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