cross-sectional sensitivity studies on high risk
populationss demonstrated that infections with HIV-1 subtype O are not reliably detected
byy anti-HIV ELISAs based on synthetic peptides or recombinant antigens [7-12]. Anti-
HIV-1/22 ELISAs based on whole-virus lysate, however, appear to detect a broader
repertoiree of HIV antibodies [7,8]. The false-negative results obtained with certain anti-
HIV-11 subtype O specimens in recombinant or peptide based ELISAs led to the
modificationn of HIV assays by the test manufacturers [12].
Wee report here the results of an evaluation study of two new third generation anti-
HIV-1/22 ELISAs enhanced for subtype O detection: HIV-l/HIV- 2 Plus ELISA ([Abbott
Plus],, Abbott, Wiesbaden, Germany) and HIV-l/HIV- 2 Enhanced ELISA ([Ortho
Enhanced],, Ortho, Raritan, NJ). This study was designed to ensure that the adaptation
aimedd at increasing the sensitivity for anti-HIV-1 subtype O would not result in a loss of
sensitivityy for more common HIV types or a loss of specificity in routine practice.
30 0
Materialss and methods
SerumSerum panels
Panell A, used for determining sensitivity, was composed of 91 Western blot-
confirmed,, anti-HIV-1-positive samples. The blood samples were drawn in 1987 from
HTV-1-infectedd Dutch (68) and Tanzanian (23) patients.
Panell B, used for determining analytical sensitivity, was composed of log3 serial
dilutionss of samples from 20 Western blot-confirmed, anti-HIV-1-positive Dutch patients.
Thesee samples reacted strongly in all three ELISAs and were selected from samples sent
too our laboratory for diagnosis of HTV infection. Human AB serum that was negative for
hepatitiss B surface antigen and for antibodies to HTV-1/2, human T-lymphotropic virus
typee I/II , hepatitis C virus and Treponema pallidum was used to prepare the serial
dilutions. .
Panell C, a specificity panel, consisted of sera from 1463 consecutive
nonremuneratedd volunteer blood donors in The Netherlands.
Anti-HIV-1/2Anti-HIV-1/2 ELISAs
Alll samples from panels A, B and C were tested in the Abbott Plus and Ortho
Enhancedd ELISAs and compared with the currently used third generation recombinant
HIV-l/HIV- 22 ELISA ([Abbott Recombinant], Abbott). Both Abbott ELISAs are
recombinant-based,, double-antigen-sandwich ELISAs. The solid phase, a polystyrene
bead,, is coated with recombinant HIV-1 env and gag and HIV-2 em proteins. In the
Abbottt Plus ELISA, an HIV-1 synthetic envelope peptide labeled with horseradish
peroxidasee is added to the antigen conjugate. The Ortho Enhanced ELISA is based on the
indirectt ELISA principle. Microwells coated with a combination of recombinant HIV-1
andd HIV-2 env proteins and inactivated HIV-1 viral lysate antigens are utilized.
Thee double-antigen-sandwich, recombinant synthetic peptide-based HTV 1+2
ELISAA (Wellcozyme, Murex Diagnostics Ltd., Dartford, UK), the indirect viral lysate-
andd synthetic peptide based ELISA (Vironostika HTV Mixt, Organon Teknika, Boxtel, The
31 1
Netherlands),, and the double-antigen, direct, recombinant peptide-based ELISA
(Vironostikaa HIV Uniform II, Organon Teknika) were used for additional testing. All
ELISAss were performed and interpreted according to the manufacturers' instructions.
ConfirmatoryConfirmatory assays
Confirmatoryy retesting on Panel A samples was performed with two immunoblot
systemss (HIV blot 2.2, Diagnostics Biotechnology Ltd., Singapore; and LiaTek HIV 1+2,
Organonn Teknika). Samples from panel B were also confirmed as anti-HTV-1 -positive by
HIVV blot 2.2.
Initiallyy reactive samples from Panel C were retested in duplicate. Repeatably
reactivee samples were confirmed by Western blot (HIV blot 2.2) at an external
confirmatoryy laboratory (Central Laboratory of The Netherlands Red Cross Blood
Transfusionn Service, Amsterdam, The Netherlands).
Statistics Statistics
Probitt analysis was used to estimate the analytical sensitivity of the assays in panel
B,, and the result was expressed as the relative potency [13]. The relative potency is the
ratioo of the two values of the calculated dilution at which 50 percent of the serum samples
aree estimated to be positive [13].
Results s
PanelPanel A
Ass summarized in Table 1, 2 (2.2%, sample M55 and M56) of the 91 samples from
Panell A were negative in the Abbott Recombinant ELISA but were positive in the Abbott
Pluss and Ortho Enhanced ELISAs. The results with these two samples in Abbott
Recombinantt and Abbott Plus ELISAs were confirmed independently by a second
laboratory.. These samples were obtained from two Tanzanian patients suspected of having
AIDS.. These two discrepant samples had earlier (1988) been positive in lysate-based
32 2
ELISAss and in Western blot. Additional testing with other ELISAs revealed that both
sampless were negative in the HIV Uniform II ELISA and that one of these two was
negativee and the other positive in the Wellcozyme ELISA. In the HIV Mixt ELISA, both
sampless were positive. Confirmatory retesting of these two samples was performed with
twoo immunoblot systems; it showed that both samples were positive for anti-HTV-l and
hadd a complete antibody recognition pattern. Further analysis using other anti-HTV tests
andd / or a molecular amplification method were not possible because of the exhaustion of
thee two discrepant samples.
TableTable 1. ELISA reactivity of two discrepant Western blot-confirmed anti-HIV-1-positive
samplessamples from serum Panel A
ELISAss OD:CO *
Sample e
M55 5
M56 6
** Results
Abbott t
Recombinant t
0.45 5
0.67 7
aree expressed as
Abbott t
Plus s
1.22 2
1.36 6
Orthoo Wellcozyme
Enhanced d
1.322 1.57
1.455 0.68
opticall density-to-cutoff ratio (<1 =
HIV V
Mixt t
3.55 5
4.83 3
:: negative; £1
HIV V
Uniformm II
0.90 0
0.15 5
== positive).
PanelPanel B
Thee calculated dilutions (Probit analysis) at which 50 percent of the log} serial-
dilutedd samples were positive were 1 in 3,060 (Ortho Enhanced ELISA); 1 in 111,019
(Abbottt Recombinant ELISA); and 1 in 213,321 (Abbott Plus ELISA) (Fig. 1). The
relativee potency of Abbott Recombinant and Abbott Plus was calculated at 0.52 (95% CI,
0.25-1.01,, not significant). The relative potency of Ortho Enhanced and Abbott
Recombinantt was calculated at 0.028 (95% CI, 0.006-0.068, significant). The relative
potencyy of Ortho Enhanced and Abbott Plus was calculated at 0.014 (95% CI, 0.002-
0.053,, significant).
33 3
FigFig 1. Comparison ofOrtho Enhanced ELISA, Abbott Recombinant ELISA and Abbott
PlusPlus ELISA in log3 serial diluted samples from 20 Western blot-confirmed HIV-1-positive
patients patients
25 5
20 0 a a > > || 15 L. .
|| 10 3 3 z z
5 5
0 0 11 2 3 4 5 6 7 8 9 10 11 12 13
Log33 serial dilution
Orthoo Enhanced ELISA t Recombinant ELISA —*—— Abbott Plus ELISA
PanelPanel C
Alll 1463 samples from Panel C were negative in the Abbott Recombinant ELISA,
andd one sample (0.07%) was repeatably reactive in the Abbott Plus ELISA. In the Ortho
Enhancedd ELISA, two additional samples (0.14%) were repeatably reactive. All three
repeatablyy reactive samples were negative in Western blot. The specificities for Abbott
Recombinant,, Abbott Plus, and Ortho Enhanced ELISAs were 100, 99.93 and 99.86
percent,, respectively.
Discussion n
Thiss study shows that two new HIV-1 subtype O-enhanced third-generation
ELISAss detected 2 samples (2.2%) among those from 91 HIV-1 infected individuals that
weree missed by a currently used recombinant ELISA. These samples were obtained in
19877 from two Tanzanian patients suspected of AIDS. Testing of these two samples with
otherr ELISAs revealed additional discrepant results. Unfortunately, sequencing and
34 4
serologicc studies to establish the subtype of these samples were not possible because of
specimenn exhaustion. A review of the ELISA and Western blot patterns of the subtype O
sampless in French [7,10], United States [9], and German [11] studies did not provide any
indicationn as to whether these samples would be linked to subtype O. Serologic tests for
thee detection of HTV antibodies have been modified continuously during the last decade
too improve both sensitivity and specificity by an increased use of recombinant and peptide
(virall and synthetic) proteins. The isolation and characterization of new HTV variants,
suchh as the highly divergent Cameroonian strains ANT-70 and MVP-5180, resulted in
modificationss of HIV antibody tests to ensure the safety of blood components with regard
too HIV [1-3]. The discovery and existence of such variant demand active global
surveillancee for other HIV variants. Timely modification of HIV antibody tests will be
necessaryy to correct possible future deficiencies in sensitivity [8,12]. Dondero et al.[9]
reportedd that it would be prudent to evaluate new and / or modified diagnostic tests with
seraa from as geographically broad an area as possible, even for the more familiar HIV-1
subtypes.. In addition, new tests should be evaluated to ensure that alterations do not result
inn a loss of sensitivity in detecting the more prevalent HIV-1 strains. The recent problems
withh the third-generation assay (IMx HrV-l/HIV- 2 Plus, Abbott) in Europe further reveal
thee necessity for continuous monitoring of the specificity and sensitivity of current and
newlyy introduced HIV assays [14,15].
Itt was demonstrated in serial dilutions that the Ortho Enhanced ELISA was
significantlyy less sensitive than the Abbott Recombinant and Abbott Plus ELISAs, with
thee latter two being of comparable sensitivity. The use of dilution panels provides
informationn on the analytical sensitivity rather than the clinical sensitivity in anti-HIV
undilutedd and / or seroconversion samples.
Thee specificities of the Abbott Recombinant, Abbott Plus, and Ortho Enhanced
ELISAss were comparable. Although the donor population might previously have been
selectedd by routine use of Abbott Recombinant ELISA, after this study, we began to use
thee Abbott Plus ELISA for routine blood donor screening. From February through August
1996,, 29,102 donations were tested. The repeatably reactive rate was 0.07 percent, which
35 5
demonstratedd that implemented modifications did not result in a significant loss of
specificityy for blood donor HIV-screening purposes. In 1995 we tested 58,661 donations
withh the Abbott Recombinant ELISA and found a repeatably reactive rate of 0.03 percent.
Wee conclude that the specificities of Abbott Recombinant, Abbott Plus, and Ortho
Enhancedd ELISAs were comparable. Anti-HIV-1/2 subtype O-enhanced ELISAs and
lysate-basedd ELISAs are apparently more sensitive than recombinant-based ELISAs. The
heterogeneityy of HTV demands continuous global monitoring and timely adaptation of
HIVV antibody tests.
Acknowledgment t
Thee authors are indebted to Nol Teunissen (Red Cross Blood Bank Utrecht) for
technicall assistance.
References s
1.. Vanden Haesevelde M, Decourt J-L, De Leys RJ, et al. Genomic cloning and complete
sequencee analysis of a highly divergent African human immunodeficiency virus
isolate.. J Virol 1994;68:1586-96.
2.. Giirtler LG, Hauser PH, Eberle J, et al. A new subtype of human immunodeficiency
viruss type 1 (MVP-5180) from Cameroon. J Virol 1994;68:1581-5.
3.. De Leys RJ, Vanderborght B, Vanden Haesevelde M, et al. Isolation and partial
characterizationn of an unusual human immunodeficiency retrovirus from two persons
off west-central African origin. J Virol 1990;64:1207-16.
4.. Myers G, Korber B, Wain-Hobson S, et al. Human retroviruses and AIDS: a
compilationn and analysis of nucleic acid and amino acid sequences. Los Alamos, NM:
Loss Alamos National Laboratory, 1993; acquired immune deficiency syndrome,
periodicall software, available via the Internet.
36 6
5.. Lelie PN, van der Poel CL, Reesink HW, et al. Efficacy of the latest generation of
antibodyy assays for (early) detection of HTV 1 and HTV 2 infection. Vox Sang
1989:56;59-61. .
6.. Zaaijer HL, van Exel-Oehlers P, Kraaijeveld T, et al. Early detection of antibodies to
HIV-11 by third generation assays. Lancet 1992;340:770-2.
7.. Loussert-Ajaka I, Ly TD, Chaix ML, et al. HIV-l/fflV- 2 seronegativity in HIV-1
subtypee O infected patients. Lancet 1994;343:1393-4.
8.. Schable C, Zekeng L, Pau C-P, et al. Sensitivity of United States HTV antibody tests
forr detection of HTV-1 group O infections. Lancet 1994;344:1333-4.
9.. Dondero TJ, Hu DJ, George JR. HrV-1 variants: yet another challenge to public health
(comment).. Lancet 1994;343:1376.
10.. Simon F, Ly TD, Baillou-Beaufils A, et al. Sensitivity of screening kits for anti-HTV-1
subtypee O antibodies (letter). AIDS 1994;8:1628-9.
ll.Giirtlerr LG, Zekeng L, Simon F, et al. Reactivity of five anti-HTV-1 subtype O
specimenss with six different anti-HIV screening ELISAs and three immunoblots. J
Viroll Methods 1995;51:177-83.
12.Gürtlerr L. Difficulties and strategies of HTV diagnosis. Lancet 1996;348:176-9.
13.. Finney DJ. Probit analysis. Cambridge, UK: Cambridge University Press, 1971.
14.Chooo V. False-negative FflV test leads to widespread retesting (news). Lancet
1996;347:1031. .
15.Banatvalaa JE. fflV testing over Easter. Lancet 1996;347:1058-9.
37 7
38 8
Chapterr 3
Neww hepatitis B virus mutant form in a blood donor
thatt is undetectable in several hepatitis B surface
antigenn screening assays
J.M.. Jongerius1, M. Wester2, H.T.M. Cuijpers2, W.R. van Oostendorp1, P.N. Lelie2
C.L.. van der Poel1, E.F. van Leeuwen'
11 Blood Bank Midden-Nederland, Utrecht 22 Central Laboratory of The Netherlands Red Cross Blood Transfusion Service,
Amsterdam m
Publishedd in: Transfusion 1998;38:56-9
39 9
40 0
Abstract t
Envelopee mutant forms of hepatitis B virus (HBV), impairing HBV antibody
recognition,, have been reported with mutations in single or multiple sites of the hepatitis
BB surface antigen (HBsAg) group specific "a" determinant. Blood donors infected with
suchh an HBsAg mutant form of HBV may escape detection by HBsAg screening assays
andd therefore may affect the safety of the blood supply.
AA repeat blood donor became HBsAg- reactive in an enzyme immunoassay.
Confirmatoryy testing yielded negative results for HBsAg in a radioimmunoassay and in
fourr enzyme immunoassays used in blood donor screening. The specificity of the HBsAg
reactivityy in the first enzyme immunoassay was confirmed by HBsAg neutralization with
antibodyy to HBsAg. Additional HBV confirmatory test results were positive for antibody
too hepatitis B core antigen and antibody to hepatitis B e antigen; negative for antibody to
HBsAgg and for hepatitis B e antigen; and positive for HBV DNA. DNA sequence analysis
off the 'a' determinant region of HBsAg revealed amino acid substitutions from Q (Gin)
too R (Arg) at codon 129 and from M (Met) to T (Thr) at codon 133.
Thiss case illustrates the presence of HBsAg mutant forms of HBV in a West-
Europeann blood donor population that were undetected by several HBsAg screening
assays.. Adaptation of HBsAg screening is indicated to overcome deficiencies in sensitivity
inn detecting HBsAg mutant forms of HBV. Screening for antibody to hepatitis B core
antigenn or HBV DNA may also detect blood donors infected with HBsAg mutant forms
off HBV.
41 1
Introduction n
Hepatitiss B surface antigen (HBsAg) screening of blood donations does not totally
eliminatee the risk of hepatitis B virus (HBV) infection by blood transfusion [1]. HBsAg
testss may be negative in the very early phase (window phase) of HBV infection, in the
earlyy convalescence phase (core window) of HBV infection, and in chronic HBV
infectionn with very low levels of HBsAg [2-4]. In addition, the existence of HBsAg
mutantt forms of HBV, with mutations in single or multiple sites of the HBsAg "a"
determinant,, caused difficulties with sensitivity in HBsAg testing [5]. Blood donors
infectedd with HBsAg mutant forms of HBV may escape detection by the currently
availablee HBsAg screening assays, and the safety of the blood supply can be affected [6].
Individualss infected with an HBsAg mutant form of HBV can be identified by HBV DNA
hybridizationn or polymerase chain reaction (PCR) assays and/or by testing for anti-HBc
[6,7].. In the United States, routine anti-HBc testing of blood donors was introduced in the
mid-1980ss as a surrogate marker by which to screen blood donations at risk of transmit-
tingg non-A, non-B hepatitis [8]. After the introduction of screening for antibodies to
hepatitiss C virus, the continuation of surrogate anti-HBc blood donor testing was
questionedd [9,10]. However, it was advocated by others in view of further improvement
off the safety of blood and plasma components with regard to HBV transmission [7,11].
Inn addition, the use of nucleic acid amplification methods such as PCR might increase the
sensitivityy of virus detection, but such techniques are not yet available for high-throughput
routinee testing of individual blood donations [7]. At present, PCR testing of blood
donationss in small sample pools ("minipools"), which may further increase the safety of
bloodd components, is under development.
Wee report on a repeat blood donor who was found positive for HBsAg at routine
bloodd donor screening but had inconsistent results in a number of HBsAg screening
assays,, due to an infection with an HBsAg mutant form of HBV.
42 2
Casee Report
Inn March 1993 a blood donation from a 42-year-old black female repeat blood
donor,, a native of Surinam (South-America) who immigrated to The Netherlands in 1968,
wass found for the first time to be repeatably reactive for HBsAg in a third- generation
enzymee immunoassay (EIA) (AUSZYME monoclonal EIA, Abbott Laboratories, North
Chicago,, IL) with an optical density-to-cutoff (OD/CO) ratio of 2.2 on her third donation
(Tablee 1). Samples and results of testing from this donation are designated as 0-month
sampless and results, respectively. Other samples and results are designated by the time of
theirr collection and report date in relation to the 0-month sample. However, supplemental
testingg with an HBsAg radioimmunoassay (RIA) (AUSRIAII, Abbott) was negative. On
follow-upp research samples (Table 1), the serologic results for HBV markers were positive
inn HBsAg EIA (AUSZYME) (+ 35- month sample was negative, see Table 1), anti-HBc
RIAA (CORAB, Abbott) and anti-HBe RIA (HBe [rDNA] , Abbott), but negative in
antibodyy to HBsAg (anti-HBs) RIA (AUSAB, Abbott) and HBeAg RIA (HBe [rDNA]) .
HBVV DNA PCR testing with HBV-specific primers from the core region was positive
(assayy performed in triplicate) [12].
Extractedd DNA of the donors' +15-months sample (Table 1) was analyzed with
directt sequencing, using dye primer technology (Prism 21 Ml3 dye primer kit, Perkin-
Elmer,, Foster City, CA). First-round PCR was performed with the primers 5'-
CCTGCTGCTATGCCTCATCTTC-3'' and 5'-CAAATGGCACTAGTAAACTGAG-3' and
secondd round was performed with the primers S'-GTATGTTGCCCGTTTGTCCTCT-S'
andd 5'-GCCAGGAGAAACGGACTGAAGC-3'. The second- round primers were used,
extendedd at the 5'-site with the 21 Ml3 universal primer sequence, to obtain appropriate
ampliconss for sequencing. Sequencing was performed in both directions. Sequence
analysiss of the "a" determinant of HBsAg, which is formed by a hydrophilic region from
aminoo acids 124 to 147, and comparison with normal "a" determinants without impaired
antigenicity,, revealed two amino acid changes; from Q (Gin) to R (Arg) at codon 129 and
fromfrom M (Met) to T (Thr) at codon 133. Sequencing was repeated, from an independent
43 3
extractionn and amplification, to confirm the findings.
Thee results of additional serological testing of the +22-month sample (Table 1) of
thee donor with AUSZYME, AUSRIA II and other HBsAg screening assays were as
follows:: positive in AUSZYME with an OD:CO ratio of 4.8; negative in AUSRIA II with
aa ratio of sample net counts per minute to cutoff counts per minute of 0.2; positive in a
third-generationn microparticle HBsAg EIA (IMx, Abbott) with a ratio of sample rate to
modee 1 calibrator rate of 2.8; positive in a third-generation microparticle HBsAg EIA
(AxSYM,, Abbott) with a ratio of sample rate to index calibrator mean rate of 5.0; positive
inn a chemiluminescent immunoassay for HBsAg (PRISM, Abbott) with a ratio of sample
nett counts to cutoff of 3.3; negative in an EIA for HBsAg (Hepanostika HBsAg Uniform
II ,, Organon Teknika, Boxtel, The Netherlands) with an OD:CO ratio of 0.3; negative in
aa third-generation EIA for HBsAg (System III , Ortho, Raritan, NJ) with an OD:CO ratio
off 0.5; negative in a second-generation EIA for HBsAg (Monolisa second- generation,
Pasteur,, Marnes-la-Coquette, France) with an OD:CO ratio of 0.3; negative in an EIA for
HBsAgg (Wellcozyme VK20/21, Murex Diagnostic Ltd., Dartford, UK) with an OD:CO
ratioo of 0.1; and positive in an enhanced EIA for HBsAg (GE14/15/16 Murex, "enhanced"
forr detection of HBsAg mutant forms) with an OD:CO ratio of 1.6. The specificity of the
HBsAgg reactivity in the AUSZYME was confirmed by HBsAg neutralization with
antibodyy to HBsAg.
Att interview, the donor reported sharing a household in Surinam in December 1992
withh three siblings and her father, who died during this time of chronic liver disease. After
herr March 1993 blood donation, the donor experienced a period of fatigue, nausea and
rightt abdomen pain, without clinical jaundice. She had not been vaccinated against HBV.
Inn addition, she reported that, in early 1993, two of her three siblings had experienced a
periodd of jaundice, and the third was diagnosed with HBV infection. Unfortunately, the
siblingss did not respond to repeated invitation to donate a blood sample for HBV testing.
Thee donor was permanently deferred in 1993, but she donated several follow-up research
sampless (Table 1).
44 4
TableTable 1. HBVtest results* in a carrier ofHBsAg mutant form of HBV during 50 months
ofof follow-up
Monthh HBsAg HBsAg Anti-HBs Anti-HBc HBeAg Anti-HBe HBV DNA
E1AA RIA RIA RIA RIA RIA
(AUSZYME)) (AUSRIAII) (AUSAB) (CORAB) {HBe [rDNA] ) (HBe [rDNA] ) (PCRt)
-- 15
-- 8
0 0
++ 7
++ 14
++ 15
++ 2211
++ 35
0.7 7
0.7 7
2.2 2
2.8 8
2.0 0
2.0 0
4.8 8
0.7 7
NT T
NT T
+ + + +
--------
--
NT T
NT T
NT T
----
NT T
--
--
NT T
NT T
NT T
+§II I
+ +
NT T
+ +
+** *
NT T
NT T
NT T
_ _
--NT T
--_** *
NT T
NT T
NT T
+ +
+ +
NT T
+ +
+ +
NT T
NT T
NT T
NT T
NT T
+ +
+ +
+ +
** = results in AUSZYME are expressed as OD:CO ratio (<1 = negative; >1 = positive).
tt = in-house PCR.
%% = negative.
§§ = positive.
III = IgM anti-HBc-negative (CORAB-M RIA, Abbott).
\\ = sample also tested in other HBsAg assays (see text).
*** = microparticle E1A (Abbott).
Discussion n
Thiss case describes the serologic and molecular biologic profile of a repeat blood
donorr who acquired an HBsAg mutant form of HBV infection and whose HBsAg was
undetectablee in 5 (50%) of 10 HBsAg screening assays. The history of household
exposuree to a patient with chronic liver disease and the subsequent clinical symptoms,
togetherr with the occurrence of at least one other case of HBV infection within the family,
suggestt that HBV infection was acquired during her stay in Surinam in late 1992. Serum
sampless from all donations were routinely archived as of March 1993. Retrospective
testingg for HBV markers on the donor's seronegative 1992 donations (-8 and -15 months,
45 5
Tablee 1) was not possible.
HBsAgg mutant forms of HBV, which impair detection by HBsAg assays, have been
foundd in several countries [5]. HBsAg (escape) mutant forms of HBV were initially
observedd in vaccinees and patients treated with monoclonal antibody [5]. The most
commonn vaccine-associated HBsAg escape mutant form of HBV had a R (Arg)
replacementt of G (Gly) at position 145 of the envelope protein. Other replacements
includee E (Glu) at position 141 and N (Asn) at position 126. In the case reported here,
genomicc sequencing revealed amino acid changes from Q (Gin) to R (Arg) at position 129
andd from M (Met) to T (Thr) at position 133. This combination has not been described to
datee [13]. Probably one or both of the mutations have influenced the structure of the 'a'
determinantt in such a way that epitope changes resulted in false-negative results in several
HBsAgg screening assays. Differences in the inherent sensitivities of the HBsAg screening
assayss also can contribute to undetectability in some of the assays.
Carmann et al. [14] reported on an Indonesian patient infected with an HBV mutant
formm that was undetectable by the monoclonal antibody-based AUSZYME but was
detectedd by the polyclonal antibody-based AUSRIAII. The reverse occurred in our case.
Waterss et al. [15] reported that the use of even polyclonal antibody-based assays does not
guaranteee full sensitivity for detection of HBsAg mutant forms of HBV. Therefore, the
incorporationn of specific antibodies against HBsAg mutant forms of HBV in commercially
availablee HBsAg screening assays may be considered to increase the sensitivity for
HBsAgg mutant forms of HBV. This is illustrated by our results with the improved version
off the GE14/15/16 EIA (Murex). That EIA is "enhanced" for the detection of HBsAg
mutantt forms of HBV using multiple monoclonal and polyvalent antibodies directed
againstt the major HBsAg "a" determinant epitopes.
Mostt chronic carriers of HBsAg mutant forms of HBV appear to be detected by
anti-HBcc testing and/or HBV DNA PCR. Anti-HBc testing is not performed in The
Netherlands.. Protocols may soon become available for routine HBV DNA PCR minipool
testingg that also have the potential to detect HBsAg mutant forms of HBV [16]. However,
carrierss of HBsAg mutant forms of HBV with a low titer will escape detection in
46 6
screeningg systems of pooled samples. At present, the prevalence and incidence of HBsAg
mutantt forms of HBV in the blood donor population are unknown.
Wee conclude that several HBsAg screening assays failed to identify a blood donor
infectedd with an HBsAg mutant form of HBV. Improvement of HBsAg screening assays
iss indicated to correct these deficiencies in detection. The value and cost benefit of
additionall screening assays such as anti-HBc and / or HBV DNA PCR for the detection
off HBsAg mutant forms of HBV in blood donations have to be established by further
studies. .
Acknowledgment t
Thee authors thank Wilma Paulij, PhD (Organon Teknika Research and Development,
Boxtel,, The Netherlands); Anton van Loon, PhD (Department of Virology, Academic
Hospitall Utrecht, Utrecht, The Netherlands); Hans Molijn (Red Cross Blood Bank
Rotterdam,, Rotterdam, The Netherlands); and Bob Schelstraete, PhD (Blood Bank
Louvain,, Belgium) for their contributions to the study.
References s
1.. Schreiber GB, Busch MP, Kleinman SH, et al. The risk of transfusion-transmitted viral
infections.. The Retrovirus Epidemiology Donor Study. N Eng J Med 1996;334:1685-
90. .
2.. Jagodzinski L, Kraus F, Garrett P, et al. Detection of hepatitis B viral sequences in
earlyy HBV infection. Transfusion 1994;34 (Suppl):37S.
3.. McMahon BJ, Bender TR, Berquist KR, et al. Delayed development of antibody to
hepatitiss B surface antigen after symptomatic infection with hepatitis B virus. J Clin
Microbioll 1981;14:130-4.
47 7
4.. Bréchot C, Degos F, Lugassy C, et al. Hepatitis B virus DNA in patients with chronic
liverr disease and negative tests for hepatitis B surface antigen. N Eng J Med
1985;312:270-6. .
5.. Carman WF. Vaccine-associated mutants of hepatitis B virus. In: Nishioka K, Suzuki
H,, Mishiro S, Oda T, eds. Viral hepatitis and liver disease: proceedings of the
Internationall Symposium on Viral Hepatitis and Liver Disease. Tokyo: Springer-
Verlag,, 1994:243-7.
6.. Carman WF, Zanetti AR, Karayiannis P, et al. Vaccine-induced escape mutant of
hepatitiss B virus. Lancet 1990;336:325-9.
7.. Roberts P. Virus safety of plasma products. Med Virol 1996;6:25-38.
8.. Koziol DE, Holland PV, Ailing DW, et al. Antibody to hepatitis B core antigen as a
paradoxicall marker for non-A, non-B hepatitis agents in donated blood. Ann Intern
Medd 1986;104:488-95.
9.. Blajchman MA, Bull SB, Feinman SV. Post-transfusion hepatitis: impact of non-A,
non-BB hepatitis surrogate tests. Canadian Post-Transfusion Hepatitis Prevention Study
Group.. Lancet 1995;345:21-5.
10.. Mosley JW, Stevens CE, Aach RD, et al. Donor screening for antibody to hepatitis B
coree antigen and hepatitis B virus infection in transfusion recipients. Transfusion
1995;35:5-12. .
11 l.NIH consensus statement on infectious disease testing for blood transfusion. Bethesda,
MD:: National Institutes of Health, 1995.
12.. Liang TJ, Isselbacher KJ, Wands JR. Rapid identification of low level hepatitis B
relatedd viral genome in serum. J Clin Invest 1989;84:1367-71.
13.Norderr H, Hammas B, Lee SD, et al. Genetic relatedness of hepatitis B viral strains
off diverse geographical origin and natural variations in the primary structure of the
surfacee antigen. J Gen Virol 1993;74:1341-8.
14.. Carman WF, Korula J, Wallace L, et al. Fulminant reactivation of hepatitis B due to
envelopee protein mutant that escaped detection by monoclonal HBsAg ELISA. Lancet
1995;345:1406-7. .
48 8
15.. Waters JA, Kennedy M, Voet P, et al. Loss of the common "A" determinant of
hepatitiss B surface antigen by a vaccine-induced escape mutant. J Clin Invest
1992;90:2543-7. .
16.. Rogers PM, Saldanha J, Allain JP. Report of EPFA/NIBSC workshop "nucleic acid
amplificationn tests (NAT) for the detection of blood borne viruses" held on 31 October
19966 in Amsterdam, The Netherlands. Vox Sang 1997;72:199-206.
49 9
50 0
Chapterr 4
AA simple strategy to look back on posttransfusion
hepatitiss B in a multitransfused patient
J.M.. Jongerius, C.L. van der Poel, E.F. van Leeuwen
Bloodd Bank Midden-Nederland, Utrecht
Publishedd in: Vox Sang 1998;75:66-9
52 2
Abstract t
Inn January 1996, a case of hepatitis B virus (HBV) seroconversion in a
multitransfusedd patient was reported to the blood bank. From March through October
1995,, the patient had received 23 units of red cells and 30 units of pooled platelet
concentrates,, encompassing an exposure to a total of 200 whole blood donations.
Inn order to trace hepatitis B surface antigen (HBsAg)-negative but HBV infectious
bloodd donation(s), we tested samples of the donors obtained £ 3 months after the
implicatedd donations for anti-HBc (Corezyme EIA, Abbott). From 172/200 donors,
archivedd samples of subsequent donations were available for this purpose. The remaining
288 donors were reinvited to the blood bank to obtain an additional blood sample for anti-
HBcc testing.
1/2000 follow-up donor samples was anti-HBc positive. Retrospective testing of the
implicatedd HBsAg-negative blood donation of this donor revealed anti-HBc-negative and
HBV-DNA-positivee results. The patient was transfused with the platelets of the HBV-
infectiouss donation. On looking back, the other blood products prepared from this HBV-
infectiouss donation caused posttransfusion HBV infection (PT-HBV) in 2 additional
patients. .
Anti-HBcc testing on mainly archived follow-up samples of 200 donors implicated
inn PT-HBV was a rapid, simple, cost-effective and donor-friendly method to identify an
infectiouss but HBsAg-negative, anti-HBc-negative and HBV-DNA PCR-positive blood
donation.. Routine anti-HBc screening would not have prevented this HBV transmission.
53 3
Background d
Hepatitiss B Virus (HBV) infection was a common complication of blood
transfusionn prior to mandatory routine blood donor screening for Hepatitis B surface
antigenn (HBsAg) [1]. Over the years, commercially available HBsAg screening tests have
beenn adapted to increase sensitivity for HBsAg detection. However, HBsAg screening
doess not totally eliminate the risk of HBV infection by blood transfusion. HBsAg
screeningg may be negative in the very early phase (window phase) of HBV infection [2],
inn the early convalescence phase (core window) of HBV infection [3], in chronic HBV
infectionn with very low levels of HBsAg [4] and in infections with HBsAg mutant forms
off HBV [5,6].
Inn The Netherlands it is mandatory, for look-back procedures, to archive a serum
orr plasma sample of each blood donation for at least 2 years. There are no specific
retestingg strategies regulated to retrospectively establish the infectivity of blood products
inn case of transfusion-associated infectious diseases.
Wee describe here a simple and cost-effective look-back strategy in a case of
posttransfusionn HBV infection with multiple and repeated donation exposures. Instead of
performingg expensive polymerase chain reaction (PCR) testing on 200 archived samples,
anti-HBcc testing on mainly archived samples of subsequent blood donations of the
implicatedd donors identified a donor of an HBsAg negative, anti-HBc negative, HBV-
DNA-positivee blood donation.
Casee Report
Inn January 1996, a case of HBV seroconversion in a multitransfused hemato-
oncologicc patient was reported by a hospital to the regional blood bank. In June 1995, the
patient'ss serum had been nonreactive in routine tests for HBsAg (IMx, Abbott
Laboratories,, North Chicago, Illinois, USA) and for anti-HBc (IMx, Abbott). Screening
off the patient for HBV (IMx, Abbott) in January 1996 revealed HBsAg-positive, HBeAg-
54 4
positive,, anti-HBc-positive, anti-HBe-negative and anti-HBsAg-negative results. The
patientt was not tested for IgM anti-HBc. From March 3, 1995 through October 13, 1995,
thee patient had received 23 units of red cells (RBC) and 30 pooled platelet concentrates
off 5-6 donor units, encompassing an exposure to 200 whole blood donations.
Methods s
Thee implicated whole blood donations (n=200) and corresponding donors (n=200)
weree traced in order to find the possible source of the HBV infection. All implicated
donationss were tested for HBsAg in the static procedure of the Abbott Auszyme EIA.
Ann archived plasma sample (800 uL) from all blood donations is kept for at least
22 years at -23°C in disposable cryotubes in microliter plate format. In order to trace
HBsAg-negative,, HBV-infectious blood donations, and to avoid expensive testing of 200
archivedd samples by HBV-DNA PCR, we tested samples of subsequent donations from
thee 200 implicated donors obtained > 3 months after the implicated donations for anti-
HBcc by enzyme immunoassay (EIA; Corezyme EIA, Abbott). Archived samples of
172/2000 blood donors fulfillin g these criteria were available. The remaining 28 donors did
nott donate blood subsequently. All 28 were reinvited to the blood bank and after informed
consentt a blood sample for anti-HBc testing was drawn.
Thee archived samples, of the implicated donations transfused to the patient, were
reservedd for anti-HBc and HBV-DNA PCR testing (Central Laboratory of The
Netherlandss Red Cross Blood Transfusion Service, Amsterdam, The Netherlands) in case
ann anti-HBc-positive donor would be found at follow-up.
Results s
Alll implicated donations were negative for HBsAg. One (0.5%) of the 200 follow-
upp samples was found to bee anti-HBc-positive. It was a sample drawn 7 months after the
implicatedd donation. The archived sample from the implicated HBsAg-negative donation
55 5
itselff was then thawed and tested for both anti-HBc and HBV-DNA. Anti-HBc was not
detectedd but the HBV-DNA PCR was repeatedly positive with a consistently weak signal.
Anotherr archived sample of this donor, drawn 6 months before the implicated donation,
wass tested and found to be anti-HBc and HBV-DNA PCR-negative.
Thee corresponding blood donor, a 37-year-old male Dutch Caucasian was invited
forr counseling. No risk factors for HBV infection were reported. The longitudinal
serologicall profile of the infectious donor is summarized in Table 1.
TableTable 1. HBV test results in an HBsAg-negative donor who caused posttransfusion HBV
infectioninfection in 3 patients
Date e
Februaryy 13, 1995
Augustt 3, 1995b
Marchh 4, 1996
Mayy 13, 1996
MEIAA = microparticle
NTT = not tested. aa = in-house PCR.
HBsAg g
EIA A
Abbott t
--
--
--
--
Anti-HBs s
RIA A
Abbott t
NT T
NT T
+ +
+ +
enzymee immunoassay,
bb = suspected donation.
Anti-HBc c
RIA A
Abbott t
--
--
+ +
+ +
++ = positive
HBeAg g
MEIA A
Abbott t
NT T
NT T
NT T
--
Anti-HBe e
MEIA A
Abbott t
NT T
NT T
NT T
+ +
,, - = negative.
HBV--
DNA A
PCRa a
--
+ +
NT T
--
Thee HBV-infectious donation had been separated into platelets, RBC and fresh
frozenfrozen plasma (FFP) by standard procedures. The index patient was transfused with the
platelets.. Look-back procedures on the recipients transfused with the corresponding RBC
andd FFP revealed HBV transmission to both. Blood samples of the recipient of the RBC
takenn 3 and 10 months after transfusion with the RBC tested positive for HBsAg. The first
samplee was anti-HBc-negative and the second sample anti-HBc-positive. The recipient
off the FFP was shown to become HBsAg-positive after previously being HBsAg-negative.
56 6
Discussion n
Thiss case illustrates a strategy to look back on posttransfusion hepatitis B in
patientss exposed to multiple blood donations and the value of sample archiving for look-
backk purposes. Without expensive HBV-DNA PCR testing on archived samples of 200
bloodd donations and unnecessary notification of all 200 corresponding donors, an HBsAg-
negative,, but infectious blood donation was traced. The look-back strategy based on
detectionn of anti-HBc antibody seroconversion was rapid, simple, cost-effective and
donor-friendly.. The testing of archived samples retrospectively documented anti-HBc
seroconversionn and HBV-DNA positivity in 1/200 donors. Apparently, the infectious
bloodd donation took place during the HBsAg-negative, HBV-DNA-positive early window
phasee of HBV infection. Carryover during the preparation of sample archives can result
inn a weak signal in the HBV-DNA PCR. The look-back strategy we describe is based on
detectionn of anti-HBc antibody seroconversion and not on HBV-DNA PCR testing. HBV-
DNAA PCR testing is used as a confirmatory marker. A further look-back on the recipients
off the two other blood products of the infectious donation also revealed HBV
transmission. .
Ass illustrated HBV-infected donors may escape HBsAg screening if HBsAg
positivityy occurs in the interval between two consecutive blood donations. The use of anti-
HBcc screening as an additional serological HBV marker can play a role in further
preventingg HBV transmission by blood transfusion [7,8]. Anti-HBc can usually be
detectedd 3-5 weeks after the appearance of HBsAg in the blood and remains detectable
forr several years. Currently, HBsAg is the only mandatory marker for HBV blood donor
screeningg in The Netherlands. Anti-HBc screening of blood donations is not performed,
sincee no significance for prevention of posttransfusion hepatitis was demonstrated in a
controlledd prospective Dutch study [9]. In our case additional anti-HBc screening would
nott have prevented HBV transmission.
Sincee January 1996 one of the obligatory methods to further diminish the risk of
virall transmission by FFP in The Netherlands is the 'quarantine method'. FFP may only
57 7
bee released if both the initial blood donation and a second blood donation obtained from
thee corresponding donor at least 6 months later are tested negative for HBsAg, anti-HCV
andd for anti-HIV. The safety of quarantine FFP regarding HBV infection might benefit
fromfrom additional screening of the second blood donation for anti-HBc.
Thee use of nucleic acid amplification methods such as PCR might increase the
sensitivityy of virus detection during the early window period [2,8]. At present PCR testing
off blood donations in pools of a restricted number of samples, so called minipools, is
underr consideration. Possibly this method could further increase the safety of blood
products.. However, given the weak signal in our HBV infectious donation, it is unlikely
thatt it would have been detected by HBV-DNA testing in (mini)pools. PCR testing on
singlee donation would probably have detected this early window donation.
Inn conclusion, our case illustrates the limitation of HBsAg screening and the
usefulnesss of sample archiving. Anti-HBc testing on mainly archived follow-up samples
off the implicated donors was a rapid, simple, cost-effective and donor-friendly method to
identifyy an infectious but HBsAg-negative, anti-HBc-negative, HBV-DNA PCR-positive
bloodd donation. Routine anti-HBc testing would not have prevented this HBV
transmission. .
Acknowledgment t
Wee thank Dr. H.C. van Prooijen, Academic Hospital Utrecht, The Netherlands,
FJ.L.M.. Haas, St. Antonius Hospital, Nieuwegein, The Netherlands and Dr. A.B.J.
Prakken,, Wilhelmina Children's Hospital, Utrecht, The Netherlands for clinical data; Dr.
H.T.M.. Cuijpers and Dr. P.N. Lelie at the Central Laboratory of The Netherlands Red
Crosss Blood Transfusion Service Amsterdam, The Netherlands for helpful suggestions;
W.R.. van Oostendorp, A. Teunissen and A.H. de Vries at the Blood Bank Midden-
Nederlandd for their assistance.
58 8
References s
1.. Walker RH: Technical Manual, ed 11. Bethesda, American Association of Blood
Banks,, 1993, p82.
2.. Jagodzinski L, Kraus F, Garrett P, Schumacher R, Manak M: Detection of hepatitis
BB viral sequences in early HBV infection. Transfusion 1994;34 (suppl):S148.
3.. McMahon BJ, Bender TR, Berquist KR, Schreeder MT, Harpster AP: Delayed
developmentt of antibody to hepatitis B surface antigen after symptomatic infection
withh hepatitis B virus. J Clin Microbiol 1981;14:130-4.
4.. Bréchot C, Degos F, Lugassy C, Thiers V, Zafrani S, Franco D, Bismuth H, Trépo C,
Benhamouu J-P, Wands J, Isselbacher K, Tiollais P, Berthelot P: Hepatitis B virus
DNAA in patients with chronic liver disease and negative tests for hepatitis B surface
antigen.. N Engl J Med 1985;312:270-6.
5.. Carman WF, Korula J, Wallace L, MacPhee R, Mimms L, Decker R: Fulminant
reactivationn of hepatitis B due to envelope protein mutant that escaped detection by
monoclonall HBsAgELISA. Lancet 1995;345:1406-7.
6.. Jongerius JM, Wester M, Cuijpers HTM, van Oostendorp WR, Lelie PN, van der Poel
CL,, van Leeuwen EF: New hepatitis B mutant form in a blood donor undetectable by
severall HBsAg screening assays. Transfusion 1998;38:56-9.
7.. NIH Consensus Statement on Infectious Disease Testing for Blood Transfusion.
Bethesda,, National Institutes of Health, 1995.
8.. Roberts P: Virus safety of plasma products. Med Virol 1996;6:25-38.
9.. Reesink HW, Leentvaar-Kuypers A, van der Poel CL, Lelie PN, Pietersz RNI, Mulder-
Folkertss DKF, Pieters T, van den Ende A, Schaasberg W, Coutinho RA: Non-A, non-
BB posttransfusion hepatitis in open hart surgery patients in The Netherlands:
Preliminaryy results of a prospective study; in: Zuckerman AJ (ed): Viral Hepatitis and
Liverr Disease. New York: Liss, 1988, pp 558-60.
59 9
60 0
Chapterr 5
Evaluationn of automated nucleic acid extraction
devicess for application in HCV NAT
J.M.. Jongerius1, M. Bovenhorst1, C.L. van der Poel1, J.A. van Hilten2,
A.C.M.. Kroes3, J.A. van der Does2, E.F. van Leeuwen1, R. Schuurman4
11 Blood Bank Midden-Nederland, Utrecht 22 Blood Bank Leiden-Haaglanden, Den Haag 33 Leiden University Medical Center, Leiden
44 University Hospital, Utrecht
Publishedd in: Transfusion 2000;40:871-4
61 1
62 2
Abstract t
Too further improve the safety of the blood supply, various national blood
transfusionn organizations presently use or are in the process of implementing routine HCV
NATT in minipools. According to the Committee for Proprietary Medicinal Products
(CPMP)) of the European Union, the HCV NAT detection limit of the assay should be 100
IUU per mL (270 geq/mL) for testing initial plasma pools. Paul Ehrlich Institute (PEI)
regulationss stipulate that 5,000 IU per mL (13,500 geq/mL) must be detected to calculate
thee amount contributed by individual donations composing the minipool. The sensitivity
forr HCV RNA extraction achieved by three commercially available laboratory kits was
compared. .
Nucleicc acids from l-in-3 serial dilutions of an HCV RNA run control (PeliSpy,
CLB)) were extracted with three kits (Cobas Amplicor, Roche Diagnostic Systems;
BioRobott 9604, Qiagen; and NucliSens Extractor, Organon Teknika). HCV PCR of all
extractss was performed using a second generation Cobas Amplicor HCV test and the
Cobass Amplicor analyzer.
Thee manual Cobas Amplicor, the BioRobot 9604, and the NucliSens Extractor
setupss allow a 95-percent HCV RNA detection limit of 129, 82, and 12 geq per mL,
respectively.. The maximal pool size for the manual Cobas Amplicor, the BioRobot 9604
andd the NucliSens Extractor kits that would still meet the PEI criteria for HCV NAT in
minipoolss was calculated at 104, 164, and 1125 donations, respectively.
Alll three HCV NAT kits evaluated meet the criteria set by CPMP and PEI. The highest
sensitivityy for HCV NAT screening can be achieved with the high-volume NucliSens
Extractorr method in combination with the Cobas Amplicor HCV v2.0 test on the Cobas
Amplicorr analyzer.
63 3
Introduction n
Too further improve the safety of plasma derivatives, the Committee for Proprietary
Medicinall Products (CPMP) of the European Union, as of January 1999, requires HCV
NATT of manufacturing plasma pools by a gene-amplification procedure such as PCR [1].
Onlyy products derived from HCV NAT-negative pools can be released. Positive HCV
NATT test results in a manufacturing plasma pool, often consisting of up to 6,000
donations,, will result in an excessive loss of plasma. Therefore, the CPMP recommended
thee use of so-called minipool screening [1]. Minipools may consist of up to 100 donations,
dependingg on the sensitivity of the system used [2-5]. According to the CPMP, the
detectionn limit of HCV NAT should be at least 100 IU per mL, which is equivalent to 270
geqq per mL [6], When testing initial plasma pools, the CPMP does not require the
resolutionn of positive HCV NAT minipools to the single donation level [1]. Paul Ehrlich
Institutee (PEI) regulations stipulate that at least 5,000 IU per mL, equivalent to 13,500 geq
perr mL, must be detected to calculate the amount contributed by the individual donations
composingg the minipool [6]. HCV NAT screening in a minipool setting will result in
positivee reactions with a frequency that depends on the incidence of HCV infection in the
donorr population and the specificity of the test method used [2].
Inn The Netherlands, routine minipool HCV NAT screening of all blood donations
hass been obligatory since July 1999 and is included in the release procedures for plasma,
RBCs,, and platelets. In case of a positive HCV NAT result in a minipool, the individual
donationn responsible for that result must be identified. Therefore, the magnitude of the
minipooll is also determined by the logistics for the provision of products with a short shelf
life,, such as platelet concentrates.
Automatedd HCV NAT equipment is not yet available for high-throughput routine
testingg of single blood donations. Recently, two automated devices for silica-based nucleic
acidd purification became commercially available (BioRobot 9604, Qiagen, Hilden,
Germany;; and NucliSens Extractor, Organon Teknika, Boxtel, The Netherlands). We
reportt the results of a study on HCV RNA extraction with both currently available
64 4
automatedd extraction devices and a manual extraction method (Cobas Amplicor, Roche
Diagnosticc Systems, Branchburg, NJ), all used in combination with a second-generation
HCVV test (Cobas Amplicor HCV v2.0) on an analyzer (Cobas Amplicor).
Materialss and Methods
HCVHCV RNA serial dilutions
Ann HCV RNA run control (PeliSpy, CLB, Amsterdam, The Netherlands) is a 1-in-
10000 dilution of the EUROHEP HCV genotype 1 plasma standard, which has been
characterizedd in two collaborative studies [7,8]. The PeliSpy HCV RNA run control
containss approximately 3,600 geq per mL and was used to prepare l-in-3 serial dilutions
inn normal human plasma that was negative for HCV RNA, HBsAg, and antibodies to
HCV,, HIV-1/2, HTLV-I/I I and Treponema pallidum [7,8], For each nucleic acid
extractionn technology, 10 replicates per dilution were tested in five runs for the detection
off HCV RNA.
HCVHCV RNA extraction
Thee manual Cobas Amplicor method. Viral RNA was extracted and purified from
2000 uL of plasma using the manual Cobas Amplicor HCV v2.0 test according to the
instructionss of the manufacturer. After completion of the extraction procedure, nucleic
acidss were dissolved in 200 uL of HCV v2.0 diluent.
Thee BioRobot 9604 method. Viral RNA was extracted and purified from 200 ^L
off plasma using a viral RNA test kit (QIAamp 96, Qiagen) for use with the BioRobot 9604
accordingg to the instructions of the manufacturer. After purification, nucleic acids were
elutedd in a volume of 80 uL of RNase-free elution buffer. The QIAamp 96 viral RNA test
kitt uses the selective binding properties of a silica-based membrane with a high
throughputt 96-well format and is designed for automated, simultaneous processing of
multiplee samples on the BioRobot 9604. With this device, up to 96 samples of 200 uT
eachh can be processed in less than 2 hours.
65 5
Thee NucliSens Extractor method. Viral RNA was extracted and purified from 1.0
mLL of plasma using the NucliSens Extractor according to the instructions of the
manufacturer.. After purification, nucleic acids were eluted in a 50-uL volume of elution
buffer.. The NucliSens Extractor uses the silica-based extraction method described by
Boomm et al. [9] for the purification of nucleic acids. The extractor allows sample volumes
upp to 2.0 mL and has a capacity of 10 samples per run with a processing time of less than
600 minutes.
InternalInternal control
Thee HCV internal control supplied with the Cobas Amplicor HCV v2.0 kit is an
RNAA transcript with primer-binding regions identical to those of the HCV target sequence
andd a unique probe-binding region that differentiates the internal control from the target
amplicon.. According to the instructions of the manufacturer, the internal control was
introducedd into each specimen with the lysis buffer and, as such, it monitored the
performancee of the laboratory procedures from extraction to amplification and detection.
Thee amount of internal control used in both automated extraction methods was guided by
thee prescribed amount in the manual Cobas Amplicor method and by the elution volumes
off the specific automated extraction procedures. Accordingly, the amount of internal
controll used as input in the amplification reactions was identical for each extraction kit.
DetectionDetection of HCV RNA
Immediatelyy upon extraction, 50 uL of eluted and purified nucleic acids was
amplifiedd and detected automatically using the Cobas Amplicor HCV v2.0 kit and
analyzer,, according to the manufacturer's instructions. The assay utilizes a nucleic acid
amplificationn method, RT-PCR, to generate amplified product material from HCV RNA
combinedd with nucleic acid hybridization to produce a colorimetric product for the
detectionn of HCV RNA [10]. A sample was HCV RNA-positive if the absorbance read
att 660 nm was £0.2, regardless of the internal control result. A sample was HCV RNA-
negativee if the absorbance at 660 nm was <0.1 and the internal control result was positive
66 6
(absorbancee at 660 nm was £0.2). A sample was HCV RNA-invalid if the absorbance at
6600 nm was <0.1 and the internal control result was negative (absorbance at 660 nm
<0.2).. Specimens with absorbance at 660 nm in the gray zone (£0.1 and <0.2) must be
testedd again in duplicate.
PlasmaPlasma equivalents analyzed in each extraction kit
Calculatedd from the extraction input plasma volumes, the elution volumes, and the
inputt volumes of the eluate in the RT-PCR method, the plasma equivalents analyzed in
thee manual Cobas Amplicor, the BioRobot 9604 and the NucliSens Extractor kits were,
respectively,, 50, 125, and 1000 uL.
Statistics Statistics
Thee detection limit for each of the assay kits was estimated using Probit analysis
[11].. Thus, the HCV RNA concentration that generates a positive PCR result in 95 percent
off cases (ED95), was calculated for each laboratory procedure.
Results s
Thee percentage of HCV NAT-positive replicates in the PeliSpy HCV RNA run
controll dilution series with manual Cobas Amplicor method, BioRobot 9604, and
NucliSenss Extractor, all combined with the HCV v2.0 test on the analyzer, is shown in
Tablee 1. By use of the manual Cobas Amplicor and the BioRobot 9604 laboratory
procedures,, HCV RNA was detected in 10 (100%) of 10 replicates tested at a
concentrationn of 133 geq per mL or higher, whereas, with the NucliSens Extractor kit,
HCVV RNA was detected in 10 (100%) of 10 replicates tested at a concentration of 44 geq
perr mL or higher. When HCV RNA concentrations declined, the number of positive
sampless dropped for all three methods. With the NucliSens Extractor kit, one invalid
resultt was seen in a replicate containing 1.6 geq per mL.
67 7
Thee HCV RNA ED95 for the manual Cobas Amplicor, the BioRobot 9604, and the
NucliSenss Extractor kits was calculated to be 129 geq per mL (95% CI, 70-590), 82 geq
perr mL (95% CI, 43-360), and 12 geq per mL (95% CI, 7-41), respectively. Extrapolated
forr minipools of 48 donations as used in The Netherlands, those three kits allow a 95-
percentt HCV RNA detection limit of 6192, 3936, and 576 geq per mL, respectively, as
calculatedd for the amounts from individual donations. The maximal pool size for the three
kitss that would still meet the PEI criteria for HCV NAT testing in minipools was
calculatedd at 104, 164 and 1125 donations, respectively.
TableTable 1. The percentage of HCV NAT-positive replicates in PeliSpy HCV RNA genotype
11 run control dilution series
Concentrationn of
HCVV RNA
(geq/mL) )
1200 0
400 0
133 3
44 4
15 5
5 5
1.6 6
0.5 5
0.2 2
** Cobas Amplicor.
HCVV RNA extraction setupp in combination with Cobas Amplicor
HCVV v2.0 test on
Manuall *
(nn =
100 0
100 0
100 0
50 0
30 0
0 0
10 0
10 0
0 0
10) )
% %
% %
% %
% %
% %
% %
% %
% %
% %
tt n = 9 due to an invalid test result.
thee Cobas
BioRobott 9604
(nn =
100 0
100 0
100 0
90 0
20 0
20 0
0 0
0 0
0 0
10) )
% %
% %
% %
% %
% %
% %
% %
% %
% %
analyzer r
NucliSens s
(n=10) )
1000 %
1000 %
1000 %
1000 %
900 %
900 %
111 %t
00 %
00 %
68 8
Discussion n
Thiss study shows that all three evaluated HCV NAT extraction systems combined
withh the HCV v2.0 test on an analyzer meet the criteria defined by the CPMP and PEI
whenn minipools of up to 104 donations are used. It is also shown that, to date, the
NucliSenss Extractor may provide the most sensitive HCV RNA detection in combination
withh the HCV v2.0 test on the analyzer.
Accordingg to the CPMP, the HCV NAT detection limit of the assay must be at least
1000 IU per mL. It has been reported by Lelie et al. [6] that 100 IU per mL is equivalent
too 270 geq per mL. Our results show that the manual Cobas Amplicor method, the
BioRobott 9604, and the NucliSens Extractor, all in combination with the Amplicor HCV
v2.00 test on the Cobas Amplicor analyzer, allow sufficiently sensitive detection to meet
CPMPP regulations.
PEII regulations stipulate that at least 5,000 IU per mL, which equals 13,500 geq
perr mL, must be detected to calculate the amount contributed by the individual donations
composingg the minipool [6]. Given the logistics for the production and release of platelet
concentrates,, minipools of a maximum of 48 donations for the HCV NAT screening
programm are used in The Netherlands. Extrapolated for use with such minipools, the
manuall Cobas Amplicor method, the BioRobot 9604, and the NucliSens Extractor allow
aa 95-percent HCV RNA detection limit of 6192, 3936, and 576 geq per mL, respectively,
ass calculated for the amounts from individual donations. Thus, HCV RNA extraction by
alll three kits provides sufficiently sensitive amplification and detection to meet PEI
regulations. .
Itt has been reported that silica extraction of RNA will improve the sensitivity of
NATT by allowing the concentration of genomicc target material [10]. In that study, an input
samplee volume of 50 uL was used [10]. The BioRobot 9604 and the NucliSens methods,
bothh based on the principle of silica extraction, employ input sample volumes of 200 uL
andd 1 mL, respectively. Our results demonstrate that, in comparison with the BioRobot
96044 method, the NucliSens method results in an automated HCV NAT setup
69 9
approximatelyy seven times more sensitive. Thus, the difference in the sensitivity of the
automatedd laboratory procedures is in line with the difference in plasma equivalents
analyzedd in the Cobas Amplicor HCV PCR. Given the fact that the sample input volume
mayy be enlarged up to 2 mL with the use of the NucliSens kit, a 90-percent lower HCV
RNAA detection limit may theoretically be achieved in comparison with the BioRobot 9604
kit. .
Thee sensitivity of the HCV NAT, logistics of scale, and blood component provision
playy a role in the decision as to which NAT system to choose. We conclude that all
evaluatedd kits meet the criteria for (minipool) HCV NAT set by the CPMP and PEL The
highestt sensitivity for HCV NAT screening can be achieved using a high-volume
NucliSenss Extractor method in combination with the Cobas Amplicor HCV v2.0 test on
thee Cobas Amplicor analyzer.
Acknowledgment t
Thee authors thank Prof Dr. W.G. van Aken at the Central Laboratory of The
Netherlandss Red Cross Blood Transfusion Service (Amsterdam) for review of the
manuscript;; W.T. Schaasberg for his support with the statistical analysis; and M.J. van
Bussell and E.C.J. Claas, PhD, at Leiden University Medical Center, Leiden, The
Netherlands,, for technical assistance.
References s
1.. Flanagan P, Snape T. Nucleic acid technology (NAT) testing and the transfusion
service:: a rationale for the implementation of minipool testing. Transfus Med
1998;8:9-13. .
2.. Cardoso MS, Koerner K, Kubanek B. Mini-pool screening by nucleic acid testing for
hepatitiss B virus, hepatitis C virus, and HIV: preliminary results. Transfusion
1998:38:905-7. .
70 0
3.. Yerly S, Pedrocchi M, Perrin L. The use of polymerase chain reaction in plasma pools
forr the concomitant detection of hepatitis C virus and HTV type 1 RNA. Transfusion
1998:38:908-14. .
4.. Lefrère JJ, Coste J, Defer C, et al. Screening blood donations for viral genomes:
multicenterr study of real-time simulation using pooled samples on the model of
hepatitiss C virus RNA detection. Transfusion 1998;38:915-23.
5.. Roth WK, Weber M, Seifried E. Feasibility and efficacy of routine PCR screening of
bloodd donations for hepatitis C virus, hepatitis B virus, and HTV-1 in a blood-bank
setting.. Lancet 1999;353:359-63.
6.. Lelie PN, Cuijpers HT, van Drimmelen AA. Quality assessment of hepatitis C virus
nucleicc acid amplification methods. Infusionther Transfusionmed 1998;25:102-10.
7.. Zaaijer HL, Cuijpers HT, Reesink HW, et al. Reliability of polymerase chain reaction
forr detection of hepatitis C virus. Lancet 1993;341:722-4.
8.. Damen M, Cuijpers HT, Zaaijer HL, et al. International collaborative study on the
secondd EUROHEP HCV-RNA reference panel. J Virol Methods 1996:58:175-85.
9.. Boom R, Sol CJ, Salimans MM, et al. Rapid and simple method for the purification
off nucleic acids. J Clin Microbiol 1990;28:495-503.
10.. Young KK. Resnick RM, Myers TW. Detection of hepatitis C virus RNA by a
combinedd reverse transcription polymerase chain reaction assay. J Clin Microbiol
1993;31:882-6. .
11.. Finney DJ. Probit analysis. Cambridge, UK: Cambridge University Press, 1971.
71 1
72 2
Chapterr 6
Evaluationn of HIV nucleic acid tests in combination
withh NucliSens Extractor for application in minipool
screening g
J.M.. Jongerius1, M. Sjerps2, R. van Dijk2, P.N. Lelie2, A.A.J. van Drimmelen2,
C.L.. van der Poel3, H.T.M. Cuijpers2
11 Sanquin Blood Bank Midden-Nederland, Utrecht 22 Viral Diagnostic Laboratory, Sanquin-CLB, Amsterdam
33 Sanquin Blood Supply Foundation, Amsterdam
Preliminaryy study for chapter 7
73 3
74 4
Introduction n
Routinee HCV NAT minipool screening (max 48 donations) with negative NAT as
releasee criterion for all blood components was implemented in July 1999 in The
Netherlandss [1]. Since that time HCV NAT screening is performed in four identical NAT
laboratoriess using the NucliSens Extractor (Organon Teknika, Boxtel, The Netherlands)
inn combination with the Cobas Amplicor HCV v2.0 test (Roche Diagnostic Systems,
Branchburg,, NJ) [1]. Late 1999, the Dutch blood supply foundation decided to expand the
HCVV NAT minipool screening program with routine HTV NAT. The NucliSens HTV-1 QL
testt (Organon Teknika) and the AmpliScreen HTV-1 vl .5 test (Roche Diagnostic Systems),
bothh combined with NucliSens Extractor, were optional for future implementation.
Thee aim of the present study was to compare the specificity, robustness and
sensitivityy of the NucliSens HTV-1 QL test and the AmpliScreen HTV-1 test, both
combinedd with NucliSens Extractor, to allow proper selection.
Materialss and Methods
ReferenceReference materials
Specificityy of the HTV NAT tests was determined by analyzing at least 100 HTV-1
RNA-negativee human plasma minipools of 48 donations each.
Robustnesss of the HTV NAT tests was determined by alternate testing of 20 HTV-1
RNA-high-positivee samples (106 geq per mL HIV-1 RNA) and 20 HIV-1 RNA-negative
sampless in the first two and last two extraction runs of the study. Furthermore, run
controlss containing varying levels of HIV-1 RNA (PeliSpy S2082, Sanquin-CLB,
containingg 375 geq per mL HTV-1 RNA, and 1:10 diluted PeliSpy containing 38 geq per
mLL HIV-1 RNA, and a Organon Teknika VQA run control containing 250 geq per mL
HIV-11 RNA) were analyzed.
Sensitivityy of the HIV NAT tests was determined by analyzing HIV-1 RNA
genotypee B and genotype E reference panels (PeliCheck S2092 and S2094, Sanquin-CLB,
75 5
Amsterdam,, The Netherlands). For each HIV NAT test, up to 8 replicates per dilution
weree analyzed.
NucleicNucleic acid extraction protocol
Thee NucliSens Extractor, which uses the silica based extraction method described
byy Boom et al. [2], allows sample volumes up to 2 mL. The capacity of the NucliSens
Extractorr is 10 samples per run with a processing time of approximately 45 minutes. In
ourr study, viral RNA was extracted and purified from 2 mL EDTA-plasma samples using
NucliSenss automated isolation reagents and the NucliSens Extractor according to
instructionss of the manufacturer [2]. After completion of the extraction procedure, nucleic
acidss were eluted and dissolved in a mean volume of 45.4 uL (SD = 4.2 uL) elution
buffer. .
InternalInternal control solutions
Thee system control solution for the NucliSens HIV-1 QL test was used according
too the instructions of the manufacturer. The HIV-1 RNA copy number in this system
controll solution is not provided by the manufacturer.
Forr the AmpliScreen HIV-1 vl.5 test procedure, 7 uL off AmpliScreen Multiprep
Internall Control (84 copies of a HIV-1 RNA transcript) was added to the plasma lysis
reagentt mixture.
AmplificationAmplification and detection of HIV-1 RNA
NucliSenss HIV-1 QL test: immediately upon extraction, eluted and purified
nucleicc acids in samples of 5 uL were amplified and detected automatically using
respectivelyy NASBA amplification technology in combination with
electrochemiluminescencee (ECL) detection technology (Organon Teknika) according to
thee instructions of the manufacturer. The ECL results were automatically measured by the
NucliSenss reader and interpreted by the NucliSens user software. An amplification
reactionn was qualified as valid or invalid depending on the ECL signals obtained for the
76 6
systemm control HTV-1 RNA. A sample for which the amplification reaction was qualified
ass valid, but for which the wild type (WT) HIV-1 RNA ECL signal was below the
calculatedd cut-off value was considered HTV-1 RNA-negative. A sample was considered
HIV-11 RNA-positive if the WT HIV-1 RNA ECL signal was equal or higher than the
calculatedd cut-off value in combination with a valid amplification reaction.
AmpiiScreenn HIV-1 vl.5 test: Immediately upon extraction, 25 uL of eluted and
purifiedd nucleic acids were amplified and detected automatically using the AmpliScreen
HIV-11 vl.5 test on the Cobas analyzer according to the instructions of the manufacturer.
Thee test utilizes a nucleic acid amplification method, RT-PCR, to generate amplified
productt material from HIV-1 RNA combined with nucleic acid hybridization to produce
aa colorimetric product for the detection of HIV-1 RNA. A sample was considered HIV-1
RNA-positivee if the WT absorbance at 660 nm was > 0.2 regardless of the internal control
(IC)) result and negative if the WT absorbance at 660 nm was < 0.2 in combination with
ann IC absorbance at 660 nm > 0.2. The amplification and detection procedure was
consideredd invalid if the WT absorbance at 660 nm was < 0.2 in combination with an IC
absorbancee at 660 nm < 0.2.
PlasmaPlasma equivalents analyzed
Calculatedd from the extraction input plasma volumes, the mean NucliSens
Extractorr elution volumes, and the input volumes of the eluate in the PCR, the plasma
equivalentss analyzed in the NucliSens HIV-1 QL and the AmpliScreen HIV-1 vl.5 test
were,, respectively, 220 and 1100 uL.
Statistics Statistics
Thee 50% and 95% detection limits for each of the HIV NAT tests in combination
withh NucliSens extraction were estimatedd using Probit analysis [3], Thus, the HTV-1 RNA
concentrationss which generate a positive PCR result in respectively 50 and 95 percent of
casess (ED50 and ED95) were calculated for each HIV NAT testing procedure [4].
77 7
Results s
SpecificitySpecificity and robustness
Thee results of the specificity and run control testing are summarized in Table 1.
TableTable 1. Specificity and run control testing ofNucliSens extraction in combination with
thethe NucliSens HIV-1 QL or with the AmpliScreen HIV-1 vl.5 test
Resultss of HIV-1 RNA testing
HIVV NAT test Type of samples
NucliSenss Minipools *
HIV-11 QL test PeliSpy t
RC-VQAA %
AmpliScreenn Minipools *
HIV-11 vl.5 test PeliSpy t
PeliSpyy 1:10 §
n n
112 2
14 4
10 0
103 3
10 0
10 0
Negative e
112 2
0 0
0 0
101 1
0 0
0 0
Positive e
0 0
14 4
10 0
22 **
10 0
10 0
Invalid d
0 0
0 0
0 0
0 0
0 0
0 0
** = HIV-1 RNA-negative human plasma minipools of 48 donations each.
tt = PeliSpy run control S2082 containing 375 geq per mL HIV-1 RNA.
XX = Organon Teknika VQA run control containing 250 geq per mL HIV-1 RNA.
§§ = 1:10 diluted PeliSpy run control containing 38 geq per mL HIV-1 RNA.
*** = repeat amplification and detection of one of these minipools revealed a negative result. Due
too sample exhaustion, repeat testing of the second minipool was not possible.
Duringg cross contamination testing in the first two and last two extraction runs of
thee study, 1 invalid test result was observed using the NucliSens HIV-1 QL test. Using
AmpliScreenn HTV-1 vl.5 test, 1/20 (5%) HIV-1 RNA-negative samples tested positive.
Repeatt amplification and detection of the corresponding eluate revealed a negative result
forr HIV-1 RNA.
78 8
Sensitivity Sensitivity
Thee percentage of HIV-1 NAT-positive replicates in PeliCheck HIV-1 RNA
genotypee B and E reference panels is shown in Table 2. No invalid test results were
observed. .
TableTable 2. The percentage of HIV-1 NAT-positive results in PeliCheck HIV-1 RNA genotype
BB and E panels using NucliSens extraction in combination with the NucliSens HIV-1 QL
oror with the AmpliScreen HIV-1 vl.5 test
NucliSenss HIV-1 QL AmpliScreen HIV-1 vl .5
(nn = 8) (n = 8)
Concentrationn of
HIV-11 RNA
(geq/mL) )
750 0
250 0
75 5
25 5
7.5 5
2.5 5
0.75 5
0.25 5
0.075 5
HIV-1 1
Genotypee B
%% positive
100 0
100 0
100 0
62.5 5
37.5 5
0 0
12.5 5
0 0
NTT J
HIV-1 1
Genotypee E
%% positive
100 0
100 0
87.5 5
50 0
40* *
500 t
Of f
NT T
NT T
HIV-1 1
Genotypee B
%% positive
100 0
100 0
100 0
100 0
100 0
37.5 5
12.5 5
12.5 5
NT T
HIV-1 1
Genotypee E
%% positive
100 0
100 0
100 0
100 0
87.5 5
25 5
37.5 5
0 0
25 5
* nn = 5.
t nn = 2.
JJ = not tested.
Thee results of Probit analysis (ED50 and ED95) for NucliSens extraction combined with
thee NucliSens HIV-1 QL and the AmpliScreen HIV-1 vl .5 test are shown in Table 3.
79 9
TableTable 3. Results ofProbit analysis for NucliSens extraction combined with the NucliSens
HIV-1HIV-1 QL and the AmpliScreen HIV-1 vl.5 test
HTV-11 test
NucliSens s
HIV-11 QL
AmpliScreen n
HIV-11 vl.5
Detectionn limit
ED50* *
ED95t t
ED50 0
ED95 5
** = HIV-1 RNA concentration that
tt = HIV-1 RNA concentration that
HIV-11 genotype B
geq/mLL (95% CI)
111 (6-23)
1022 (42-670)
22 (1-4)
133 (6-93)
generatess a positive result
generatess a positive result
HIV-11 genotype E
geq/mLL (95% CI)
122 (3-29)
1900 (67-5040)
22 (0.4-8)
399 (8-16634)
inn 50-percent of cases.
inn 95-percent of cases.
Discussion n
Thiss validation study shows that NucliSens extraction in combination with the
AmpliScreenn HTV-1 vl.5 test may provide the most sensitive HIV-1 RNA genotype B and
EE detection of the two test options. The AmpliScreen HTV-1 test allows a 95 percent HTV-
11 RNA detection limit of- 625 geq per mL for HIV-1 genotype B and - 1875 geq per mL
forr HIV-1 genotype E per individual donation in a pool composed of 48 donations. The
NucliSenss HTV-1 QL test allows a 95-percent HTV-1 RNA detection limit of- 4900 geq
perr mL for HIV-1 genotype B and -9125 geq per mL for HIV-1 genotype E per
individuall donation in a pool composed of 48 donations. Thus, compared to the NucliSens
HTV-11 QL test, the AmpliScreen HIV-1 vl.5 test is approximately eight times more
sensitivee in detecting HIV-1 RNA genotype B and five times more sensitive in detecting
HTV-11 RNA genotype E. The difference in the sensitivity of the methods is in line with
thee difference in plasma equivalents analyzed.
Alll 112 minipools analyzed with the NucliSens HIV-1 QL test were found to be
HIV-11 RNA-negative, whereas 2/103 (1.9%) minipools analyzed with the AmpliScreen
HTV-11 vl.5 test were initially false-positive. An additional false-positive result was
80 0
observedd with the AmpliScreen HTV-1 vl .5 test during alternate testing of HTV-1 RNA-
high-positivee and negative samples. All three false-positive test results were found at the
beginningg of the study. False-positivity was probably caused using a contaminated pipette
set.. No more false-positive results were found after replacement of this set. Repeat
amplificationn and detection of 2/3 false-positive samples revealed negative results.
Unfortunately,, due to sample exhaustion, repeat testing of the third false-positive sample
wass not possible. Thus, in comparison with the AmpliScreen HIV-1 vl.5 test, the
NucliSenss HIV-1 QL test may provide better specificity.
Inn our study, during alternate testing of HIV-1 RNA-high-positive and negative
samples,, 1 false-positive result was observed using the AmpliScreen HTV-1 test. Repeat
amplificationn and detection of the corresponding eluate was negative for HTV-1 RNA.
Bothh HIV NAT tests detected all HIV-1 RNA-positive run controls. Failure of internal
controll detection (invalid result) was observed in 1/307 (0.3%) samples analyzed with the
NucliSenss HIV-1 QL test. No internal control failure was observed in analyzing 299
sampless with the Ampliscreen HTV-1 vl.5 test. Thus, the robustness of both HTV NAT
testt options is comparable.
Inn conclusion, NucliSens extraction in combination with the AmpliScreen HIV-1
vl.55 test may provide better sensitivity and sufficient specificity and robustness for
applicationn in routine HTV NAT minipool screening. NucliSens extraction in combination
withh the AmpliScreen HIV-1 vl.5 test was more extensively studied before actual
implementationn by the four Dutch NAT laboratories was mandated [5].
Acknowledgment t
Thee authors thank Prof. Dr. W.G. van Aken at the Sanquin-CLB, Amsterdam, The
Netherlands,, and Dr. E.F. van Leeuwen at the Sanquin Blood Bank Midden-Nederland,
Utrecht,, The Netherlands, for review of this chapter.
81 1
References s
1.. Cuijpers HTM, Molijn MHJ, Bos HJ, et al. Validation of NucliSens Extractor in
combinationn with the hepatitis C virus Cobas Amplicor 2.0 assay in four laboratories
inn The Netherlands utilizing nucleic acid amplification technology for blood screening.
Voxx Sanguinis 2001 ;81:12-20.
2.. Boom R, Sol CJA, Salimans MMM, et al. Rapid and simple method for purification
off nucleic acids. J Clin Microbiol 1990;28:495-503.
3.. Finney DJ. Probit analysis. Cambridge, UK: Cambridge University Press, 1971.
4.. Jongerius JM, Bovenhorst M, van der Poel CL, et al. Evaluation of automated nucleis
acidd extraction devices for application in HCV NAT. Transfusion 2000;40:871-4.
5.. Jongerius JM, Sjerps M, Cuijpers HTM, et al. Validation of the AmpliScreen HIV-1
vl.55 and HCV v2.0 test combined with NucliSens Extractor for application in
minipooll screening. Transfusion, accepted for publication.
82 2
Chapterr 7
Validationn of the NucliSens Extractor combined with
thee AmpliScreen HIV vl.5 and HCV v2.0 test for
applicationn in NAT minipool screening
J.M.. Jongerius1, M. Sjerps2, H.T.M. Cuijpers2, A.A.J, van Drimmelen2,
C.L.. van der Poel3, H.W. Reesink2, M.H.J. Molijn4, G.A.H. Peeters5,
A.P.W.. Peeters6, P.N. Lelie2
11 Sanquin Blood Bank Midden-Nederland, Utrecht 22 Sanquin-CLB, Viral Diagnostic Laboratory, Amsterdam
33 Sanquin Blood Supply Foundation, Amsterdam 44 Sanquin Blood Bank Zuid-West Nederland, Rotterdam
55 Sanquin Blood Bank Geldersche Rivieren, Nijmegen 66 Sanquin Blood Bank Noord Nederland, Groningen
Transfusion,, accepted for publication
83 3
84 4
Abstract t
Routinee HCV NAT minipool screening (48 donations) of all blood donations was
implementedd in July 1999 and was combined with HIV NAT in November 2000. This
reportt describes the validation of the HIV and HCV NAT method. In addition, the results
observedd during the first eleven months of routine HIV and HCV NAT screening are
reported. .
Nucleicc acid was extracted from 2 mL plasma samples using an automated silica-
basedd extraction method (NucliSens Extractor, Organon Teknika). Eluates were tested
withh RT-PCR (AmpliScreen HIV-1 vl.5 and AmpliScreen HCV v2.0 test, Roche
Diagnosticc Systems). HIV-1 and HCV RNA reference panels and run controls (PeliCheck
andd PeliSpy, respectively, Sanquin-CLB), and human plasma minipools were used for
validationn of the NAT methods.
Thee 95-percent detection limit (and 95% confidence interval) for HIV-1 RNA
genotypee B, HIV-1 RNA genotype E and HCV RNA genotype 1 was 32 (19-76), 30 (17-
72)) and 21 (13-44) geq per mL, respectively. At analyzing 2332 samples for HtV-1 RNA
andd 2644 samples for HCV RNA during an initial validation phase, 13 (0.56%) and 12
(0.45%)) invalid test results were found, respectively. The run controls of either 140 geq
perr ml or 38 geq per mL were found positive in 256/257 (99.6%) and 74/78 (94.9%) of
HIV-11 RNA tests, respectively, and in 319/322 (99.1%) and 50/50 (100%) of HCV RNA
tests,, respectively. Thereafter, over 19600 samples (minipools and run controls) were
analyzedd during the first eleven months of routine HIV and HCV screening. Invalid test
resultss for HIV-1 RNA and HCV RNA were found in 1.1% and 1.07% of the samples
tested,, respectively. HIV-1 RNA minipool testing resulted in 27 (0.16%) initial false-
positivee results and 3 (0.02%) confirmed positive results. HCV RNA minipool testing
resultedd in 4 (0.02%) initial false-positive results and 5 (0.02%) confirmed positive
results.. No seronegative window donations were detected. The run controls of either 140
85 5
geqq per ml or 38 geq per mL were respectively found positive in 2201/2219 (99.2%) and
534/5600 (95.4%) of the HTV-1 RNA tests and in 2201/2218 (99.2%) and 534/565 (94.5%)
oftheHCVV RNA tests.
Routinee HIV and HCV NAT minipool screening of 48 donations, utilizing
NucliSenss Extractor combined with the AmpliScreen HIV-1 vl.5 and AmpliScreen HCV
v2.00 test, meets the sensitivity criteria set by the regulatory bodies and provides sufficient
specificityy and robustness for timely release of blood donations.
Introduction n
Fromm July 1999, routine HCV NAT screening of minipools of 48 donations was
startedd in The Netherlands in four NAT laboratories [1,2]. The HCV NAT screening
methodd is supported by a custom built Pooling Management System (PMS) software
packagee [1]. Plasma minipools are prepared using robotic samplers (Tecan Genesis, Tecan
Groupp Ltd, MSnnedorf, Switserland) [1]. HCV RNA is isolated using an automated silica-
basedd extraction method (NucliSens Extractor, Organon Teknika, Boxtel, The
Netherlands)) and amplified and detected using an automated RT-PCR system (Cobas
Amplicorr HCV v2.0, Roche Diagnostic Systems, Branchburg, NJ) [1]. The overall 95-
percentt detection limit of this test system was reported to be ~ 30 geq per mL (~ 8 IU per
mL)) per tested pool and - 1500 geq per mL (~ 400 IU per mL) per individual donation
[1].. This detection limit lies far below the detection limit of 5000 IU per mL and 5000 geq
perr mL per individual donation required by the Paul Ehrlich Institute (PEI) and the Food
andd Drug Administration (FDA), respectively [2-6].
Too diminish the risk of HTV transmission by blood donated during the seronegative
windoww period, Sanquin, the Dutch blood supply foundation, decided to expand the HCV
NATT minipool screening programme with HIV NAT testing. Both the AmpliSreen HIV-1
vl.55 test (Roche Diagnostic Systems) and the NucliSens HIV-1 QL test (Organon
Teknika)) could be combined with the NucliSens Extractor and were thus considered for
86 6
implementation.. In an initial validation study it was shown that the AmpliScreen HTV-1
vll .5 test was more sensitive than the NucliSens HIV-1 QL test in detecting HIV-1 RNA
standardd dilutions [7]. Therefore it was decided to combine the NucliSens Extractor with
thee AmpliScreen HIV-1 vl.5 test for a more extensive validation in the Sanquin NAT
screeningg laboratories. In addition, the routinely used Amplicor HCV v2.0 test was
replacedd by the AmpliScreen HCV v2.0 test and consequently validated.
Thiss report describes the validation of the NucliSens Extractor combined with the
AmpliScreenn HIV-1 vl.5 and the AmpliScreen HCV v2.0 test for application in routine
NATT screening (this test system is further called the NucliSens-AmpliScreen method). In
addition,, the (quality control) results observed during the first eleven months of routine
HIVV and HCV NAT minipool screening are reported.
Materialss and Methods
StudyStudy design
Referencee panels (PeliCheck HIV-1 RNA genotype B, HIV-1 RNA genotype E,
andd HCV RNA genotype 1, Sanquin-CLB, Amsterdam, The Netherlands) were tested 24
timess in separate runs to determine the sensitivity for HIV-1 and HCV RNA.
Robustnesss and specificity was determined in an initial validation phase by
alternatee testing of human plasma minipools with multi-marker run controls (PeliSpy,
Sanquin-CLB,, Amsterdam, The Netherlands) containing either 140 or 38 geq per mL
HIV-11 and HCV RNA (further called the RC140 and the RC38, respectively).
Thee (quality control) results obtained during the first eleven months of routine
NATT minipool screening were analyzed. The samples obtained from blood donations were
processedd without pre-selection for serological test results. A criterion stipulated by
Sanquinn was that at least 98% valid results must be found at initial testing. Confirmation
off NAT-positives was performed at the Viral Diagnostic Laboratory of Sanquin-CLB
usingg the mandatory test algorithms for the confirmation of HIV and or HCV infection
[8].. For routinely monitoring the sensitivity of test runs, a RC140 is included as a "stop
87 7
orr go" control in each Cobas A-ring, whereas, at regular intervals, also a RC38 is tested.
Thee performance of both the RC140 and the RC38 was examined. Additionally, for the
RC38,, the geometrical mean of the optical detection to cutoff ratio (OD:CO) for both
wild-typee (WT) and internal control (IC) was monitored. If the IC result was <2.5, the
individuall test was considered to be of too low sensitivity and the corresponding RC38
responsee values were excluded from the calculations.
HIVHIV and HCV RNA extraction
Twoo mL aliquots of plasma pools were mixed with 9 mL of lysis buffer using the
NucliSenss isolation kit (Organon Teknika). Thereafter, 11 JOL of HIV-1 and HCV IC
(equalss ~ 130 copies) was added to the sample lysis mixture. The IC is part of the
Amplii Screen multiprep specimen preparation reagents kit (Roche Diagnostic Systems) and
servess as an extraction and amplification control for each processed sample and control.
Nucleicc acid was released by incubating the sample lysis mixture for 30 minutes at 37 °C.
Nucleicc acid was then absorbed to 50 ul of added silica suspension using end-over-end
inversionn for 5 minutes. After pelleting the silica particles by centrifugation for 3 minutes
att 1500 #,10 mL of the plasma lysis buffer mixture was removed (so-called 'decant'
method)) [1]. The silica pellet was then resuspended in the residual 1 mL content of the
lysiss tube and transferred to a plastic NucliSens Extractor cartridge. Automated NucliSens
Extractorr (Organon Teknika) EluHigh procedure, which is based on the Boom silica
extractionn method, was used for purification of nucleic acids [1-3,9], With this procedure,
thee silica particles were subjected to a number of washing cycles with buffer and organic
solutionn by an airpump mechanism, while the silica particles were captured by a
membranee in the cartridge. The silica particles were then dried at 56 °C and nucleic acid
wass eluted and dissolved in a mean volume of 68 uL elution buffer (SD = 6.1 uL). The
capacityy of NucliSens Extractor is 10 samples per run with a processing time of
approximatelyy 45 minutes.
88 8
AmplificationAmplification and detection ofHIV-1 RNA and HCVRNA
Accordingg to the instructions of the manufacturer, 25 uL of eluate containing
purifiedd nucleic acid, equivalent to an input of isolated RNA from ~ 735 uL of the original
plasmaa sample, was amplified and detected automatically using the AmpliScreen HIV-1
vl.55 test and the AmpliScreen HCV v2.0 test on the Cobas analyzer (Roche Diagnostic
Systems).. A sample was considered positive for the target RNA if the WT absorbance at
6600 nm was > 0.2 regardless of the IC result. A sample was considered negative for the
targett RNA if the WT absorbance at 660 nm was < 0.2 in combination with a IC
absorbancee at 660 nm > 0.2. The procedure was considered invalid if the WT absorbance
att 660 nm was < 0.2 in combination with an IC absorbance at 660 nm < 0.2.
PootingPooting Management System (PMS)
PMSS (Tecan software in collaboration with Organon Teknika) was earlier
developedd for the German blood banks and adapted for the needs of the Dutch blood
bankss to introduce HCV NAT minipool screening [1]. Over time, PMS, was modified in
suchh a way that it had complete positive sample identification and sample tracking from
thee individual donation tubes to primary and secondary plasma pools, archive plates and
alll steps from the extraction process to the interpretation and transfer of test results [1].
Thee manual transfer of material from individual donations, plasma pools, lysis tubes,
NucliSenss Extractor cartridges, eluates and Cobas A rings is controlled and registered by
barcodee transition protocols in PMS [1]. The AmpliScreen results are automatically
transferredd to the PMS database. Due to the decision to expand the HCV NAT screening
program,, it was decided to upgrade PMS with functionality's for future NAT (HIV and
HBV).. The PMS resolution algorithm for HCV NAT screening, as previously described
byy Cuijpers et al. [1], is adopted for simultaneous HIV and HCV NAT minipool
screening.. After acceptance of this upgraded multifunctional PMS version (v3.0) it was
implementedd in October 2000 to support HIV and HCV NAT minipool screening of all
bloodd donations.
89 9
Statistics Statistics
Thee 50, 95 and 99-percent detection limits for each test were estimated by probit
analysiss [10]. Thus, the RNA concentrations which generate a positive PCR result in 50,
955 and 99 percent of cases (ED50, ED95 and ED99) were calculated for both HIV and HCV
NAT. .
Performancee of the RC38 was analyzed by calculating the geometrical mean of the
OD:COO for WT and IC. Results with an IC absorbance < 2.5 were excluded from the
calculations. .
Results s
SensitivitySensitivity of HIV and HCV NA T
Tablee 1 shows the percentages ofNAT-positive results in HIV-1 RNA genotype
BB and HIV-1 RNA genotype E and HCV RNA genotype 1 panels and the corresponding
detectionn limit estimates (ED50, ED95 and ED99) of the NucliSens-Ampliscreen method.
Thee sensitivity for HIV-1 RNA genotype B and E was comparable.
SpecificitySpecificity and robustness of HIV and HCVNA T in initial validation phase
Testingg of 2332 samples (1997 plasma minipools of 48 donations and 335 run
controls)) for HIV-1 RNA showed 13 (0.56%) invalid test results: 5 (0.21%) due to an IC
absorbancee at 660 nm < 0.2 and 8 (0.34%) resulting from (technical) problems with either
NucliSenss extraction (n = 5, 0.21%) or Cobas amplification and detection (n = 3,
0.13%).. Initial false-positive results were observed in 15/1997 (0.75%) minipools tested.
Retestingg of these 15 minipools showed negative results. Of 257 RC140, 256 (99.6%)
testedd positive and 1 (0.4%) invalid. Of 78 RC38 samples, 74 (94.9%) tested positive and
44 (5.1%) negative.
Testingg of 2644 samples (2272 minipools of 48 donations and 372 run controls)
forr HCV RNA showed 12 (0.45%) invalid test results: 7 (0.26%) due to an IC absorbance
att 660 nm < 0.2 and 5 (0.19%) due to (technical) problems with NucliSens extraction. All
90 0
22722 minipools were found to be negative for HCV RNA. Of 322 RC140, 319 (99.07%)
testedd positive and 3 (0.93%) invalid. All 50 RC38 samples tested positive.
TableTable 1. The percentage of NAT-positive results in Pelicheck HIV-1 RNA genotype B,
HIV-1HIV-1 RNA genotype E, and HCV RNA genotype 1 panels and the corresponding
detectiondetection limit estimates using NucliSens EluHigh extraction in combination with the
AmpliScreenAmpliScreen HIV-1 vl.5 and HCV v 2.0 test
Concentrationn of
HIV-11 RNA
(geq/mL) )
75 5
25 5
7.5 5
2.5 5
0.75 5
0.25 5
0.075 5
AmpliScreen n
HIV-11 vl.5 test
HIV-1 1
Genotypee B
(nn = 24)
1000 %
966 %
588 %
255 %
133 %
00 %
00 %
HIV-1 1
genotypee E
(nn = 24)
1000 %
966 %
677 %
466 %
44 %
88 %
00 %
Concentrationn of
HCVV RNA
(geq/mL) )
270 0
140 0
100 0
38 8
11 1
4 4
1 1
0.4 4
0.1 1
AmpliScreen n
HCVV v2.0 test
HCV V
Genotypee 1
(nn = 24)
1000 %
1000 %
1000 %
1000 %
799 %
500 %
88 %
00 %
00 %
Probitt analysis * Probit analysis *
ED500 (95% CI) 5(3-7) 3(2-5) ED50 (95% CI) 4(3-6)
ED95(95%CI)) 32(19-76) 30(17-72) ED95 (95% CI) 21(13-44)
ED99(95%CI)) 72(37-222) 74(36-236) ED99 (95% CI) 40(22-108)
** results expressed in geq/mL.
91 1
PerformancePerformance of the NucliSens-AmpliScreen method in routine screening
Thee results obtained during the first eleven months of routine HTV and HCV NAT
minipooll screening using the NucliSens-AmpliScreen method are summarized in Table
2.. Invalid test results for HIV-1 RNA and HCV RNA were found in 1.1% and 1.07% of
thee minipools tested, respectively. HIV-1 RNA testing revealed 0.16% initial false-
positivee and 0.02% confirmed positive results. HCV RNA testing revealed 0.02% initial
false-positivee and 0.03% confirmed positive results. No seronegative window period
donationss were found.
PerformancePerformance of the run controls in routine screening
Ass shown in Table 2, the RC140 "stop or go" run control was positive in 99.2%
off both HIV-1 RNA and HCV RNA tests. The Multi Marker RC38 was chosen to monitor
thee sensitivity of the NucliSens-AmpliScreen test runs. In accordance with the observed
95-percentt detection limit for HIV-1, the RC38 was found positive 95.4% of cases. The
performancee of the RC38 was below the expected detection rate for HCV (actual 94.5%,
expectedd near 99%). The performance of the RC38 was also monitored by calculating the
geometricall mean of OD:CO for both WT and IC. The results are summarized in Table
3. .
92 2
TableTable 2. Results obtained during the first eleven months of routine HIV and HCV NAT
screeningscreening using NucliSens EluHigh extraction in combination with the AmpliScreen HIV-1
vl.5vl.5 and the AmpliScreen HCV v2.0 test
Numberr of samples analyzed *
Invalidd test results:
NucliSenss Extractor failure (%) t
Cobass Analyzer failure (%) %
Internall control negative (%)
Operationall failure (%)
Totall invalid (%)
Multii Marker run control 140 geq per mL:
Numberr analyzed
Positivee (%)
Negativee (%)
Invalidd (%)
Multii Marker run control 38 geq per mL:
Numberr analyzed
Positivee (%)
Negativee (%)
Invalidd (%)
Minipools: :
Numberr analyzed
False-positivee (%)
Confirmedd positive (%)
AmpliScreen n
HIV-11 vl.5 test
19672 2
1188 (0.6)
166 (0.08)
611 (0.31)
222 (0.11)
2177 (1.1)
2219 9
22011 (99.2)
33 (0.1)
155 (0.7)
560 0
5344 (95.4)
244 (4.3)
22 (0.4)
16893 3
277 (0.16)
33 (0.02)
AmpliScreen n
HCVV v 2.0 test
19632 2
1188 (0.6)
288 (0.14)
455 (0.23)
200 (0.1)
2111 (1.07)
2218 8
22011 (99.2)
33 (0.2)
144 (0.6)
565 5
5344 (94.5)
299 (5.1)
22 (0.4)
16849 9
44 (0.02)
55 (0.03)
** minipools of maximum 48 donations and run controls.
tt both technical failures and not enough eluate available for further processing.
XX failures that result in repeat amplification and detection.
93 3
TableTable 3. The geometrical mean* of the optical detection to cutoff ratio 's (OD.CO) for
wild-typewild-type and internal control of multi-marker RC38 testing during the first eleven months
ofof routine HIV and HCV NAT screening using NucliSens EluHigh extraction in
combinationcombination with the AmpliScreen HIV-1 v 1.5 and HCV v 2.0 test
HIVV NAT HCV NAT
Monthh n t Geometrical mean n Geometrical mean
OD:COO OD:CO
Novemberr 2000
December r
Januaryy 2001
February y
March h
April l
May y
June e
July-September r
52 2
63 3
46 6
50 0
57 7
47 7
39 9
51 1
151 1
WTJ J
15.7 7
12.6 6
15.9 9
18.5 5
11.7 7
12.4 4
19.1 1
9.8 8
13.8 8
IC§ §
19.2 2
19.3 3
19.2 2
19.2 2
19.1 1
19.0 0
19.4 4
19.2 2
19.1 1
56 6
61 1
47 7
50 0
56 6
48 8
41 1
49 9
151 1
WT T
11.3 3
12.0 0
10.3 3
11.4 4
5.6 6
9.0 0
13.4 4
14.3 3
13.3 3
IC C
19.4 4
19.4 4
19.2 2
19.0 0
19.0 0
19.0 0
18.9 9
19.4 4
19.4 4
** results with an IC absorbance < 2.5 were excluded from calculation (in total: 4 for HIV NAT
andd 6 for HCV NAT).
tt number of RC38 results used for calculating the geometrical mean of OD:CO.
%% = wild-type.
§§ = internal control.
Discussion n
Ourr validation study shows that the 95-percent detection limit of the NucliSens-
AmpliScreenn method, was 32 geq per mL for HIV-1 RNA genotype B and 30 geq per mL
forr HTV-1 RNA genotype E. Currently, in contrast to HCV NAT, there are no European
94 4
sensitivityy requirements for HIV NAT screening of blood donations. Busch et al. [11]
statedd that analysis of the distribution of HIV-1 RNA levels during the pre-seroconversion
periodd of infection indicate that at least 95-percent of viremic, anti-HIV-negative units
couldd be detected if a minipool test system has a sensitivity of 5000 HTV-1 RNA copies
perr mL per individual donation tested. The NucliSens-AmpliScreen method allows a 95-
percentt HTV-1 RNA detection limit of- 1500 geq per mL per individual donation in a
minipooll of 48 donations. Assuming that 5000 copies HTV-1 RNA per mL is equivalent
too 5000 geq per mL HTV-1 RNA, our results show, that this level of sensitivity is easily
fulfilledd utilizing a pool size of 48 donations.
Thee NucliSens Extractor uses the silica-based extraction method as described by
Boomm et al. [9] for non specific extraction and purification of nucleic acids. Since, the
previouslyy used Amplicor HCV v2.0 and the new AmpliScreen HCV v2.0 test use the
samee amplification and detection procedures, it was decided, to restrict the HCV
sensitivityy study to genotype 1 [12]. It was shown that the NucliSens-AmpliScreen method
hass a 95-percent HCV RNA genotype 1 detection limit of 21 geq per mL (-5.5 IU per mL)
forr the test and -1000 geq per mL (-260 IU per mL) per individual donation when
minipoolss of 48 donations are applied. Thus, the sensitivity of the NucliSens-AmpliScreen
methodd is comparable with the sensitivity of the previously used NucliSens Extractor-
Cobass Amplicor HCV v2.0 combination, even though the input of isolated RNA analyzed
wass reduced from an equivalent of - 1100 uL to - 735 uL of the original 2 mL plasma
volumee used for extraction [1]. According to the Committee for Proprietary Medicinal
Productss (CPMP) of the European Union, the HCV NAT detection limit of the assay
shouldd be 100 IU per mL (~ 380 geq/mL) for testing initial plasma pools [2,3,13]. PEI
regulationss prescribe that 5,000 IU per mL (~ 19,000 geq/mL) must be detected to
calculatee the amount contributed by individual donations composing the minipool
[2,3,13].. Our validation study shows that the sensitivity requirements set by these
regulatoryy bodies are easily fulfilled.
Ourr validation study also shows that the NucliSens-AmpliScreen method is robust
enoughh to pick up > 99% of the RC140 samples. The criterion to achieve at least 98%
95 5
validd results at initial testing was fulfilled whereas invalid test results for HTV-1 RNA and
HCVV RNA were found in only 0.56% and 0.45% of the minipools tested, respectively.
Initiall false-positive results for HTV-1 RNA were observed in 15/1997 (0.75%) minipools
tested.. Retesting of these pools revealed negative results; 4 of these 15 initial false-
positivee results were probably caused by contamination of an extractor channel during
maintenance.. The remaining false-positive results were found in one laboratory that could
nott identify the source of contamination. This laboratory is located near the clinical
diagnosticc laboratory. Therefore, false-positivity may have been due to cross-
contamination. .
Basedd on the results of this validation study, Sanquin decided to implement the
NucliSens-AmpliScreenn method for simultaneous HTV and HCV NAT minipool screening
(488 donations) of all blood donations as of November 2000. The turnaround time of this
methodd is approximately 8 hours for processing 24 minipools and controls. The results of
HTVV and HCV NAT screening are available within 24 hours after blood collection. During
thee first eleven months of routine screening, 98.9% and 98.93% initial valid results were
foundd for HTV and HCV NAT, respectively. Thus the criterion to achieve at least 98%
validd results at initial testing is also fulfilled during routine screening. Additionally, it was
shownn that the NucliSens-AmpliScreen method was robust enough to pick up the "go no
go"" RC140 samples under routine conditions, as 99.2% positive results were found. In
accordancee with the 95-percent detection limit for HIV-1 RNA, the RC38 was found
positivee in 95.4 percent of cases. The RC38 detection rate for HCV RNA was not quite
ass high as expected. The performance of the RC38 was also studied by calculating the
geometricall mean of OD:CO for both WT and IC. In the fifth month of routine screening,
thee RC38 WT geometrical mean of OD:CO declined for both HIV and HCV NAT. This
wass probably due to a problem with reagents used for nucleic acid extraction in one of the
NATT laboratories. Further study is necessary to reveal the value of monitoring the RC38
geometricall mean of OD:CO in relation to reagents and equipment applied in NAT.
Inn the Netherlands, ~ 900,000 blood donations are collected annually, of which ~
5%% are from first-time blood donors. The incidence for HIV and HCV infection among
96 6
repeatt blood donors is extremely low. In 1999, the seroconversion rate in repeat donors
wass 0.5 per 100,000 for both HIV and HCV; the rate of infection among first-time blood
donorss was 2.3 per 100,000 for HIV and 35 per 100,000 for HCV. Using calculations
basedd on the seroconversion rate among repeat blood donors (year 1999) and the length
off the seronegative window period (HIV: 22 days, HCV: 70 days), we estimated the
residuall risk attributed to donations given in The Netherlands during the window period
off infection as follows: 1 in 3.3 million for HTV and 1 in 1 million for HCV [11,14]. In
ourr study, during eleven months of HTV and HCV NAT minipool screening, only NAT-
positivee blood donations were found that were also seropositive.
Inn conclusion, routine HIV and HCV NAT minipool screening of 48 donations,
utilizingg automated NucliSens EluHigh extraction combined with the AmpliScreen HIV-1
vl.55 and AmpliScreen HCV v2.0 test meets the sensitivity requirements set by the
regulatoryy bodies. In addition, the test system provides sufficient specificity and
robustnesss for timely release of blood donations.
Acknowledgment t
Wee thank Prof. Dr. W.G. van Aken at the Sanquin-CLB, Amsterdam, The
Netherlands,, and Dr. E.F. van Leeuwen at the Sanquin Blood Bank Midden-Nederland,
Utrecht,, The Netherlands, for review of the manuscript; F.E. Mol let at the Sanquin-CLB
forr review of the validation protocols; H. Hopman at the Sanquin Blood Bank Geldersche
Rivieren,, Nijmegen, The Netherlands, and H. Nelis, at Univalid, Leiden, The Netherlands,
forr their contribution to the Dutch HIV NAT working party.
97 7
References s
1.. Cuijpers HTM, Molijn MHJ, Bos HJ, et al. Validation of the NucliSens Extractor in
combinationn with the hepatitis C virus Cobas Amplicor 2.0 assay in four laboratories
inn The Netherlands utilizing nucleic acid amplification technology for blood screening.
Voxx Sang 2001;81:12-20.
2.. Jongerius JM, Bovenhorst M, vann der Poel CL, et al. Evaluation of automated nucleic
acidd extraction devices for application in HCV NAT. Transfusion 2000;40:871-4.
3.. Beld M, Habibuw MR, Rebers SPH, et al. Evaluation of automated RNA-extraction
technologyy and a qualitative HCV assay for sensitivity and detection of HCV RNA in
pool-screeningg systems. Transfusion 2000;40:575-9.
4.. Flanagan P, Snape T. Nucleic acid technology (NAT) testing and the transfusion
service:: a rationale for implementing of minipool testing. Transfus Med 1998;8:9-13.
5.. Roth WK, Weber M, Seifried E. Feasibility and efficacy of routine PCR screening of
bloodd donations for hepatitis C virus, hepatitis B virus, and HIV-1 in a blood bank
setting.. The lancet 1999;353:359-63.
6.. Flanagan P, Barbara J. PCR testing in plasma pools: from concept to reality. Transfus
Medd 1999;13:164-176.
7.. Jongerius JM, Sherps MC, Cuijpers HTM, et al. Comparison of two HIV-1 nucleic
acidd tests in combination with NucliSens Extractor. Transfusion Clinique et
Biologiquee 2001 ;8 (Suppl.1): 57S.
8.. College voor de Bloedtransfusie van het Nederlandse Rode Kruis. Richtlijn voor
laboratoriumonderzoekk op infecties. 1996.
9.. Boom R, Sol CJA, Salimans MMM, et al. Rapid and simple method for purification
off nucleic acids. J Clin Microbiol 1990;28:495-503.
10.. Finney DJ. Probit analysis. Cambridge, UK: Cambridge University Press, 1971.
11.. Busch MP, Kleinman SH. Report of the interorganizational task force on nucleic acid
amplificationn testing of blood donors: Nucleic acid amplification testing of blood
donorss for transfusion-transmitted infectious diseases. Transfusion 2000;40:143-59.
98 8
12.Cuijperss HTM, van Dijk R, Viret J-F, et al. European multi-centre validation study of
NucliSenss Extractor in combination with HCV Amplicor version 2.0 for HCV-NAT
screeningg of plasmapools. Biologicals 1999;27:303-14.
13.. Saldanha J, Heath A, Lelie PN, et al. Calibration of HCV working reagents for NAT
assayss against the HCV international standard. Vox Sang 2000;78:217-24.
14.. Schreiber GB, Busch MP, Kleinman SH, et al. The risk of transfusion-transmitted viral
infections.. The Retrovirus Epidemiology Donor Study. N Eng J Med 1996 ;334:1685-
90. .
99 99
100 0
Chapterr 8
GBB Virus Type C viremia and envelope antibodies
amongg population subsets in The Netherlands
J.M.. Jongerius1, G.J. Boland2, C.L. van der Poel1, M.C. Rasch2,
J.. van der Reijden2, P.N. Friedman
E.F.. van Leeuwen', J. van Hattum2
E.. Italiaander2, J.J. van der Reijden2, P.N. Friedman3, J.J. Cockerill3,
11 Blood Bank Midden-Nederland 22 University Hospital Utrecht
Abbottt Laboratories, Abbott Park, Illinois, USA
Publishedd in: Vox Sang 1999;76:81-4
101 1
102 2
Abstract t
Twoo new flaviviruses, hepatitis G virus and GB virus type C (GBV-C), are possible
causativee agents for non-A-E hepatitis. In this study we established the prevalence of
GBV-CC markers in various population subsets in The Netherlands by assays for GBV-C
antibodiess and GBV-C nucleic acid.
Wee tested specimens from groups of patients with hepatitis of various causes,
intravenouss drug users (TVDUs), and blood donors for GBV-C RNA (LCx® GBV-C assay,
Abbottt Laboratories), and for antibodies to GBV-C envelope E2 protein (GBV-C anti-E2)
withh an enzyme immunoassay (Abbott Laboratories). Patients and donors were
representedd in one group only.
GBV-CC RNA and GBV-C anti-E2 prevalence were, respectively, 2/34 (6%) and
3/344 (9%) among patients with non-A-E hepatitis, 2/10 (20%) and 0/10 (0%) among
hepatitiss B virus patients, 10/40 (25%) and 19/40 (48%) among hepatitis C virus (HCV)
patients,, 1/8 (13%) and 0/8 (0%) among patients with autoimmune hepatitis (AIH), 24/102
(24%)) and 72/102 (71%) among IVDUs, 1/34 (3%) and 2/34 (6%) among blood donors
withh indeterminate anti-HCV recombinant immunoblot assay reactivity, and 3/250 (1.2%)
andd 8/250 (3.2%) among first-time blood donors. The profile of simultaneous GBV-C
RNAA positivity plus GBV-C anti-E2 positivity was found in 2/40 (5%) HCV patients,
4/1022 (4%) IVDUs and 1/250 (0.4%) first time blood donors.
GBV-CC infection appears not to be a major cause of non-A-E hepatitis and AIH,
butt is associated with parenteral risk. The prevalence of GBV-C viremia in first time
bloodd donors is higher than that of HCV (1.2 vs. 0.04%), but GBV-C viremia in IVDUs
iss lower than HCV (24 vs. 59%). Most IVDUs have probably previously been exposed
too GBV-C given the very high prevalence of GBV-C anti-E2 (71%). Most persons with
GBV-CC markers are GBV-C RNA-negative and anti-E2-confirmed positive, suggesting
103 3
thatt GBV-C infection is transient.
Introduction n
Simonss et al. [1] identified a novel RNA virus, designated GB virus type C (GBV-
C),, in patients with clinical evidence of hepatitis of unknown etiology. In addition Linnen
ett al. [2] independently identified a similar RNA virus, designated hepatitis G virus
(HGV),, from plasma samples of a patient with chronic hepatitis. Genomic sequencing
indicatedd that GBV-C and HGV are essentially the same. The homology between GBV-C
andd HGV at nucleotide and amino acid level is approximately 86 and 95%, respectively
[3].. Both viruses are classified in the family of Flaviviridae but are distinct from hepatitis
CC virus (HCV). The homology at the nucleotide and amino acid level between GBV-
C/HGVV and HCV is approximately 25 and 29%, respectively [3]. The GBV-C/HGV
genomee of 9,100-9,400 nucleotides encodes for one polyprotein precursor of 2,850-
2,9000 amino acids. Due to the high degree of conservation, the 5' non-translated region,
ass in HCV, is suitable for diagnostic testing with polymerase chain reaction (PCR). In
contrastt with HCV, the envelope E2 region of the genome encoding for the E2 protein is
nott hypervariable [4], a finding that may be important for the effectiveness of the humoral
immunee response against GBV-C [5].
Recently,, an experimental, single-tube assay has been developed for the qualitative
detectionn of GBV-C RNA in human serum or plasma (Abbott Laboratories, Abbott Park,
III ,, USA). This uses reverse-transcription polymerase chain reaction amplification (RT-
PCR)) technology in the LCx® probe system for the qualitative detection of GBV-C RNA
inn human serum or plasma. Experimental enzyme immunoassays (EIA) were also
developedd to determine antibody reactivity against GBV-C E2. With these assays we
studiedd the prevalence of GBV-C RNA and GBV-C anti-E2 antibodies in The Netherlands
amongg groups of patients with hepatitis of different causes, intravenous drug users
(IVDUs),, and blood donors.
104 4
Materialss and Methods
StudyStudy populations
Wee studied serum or EDTA-plasma specimens from 34 non-A-E hepatitis patients,
100 patients with hepatitis B virus infection, 40 patients with HCV infection, 8 patients
diagnosedd with autoimmune hepatitis (AIH), 102 IVDUs, 34 blood donors with
indeterminatee anti-HCV recombinant immunoblot assay (RIBA) reactivity (respectively
33 with isolated anti-c22, 13 with isolated anti-c33, 5 with isolated anti-clOO and 13 with
isolatedd anti-NS5 reactivity), and 250 first time blood donors. The specimens from both
patientss and donors were represented in one study population only and stored at -20°C
untill tested.
DeterminationDetermination of GBV-C RNA
Virall RNA was extracted and purified from each 25-uL specimen with a modified
QIAamp®® HCV kit procedure (Qiagen, Hilden, Germany). The purified RNA was tested
withh the LCx GBV-C assay (Abbott Laboratories, Abbott Park, 111., USA). The LCx GBV-
CC assay uses a nucleic acid amplification method, RT-PCR, to generate amplified product
materiall from GBV-C RNA.. The assay utilizes a primer/probe set that is specific for a
sequencee located within the highly conserved 5* non-translated region of the GBV-C
genome.. The first primer is complementary to the RNA genome and serves as both cDNA
primerr and PCR primer. The second primer is a PCR primer with the same sense as the
genome.. A third oligonucleotide, the probe, is present in the reaction mix and is
complementaryy to one of the strands of the PCR product. In the presence of GBV-C-
amplifiedd product the probe/PCR strand complex serves as the detection product in the
LCxx analyzer (Abbott Laboratories). For this study the cutoff value for the LCx GBV-C
assayy was established at 100 counts/s/s (c/s/s). All specimens with an initial LCx result
off 100-500 c/s/s were retested in duplicate. A specimen was considered GBV-C RNA-
positivee if the sample counts were greater than or equal to 100 c/s/s.
105 5
DeterminationDetermination of GBV-C anti-E2
Alll specimens were tested for GBV-C anti-E2 antibodies with an experimental
solid-phasee indirect GBV-C EIA (Abbott Laboratories). Specimens with a sample-to-
cutofff ratio (S/CO) greater than or equal to 1.0 were confirmed with a separate sandwich
GBV-CC EIA (Abbott Laboratories). If the confirmatory result was reactive (sample-to-
cutofff ratio greater than or equal to 1.0) the specimen was considered positive for GBV-C
anti-E22 antibodies.
Results s
Tablee 1 shows the prevalence rates of GBV-C markers with 95% confidence
intervall limits for GBV-C exposure among the various population subsets.
TableTable 1. Prevalence of GBV-C markers among various population subsets in The
Netherlands Netherlands
Populationn subsets
Non-A-EE hepatitis patients
Hepatitiss B patients
Hepatitiss C patients
AIHH patients
IVDUs s
Bloodd donors with
indeterminatee anti-HCV
RIBAA reactivity
First-timee blood donors
n n
34 4
10 0
40 0
8 8
102 2
34 4
250 0
GBV-C C
RNA+ve e
2 2
2 2
10 0
1 1
24 4
1 1
3 3
(5.9) )
(20) )
(25) )
(12.5) )
(23.5) )
(2.9) )
(1.2) )
GBV-C C
anti-E22 +ve
3 3
0 0
19 9
0 0
72 2
2 2
8 8
(8.8) )
(0) )
(47.5) )
(0) )
(70.6) )
(5.8) )
(3.2) )
GBV-C C
exposed d
5 5
2 2
27 7
1 1
92 2
3 3
10 0
(14.7) )
(20) )
(67.5) )
(12.5) )
(90.2) )
(8.8) )
(4) )
95%% CI for
GBV-C C
exposure e
0.05-0.31 1
0.03-0.56 6
0.51-0.81 1
0.00-0.53 3
0.83-0.95 5
0.02-0.24 4
0.02-0.07 7
Figuress in parenthesis represent percentage, +ve = positive, CI = confidence interval.
106 6
Thee prevalence rates of GBV-C RNA (viremia) and GBV-C anti-E2 antibodies in
aa high-risk (IVDUs) versus a low-risk (donor) population are in Table 2.
TableTable 2. Prevalence of GBV-C RNA and/or GBV-C anti-E2 antibodies as compared in a
high-riskhigh-risk population (IVDUs) and a low-risk population (first-time blood donors)
Serologicall GBV-C profile First-time IVDUs
Bloodd donors
GBV-CC RNA+ve / GBV-C anti-E2 +ve 1 (0.4) 4 (3.9)
GBV-CC RNA+ve / GBV-C anti-E2 -ve 2 (0.8) 20 (19.6)
GBV-CC RNA-ve / GBV-C anti-E2+ve 7 (2.8) 68 (66.7)
GBV-CC RNA-ve / GBV-C anti-E2-ve 240 (96) 10 (9.8)
Totall " 250 (100) 102 (100)
Figuress represent number with the percentage in parenthesis.
+vee = positive,
-vee = negative.
Off 108 solid-phase indirect GBV-C anti-E2 EIA-reactive specimens, 4 tested
negativee and 104 positive in the supplementary confirmatory GBV-C anti-E2 sandwich
EIA.. Of 104 confirmed GBV-C anti-E2-positive specimens, 97 were negative for GBV-C
RNA.. However, simultaneous reactivity for GBV-C RNA and GBV-C anti-E2 antibodies
wass detected in 2/40 (5%) patients with HCV infection, 4/102 (4%) IVDUs, and 1/250
(0.4%)) first-time blood donors.
Amongg a panel of 34 specimens collected from blood donors with indeterminate
anti-HCVV RIBA reactivity, 1 specimen (with isolated anti-NS5 reactivity) was GBV-C
RNA-positive.. Within the same panel, the presence of GBV-C anti-E2 antibodies was
detectedd in 2 specimens (one with isolated anti-NS5 reactivity and one with isolated anti-
c333 reactivity).
107 7
Discussion Discussion
GBV-CC infection appears not to be a major cause of non-A-E hepatitis. Among 34
non-A-EE hepatitis patients 2 (6%) were GBV-C RNA-positive and another 3 (9%) had
GBV-CC anti-E2. In a sentinel surveillance study, Alter et al. [6] reported GBV-C (HGV)
RNAA in 4/45 (9%) non-A-E hepatitis patients. In a posttransfusion hepatitis study, Alter
ett al. [7] reported GBV-C (HGV) RNA in 3/13 (23%) patients with transfusion-associated
non-A,, non-B, non-C hepatitis. GBV-C anti-E2 was not tested in either study [6,7]. The
etiologyy of the remaining hepatitis non-A-E needs further study.
Ourr data suggest parenteral transmission of GBV-C. Respectively, 90% of IVDUs
andd 68% of HCV patientss had GBV-C markers, whereas among first time donors, who are
selectedd against parenteral risk, 4% had GBV-C markers.
Off 8 patients with AIH, 1 (13%) was positive for GBV-C RNA. This is in
accordancee with a report on the presence of GBV-C RNA in 8/82 (10%) of AIH patients
[8].. The relatively low prevalence of GBV-C markers in AIH patients suggests that GBV-
CC is not a major etiologic agent of AIH.
Off the IVDUs, 90% had GBV-C markers, suggestive of previous GBV-C exposure,
butt only 24% were viremic. In this high-risk group GBV-C viremia is infrequent
comparedd with HCV viremia [59% HCV PCR-positive, Rasch: unpubl. data]. With the
transversee design of our study, it remainss unclear whether GBV-C RNA positivity reflects
aa chronic carrier state or a recent but finally resolving infection. The high prevalence of
GBV-CC anti-E2 (67%) among IVDUs without GBV-C RNA suggests that GBV-C viremia
iss transient. Simultaneous GBV-C RNA positivity and GBV-C anti-E2 positivity was
detectedd in 4% of the IVDUs. This phenomenon was also reported with the same
frequencyy among IVDUs in another study [5], Simultaneous presence of GBV-C RNA
andd GBV-C anti-E2 may represent early viral clearance, incomplete protection by
circulatingg GBV-C anti-E2, or false-positive EIA results.
Cross-reactivityy with GBV-C anti-E2 does not seem to cause indeterminate anti-
HCVV RIBA reactivity, as only 6% of the donors with indeterminate anti-HCV RIBA
108 8
reactivityy had GBV-C anti-E2. This is not significantly different from the GBV-C anti-E2
prevalencee within the group of first time blood donors (Fisher's exact).
GBV-CC RNA and GBV-C anti-E2 prevalence among first-time blood donors (250)
wass 1.2 and 3.2%, respectively. The prevalence of GBV-C viremia in first-time blood
donorss is relatively high compared with HCV viremia of 0.04% [9]. This and other studies
[3,5]] suggest a high prevalence of GBV-C infection among the general population
comparedd with other hepatotropic viral infections. Apparently GBV-C transmission occurs
oftenn enough to maintain the route of infection, since about 80% of infected persons seem
too clear the virus. Transmission may be other than parenteral, e.g., arthropod-borne GBV-
CC [10].
Att present, no suitable GBV-C antibody or antigen assay is available for blood
donorr screening. Given the apparently low pathogenicity of GBV-C [3], the use of such
screeningg would be questionable. Further studies on GBV-C could focus on transmission
routes,, duration of viremia, pathogenicity, long-term sequelae of the carrier state, and the
protectivee properties of antibodies.
Inn summary, GBV-C appears not to be a major cause of non-A-E hepatitis or AIH.
GBV-CC infection is associated with parenteral risks. The prevalence of GBV-C viremia
inn first-time blood donors is higher than that of HCV viremia. GBV-C viremia in IVDUs
iss less common than HCV. Most IVDUs have been exposed to GBV-C judging from the
highh prevalence of GBV-C anti-E2. The data suggest that GBV-C infection is transient.
Acknowledgment t
Wee thank Dr. H.T.M. Cuijpers at the Central Laboratory of The Netherlands Red
Crosss Blood Transfusion Service, Amsterdam, for providing specimens with indeterminate
anti-HCVV RIBA reactivity, and also M. Bovenhorst and the medical technicians at the
Bloodd Bank Midden-Nederland for their assistance.
109 9
References s
1.. Simons JN, Leary TP, Dawson GJ, Pilot-Matias TJ, Muerhoff AS, Schlauder GG,
Desaii SM, Mushahwar IK: Isolation of novel virus-like sequences associated with
humann hepatitis. Nat Med 1995;1:564-9.
2.. Linnen J, Wages J Jr., Zhang-Keck Z-Y, Fry KE, Krawczynski KZ, Alter H, Koonin
E,, Gallagher M, Alter M, Hadziyannis S, Karayiannis P, Fung K, Nakatsuji Y, Shih
JW-K,, Young L, Patiak M, Hoover C, Fernandez J, Chen S, Zou J-C, Morris T,
Hyamss KC, Ismay S, Lifson JD, Hess G, Foung SKH, Thomas H, Bradley D, Margolis
H,, Kim JP: Molecular cloning and disease association of hepatitis G virus: A
transfusion-transmissiblee agent. Science 1996;271:505-8.
3.. Karayiannis P, Thomas HC: Current status of hepatitis G virus (GBV-C) in
transfusion:: Is it relevant? Vox Sang 1997;73:63-9.
4.. Wiener AJ, Brauer MJ, Rosenblatt J: Variable and hypervariable domains are found
inn the regions of HCV corresponding to the ftavivirus envelope and NS1 proteins and
thee pestivirus envelope glycoproteins. Virology 1991;180:842-8.
5.. Tacke M, Kiyosawa K, Stark K, Schlueter V, Ofenloch-Haehnle B, Hess G, Engel
AM:: Detection of antibodies to a putative hepatitis G virus envelope protein. Lancet
1997;349:318-20. .
6.. Alter MJ, Gallagher M, Morris TT, Moyer LA, Meeks EL, Krawczynski K, Kim JP,
Margoliss HS, for the Sentinel Counties Viral Hepatitis Study Team: Acute non-A-E
hepatitiss in the United States and the role of hepatitis G virus infection. N Engl J Med
1997;336:741-6. .
7.. Alter HJ, Nakatsuji Y, Melpolder J, Wages J, Wesley R, Wai-Kuo Shih J, Kim JP: The
incidencee of transfusion-associated hepatitis G virus infection and its relation to liver
disease.. N Engl J Med 1997;336:747-54.
8.. Heringlake S, Tillmann HL, Cordes-Temme P, Trautwein C, Hunsmann G, Manns
MP:: GBV-C/HGV is not the major cause of autoimmune hepatitis. J Hepatol
1996;25:980-4. .
no o
9.. College voor de Bloedtransfusie van het Nederlandse Rode Kruis: Overzicht van de
bloedtransfusiee in Nederland, 1995, pp 18-9.
10.. Selvey LA, Hyland CA, Mison L, Solomon N, Gowans EJ: Is there evidence for vector
transmissionn of GBV-C? Lancet 1998;351:1104.
112 2
Chapterr 9
Post-transfusionn hepatitis infections among cardiac
surgeryy patients after the introduction of anti-HCV
screeningg of blood donations
J.M.. Jongerius1, L.M.G. van de Watering2,3, C.L. van der Poel1, M. Bovenhorst1,
R.. Meuken1, A. Brand2 3, E.F. van Leeuwen'
11 Sanquin Blood Bank Midden-Nederland, Utrecht 22 Sanquin Blood Bank Leiden-Haaglanden, Den Haag
33 Department of Immunohematology & Blood Transfusion of the Leiden University
Medicall Center, Leiden
Submittedd for publication
113 3
114 4
Abstract t
Untill the early nineteen nineties, when a test for the detection of anti-hepatitis C
viruss (anti-HCV) antibodies was developed, post-transfusion hepatitis was a major
complicationn of blood transfusion. In this study we investigated the incidence of post-
transfusionn GB virus type C (GBV-C), hepatitis C virus (HCV) and hepatitis B virus
(HBV)) infection among patients who underwent cardiac surgery after the introduction of
mandatoryy anti-HCV antibody screening of blood donations.
Storedd pre and post-transfusion blood samples of patients (n=373) were tested for
GBV-C,, HCV and HBV markers by serological as well as molecular biological tests.
Informationn was retrieved on the number and type of blood products received during the
cardiacc surgery related transfusion episode and the patients history of previous blood
transfusionss or major surgery.
Thee patients enrolled in this study received a total 3270 blood products. 41/373
(11%)) patients were found to be GBV-C exposed (GBV-C marker-positive) before the
surgeryy related transfusion episode. 15/373 (4%) patients became infected with GBV-C
(GBV-CC RNA-positive) following the surgery related transfusion episode and 6/373
(1.6%)) patients seroconverted for GBV-C anti-E2 antibodies but no GBV-C RNA was
found.. The remaining 311 (83.4%) patients were negative for both GBV-C RNA in the
earlyy post-transfusion sample and GBV-C anti-E2 in the late post-transfusion sample and
thuss not further tested. None of the 373 patients became infected with HCV and or HBV.
Thee data of this study indicate that GBV-C is transmitted by blood transfusion.
Clearancee of GBV-C RNA is associated with an anti-E2 response. GBV-C marker-
positivityy preceding the cardiac surgery related transfusion episode was found to be
associatedd with a history of previous blood transfusion or major surgery. HBV and HCV
wass not transmitted to the patients enrolled in this study.
115 5
Introduction n
Thee introduction of Hepatitis B surface antigen (HBsAg) screening of blood
donationss in the early nineteen seventies did not completely prevent the occurrence of
post-transfusionn hepatitis (PTH). In the mid-nineteen seventies, PTH, not attributed to
hepatitiss B virus (HBV) nor hepatitis A virus (HAV) infection was designated post-
transfusionn hepatitis non-A, non-B (PTH-NANB) [1 j . After cloning of parts of the HCV,
thee major causative agent of PTH-NANB, routine donation screening for the presence of
specificc antibodies to this persistent virus for the greater part abolished the risk of PTH-
NANBB [2], However, some blood recipients may still develop PTH [3,4] which result
fromfrom donations during the window period, atypical seroconversions, immunosilent
infections,, rare viral variants, new hepatotropic viruses or laboratory testing errors during
bloodd screening [5].
Inn 1995, hepatitis G virus (HGV) and GB virus type C (GBV-C), were
independentlyy identified as possible causative agents for non-A-E hepatitis [6,7]. Genomic
sequencingg indicated that GBV-C and HGV are different isolates of the same virus [8].
Thiss virus, classified in the family of Flaviviridae, hereinafter called GBV-C, is known
too be transmitted by blood transfusion, yet the pathogenicity of GBV-C remains unclear
[8,9].. We previously reported that the prevalence of GBV-C RNA and GBV-C anti-E2
antibodiess among first-time blood donors in The Netherlands was 1.2% and 3.2%,
respectivelyy [10]. Additional studies suggest that GBV-C anti-E2 antibodies may have
neutralizingg or protective properties against GBV-C transmission by blood transfusion
[10-13].. It is also shown that GBV-C infection may be transient in most cases, with
remainingg detectable anti-GBV-C antibodies [10,11],
Schreiberr et al. estimated the residual risk of transfusion transmitted HBV and
HCVV infections in the USA, due to donor blood donated during the infectious window
period,, to be 1 in 63,000 for HBV and 1 in 103,000 for HCV [14]. In Europe this risk was
assessedd to be 1 in 398,499 for HBV and 1 in 620,754 for HCV [15]. Due to the recent
introductionn of blood donation screening for transfusion-transmitted infectious diseases
116 6
usingg nucleic acid amplification technology (NAT) in several countries, these residual
riskss may be further reduced in Europe to 1 window donation in 689,655 for HBV and 1
inn 1,781,293 for HCV [15].
Wee here report the results of a study on the incidence of post-transfusion GBV-C,
HCVV and HBV infection in patients who underwent cardiac surgery and received blood
transfusionss between 1992-1994 i.e.: after the introduction of mandatory anti-HCV
antibodyy screening of blood donations in The Netherlands.
Materialss and Methods
StudyStudy populations
Clinicall data and samples used for this analysis were derived from a study to
establishh the effects of leukocyte depletion of red blood cells on HLA antibody formation,
andd occurrence of post-operative bacterial infections in patients undergoing cardiac
surgeryy [16]. Patients were included retrospectively in the present study when the
followingg criteria were met: i) pre-transfusion samples obtained <14 days before surgery
(furtherr called PreTr sample), //') early post-transfusion samples obtained within 20-90
dayss after surgery (further called EPostTr sample), Hi) late post-transfusion samples
obtainedd >120 days after surgery (further called LPostTr sample), iiii) no blood
transfusionss within the last six months preceding surgery. In addition, information was
retrievedd on the patients age, sex, previous transfusion history, previous surgery history,
andd the number and type of blood products transfused during the cardiac surgery related
transfusionn episode. Unfortunately, longitudinal pre and post-transfusion ALT testing
resultss of the patients were not available.
DeterminationDetermination of GBV-C markers
Virall RNA was extracted and purified from each 25 u.L EPostTr samples using a
modifiedd QIAamp® HCV kit procedure (Qiagen, Hilden, Germany) as described
elsewheree [10]. The purified RNA was tested with the LCx® GBV-C assay (Abbott
117 7
Laboratories,, Abbott Park, II, USA) [10]. The cutoff value used for the LCx® GBV-C
assayy was 100 counts/s/s (c/s/s). Accordingly, a sample was considered GBV-C RNA-
positivee if the sample counts were greater than or equal to 100 c/s/s.
Alll LPostTr samples were tested for GBV-C anti-E2 antibodies with a qualitative
researchh enzyme-linked immunosorbent assay (ELISA) (HGV CHOe2 ELISA Test
System,, Ortho Diagnostic Systems, Raritan, NJ, USA). Briefly, in the wells of a GBV-C
antigenn coated microplate, 10 uL of control or sample was added to 200 uL off specimen
diluentt containing bovine protein stabilizers. After 1 hour incubation at 37°C and five
washingss of the wells with wash buffer, 200 uL of conjugate (murine monoclonal anti-
humann IgG conjugated to horseradish peroxidase) was added to the wells. Following a
secondd incubation of 1 hour at 37°C and five washings with wash buffer, 200 uL of
substratee (o-phenylenediamine-2HCL) solution was added to the wells. After 30 minutes
incubationn at room temperature in the dark, 50 uL of 4N sulfuric acid was added to the
wellss to stop the reaction. The microplate was read at a wavelength of 490 nm and a cutoff
valuee was established according to the negative controls provided with the kit. Samples
withh an optical density-to-cutoff ratio (OD:CO) greater than or equal to 1.0 were re-tested
inn duplicate. A sample was considered GBV-C anti-E2-positive if re-testing revealed an
OD:COO greater than or equal to 1.0 in one or both duplicate tests.
PreTr,, EPostTr and LPostTr samples of patients with a positive result for GBV-C
RNAA in the EPostTr sample and/or a positive result for GBV-C anti-E2 in the LPostTr
sample,, were tested for GBV-C RNA and GBV-C anti-E2.
DeterminationDetermination of HCV markers
LPostTrr samples were tested for anti-HCV antibodies using a qualitative ELISA
(HCVV 3.0 ELISA Test System with Enhanced SAVe, Ortho Diagnostic Systems),
accordingg to the instructions of the manufacturer. PreTr, EPostTr and LPostTr samples
off patients with a repeat positive result for anti-HCV antibodies in the LPostTr sample,
weree tested for anti-HCV with ELISA (HCV 3.0 ELISA Test System with Enhanced
SAVe),, and recombinant immunoblot assay (RIBA) (CHIRON RIBA HCV 3.0 SIA, Ortho
118 8
Diagnosticc Systems), and for HCV-RNA with PCR (Cobas Amplicor HCV v2.0 test,
Rochee Diagnostic Systems, Inc., Branchburg, NJ, USA), all according to the instructions
off the manufacturers [17].
EPostTrr and LPostTr samples of all patients were tested for HCV RNA in a HCV
NATT minipool system set up (maximum of 48 samples in a minipool) employing manual
NucliSenss extraction (Organon Teknika, Boxtel, The Netherlands) of a plasma input of
0.55 mL, combined with amplification and detection using the Cobas Amplicor HCV v2.0
testt on the Cobas Amplicor Analyzer (Roche Diagnostic Systems) [17,18].
DeterminationDetermination of HBVmarkers
LPostTrr samples were tested for antibody to hepatitis B core antigen (anti-HBc)
usingg a qualitative enzyme immunoassay (EIA) (CORE™, Abbott Laboratories),
accordingg to the instructions of the manufacturer. Additionally, the PreTr samples of
patientss with a positive result for anti-HBc in the LPostTr sample were tested for anti-HBc
(CORE™),, and the PreTr, EPostTr and LPostTr samples for HBsAg (HBsAg Vitros,
Orthoo Diagnostic Systems), according to the instructions of the manufacturer. In case of
aa positive test result for HBsAg, the sample was additionally tested for HBV-DNA using
aa previously described "in house" HBV-DNA PCR [19].
Statistics Statistics
Thee Pearson Chi-square test was used to compare groups for variables.
Results s
StudyStudy population
Patientt and transfusion characteristics are summarized in Table 1. The patients
includedd in this study received a total 3270 blood products i.e.: 1827 red blood cell
concentratess (RBCs) and 1443 units of fresh frozen plasma (FFP).
119 9
TableTable 1. Patient and transfusion characteristics
Patients: :
-- Number of patients in study 373
-- Age in years * 64.39 9.05
Maless 271
Femaless 102
Historyy of blood transfusion or major surgery:
Patientss with a negative history 256
Patientss with a positive history 117
Transfusionn characteristics:
-- RBC transfusions per patient* 4.90 3.39
-- FFP transfusions per patient* 3.87 3.05
** = mean SD.
GBV-C GBV-C
Thee longitudinal GBV-C marker profiles observed are shown in Table 2. 41/373
(11%)) patients were GBV-C exposed (GBV-C marker-positive) before, and 15/373 (4%)
patientss became infected with GBV-C (GBV-C RNA-positive) after the cardiac surgery
relatedd transfusion episode. 6/373 (1.6%) patients seroconverted for GBV-C anti-E2
antibodiess but were negative for GBV-C RNA in all pre and post-transfusion samples.
Thee remaining 311 (83.4%) patients were negative for GBV-C RNA in the EPostTr
samplee and GBV-C anti-E2 in the LPostTr sample. Samples from patients with this latter
profilee were not further tested.
Off the 41 GBV-C pre-exposed patients, 11 were GBV-C RNA-positive while the
remainingg 30 were GBV-C anti-E2-positive in the PreTr sample. In addition, 21/41 (51%)
off the GBV-C pre-exposed patients had a history of previous blood transfusions or major
surgeryy (Table 2). Among the remaining 332 patients, 96 (29%) had a history of previous
bloodd transfusions or major surgery (Table 2). GBV-C marker-positivity preceding the
cardiacc surgery related transfusion episode was found to be associated with a history of
previouss blood transfusion or major surgery (Chi-square: P = 0.004).
120 0
TableTable 2. Longitudinal GBV-C marker profiles and history of blood transfusion (BITr) or
majormajor surgery (MS) among 3 73 patients who underwent cardiac surgery
GBV-CC marker profile
1.. Pre-exposure:
Profilee 1.1
Profilee 1.2
Profilee 1.3
Profilee 1.4
Profilee 1.5
2.. De novo infection:
Profilee 2.1
Profilee 2.2
Profilee 2.3
3.. Inconclusive:
Profilee 3.1
Profilee 3.2
4.. Not exposed:
Profilee 4.1
PreTrr = pre-transfusion
PreTr r
RNAA a-E2
+ +
+ +
+ +
+ +
+ +
--
--
--
--
--
NTT NT
sample. .
EPostTrr = early post-transfusion sample
LPostTrr = late post-transfusion sample.
++ = positive.
-- = negative.
NTT = not tested.
EPostTr r
RNA A
+ +
+ +
+ +
--
--
+ +
+ +
+ +
--
--
--
a-E2 2
--
--
+ +
+ +
--
--
--
--
+ +
--
NT T
LPostTr r
RNA A
+ +
--
--
--
--
+ +
+ +
--
--
--
NT T
a-E2 2
--
--
+ +
+ +
+ +
--
+ +
+ +
+ +
+ +
--
Historyy of
BITr r
19 9
7 7
1 1
0 0
10 0
1 1
6 6
5 5
0 0
1 1
1 1
0 0
1 1
72 2
72 2
MS S
2 2
0 0
0 0
0 0
2 2
0 0
1 1
1 1
0 0
0 0
0 0
0 0
0 0
16 6
16 6
Patients s
n n
41 1
9 9
1 1
1 1
26 6
4 4
15 5
12 2
2 2
1 1
6 6
1 1
5 5
311 1
311 1
(%) )
(11) )
(2.4) )
(0.3) )
(0.3) )
(7.0) )
(1.1) )
(4) )
(3.2) )
(0.5) )
(0.3) )
(1.6) )
(0.3) )
(1.3) )
(83.4) )
(83.4) )
HCV HCV
Off 373 LPostTr samples, 3 (0.8%) were anti-HCV-positive. The three involved
patientss were also anti-HCV-positive in the PreTr and EPostTr samples. RIBA and HCV-
RNAA testing in these three patients revealed that 2/3 were RJBA-positive, 1/3 RIBA-
negative,, and all three HCV-RNA-negative in both pre and post-transfusion samples. In
121 1
addition,, both RIBA positive patients were also found to be GBV-C pre-exposed (anti-E2-
positivee and GBV-C RNA-negative in the PreTr sample) and one anti-HBc-positive in
bothh pre and post-transfusion samples.
HCVV NAT minipool screening on all 373 EPostTr and LPostTr samples revealed
negativee results.
HBV HBV
Outt of the total of 373 LPostTr samples, 28 (7.5%) were positive for anti-HBc
antibodies.. In five of these patients the sample set was incomplete; three of these could
nott be tested for anti-HBc in the PreTr samples but were HBsAg-negative in the PreTr
samplee and anti-HBc-positive and HBsAg-negative in the EPostTr sample; the remaining
twoo patients could not be tested for anti-HBc and HBsAg in the PreTr sample but were
anti-HBc-positivee and HBsAg-negative in the EPostTr sample.
Testingg on samples of the 23 anti-HBc-positive patients with a complete sample
sett revealed that 23 were anti-HBc-positive in the PreTr sample and 22/23 HBsAg-
negativee in both pre and post-transfusion samples. 1/23 patients was found to be HBsAg-
positivee and HBV-DNA-positive in all pre and post-transfusion samples. This latter
patientt tested negative for both GBV-C and HCV markers in the complete sample set.
Discussion Discussion
Thee data of this study indicate that GBV-C is transmitted by blood transfusion. De
novoo GBV-C infection occurred in 15/373 (4%) patients who underwent cardiac surgery
andd received both RBCs and FFP. GBV-C marker-positivity preceding the cardiac surgery
relatedd transfusion episode was found to be associated with a history of previous blood
transfusionn or major surgery.
Persistencee of GBV-C viremia without detectable anti-E2 antibodies was found in
12/155 (80%) patients with de novo GBV-C infection. The remaining 3 (20%) patients
weree found to be strong positive for GBV-C anti-E2 antibodies in the LPostTr sample. Of
122 2
thesee 3 patients, one became negative for GBV-C RNA in combination with one of the
highestt OD:CO ratio's for GBV-C anti-E2 antibodies in the ELISA. Apparently this
patientt cleared the virus. The other 2 patients were both GBV-C RNA-positive and GBV-
CC anti-E2-positive in the LPostTr sample, however, with declining counts/s/s for GBV-C
RNAA in the LCxR GBV-C assay (suggestive of a declining viral load) and a strong
antibodyy response in terms of OD:CO ratio in the GBV-C anti-E2 ELISA. The detection
off simultaneous GBV-C RNA and GBV-C anti-E2-positivity has earlier been reported
withh a high prevalence (4%) among intravenous drug users in The Netherlands and may
representt early viral clearance or chronic persistent carriership [10].
Off the 11 patients with GBV-C RNA-positivity before the cardiac surgery related
transfusionn episode, one became negative for GBV-C RNA without detectable GBV-C
anti-E22 antibodies in the LPostTr sample. Theoretically this could be due to false-positive
PCRR results in both PreTr and EPostTr sample or a false-negative PCR and / or anti-E2
ELISAA test result in the LPostTr sample. Another patient within this group became
negativee for GBV-C RNA and positive for GBV-C anti-E2 in the LPostTr sample.
Apparentlyy this patient also cleared the virus. Thus, clearance of GBV-C RNA is
associatedd with an anti-E2 response [11].
6/3733 (1.6%) patients became positive for GBV-C anti-E2 antibodies after the
cardiacc surgery related transfusion episode but were negative for GBV-C RNA in all
samples.. This could be due to false-positive ELISA results, false-negative GBV-C RNA
results,, passive immunization by transfused blood products or a booster effect due to re-
infectionn without detectable GBV-C RNA levels. Additionally, these patients may have
escapedd GBV-C RNA screening if GBV-C RNA-positivity occurred in the interval
betweenn the sample collection dates. We interpreted these longitudinal GBV-C marker
profiless as inconclusive.
Inn our study none of the 373 patients became infected with HCV and or HBV
followingg the cardiac surgery related transfusion episode. In a prospective study carried
outt in The Netherlands before the introduction of anti-HCV screening, post-transfusion
HCVV infection was found in 9/383 (2.3%) patients undergoing cardiac surgery [20]. Thus,
123 3
ourr study confirms the increased safety of the blood supply in that respect. Since the
introductionn of anti-HCV screening in 1991, the theoretical risk of HBV and HCV
transmissionn by blood has declined dramatically [14,15]. To determine the current
incidencee for PTH would require prospective studies involving much larger numbers of
patients.. Alternatively, mathematical models can be used to estimate the residual risks for
thee known PTH viruses [14,15,21].
Residuall risks for PTH are expected to decrease even further due to extended donor
questioning,, improvement of donor screening methods (viral mutant detection),
implementationn of NAT (minipool) technology and the application of new techniques for
virall inactivation of blood products [15,21].
Acknowledgment t
Wee thank Prof. Dr. W.G. van Aken at the Central Laboratory of The Netherlands
Redd Cross Blood Transfusion Service, Amsterdam, for review of the manuscript; and Dr.
H.T.M.. Cuijpers and M. Sjerps at the Viral Diagnostic Laboratory of Sanquin CLB
Diagnosticc Division, Amsterdam, The Netherlands for their assistance with HCV-NAT
testing;; and Dr. H. Putter at the Leiden University Medical Center, Leiden, The
Netherlands,, for statistical analysis of the data; and the Thrombosis Services who
participatedd in the collection of the blood samples.
References s
1.. Prince AM, Brotman B, Grady GF, et al: Long incubation post-transfusion hepatitis
withoutt serological evidence of exposure to hepatitis B virus. Lancet 1974;11:241-6.
2.. Choo QL, Kuo G, Weiner AJ, et al: Isolation of a cDNA clone derived from a blood-
bornee non-A, non-B viral hepatitis genome. Science 1989;244:359-62.
124 4
3.. Alter HJ, Purcell RH, Shih JW, et al: Detection of antibody to hepatitis C virus in
prospectivelyy followed transfusion recipients with acute and chronic non-A, non-B
hepatitis.. N Engl J Med 1989;321:1494-500.
4.. Alter HJ, Jett BW, Polito AJ, et al: Analysis of the role of hepatitis C virus in
transfusion-associatedd hepatitis. In: Hollinger FB, Lemon SM, Margolis H, eds. Viral
hepatitiss and liver disease. Baltimore: Williams & Wilkins, 1991:396-402.
5.. Busch MP, Kleinman SH, report of the interorganizational task force on nucleic acid
amplificationn testing of blood donors: Nucleic acid amplification testing of blood
donorss for transfusion-transmitted infectious diseases. Transfusion 2000;40:143-59.
6.. Simons JN, Leary TP, Dawson GJ, et al: Isolation of novel virus-like sequences
associatedd with human hepatitis. Nat Med 1995;1:564-9.
7.. Linnen J, Wages J Jr., Zhang-Keck Z-Y, et al: Molecular cloning and disease
associationn of hepatitis G virus: a transfusion-transmissible agent. Science 1996;
271:505-8. .
8.. Karayiannis P, Thomas HC: Current status of hepatitis G virus (GBV-C) in
Transfusion:: is it relevant? Vox Sang 1997;73:63-9.
9.. Roth WK, Waschk D, Marx S, et al: Prevalence of hepatitis G virus and its strain
variant,, the GB agent, in blood donations and recipients. Transfusion 1997;37:651-6.
10.. Jongerius JM, Boland GJ, van der Poel CL: GB virus type C viremia and envelope
antibodiess among population subsets in The Netherlands. Vox Sang 1999;76:81-4.
11.. Tacke M, Kiyosawa K, Stark K, et al: Detection of antibodies to a putative hepatitis
GG virus envelope protein. Lancet 1997;349:318-20.
12.Niiblingg CM, Gröner A, Lower J: GB virus C/hepatitis G virus and intravenous
immunoglobulins.. Vox Sang 1998;75:189-92.
13.Cristianoo K, Bisso G, Wirz M: Do antibodies to the hepatitis G virus E2 antigen in
immunoglobulinn products reduce infectivity? Transfusion 1999;39:1271-2.
14.. Schreiber GB, Busch MP, Kleinman SH, et al., for the retrovirus epidemiology donor
study:: The risk of transfusion-transmitted viral infections. N Engl J Med
1996;334:1685-90. .
125 5
15.Müller-Breitkreuzz K, for the EPFA working group on quality assurance: Results of
virall marker screening of unpaid blood donations and probability of window period
donationss in 1997. Vox Sang 2000;78:149-57.
16.. Van de Watering LMG, Hermans J, Houbiers JGA, et al: Beneficial effects of
leukocytee depletion of transfused blood on postoperative complications in patients
undergoingg cardiac surgery, a randomized clinical trial. Circulation 1998;97:562-8.
17.. Jongerius JM, Bovenhorst M, van der Poel CL, et al: Evaluation of automated nucleic
acidd extraction devices for application in HCV NAT. Transfusion 2000;40:871-4.
18.. Boom R, Sol CJA, Salimans MMM, et al: Rapid and simple method for the
purificationn of nucleic acids. J Clin Microbiol 1990;28:495-503.
19.Zaaijerr HL, ter Borg F, Cuijpers HTM, et al: Comparison of methods for detection of
hepatitiss B virus DNA. J Clin Microbiol 1994;34:877-80.
20.. Van der Poel CL, Reesink HW, Schaasberg W, et al: Infectivity of blood seropositive
forr hepatitis C virus antibodies. Lancet 1990;335:558-60.
21.Goodnoughh LT, Brecher ME, Kanter MH, et al: Medical progress Transfusion
Medicinee first of two parts Blood Transfusion (review article). N Engl J Med
1999;340:438-47. .
126 6
Chapterr 10
Summary y
Samenvatting g
127 7
128 8
Summary y
Thee studies described in this thesis deal with the limitations of serological
screeningg assays for the detection of hepatitis B virus (HBV), human immunodeficiency
viruss (HIV), and hepatitis C virus (HCV) in blood donations. Secondly, the results of the
implementationn of molecular biological screening techniques to reduce the risk of
transmissionn of HCV and HIV via blood in The Netherlands are presented. Thirdly, the
prevalencee of a recently identified blood-borne virus, GB virus type C (GBV-C), among
severall population subsets in The Netherlands was determined. Lastly, the incidence of
post-transfusionn GBV-C, HCV and HBV infection in patients who underwent cardiac
surgeryy between 1992-1994 was determined.
Itt was demonstrated by others that HIV-1 subtype O is not reliably detected by
certainn routinely used anti-HIV-1 screening assays. Especially assays based on synthetic
peptidess or recombinant antigens fail to detect anti-HIV-1 subtype O antibodies. Assays
basedd on whole-virus lysate, however, appear to detect a broader repertoire of HTV
antibodies.. The false-negative results obtained with anti-HIV-1 subtype O samples in
recombinantt or peptide based assays led to the modification of HIV assays by the
manufacturers.. In Chapter 2, we describe a comparative study of two subtype O enhanced
anti-HIV-1/22 enzyme-linked immunosorbent assays (ELISAs) (Abbott Plus and Ortho
Enhanced)) with a routinely used anti-HIV-1/2 ELISA (Abbott Recombinant). It was
demonstratedd that two Western blot-confirmed anti-HIV-1-positive samples were missed
byy the Abbott Recombinant ELISA but detected by the subtype O enhanced ELISAs.
Apparently,, anti-HIV-1/2 subtype O-enhanced ELISAs and lysate-based ELISAs are more
sensitivee than recombinant-based ELISAs. It was also demonstrated that the analytic
sensitivityy of the Ortho Enhanced ELISA was inferior to that of both Abbott ELISAs. The
specificitiess of the Abbott Recombinant, Abbott Plus, and Ortho Enhanced ELISAs were
comparable.. The heterogeneity of HIV demands continuous global monitoring and timely
adaptationn of HIV antibody assays. We recommend that modifications to increase
129 9
sensitivityy for HIV variant detection should always be evaluated to ensure that the
modificationss in the assays included by manufacturers do not result in a loss of sensitivity
andd specificity for antibodies directed against the more common HIV types.
Hepatitiss B surface Antigen (HBsAg) screening of blood donations does not
completelyy eliminate the risk of HBV infection by blood transfusion. HBsAg screening
testss may be negative in the early window phase of HBV infection, in the early conva-
lescencee phase (core window) of HBV infection, and during chronic HBV infection when
thee level of HBsAg is very low. In addition, blood donors infected with HBsAg mutant
formss of HBV may escape detection by HBsAg screening assays. Limitations of HBsAg
screeningg are reported in Chapter 3 and Chapter 4. We studied the serological and
molecularr biological profile of a repeat blood donor who acquired an HBsAg mutant form
off HBV infection {Chapter 3). HBsAg of this donor was undetectable in 5 (50%) out of
100 routinely used HBsAg screening assays. DNA sequence analysis of the 'a1 determinant
regionn of the donor's HBsAg revealed a double mutation which affected amino acid 129
(fromm Q to R) and amino acid 133 (from M to T). This combination has not earlier been
described.. We recommend improvement of HBsAg screening in order to correct
deficienciess in the detection of HBsAg mutant forms. In addition, we conclude that
implementationn of anti-HBc and / or (minipool) nucleic acid amplification testing for
HBVV DNA may improve the detection of HBsAg mutant forms.
Inn Chapter 4, a case of posttransfusion HBV infection in a multitransfiised hemato-
oncologicc patient is described. The patient was exposed to a total of 200 blood
componentss from 200 donors. To identify HBsAg-negative but HBV infectious blood
donation(s),, we tested mainly archived follow-up samples of the involved donors obtained
££ 3 months after the implicated donations for anti-HBc. 1/200 of these follow-up samples
wass anti-HBc-positive. Retrospective testing of the implicated HBsAg-negative blood
donationn of this donor revealed anti-HBc-negative and HBV-DNA-positive results. The
indexx patient was transfused with the platelets of this donation. During look back, it was
130 0
foundd that other blood products prepared from this HBV-infectious donation caused
posttransfusionn HBV infection in 2 additional patients. This study illustrates the limitation
off HBsAg screening and the usefulness of sample archiving. Without expensive HBV-
DNAA testing on archived samples of 200 blood donations and unnecessary notification of
alll 200 corresponding donors, an HBsAg-negative, but infectious blood donation was
traced.. It has been advocated that additional screening tests such as anti-HBc and / or
(minipool)) nucleic acid amplification testing for HBV DNA (HBV NAT) might improve
thee safety of the blood supply. In this case routine anti-HBc testing would not have
preventedd the transmission of HBV. On the other hand, the safety of quarantine fresh
frozenfrozen plasma might benefit from additional screening of the second blood donation for
anti-HBc.. Given the weak HBV DNA polymerase chain reaction (PCR) signal in our
HBVV infectious donation, it is unlikely that it would have been detected by HBV NAT
testingg in (mini)pools. HBV NAT on single donations would probably have detected this
earlyy window donation.
Inn Chapter 5, we compared the sensitivity of HCV RNA extraction using three
commerciallyy available methods optional for application in minipool HCV NAT. Nucleic
acidss from l-in-3 serial dilutions of an HCV RNA run control were extracted and purified
using,, respectively, the manual Cobas Amplicor, the automated BioRobot 9604, and the
automatedd NucliSens Extractor method. HCV PCR of all extracts was performed using
thee Amplicor HCV v2.0 test. According to the Committee for Proprietary Medicinal
Productss (CPMP), the detection limit of HCV NAT should be 270 geq per mL. We found
thatt the Cobas Amplicor, the BioRobot 9604, and the NucliSens Extractor methods in
combinationn with the Amplicor HCV v2.0 testt allow a 95-percent HCV RNA detection
limitt of 129, 82, and 12 geq per mL, respectively. Paul Ehrlich Institute (PEI) stipulate
thatt at least 13,500 geq per mL should be detected to calculate the quantity of HCV RNA
contributedd by individual donations composing a minipool. The maximal pool size for the
Cobass Amplicor, the BioRobot 9604 and the NucliSens Extractor kits that would still meet
thiss PEI regulation was calculated at 104, 164, and 1125 donations, respectively. The data
131 1
showw that all three evaluated HCV NAT extraction systems combined with the Amplicor
HCVV v2.0 test meet the criteria defined by the CPMP and PEI when minipools of up to
1044 donations are used. It was also shown that the difference in the sensitivity of the
laboratoryy procedures is in line with the difference in plasma equivalents analyzed in the
PCR.. We conclude that the highest sensitivity for HCV NAT minipool screening can be
achievedd using the NucliSens Extractor in combination with the Cobas Amplicor HCV
v2.00 test.
Inn The Netherlands, routine HCV NAT minipool screening of 48 donations with
negativee NAT as release criterion for all blood components has become mandatory since
Julyy 1999. Recently, it was decided to expand the HCV NAT minipool screening program
withh HIV NAT. The NucliSens HIV-1 QL test (Organon Teknika) and the AmpliScreen
HIV-11 vl.5 test (Roche Diagnostic Systems), both combined with NucliSens Extractor,
weree optional for implementation in the NAT laboratories. In Chapter 6, HIV NAT test
optionss are compared. Sensitivity for HIV-1 genotype B and E RNA was determined by
testingg HIV RNA reference panels. For the NucliSens HIV-1 QL test, the 95 percent
detectionn limit for HIV-1 genotype B RNA and genotype E RNA was found to be 102 geq
perr mL and 190 geq per mL, respectively. For the AmpliScreen HIV-1 vl .5 test, the 95
percentt detection limit for HIV-1 genotype B RNA and genotype E RNA was found to be
133 geq per mL and 39 geq per mL, respectively. Thus, the AmpliScreen HIV-1 vl.5 test
inn comparison with the NucliSens HIV-1 QL test is approximately eight times more
sensitivee in detecting HIV-1 genotype B RNA and five times more sensitive in detecting
HIV-11 genotype E RNA. The specificity of both HIV NAT tests was determined by
analyzingg at least 100 HIV-1 RNA-negative human plasma minipools of 48 donations
each.. When 112 minipools were analyzed using the NucliSens HIV-1 QL test, no false-
positivee results were found. However, 2 of 103 (1.9%) minipools analyzed with the
AmpliScreenn HIV-1 vl.5 test reacted false-positive. An additional false-positive result
withh this test was observed during alternate testing of HIV-1 RNA high positive and
negativee samples. All false-positive test results were found at the beginning of the study.
132 2
False-positivityy was probably caused by a contaminated set of pipettes. Thus, in
comparisonn with the AmpliScreen HIV-1 vl.5 test, the NucliSens HIV-1 QL test may
providee better specificity. Alternate testing of HIV-1 RNA-high-positive and negative
sampless revealed 1 false-positive result using the AmpliScreen HIV-1 test. Repeat
amplificationn and detection of the corresponding eluate was negative for HIV-1 RNA.
Bothh HIV NAT tests options detected all HIV-1 RNA positive run controls. Failure of
internall control detection was observed in 1 of 307 (0.3%) samples analyzed with the
NucliSenss HIV-1 QL test. No internal control failure was observed upon analysis of 299
sampless using the Ampliscreen HIV-1 vl.5 test. Thus, the robustness of both HIV NAT
testt options is comparable. Based on the data of this study, we recommend further studies
off NucliSens extraction in combination with the AmpliScreen HIV-1 vl.5 test. The results
off such a study are reported in Chapter 7.
Inn The Netherlands, routine HCV NAT minipool screening (max 48 donations) of
alll blood donations was extended to include HIV NAT as of 22 November 2000. In
ChapterChapter 7, we describe the validation of NucliSens Extractor (EluHigh procedure) in
combinationn with the AmpliScreen HIV vl.5 and HCV v2.0 test on an analyzer (Cobas).
HIV-11 RNA genotype B, genotype E, and HCV genotype 1 reference panels were tested
244 times to determine the sensitivity. Robustness and specificity was determined by
alternatee testing of minipools of up to 48 donations with run controls containing 38 and
1400 geq per mL HIV-1 and HCV RNA. The 95-percent detection limits for HIV-1 RNA
genotypee B and E were found to be 32 geq per mL and 30 geq per mL, respectively. The
95-percentt detection limit for HCV genotype 1 RNA was calculated 21 geq per mL.
Analysiss of 2332 samples for HIV NAT revealed 13 (0.56%) invalid test results. Analysis
off 2644 samples for HCV NAT revealed 12 (0.45%) invalid test results. Both tests
adequatelyy detected the run controls. Furthermore, we reported on the performance
characteristicss of the NucIiSens-AmpliScreen method obtained during the first eleven
monthss of routine HIV and HCV NAT screening in which over 19600 samples (minipools
andd run controls) were tested. Invalid test results for HIV-1 RNA and HCV RNA were
133 3
foundd in, respectively, 1.1% and 1.07% of the samples tested. HIV-1 RNA minipool
testingg resulted in 27 (0.16%) initial false-positive results and 3 (0.02%) confirmed
positivee results. HCV RNA minipool testing resulted in 4 (0.02%) initial false-positive
resultss and 5 (0.02%) confirmed positive results. No seronegative window donations were
detected.. The run controls of either 140 geq per ml or 38 geq per mL were respectively
foundd positive in 2201/2219 (99.2%) and 534/560 (95.4%) of the HIV-1 RNA tests and
inn 2201/2218 (99.2%) and 534/565 (94.5%) of the HCV RNA tests. We conclude that
routinee HIV and HCV NAT minipool screening of maximum 48 donations, utilizing
NucliSenss Extractor combined with the AmpliScreen HIV-1 vl.5 and AmpliScreen HCV
v2.00 test meets the requirements for sensitivity set by a number of international regulatory
bodiess and provides sufficient specificity and robustness for timely release of blood
donations. .
Inn Chapter 8, we determined the prevalence of GB virus type C (GBV-C) markers
inn various population subsets in The Netherlands. Enzyme immunoassays (EIA) were
employedd for the detection of GBV-C anti-E2 antibodies and a reverse-transcription
polymerasee chain reaction amplification (RT-PCR) assay for the detection of GBV-C
RNA.. GBV-C RNA positivity, respectively, GBV-C anti-E2 positivity was found in 2
(6%)) and 3 (9%) of 34 patients with non-A-E hepatitis, 2 (20%) and 0 (0%) of 10 patients
withh HBV infection, 10 (25%) and 19 (48%) of 40 patients with HCV infection, 1 (13%)
andd 0 (0%) of 8 patients with autoimmune hepatitis (AM), 24 (24%) and 72 (71%) of 102
intravenouss drug users (IVDUs), 1 (3%) and 2 (6%) of 34 blood donors with
indeterminatee anti-HCV recombinant immunoblot assay reactivity, and 3 (1.2%) and 8
(3.2%)) of 250 first-time blood donors. The profile of simultaneous GBV-C RNA
positivityy plus GBV-C anti-E2 positivity was found in 2/40 (5%) HCV patients, 4/102
(4%)) IVDUs and 1/250 (0.4%) first time blood donors. This profile may represent early
virall clearance, incomplete protection by circulating GBV-C anti-E2 antibodies or false-
positivee EIA results. In The Netherlands, GBV-C appears not to be a major cause of non-
A-EE hepatitis and AIH. GBV-C infection is associated with parenteral risks: 90% of
134 4
IVDUss and 68% of the HCV patients had GBV-C markers whereas only 4% of the first-
timee blood donors had GBV-C markers. The prevalence rate of GBV-C viremia in first-
timee blood donors was higher than that of HCV viremia (1.2 vs. 0.04%). GBV-C viremia
inn IVDUs is less common than HCV viremia (24 vs. 59%). In addition, judging from the
highh prevalence of GBV-C anti-E2 (71%), most IVDUs have been exposed to GBV-C.
Furthermore,, most persons with GBV-C markers are GBV-C RNA-negative and anti-E2-
confirmedd positive, suggesting that GBV-C infection is transient.
Inn Chapter 9, the incidence of post-transfusion GBV-C, HCV and HBV infection
amongg 373 patients who underwent cardiac surgery and received 3,270 blood product
transfusionss between 1992-1994 i.e. after the introduction of anti-HCV antibody screening
off blood donors is presented. Stored pre and post-transfusion samples from these patients
weree tested for GBV-C, HCV and HBV markers. Of 373 patients, 41(11%) were GBV-C
exposedd before the surgery related transfusion episode. It was found that GBV-C marker-
positivityy preceding the cardiac surgery related transfusion episode was associated with
aa history of previous blood transfusion or major surgery. Another 15 (4%) patients became
infectedd with GBV-C (GBV-C RNA-positive) following the surgery related transfusion
episode,, and 6 (1.6%) patients seroconverted for GBV-C anti-E2 antibodies but no GBV-
CC RNA was found. The remaining 311 patients showed negative results for GBV-C
markerss in the post-transfusion samples and were not further studied. Concomitant GBV-
CC RNA positivity and GBV-C anti-E2 positivity was found in 3 patients. This profile may
representt early viral clearance, chronic persistent carriership or false-positive EIA results
andd was reported in Chapter 8. The results of this study indicate that GBV-C is
transmittedd by blood transfusion. In addition, it was found that clearance of GBV-C is
associatedd with an anti-E2 response. None of the 373 patients became infected with HCV
andd or HBV while a prospective study carried out in The Netherlands before the
introductionn of anti-HCV screening found post-transfusion HCV infection in 9/383 (2.3%)
patientss undergoing cardiac surgery. Thus, our study also shows the increased safety of
thee blood supply with respect to HCV.
135 5
136 6
Samenvatting g
Ditt proefschrift bevat studies over een aantal virale bloedoverdraagbare
infectieziekten.. Allereerst (hoofdstuk 2 tot en met 4) wordt onderzoek gepresenteerd over
dee beperkingen van het serologisch screeningsonderzoek van donorbloed op infecties met
humaann immunodeficiency virus (HIV) en hepatitis B virus (HBV). Vervolgens
(hoofdstukk 5 t/m 7) wordt de introductie en toepassing van moleculair biologisch routine-
onderzoekk van bloed op de aanwezigheid van HCV- en / of HIV-RNA beschreven. Daarna
(hoofdstukk 8) wordt de prevalentie (percentage besmettingen) van hett recent ontdekte GB
viruss type C (GBV-C) onder verschillende bevolkingsgroepen in Nederland gerapporteerd.
Tott slot (hoofdstuk 9) worden de resultaten van een studie naar de incidentie (percentage
nieuwee besmettingen) van posttransfusie GBV-C, HCV en HBV infectie bij patiënten met
eenn open hart operatie gepresenteerd.
Inn Hoofdstuk 1, de inleiding, wordt achtergrondinformatie gegeven over HBV,
HIV,, hepatitis C virus (HCV) en GBV-C. Daarnaast wordt ingegaan op het, ondanks alle
genomenn maatregelen, nog aanwezige rest-risico van HBV-, HIV- en HCV-overdracht
doorr transfusie van bloedproducten.
Uitt onderzoek bleek dat antistoffen gericht tegen een nieuwe variant van HIV
(HIV-11 subtype O) niet door alle routinematig gebruikte anti-HIV-1/2 screeningstesten
herkendd werden. Met name werd dit geconstateerd bij het gebruik van op recombinant
antigenenn en synthetische eiwitten gebaseerde tests. Door deze tekortkoming werden
fabrikantenn genoodzaakt tot snelle aanpassing / verbetering van het commercieel
verkrijgbaree pakket HIV screeningstesten. In Hoofdstuk 2, vergeleken we de gevoeligheid
enn de specificiteit van twee nieuwe 3e generatie, voor HIV-1 subtype O versterkte, anti-
HIV1/22 enzyme linked immunosorbert assays (ELISAs) (Abbott Plus en Ortho Enhanced)
enn één routinematig gebruikte 3e generatie anti-HIVl/2 ELISA (Abbott Recombinant). In
tegenstellingg tot de Abbott Plus en de Ortho Enhanced test werden 2/91 (2%), als echt
137 7
anti-HTV-1/22 positief beschouwde bloedmonsters (door de Western Blot test bevestigd),
gemistt met de Abbott Recombinant test. Blijkbaar is de klinische sensitiviteit
(gevoeligheid)) voor het aantonen van anti-HIV antistoffen van beide subtype O versterkte
ELISAss (Abbott Plus en Ortho Enhanced) beter dan die van de Abbott Recombinant
ELISA.. Uit de testresultaten verkregen met anti-HIV-1-positieve bloedmonsters in
verdunningsreeksenn bleek dat de analytische sensitiviteit van de Ortho Enhanced het
laagstt was, gevolgd door de Abbott Recombinant en de Abbott Plus test (de analytische
sensitiviteitt is geen graadmeter voor de klinische sensitiviteit). Het testen van monsters
afkomstigg van bloeddonors gaf geen verschil in specificiteit tussen de drie testen te zien.
Geconcludeerdd wordt dat de heterogeniteit van HIV om voortdurende alertheid van zowel
onderzoekerss als fabrikanten vraagt en, indien noodzakelijk (b.v. bij de ontdekking van
nieuwee varianten), direct moet leiden tot aanpassing van anti-HIV screening tests.
Aanbevolenn wordt om gemodificeerde tests altijd te (blijven) evalueren. Verbetering van
dee gevoeligheid voor nieuwe HIV varianten mag tenslotte niet leiden tot verlies van
sensitiviteitt en specificiteit voor antistoffen die gericht zijn tegen de regelmatig
voorkomendee HIV stammen.
HBVV screening van bloeddonaties in Nederland is gebaseerd op detectie van het
hepatitiss B oppervlakte antigeen (HBsAg). Tijdens de vroege fase (early window), de
herstelfasee (core-window), bij lage concentraties HBsAg en bij de aanwezigheid van
HBsAgg mutanten kan een HBV-infectie, door HBsAg-screeningtests, gemist worden.
Beperkingenn van HBsAg screening worden in Hoofdstuk 3 en Hoofdstuk 4 gerapporteerd.
Wijj onderzochten het serologisch en moleculair biologisch profiel van een HBV
geïnfecteerdee bloeddonor (Hoofdstuk 3). Het HBsAg van deze HBV geïnfecteerde donor
wass niet aantoonbaar in 5/10 (50%) routinematig gebruikte HBsAg screeningtests.
Sequentiee analyse op het Polymerase Chain Rection (PCR) amplificaat van de 'a'
determinantt van het HBsAg, leverde 2 niet eerder beschreven mutaties op in dit gebied
vann het HBV genoom, namelijk van Q (Glutamine) naar R (Arginine) in codon 129, en
vann M (Methionine) naar T (Treonine) in codon 133. Geconcludeerd wordt dat
138 8
maatregelenn nodig zijn om de gevoeligheid van HBsAg screeningstesten voor de detectie
vann HBsAg mutanten te verbeteren. Uit de resultaten van longitudinaal onderzoek van
bloedmonsterss van deze donor wordt geconcludeerd dat door toevoeging van een anti-
hepatitiss B core (anti-HBc) test en of een HBV-DNA minipool test aan het
screeningsalgoritme,, de kans op transmissie van HBsAg mutanten van HBV misschien
gereduceerdd kan worden.
Inn Hoofstuk 4, beschrijven we het onderzoek naar de oorzaak van een
posttransfusiee hepatitis B infectie bij een ontvanger van 200 bloedproducten, welke uit de
donatiess van 200 donors bereid waren. In de routine screening waren al de 200
bloeddonatiess HBsAg-negatief. Om uit te sluiten dat een mogelijk besmettelijke donatie
(b.v.. infectie in de vroege fase, infectie in de herstelfase, mutant virus) met de HBsAg
screeningg zou zijn gemist, werden bloedmonsters, die langer dan drie maanden na de 200
verdachtee donaties waren afgenomen, getest op anti-HBc. Bij 172 donors werd dit
onderzoekk uitgevoerd op een bewaard bloedmonster (verplichting in Nederland) van een
lateree donatie; bij de overige 28 donors werd hiertoe een nieuw bloedmonster afgenomen.
Bijj het testen van de vervolgmonsters werd één monster anti-HBc positief bevonden.
Retrospectieff onderzoek op een bewaard bloedmonster van de corresponderende verdachte
HBsAg-negatievee donatie gaf de volgende resultaten te zien: anti-HBc-negatief en HBV-
DNA-positieff (zwak signaal). De ontvanger was met de trombocyten uit deze HBV
geïnfecteerdee donatie getransfundeerd. De erytrocyten en het plasma afkomstig van deze
donatiee was aan 2 andere ontvangers toegediend. Beide bleken (look back studie), door
transfusiee met deze producten, met HBV geïnfecteerd te zijn. Zonder grootschalig gebruik
tee maken van een kostbare en arbeidsintensieve HBV-DNA polymerase chain reaction
(PCR)) test werd, met de door ons bedachte strategie, een HBsAg-negatieve maar HBV
geïnfecteerdee bloeddonatie getraceerd. Uit longitudinaal onderzoek bleek dat deze donatie
afkomstigg was van een donor in de vroege fase van HBV-infectie. Het routinematig
screenenn van bloeddonaties op de aanwezigheid van anti-HBc antistoffen had de
beschrevenn posttransfusie HBV infecties echter niet kunnen voorkomen. Vanwege het
139 9
zwakkee signaal in de HBV-DNA PCR wordt geconcludeerd dat het routinematig screenen
vann bloeddonaties met behulp van een HBV nucleïnezuur amplificatie test (HBV NAT)
inn minipools beschreven casuïstiek niet had kunnen voorkomen. Mogelijk had een HBV
NATT op onverdunde bloedmonsters dit wel gekund.
Hett routinematig toepassen van moleculair biologisch onderzoek op
bloedoverdraagbaree virale infecties kan de veiligheid van de bloedvoorraad ten goede
komen.. Met name het screenen van bloeddonaties op HCV-RNA en HIV-RNA lijkt
geschiktt om donors in de vroege fase (window-fase) van respectievelijk een HCV of een
HIVV infectie op te sporen. In hoofdstuk 5 t/m 7 worden de resultaten van onderzoek over
dee toepassing en introductie van HCV- en HIV-RNA screeningstesten gerapporteerd. In
HoofdstukHoofdstuk 5, vergeleken we de sensitiviteit van één manuele en twee automatische
nucleïnezuurr extractie methodes (respectievelijk, Cobas Amplicor, BioRobot 9604 en de
NucliSenss Extractor) in combinatie met de Amplicor HCV v2.0 test. HCV RNA extractie
inn combinatie met een nucleïnezuur amplificatie test (NAT) wordt internationaal HCV
NATT genoemd. De voor dit onderzoek benodigde extractieprocedures werden uitgevoerd
opp HCV-RNA genotype 1 verdunningsseries. De Committee for Proprietary Medicinal
Productss (CPMP) van de Europese Unie schrij ft een detectielimiet van 270 geq/mL voor
eenn HCV NAT test voor. We vonden een 95% HCV RNA detectielimiet van 129 geq/mL
voorr Cobas Amplicor, 82 geq/mL voor BioRobot 9604 en 12 geq/mL voor NucliSens
Extractor.. Het verschil in de sensitiviteit van de testcombinaties bleek ongeveer evenredig
tee zijn met de hoeveelheid van het oorspronkelijk geëxtraheerde plasma dat in de NAT
onderzochtt wordt. De HCV NAT test wordt meestal uitgevoerd op minipools
samengesteldd uit meerdere donaties. Het Paul Ehrlich Instituut (PEI) hanteert een
detectielimiett van 13.500 geq/mL voor bloedmonsters afkomstig van individuele donaties
inn een minipool. De maximale poolgrootte, om aan dit PEI voorschrift te voldoen, werd
doorr ons berekend op 104, 164 en 1125 donaties voor respectievelijk Cobas Amplicor,
BioRobott 9604 en NucliSens Extractor. Geconcludeerd wordt dat de drie
extractiemethodieken,, in combinatie met de Cobas Amplicor HCV v2.0 test, aan zowel
140 140
CPMPP en PEI voorschriften voldoen indien minipools samengesteld uit maximaal 104
donatiess worden getest. Verder wordt geconcludeerd dat de Cobas Amplicor HCV v2.0
testt gecombineerd met de NucliSens Extractor de gevoeligste van de drie onderzochte
HCVV NAT methodieken is.
Inn juli 1999 werd HCV NAT screening op minipools, samengesteld uit
bloedmonsterss afkomstig van maximaal 48 donaties, in Nederland ingevoerd. De
NucliSenss Extractor gecombineerd met de Cobas Amplicor HCV v2.0 test (zie conclusie
HoofdstukHoofdstuk 5) werd daartoe in de 4 NAT laboratoria van Sanquin geïmplementeerd.
Vervolgenss werd besloten, uitgaande van de aanwezige HCV NAT infrastructuur, om in
dee loop van het jaar 2000 HTV NAT screening te implementeren. De HTV-1 RNA modules
vann twee fabrikanten kwamen hiervoor in aanmerking namelijk de NucliSens HIV-1 QL
testt van Organon Teknika en de AmpliScreen HIV-1 vl.5 test van Roche Diagnostic
Systems.. In Hoofdstuk 6, vergeleken we de sensitiviteit, specificiteit en robuustheid van
beidee modules in combinatie met de NucliSens Extractor (Organon Teknika). De 95%
detectielimietenn voor HTV-1 RNA genotype B en E werden vastgesteld door het testen van
HIV-11 RNA genotype B en E verdunningsseries. We vonden 95% HIV-1 RNA genotype
BB en E detectielimieten van respectievelijk 102 en 190 geq/mL voor de HIV-1 QL test en
133 en 39 geq/mL voor de AmpliScreen HIV-1 vl.5 test. De specificiteit en robuustheid
werdenn onderzocht met behulp van HIV-1 RNA-negatieve minipools en positieve
monsterss met een hoge of een lage concentratie HIV-1 RNA. Bij gebruik van de
AmpliScreenn test werden 3 vals-positieve resultaten, waarschijnlijk veroorzaakt door een
gecontamineerdee pipettenset, in de eerste 4 runs gevonden. Met beide tests werden alle
HIV-11 RNA-positieve controle monsters gedetecteerd. Met de NucliSens HIV-1 QL tests
zijnn totaal 307 monsters getest en is 1 (0,3%) invalid resultaat (resultaat voldoet niet aan
doorr de fabrikant gestelde eisen) waargenomen. Met de AmpliScreen test zijn totaal 299
monsterss getest. Hierbij werd geen invalid resultaat waargenomen. Geconcludeerd wordt
datt de AmpliScreen test in vergelijking met de HIV-1 QL tests ongeveer acht keer
gevoeligerr is in het detecteren van HIV-1 RNA genotype B en vijf keer in het detecteren
141 1
vann HIV-1 RNA genotype E. De HIV-1 QL test geeft betere resultaten in het
specificiteitonderzoekk te zien. De robuustheid van beide tests is vergelijkbaar. Op basis
vann deze data werd aan Sanquin geadviseerd om een aanvullende onderzoek {Hoofdstuk
7)7) te starten waarbij de AmpliScreen HIV-1 vl.5 tests binnen de huidige HCV NAT
configuratiee voor minipool screening van bloeddonaties wordt geïntegreerd en getest.
Inn Hoofdstuk 7, valideerden we de NucliSens Extractor (EluHigh procedure) in
combinatiee met de AmpliScreen HTV vl .5 en HCV v 2.0 test. Ter beperking van het aantal
extractieproceduress werd de test opzet zo gekozen dat voor uitvoering van HIV NAT
hetzelfdee eluaat gebruikt kon worden dat al voor HCV NAT minipool screening werd
bereid.. Voor het vaststellen van de sensitiviteit werden HIV-1 RNA genotype B, HIV-1
RNAA genotype E en HCV RNA genotype 1 referentiemonsters in verdunningsseries
getest.. Specificiteit en robuustheid werden onderzocht door het testen van plasma
minipoolss en zogenaamde spion-controles (monsters met een concentratie van 38 geq/mL
enn 140 geq/mL voor zowel HIV-1 als HCV RNA). We vonden 95% detectielimieten voor
HIV-11 RNA genotype B, HIV-1 RNA genotype E en HCV RNA genotype 1 van
respectievelijkk 32, 30 en 21 geq/mL. Met HIV NAT werden totaal 2332 monsters getest
enn 13 (0,56%) invalid resultaten waargenomen. Met HCV NAT werden totaal 2644
monsterss getest. Hierbij werden 12 (0,45%) invalid resultaten waargenomen. De spion-
controless werden adequaat gedetecteerd. In hoofdstuk 7 analyseerden we tevens de
resultatenn welke, gedurende de eerste elf maanden routinematig HIV en HCV NAT
screenenn van bloeddonaties (~ 800.000 donaties getest in ~ 17.000 minipools), door de
44 NAT laboratoria centraal gerapporteerd werden. Gedurende deze periode werden geen
serologjschh negatieve, HCV-RNA en of HIV-RNA-positieve bloeddonaties gevonden. Met
HIVV NAT werden 1,1% invalid resultaten gevonden en met HCV NAT 1,07%. HIV NAT
screeningg gaf 0,16% initieel vals-positieve resultaten te zien, HCV NAT 0,02%. De spion-
controless met een concentratie van 140 geq/mL en 38 geq/mL werden respectievelijk
positieff gevonden in 2201/2219 (99,2%) en 534/560 (95,4%) van HIV-1 RNA tests en in
2201/22188 (99,2%) en 534/565 (94,5%) van de HCV RNA tests. Geconcludeerd wordt
!42 2
datt het routinematig minipool screenen van bloeddonaties op HIV-1 en HCV RNA,
gebruikmakendd van de NucliSens Extractor in combinatie met de AmpliScreen HIV-1 vl .5
testt en de AmpliScreen HCV v2.0 test, voldoet aan zowel internationaal als door Sanquin
gesteldee criteria voor sensitiviteit, specificiteit en robuustheid.
Mett behulp van moleculair biologische technieken werden recent twee nieuwe
bloedoverdraagbaarr virussen, respectievelijk GB virus type C (GBV-C) en hepatitis G
viruss (HGV) genoemd, ontdekt. Uit aanvullend onderzoek kwam vast te staan dat het om
hetzelfdee virus, verder GBV-C genoemd, ging. Ook werd een genetisch verwantschap
tussenn GBV-C en HCV aangetoond. In Hoofdstuk 8, onderzochten we de prevalenties van
GBV-CC RNA en GBV-C anti-E2 onder verschillende bevolkingsgroepen in Nederland met
behulpp van respectievelijk een NAT module en ELISAs. GBV-C RNA, respectievelijk,
GBV-CC anti-E2 werd gevonden bij 2 (6%) en 3 (9%) van de 34 patiënten met non-A-E
hepatitiss (niet met hepatitis A, B, C, D, E geïnfecteerd), 2 (20%) en 0 (0%) van de 10
patiëntenn met een HBV infectie, 10 (25%) en 19 (48%) van de 40 patiënten met een HCV
infectie,, 1 (13%) en 0 (0%) van de 8 patiënten met een autoimmuun hepatitis (AIH), 24
(24%)) en 72 (71%) van de 102 intraveneuze druggebruikers (IVDUs), 1 (3%) en 2 (6%)
vann de 34 bloeddonors met een dubieuze anti-HCV recombinant immunoblot assay
reactiviteit,, en 3 (1,2%) en 8 (3,2%) van de 250 nieuwe bloeddonors. Een positief
resultaatt voor zowel GBV-C RNA als GBV-C anti-E2 antistoffen werd gevonden bij 2
(5%)) van de 40 HCV patiënten, 4 (4%) van de 102 IVDUs en 1 (0,4%) van de 250 nieuwe
bloeddonors.. Klaren van de GBV-C infectie, chronisch dragerschap of vals-positieve
ELISAA uitslagen kan een verklaring voor dit profiel zijn. Bij 90% van de IVDUs en 68%
vann de HCV patiëntenn werden GBV-C markers aangetoond. Dit in tegenstelling tot 4%
bijj nieuwe bloeddonors. De prevalentie van GBV-C RNA bij nieuwe bloeddonors was
relatieff hoog (1,2%) in vergelijking tot HCV RNA (0,04%). Daarentegen was de
prevalentiee van GBV-C RNA bij de groep IVDUs relatief laag (24%) in vergelijking tot
HCVV RNA (59%). De bevinding dat de meeste individuen GBV-C RNA-negatief en
GBV-CC anti-E2-positief zijn, is suggestief voor een beschermende werking van GBV-C
143 3
anti-E22 antistoffen. Geconcludeerd wordt dat de meeste gevallen van non-A-E hepatitis
enn AIH niet te verklaren zijn door een infectie met GBV-C. Ook wordt geconcludeerd dat
eenn GBV-C infectie geassocieerd is met parenterale infectie risico's.
Inn Hoofdstuk 9, onderzochten we de incidentie van posttransfusie GBV-C, HCV
enn HBV infectie bij 373 patiënten. Deze patiënten werden tezamen, in de periode 1992-
19944 na de invoering van de anti-HCV screening van bloeddonaties, tijdens en na een
hartoperatie,, met 3.270 bloedproducten getransfundeerd. Opeenvolgende bloedmonsters,
vann voor en na aan de hartoperatie gerelateerde transfusie-episode, werden bewaard en
vervolgenss met behulp van serologische en moleculair biologische technieken onderzocht
opp de aanwezigheid van GBV-C, HCV en HBV markers. 41/373 (11%) patiënten bleken
reedss voor de transfusie episode met GBV-C in aanraking geweest te zijn (GBV-C RNA
off anti-E2 positief). Een voorgeschiedenis van bloedtransfusies of zware operaties was
eenn belangrijke factor voor de aanwezigheid van GBV-C markers bij deze patiënten. Bij
15/3733 (4%) patiënten werd na de transfusie episode, al dan niet persisterend, een GBV-C
viremiaa vastgesteld. 6/373 (1,6%) patiënten vertoonden een seroconversie voor anti-E2
maarr bleven in de volledige serie monsters negatief voor GBV-C RNA. De overige
patiëntenn (311) waren negatief voor GBV-C markers in de posttransfusie monsters en
werdenn derhalve niet verder onderzocht. Gecombineerde aanwezigheid van GBV-C RNA
enn GBV-C anti-E2 antistoffen werd vastgesteld in bloedmonsters van 3 patiënten. Dit
profiell (klaren van GBV-C, chronisch dragerschap of vals-positieve testresultaten) werd
ookk in Hoofdstuk 8 beschreven. Geen van de 373 patiënten bleek na de transfusie-episode
mett HBV of HCV besmet te zijn. In een Nederlandse prospectieve studie naar de
incidentiee van posttransfusie hepatitis non-A, non-B (hepatitis, maar geen hepatitis A en
off B), voor de introductie van anti-HCV screening, werd bij 9 (2,3%) van de 383
patiënten,, na een hartoperatie gerelateerde transfusie-episode, een HCV infectie
vastgesteld.. Geconcludeerd wordt dat GBV-C overdraagbaar is door bloedtransfusie en
datt veiligheid van bloedproducten met betrekking tot HCV transmissie, in Nederland,
belangrijkk is toegenomen.
144 4
Dankwoord d
Eindd 1996, na deelname aan de jaarlijkse bijeenkomst van de American Association
off Blood Banks, vroeg Noor van Leeuwen of ik belangstelling had om reeds verricht,
lopendd en toekomstig laboratoriumonderzoek van de toenmalige Bloedbank Utrecht te
verwerkenn in een "boekje", verder proefschrift genoemd. De lezer van dit proefschrift
begrijptt dat ik op het voorstel van Noor ben ingegaan. Mijn liefde voor onderzoek op het
gebiedd van de immuno-hematologie (lees bloedgroepen) was bekend maar door het
enthousiasmee van Cees van der Poel werd tevens belangstelling gewekt voor onderzoek
opp het gebied van bloedoverdraagbare virale infectieziekten. In een periode van 4-5 jaar
zijnn op dit laatste gebied, ondanks vaak beperkt beschikbare tijd, acht studies verricht.
Veell mensen hebben een bijdrage geleverd aan genoemde studies en derhalve aan het tot
standd komen van dit proefschrift. Dank en lof daarvoor. Een aantal personen wil ik
speciaall noemen en bedanken.
Mijnn dank gaat vooral uit naar:
-- Pim van Aken, mijn promotor, voor het in mij gestelde vertrouwen, het kritisch
beoordelenn van - en de geweldige steun bij de realisatie van dit proefschrift;
-- Noor van Leeuwen en Cees van der Poel, mijn co-promotores, voor het in mij gestelde
vertrouwen,, voor de ideeën die de basis vormden voor een groot deel van de studies,
dee discussies en het kritisch commentaar op de vele manuscripten;
-- Theo Cuijpers en Nico Lely, voor julli e gastvrijheid, enthousiasme, geduld, daadkracht
enn last but certainly not least geweldige kennis en deskundigheid;
-- Nol Teunissen, Merjo Bovenhorst, Rainier Meuken, Wim van Oostendorp, Margret
Sjerpss en Irene Winkel voor het (laboratoriumonderzoek dat julli e hebben verricht.
Nol,, daarnaast ook bedankt voor het "uit de wind houden" en Irene, voor je humor;
-- Ruud van den Akker, Ton van Loon, Rob Schuurman, Ton Kroes, Jan van Hattum,
Greett Boland, Erna Italiaander, Johan van der Reijden, Anneke Brand, Leo van de
Watering,, Johan van der Does en Joost van Hilten, voor de gastvrijheid,
145 5
samenwerking,, hulp bij het onderzoek en kritisch commentaar op de manuscripten;
-- Sanquin HTV-NAT projectgroep bestaande uit Theo Cuijpers, Henk Hopman, Nico
Lely,, Hans Molijn, Frans Mollet, Geert Peeters, Ton Peeters, Cees van der Poel en
Henkk Reesink (voorzitter), voor de samenwerking;
-- Wim Schaasberg, Harry van Drimmelen en Hein Putter, voor de hulp met de statistiek;
-- Alexander Vreede, voor je belangstelling en de broodnodige zetjes in de rug;
-- Martin Smid, voor je enthousiasme, advies en suggesties;
-- De analisten uit het "dreamteam" van de BBMN, voor julli e zelfstandigheid;
-- Janny en Theo de Wildt, voor de morele steun en hulp met de lay-out van het
proefschrift; ;
-- Debby Jongerius, voor het kritisch beoordelen van omgezette tekst op fouten;
-- Frits Fijen, voor je advies en hulp met de lay-out en omslag van het proefschrift;
-- Mijn familie, vrienden en kennissen, waar ik in de nabije toekomst hopelijk meer tijd
voorr vrij kan maken;
-- Mijn ouders en schoonouders, voor de steun en het feit dat ze veel te lang op de
uiteindelijkee realisatie van dit proefschrift hebben moeten wachten.
Debby,, Michael en Danny, dit proefschrift is aan julli e opgedragen. Bedankt dat ook julli e
volgehoudenn hebben.
John. John.
146 6
Otherr publications
1.. Smit HF, Jongerius JM. Swine dysentery diagnosis: To isolate or to fluoresce? The
Veterinaryy Record 1982;343.
2.. Smit HF, van Leengoed AMG, Jongerius JM. Swine dysentery in a sow herd. The
Veterinaryy Quarterly 1985;7:150-3.
3.. Jongerius JM. Erytype, een bruikbaar alternatief voor de controle van ABO-
bloedgroepp en Rhesusfactor bij donoren. Analyse 1986;7:174-5.
4.. Jongerius JM, Teunissen N. Identificatie van allo-antistoffen tegen erytrocyten.
Analysee 1991;3:52-5.
5.. Jongerius JM, Teunissen N. Een immunologisch probleem. Analyse 1991; 5:118.
6.. Jongerius JM, Teunissen N, van Woerkom JW. Microtype. Analyse 1993; 4:84-5.
7.. Jongerius JM, Teunissen N, de Bock H. Automatisering van het bloedbank
laboratorium.. LAB 1994;5:104-7.
8.. Jongerius JM, Rijksen H, Overbeeke MAM, Derks WHJ, van Leeuwen EF.
Ontdekkingg van een nieuw laag frequent bloedgroepantigeen. Analyse 1994;12:272-4.
9.. Jongerius JM, Rijksen H, Overbeeke MAM, Derks WHJ, van Leeuwen EF.
Ontdekkingg van een nieuw laag frequent bloedgroepantigeen. Ned Tijdschr Geneesk
1995;139(9):472. .
10.. Jongerius JM, Daniels GL, Overbeeke MAM, Petty AC, Reid M, Oyen R, Rijksen H,
vann Leeuwen EF. A new low-incidence antigen in the Kell blood group system: VLAN
(KEL25).. Vox Sang 1996;71:43-7.
11.. Jongerius JM, Daniels GL, Overbeeke MAM, Petty AC, Reid M, Oyen R, Rijksen H,
vann Leeuwen EF. KEL25, een nieuw laag frequent antigeen in het Kell-
bloedgroepensysteem.. Analyse 1996;8:216-8.
12.. Jongerius JM, Teunissen N, Wulffraat NM, van Leeuwen EF. Twee patiënten met
positievee directe antiglobulinetest tijdens behandeling met intraveneuze
immunoglobuline.. Ned Tijdschr Geneesk 1997; 141(22): 1126-7.
147 7
13.. Jongerius JM, Wester M, van Oostendorp WR, Cuijpers HTM, Lelie PN, van der Poel
CL,, van Leeuwen EF. Hepatitis B virus mutant gemist door HBsAg screenings-testen.
Nedd Tijdschr Geneesk 1997; 141 (22): 1128.
14.. Jongerius JM, Boland GJ, Rasch MC, Italiaander E, van de Reijden JJ, van der Poel
CL,, van Leeuwen EF en van Hattum J. Prevalentie van hepatitis GB-virus-C in
verschillendee populaties. Ned Tijdschr Geneesk 1999; 143(31): 1638.
15.. Jongerius JM, Boland GJ, Rasch MC, Italiaander E, van de Reijden JJ, van der Poel
CL,, van Leeuwen EF en van Hattum J. De prevalentie van hepatitis-GB-virus-C anti-
E2-antistoffenn in bloeddonors en intraveneuze druggebruikers. Ned Tijdschr Geneesk
1999;; 143(31): 1639.
16.. Jongerius JM, van der Poel CL, van Prooijen HC, Cuijpers HTM, Lelie PN en van
Leeuwenn EF. Eenvoudige opsporingsmethode bij een hepatitis-B(HB)-virusinfectie
doorr transfusie met HB-oppervlakte antigeen negatief donorbloed. Ned Tijdschr
Geneeskk 1999; 143(31): 1636.
148 8