cytomegalovirus infection in immunocompetent and ...172666/fulltext01.pdf · cytomegalovirus...

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2008

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 386

Cytomegalovirus Infection inImmunocompetent andRenal Transplant Patients

Clinical Aspects and T-cell Specific Immunity

FREDRIK SUND

ISSN 1651-6206ISBN 978-91-554-7309-9urn:nbn:se:uu:diva-9324

�� !�� "� ��# ��"$%& '� �(� �� ' �� ' )(��(� *+�� ' ,��-. /(� �� 0�� 0��(.

��

�� +. ��#. 1�� 2'�� 2�� /��)��. 1�� 3�� /4�� '�� 2��. 3�� .�� 5#6. #7 ��. ��. 2 8! "7#4"%4&&9475�"4".

1�� *1,:- �� ;4(�� (�� '�� '�� (�� '�4�� (� (�� (��. � � "�< ' (�� '�� 0��( 1,:� �(�� 1,:4��'�� 1�#= �� 1�9= / ��. 2 �� 0��( � �� (0�� 1,: �'�� '�4�(��. /(�� '� �� '�� (��(4�� . 2 �(�� (�� 0� (�� 04�� (�� 1,:4��'�� / �� (��(� �� .2 �� (� 1,: �� ' ��( �(� �� *�- �� (� �� *�-

(�� (� �� ' �� 1,: ��. 2 �(� (��(4�� =>�4 ��?&�< �� 1,: �� (� �� ' ��(�� >� ��4��(��. @� (�� 04�� (�� '� (��(4�� '� �� & �� '�� 0�� '� �(�� (�� . /(� �� ' 1,: �� '�� (� '�� '�� 0�� (�� (�� 0��(�(�� 0��(�� (� �� ''�� '� �(�� (��.2 �(� �=>�= �� 0��( � %&45&< �� ' �� 1,: ��

��'� �(�� (�� (��(�� ' �� '�� (�� (�1,: ��. @� �(��'�� 0��( �� 1,: �'��. 2�� 0��( � �� 1,: �'�� 1,:4��'�� 1�#= / �� 0��(� � ��( �'�� ' �� . 2 �� 0�'�� (�� (� �� 2+!A4�� 1,:4��'�� 1�9= / �� (��4�� (��(4�� 1,: �!3��.

� �� 1�� (�� (�� 1�9 / �� 1�# / ��

�� !�� "#$%&'% ��

B +�� #

2 ! %6&%46��62 8! "7#4"%4&&9475�"4"��$�$��$��$��4"5�9 *(��$>>��.��.��>��C��D��$�$��$��$��4"5�9-

To my family

Papers

This thesis is based on the following papers, which will be referred to in the

text by their Roman numerals:

I Sund F, Wahlberg J, Eriksson BM: CMV disease in CMV-

mismatched renal transplant recipients with prophylactic low-

dose valacyclovir. J Clin Virol 23(1-2):107-11, 2001

II Lidehäll AK, Sund F, Lundberg T, Eriksson BM, Tötterman TH,

Korsgren O: T cell control of primary and latent cytomegalovirus

infections in healthy subjects. J Clin Immunol 25:473-81, 2005

III Sund F, Lidehäll AK, Claesson K, Foss A, Tötterman TH, Kors-

gren O, Eriksson BM: CMV-specific T-cell immunity, viral load,

and clinical outcome in seropositive renal transplant recipients.

Manuscript submitted

IV Sund F, Tufveson G, Döhler B, Opelz G, Eriksson BM: Clinical

outcome with low-dose valacyclovir prophylaxis in high-risk re-

nal transplant recipients: A 10-year experience.

Manuscript in preparation

Papers are reprinted with permission from the publishers.

I © 2001 Elsevier Science B.V.

II © 2005 Springer Science and Business Media

CONTENTS

INTRODUCTION.............................................................................. 11 Epidemiology...........................................................................................12

CMV replication .....................................................................................13

Immune response to CMV infection .....................................................14

Immune evasion ......................................................................................16

Immunological methods .........................................................................16

CMV diagnosis........................................................................................17

Clinical symptoms of CMV infection – direct effects ..........................19

Clinical symptoms of CMV infection – indirect effects.......................21

Prevention and treatment of CMV disease ..........................................21

CMV resistance.......................................................................................26

Vaccine development..............................................................................27

Renal transplantation and allograft rejection......................................28

Immunosuppressive treatment in renal transplantation ....................31

AIMS .................................................................................................. 37 General aim.............................................................................................37

Specific aims............................................................................................37

METHODS ........................................................................................ 39 Ethics .......................................................................................................39

Subject inclusion (I-IV) ..........................................................................39

Blood sampling (II, III) ..........................................................................40

MHC I tetramer analysis (II, III)..........................................................41

CMV-specific T-cell stimulation and intracellular cytokine staining (II, III)......................................................................................................42

Absolute counting of lymphocyte subsets (II, III)................................44

Statistics...................................................................................................44

RESULTS........................................................................................... 46 CMV disease in renal transplant patients (I, III, IV) ..........................46

CMV resistance (IV)...............................................................................48

Rejection and graft function in renal transplant patients (I, III, IV) 48

CMV DNAemia and graft function in renal transplant patients (III)..................................................................................................................49

Long-term clinical outcome in D+/R- renal transplant patients (IV) 50

Neurotoxic adverse effects of valacyclovir prophylaxis (I, IV)...........51

Frequencies of CMV-specific CD8+ T cells (II, III) ...........................52

CD4+ and CD8+ T-cell function (II, III) ..............................................53

Frequency and function of CMV-specific CD8+ T cells (II)...............55

CMV-specific CD4+ T cells and CMV DNAemia (III)........................56

Immunodominance (II) ..........................................................................57

DISCUSSION .................................................................................... 58 VACV prophylaxis in the high-risk (D+/R-) population.....................58

CMV resistance.......................................................................................60

Strategies in non-high-risk populations................................................60

CMV-specific immunity .........................................................................61

CONCLUSIONS ................................................................................ 64 Paper I .....................................................................................................64

Paper II....................................................................................................64

Paper III ..................................................................................................65

Paper IV ..................................................................................................65

FUTURE PERSPECTIVES.............................................................. 66 ACKNOWLEDGMENTS .................................................................. 68 REFERENCES.................................................................................. 72

Abbreviations

aa amino acids

ACV acyclovir

CD cluster of differentiation

CD4/Th T helper cell, Th

CD8 T killer cell, Tk

CFC cytokine flow cytometry

CMV cytomegalovirus

CTL cytotoxic lymphocytes

D+/- donor CMV serostaus

EBV Epstein-Barr virus

FITC fluorescein isothiocyanate

gB/H/M/N glycoprotein B/H/M/N

GCV ganciclovir

HHV human herpes virus

HIV human immunodefiency virus

HLA human leukocyte antigen

HSV herpes simplex virus

IE immediate-early

IgG/M immunoglobulin G/M

IFN interferon

IFNγCD4/8 interferon-gamma-producing CD4/8 T cell

kbp kilobase pair

MHC major histocompatibility complex

NK natural killer cell

ORF open reading frame

pp phosphoproteins

PE phycoerythrin

PTLD post transplant lymphoproliferative disease

R+/- recipient CMV serostatus

SCT stem cell transplantation

SOT solid organ transplantation

TetraCD8 tetramer+CD8 T cell

UL unique long domain

VACV valacyclovir

VGCV valganciclovir

VZV varicella-zoster virus

11

INTRODUCTION

Cytomegalovirus (CMV) is a member of the human herpesvirus family that

consists of 8 viruses: herpes simplex virus 1 (HSV-1), herpes simplex virus 2

(HSV-2), varicella-zoster virus (VZV), Epstein-Barr virus (EBV), CMV,

human herpes virus 6 A and B (HHV-6A and B), human herpes virus 7

(HHV-7), and human herpes virus 8 (HHV-8). Herpesviruses can be further

divided into subfamilies: α−herpesviruses (HSV-1, HSV-2 and VZV), β-

herpesviruses (CMV, HHV-6A/B and HHV-7) and γ-herpesviruses (EBV

and HHV-8). Herpesviruses are enveloped viruses with an icosahedral cap-

sid that encloses a double-stranded DNA genome (Fig.1). CMV is the largest

member of the human herpesvirus family, with a genome of 236 kbp and

more than 200 open reading frames (ORFs) [1] encoding more than 80 viral

proteins, including glycoproteins (e.g., gB), phosphoproteins (e.g., pp65),

and other transcription/replication proteins.

Figure 1. A schematic picture of the CMV structure (Reproduced with the permission of Dr. Marko Reschke, http://www.biografix.de/)

12

Genome analysis has indicated that mammalian CMV have co-speciated

with their respective hosts over the last 80 million years [2]. This prolonged

period of co-evolution has resulted in a high level of co-adaptation between

the virus and its host and, like other herpesviruses, CMV establishes latency

after primary infection. CMV has been shown to persist in monocytes, bone

marrow progenitor populations and endothelial cells [3-9]. During latency,

CMV cannot be eliminated by host defenses but the immune system keeps

the virus under close surveillance, giving it little chance to reactivate and

cause symptomatic disease.

The latency and reactivation phases have developed through evolution

and are maintained by a number of genes that allow the manipulation of

different cellular processes involved in cell-cycle control. These processes

ensure for example that cellular precursors could be used for viral replica-

tion. If the host’s immune system is impaired for any reason, CMV can reac-

tivate and cause transient episodes of viremia that are generally asympto-

matic in immunocompetent individuals [10]. In immunocompromised sub-

jects, such as transplant recipients and HIV/AIDS patients, however, the

consequences of a reactivation may be severe and even fatal.

Epidemiology

CMV is one of the most successful of human pathogens, since it can be

transmitted both vertically and horizontally, usually with little effect on the

host. CMV is shed in body fluids (e.g., saliva, urine, semen, breast milk).

Transmission usually occurs from person to person, including through the

intrauterine route, but it can also be spread during transfusion or organ trans-

plantation. Globally, between 60 and 90% of the general population is in-

fected with CMV [11], with generally higher rates in developing countries.

The high seroprevalence rate in Sweden (70-80%) is believed to be related to

the high rate of breastfeeding and the widespread use of daycare facilities for

children [12]. CMV is usually acquired during early childhood, with 40% of

all individuals being seropositive at 4 years of age [13]; the prevalence of

infection increases with age after infancy and ongoing seroconversion is

seen through out life [14].

13

CMV replication

When CMV enters the human body, it begins to rapidly and effectively

penetrate virtually all types of cells, including monocytes/macrophages, neu-

trophils, endothelial cells, epithelial cells, smooth muscle cells, fibroblasts,

stromal cells, neurons and hepatocytes [9]. This penetration is accomplished

by a pH-independent fusion in the case of most cells, such as fibroblasts, but

in epithelial and endothelial cells it occurs through endocytosis and fusion at

low-pH [15]. It is important, however, to note that only certain cells, such as

fibroblasts and fully differentiated macrophages, allow a fully permissive

replication cycle that results in the production of infectious virus. Other cells

such as monocytes are non-permissive, meaning that the CMV infection is

restricted to early events of gene expression and does not result in the pro-

duction of complete virus. Reactivation of latent CMV in vivo is also re-

stricted to certain cell types and has thus far been shown only in fully differ-

entiated monocytes/macrophages [16]. After the virus penetrates the cell, it

replicates and matures in the nucleus, and, as for other herpes-viruses, this

process occurs in three relatively distinct phases:

Immediate-early (IE) phase

This phase begins with the transcription of the IE (alpha) genes during the

first 4 hours after viral entry. The transcription event is dependent only on

cellular factors and does not require de novo viral protein synthesis. Non-

structural proteins appear in the nucleus within 4 hours after infection. These

proteins are essential for the regulation of the expression of the early- and

late-phase genes, and also for manipulating various cellular processes.

Early (E) phase

The transcription of the E (beta) genes is dependent on the expression of

functional IE gene products. In this phase a variety of essential proteins in-

volved in viral replication are produced, including DNA polymerase and

helicase-primase.

Late (L) phase

The transcription of L (gamma) genes occurs approximately 24 hours after

infection and includes structural and maturation proteins. This synthesis is

highly dependent on viral DNA replication and can be blocked by inhibitors

of viral DNA polymerase such as ganciclovir. The viral products are then

assembled and packaged in the nucleus whereas the final maturation and

budding of the virus take place in a Golgi-derived vacuole [17] from which

the virus is released. CMV replication in vivo is a very dynamic process with

14

doubling times of 12-24 hours [18] and a maximum virus release at 72 to 96

hours post-infection. The replication continues for several days until cell

lysis occurs.

Immune response to CMV infection

CMV infection triggers a forceful immune reaction in the human body, in-

cluding both antibody- and T-cell-mediated responses. Because of its effec-

tive immune evasion strategies, CMV is able to establish latency despite the

strong immune response by the host. The primary infection usually resolves

without complications, leaving the immune system to maintain an effective

balance and avoid reactivation of the virus.

B-cell immune responses

Primary CMV infection elicits a transient IgM response within 1-3 weeks

that is followed by the development of persistent IgG antibodies. These anti-

bodies play a minor role in clearing the infection but are believed to play an

important role in reducing the severity of CMV disease in adults and protect-

ing the fetus from congenital infection. The antibodies are directed against at

least 15 different proteins; the most immunogenic is pp150, against which

nearly 100% of the CMV-seropositive individuals have antibodies [19]. An-

other important immunogenic protein is pp65; the antibody response against

this protein is very high during the acute phase of the infection. The best-

characterized protein, however, is glycoprotein B (gB), and up to 50-70% of

the host’s neutralizing antibody response is accounted for by the response to

his protein [19].

T cell immune responses

Cellular immunity against CMV is multifaceted, and both CD8 (T kil-

ler/CTL) and CD4 (T helper) T cells, as well as natural killer (NK) cells and

γδT cells are involved. A substantial proportion of the T-cell immune sys-

tem, in both the CD8 and CD4 T-cell populations, is dedicated to controlling

CMV replication, and it has been demonstrated that the majority of the CMV

proteome is targeted by the T-cell arm of the immune system [20].

CD8 T-cell responses After primary infection, the virus-infected cells display viral peptides to-

gether with MHC I, and these combinations are recognized by the T-cell

receptors of the CD8+ CMV-specific T cells. Together with cytokines pro-

duced by CMV-specific CD4 T cells, the CD8 T cells become activated and

start to produce cytokines (e.g., IFNγ and TNFα) and effector proteins (e.g.,

15

granzyme and perforin) to destroy the infected cell. The main targets of the

CD8 T cells are pp65, IE1, and gB, and it is possible that the responses to

pp65 and IE1 vary from one phase of the infection to another. The pp65

response may be of importance during the initial phase in newly infected

cells before IE1 antigen is expressed, whereas IE1 responses may be ex-

panded during reactivation from latency, since the IE1 response appears

before the pp65 response in the event of a reactivation [21]. During the rep-

licative phase of the infection, CMV-specific CD8 T cells display a uniform

phenotype with high proliferation and expression of CD45RO and CD27.

When the infection resolves into the latent phase, effector populations con-

tract by apoptosis and CMV-specific T cells differentiate into resting primed

memory cells (e.g., CD45RA+CD27-), but there is a large variation in the

different subsets in the CD8+ population [22]. This central role for CD8 T

cells in protection from disease has been substantiated by the inverse correla-

tion observed between CMV disease and CD8 T-cell reconstitution after

stem cell transplantation [23].

CD4 T-cell responses In recent years, the role and importance of CD4 T cells in controlling the

CMV infection has been recognized. Their involvement was first demon-

strated in bone marrow transplant patients who received adoptive transfer of

CMV-specific CD8 T cells, with patients deficient in virus-specific CD4 T

cells displaying a lower level of CD8 anti-CMV T-cell activity [24]. This

situation has also been observed in other populations with reduced capacity

to mount a CMV-specific CD4 T-cell response. Deficiencies in CD4 T cells

in children appear to correlate with prolonged viral shedding [25], whereas

CD8 T-cell deficiencies do not [26]. In HIV patients, significant reductions

in CMV-specific CD4 T cells are associated with increased CMV replication

despite the presence of CD8 T cells, as well as a diminished ability of the

CD8 T cells to secrete IFNγ after peptide stimulation [27]. A similar pattern

has been demonstrated in the case of organ transplantation and will be dis-

cussed in more detail below.

NK cell responses

NK cells are members of the innate immune system and recognize pathogen-

infected cells and tumors. These cells are vital in the event of a viral infec-

tion, and if NK cells are depleted, host resistance against CMV infection is

markedly reduced [28].

16

Immune evasion

The capacity of CMV to establish latency and to co-exist with the host may

be the result of multiple immune evasion strategies, and increasing evidence

exists that both the innate and adaptive immune responses are affected.

CMV has been shown to downregulate both MHC class I and II expression

in infected cells and to interfere with MHC class II expression in several

ways [29-31]. The balance between host defenses and virus is maintained

through an intricate series of pathways; for example, although downregula-

tion of MHC class I antigens would make the cells more vulnerable to NK

cell-mediated lysis, other viral products (UL 16, UL 18) are responsible for

helping the virus resist NK cells [32]. One potential clinical relevance of

CMV’s immune evasion has been proposed by Humar et al.: CMV immune

evasion gene expression levels in patients with CMV infection decline expo-

nentially with antiviral therapy, and can be independently correlated with

clinical outcome [33].

Immunological methods

The functional status of T cells in organ transplant recipients is of great im-

portance, and the development of techniques such as ELISPOT, cytokine

flow cytometry (CFC), and major histocompatibility complex (MHC) tetra-

mers has facilitated the exploration of CMV-specific immunity [34].

• ELISPOT is based on the detection of a cytokine produced by single cells

after stimulation with antigens [35]. The secreted cytokine is then de-

tected by specific monoclonal antibodies that generate spots reflecting the

number of cytokine-secreting cells. This method therefore measures both

function and frequency, but one potential disadvantage is subjectivity.

• CFC is based on the detection of secreted cytokines from lymphocytes.

One approach to detection is to create an artificial affinity matrix with an-

tibodies that cover the cell surface and captures the secreted cytokines

(upon stimulation), making it possible to label and measure the number of

activated cells [36]. An alternative approach involves the intracellular

staining of cytokines and requires the use of inhibitors of intracellular

transport such as brefeldin [37]. This method allows for categorization of

T-cell subsets, such as Th1 or Th2. The advantages of this method are that

there is no need for HLA-specific epitopes, and the total capacity to re-

spond to CMV antigen is measured. The drawbacks are that the method is

time-consuming, and the samples need urgent attention.

17

• MHC tetramers consist of four MHC peptide complexes linked together

by biotin and avidin with an attached fluorochrome [38]. The tetrameriza-

tion increases the avidity of the peptides and produces a consistent bind-

ing to the T cells that facilitates the identification of the CMV-specific T

cells by flow cytometry. The advantages of this method are its very high

sensitivity and specificity and the fact that it requires very little blood;

however, the obvious disadvantage is that HLA-specific epitopes are re-

quired.

CMV diagnosis

Historically, diagnosis of CMV disease has been made by histopathology,

whih large ”protozoan-like” cells (cytomegalia [39]), with intranuclear in-

clusions being identified in biopsy material (Fig.2). This method is still used

in conjunction with immunochemistry for the diagnosis of tissue-invasive

CMV infection, but it has been superseded by serological and virological

assays for determining the level of CMV replication and serostatus prior to

transplantation.

Figure 2. Typical owl-eye inclusions in CMV disease

(Reproduced with the permission of Dr. Dan Wiedbrauk)

Serological assays

Serological assays involving the detection of antibodies (IgM and IgG)

against CMV are highly specific and sensitive in immunocompetent indi-

viduals; the presence of anti-CMV antibodies indicates previous infection

but does not indicate the level of immunity. The most commonly available

assay is ELISA [40], but other methods such as latex agglutination, radio-

immunoassay, complement fixation, and immunofluorescence tests are also

available. One area in which serological analysis is of particular interest is in

pregnant women, for whom the introduction of IgG avidity assays has im-

proved the accuracy of diagnosis [41]. On the other hand, serological assays

18

cannot be recommended for monitoring CMV infection in immunocompro-

mised individuals, since the results can be unreliable because of the reduced

level of antibody production in these individuals. In transplant populations,

serological assays can be used before transplantation in both the donor and

recipient to identify subgroups with high or low risk of contracting CMV

disease.

Virological assays

Virological assays have been developed for use in immunocompromised

patients, in whom both rapid diagnosis and the ability to monitor CMV in-

fection and the effects of antiviral therapy are vital for optimal patient man-

agement.

Isolation of CMV and immunohistochemistry Viral culture has been considered the ”gold standard” for many years. The

endpoint is the characteristic cytopathic effect produced by productive repli-

cation of CMV in cell cultures such as human fibroblasts and routinely takes

2-4 weeks to reach. In the 1980s, a rapid cultured-based assay was devel-

oped that combined the inoculation of fibroblasts with the clinical sample

followed by a 16- to 24-hour incubation followed by immunohistochemical

detection of CMV proteins [42]. The disadvantage of this assay is that it is

not possible to quantify the viral load; thus, the capacity to monitor the ef-

fects of treatment is greatly decreased [43]. The use of culture-based assays

is still of importance, however, in epidemiological studies and in resistance

testing.

Antigenemia assays Antigenemia assays are based on the use of monoclonal antibodies against a

viral protein, usually pp65, that is present in peripheral blood leukocytes

during active CMV infection [44]. The antigenemia assay is highly specific

and sensitive in diagnosing CMV infection, and has superior prognostic

value when compared to cell culture-based methods. Other advantages are its

reliability and rapidity, together with the fact that the viral load can be esti-

mated from the number of pp65-positive leukocytes. It does, however, have

some disadvantages when compared to PCR (see below): The major disad-

vantage is the need for a sufficient number of leukocytes (the absolute neu-

trophil count must be >0.2x109 cells/ml) [45], a requirement that limits the

use of antigenaemia assays in neutropenic populations. Other limitations are

the need for rapid analysis within 6-8 hours of veni-puncture [46], and its

difficulties to process large numbers of samples.

19

DNAemia assays Since the early 1990s, DNAemia assays have made use of PCR or other hy-

bridization techniques for both qualitative and quantitative detection of viral

DNA. CMV DNA assays can be used on all body fluids, including blood

components, and since CMV is a strictly cell-associated virus, the most sen-

sitive blood compartment for detecting DNA is whole blood rather than

plasma or leukocytes [47-50]. CMV DNA testing of whole blood reflects

both the content of CMV in leukocytes and the cell-free virus; this testing of

whole blood generally gives higher values than plasma does [43, 49, 51].

Quantitative DNAemia has been shown to correlate with virus replication

and clinical symptoms [52, 53] and is used in the clinical setting to identify-

ing patients at risk of developing CMV disease and to monitor the response

to antiviral therapy. CMV DNA assays can also be performed on dried blood

spots on Guthrie cards as a means of diagnosing congenital infection retro-

spectively [54].

RNAemia assays RNAemia assays are used to detect a viral product, the pp67 mRNA tran-

script, which is more directly related to virus replication than is DNA. Nu-

cleic acid sequence-based amplification (NASBA) is used to amplify the

viral mRNA. The low sensitivity of this method, however, limits its useful-

ness in guiding of pre-emptive therapy [55].

Clinical symptoms of CMV infection – direct effects

Immunocompetent individuals

In immunocompetent persons of all ages, primary CMV infection is usually

subclinical or associated with mild clinical symptoms such as fever, head-

ache, and fatigue. Some healthy individuals, most often adolescents, may

experience a mononucleosis-like syndrome, often called CMV syndrome,

with prolonged fever, myalgia, splenomegaly, and a mild hepatitis with ele-

vated liver transaminases that may continue for several weeks [10]. The in-

fection is usually benign, without sequelae, and needs no treatment; how-

ever, on rare occasions, CMV can cause meningoencephalitis, Guillan-Barré

syndrome, pneumonitis, or myocarditis. Reactivation of CMV in immuno-

competent individuals is rather uncommon, and in two studies of healthy

blood donors, the prevalence of DNA in the blood and plasma has been

found to be <0.5% [56, 57].

20

Neonates

CMV is the most common congenital infection and is an important cause of

severe congenital disease, including premature birth and death. The propor-

tion of congenitally infected individuals is between 0.1 and 2.4% [58]. Con-

genital CMV infection can follow both primary and recurrent maternal infec-

tion but severe consequences of the disease are mainly related to primary

infection [58]. Primary infection of a seronegative mother is associated with

a 30-40% risk of intrauterine transmission, as compared to approximately

1% for seropositive mothers with a reactivated CMV infection. If the virus is

transmitted, approximately 10% of the babies display symptoms of CMV

disease at birth [58, 59]. These symptoms include microcephaly, hepa-

tosplenomegaly, petechiae, hearing loss, and intracranial calcifications, and

they are associated with a poor prognosis [60] and risk of long-term seque-

lae.

Immunocompromised individuals

In patients with a reduced T-cell capacity, CMV infection and disease are

likely to cause more severe clinical consequences than in immunocompetent

individuals. This population, which include solid organ transplant (SOT)

recipients, stem cell transplant (SCT) recipients, HIV-infected patients, and

other patients with heavy immunosuppression such as rheumatologic pa-

tients, is at high risk of developing invasive disease, with high morbidity and

mortality. The risk of symptomatic disease in these individuals is variable

and depends on many factors, including the immunosuppressive protocol

used, the type of transplant, various host factors (e.g., age, neutropenia, co-

morbidity), and the serostatus of both the donor and recipient. Among the

different SOT types, heart-lung transplants are associated with the highest

risk of symptomatic disease (39-41%), followed by pancreas, liver, and kid-

ney transplants (8-32%) [61].

In renal transplant populations, the CMV serostatus of both the recipient

(R) and donor (D) has a major impact on the risk of developing CMV dis-

ease. The risk ranges from 5 to 10% in the D-/R- population, provided that

these patients are given CMV-negative or leukocyte-depleted blood products

[62]. The risk is intermediate (15-35%) in the D+/R+ and D-/R+ groups, and

highest in the high-risk D+/R- population who lack both cellular and hu-

moral immunity: Greater than 50% of these individuals are likely to develop

CMV disease in the absence of prophylaxis and/or pre-emptive therapy [63].

Another high-risk population is those patients who receive depleting anti-

bodies such as ATG and OKT-3 for the treatment of organ rejection. With

such treatment protocols, CMV disease can develop in up to 65% of the

treated individuals, as compared to 15-25% when these antibodies are used

for induction therapy [64]. CMV disease in immunocompromised patients

21

can range from the previously described CMV syndrome (together with

CMV DNAemia) to end-organ diseases such as pneumonia, colitis, hepatitis,

encephalitis, myelitis, and retinitis.

Clinical symptoms of CMV infection – indirect effects

In addition to the direct effects of invasive CMV infection, this virus is also

associated with a number of indirect effects that result in part from the influ-

ence on the host immune response. These indirect effects have been sug-

gested to be a result of persistent low-level CMV infection [65] and have

been associated with transplant-associated vasculopathy and atherosclerosis,

posttransplantation diabetes, an increased risk of opportunistic infections

(mainly bacterial and fungal), development of cancer, and decreased graft

and patient survival [52, 53, 65-69]. An association between continuous dys-

regulation of the immune system and increasing age has also been proposed.

This could be due to the persistence of CMV as a chronic antigenic stressor

and would mainly affect the shrinking T-cell antigen repertoire and poten-

tially contribute to the increased incidence of infections in the elderly [70,

71], and perhaps even depression and anxiety [72].

Prevention and treatment of CMV disease

Strategies for the prevention and treatment of CMV infection and disease in

the transplant community include:

• Prophylaxis

• Pre-emptive therapy

• Deferred therapy; treatment for established CMV disease

Prior to the availability of therapy with foscarnet and ganciclovir (developed

in the 1980s), CMV disease was often fatal, with up to 80% succumbing to

SCT-associated CMV pneumonitis [73]. During the past few decades, how-

ever, the options for treating CMV disease have expanded, with the various

antivirals, immunoglobulins and combinations thereof making CMV-

associated mortality rare. CMV morbidity, on the other hand, is still a source

of great concern, and the question of prevention versus pre-emptive or de-

ferred therapy is a matter of debate. The approaches vary widely among

different transplant programs as a result of the lack of large multicenter ran-

domized trials and the difficulties involved in comparing the existing studies.

22

Active CMV infection(viremia,invasion)

Latent CMV infection

Indirect effects

Cellular effects

Allograftinjury/rejection

Acute Chronic

Inflammation(cytokines - TNF�, growth factors,intracellular messengers)

Graft rejection

Anti-lymphocyteantibodies (ALA)

Infection

Direct effects

End organ disease CMV syndrome

Vasculopathy Immunsuppression

Opportunistic infection(Bacteria, Fungi)

Atherosclerosis Neoplasia(Cancer, PTLD)

Figure 3. A schematic overview of CMV reactivation and the direct/indirect effects (Adapted from Fishman JA and Rubin RH, N Engl J Med. 1998 Jun 11;338(24):1741-51)

Prophylaxis Antiviral prophylaxis is used in patients at risk of CMV infection/disease

before any sign of an active CMV infection can be detected. Prophylaxis is

administered over a defined period, usually correlated with the time during

which the patient is most intensly immunosuppressed (commonly 100 days

following transplantation), but longer periods (up to 6-12 months) are being

investigated in certain high-risk groups such as D+/R- recipients [74]. The

advantages of prophylaxis are a reduced morbidity and mortality related to

the direct effects of the CMV infection, as well as a reduction in the indirect

effects of CMV infection [75-79]. It has also been shown in renal transplant

populations that a prophylactic strategi using ganciclovir is less costly than

pre-emptive therapy [80, 81].

23

The disadvantages of prophylaxis may include a prolonged antiviral drug

exposure that can be associated with toxicity and the emergence of resis-

tance, the development of late-onset CMV disease (CMV disease occuring

after cessation of the prophylaxis), and higher drug costs. The over-treatment

associated with a prophylactic strategy can be substantial and in SCT, up to

65% of the patients have been reported to receive unnecessary antiviral ther-

apy [82]. It is therefore important to consider the necessity for prophylaxis in

low-risk populations. The risks and symptoms of late-onset disease are also a

matter of discussion, and there is still a need for a strategy to manage this

aspect of CMV disease.

The drugs most commonly used for antiviral prophylaxis are: • Acyclovir (ACV) – a cyclic nucleoside homologue of guanosine that is

phosphorylated by viral thymidine kinase (UL97) and cellular kinases to a

triphosphate form that inhibits the viral DNA polymerase and act as a

chain terminator of CMV DNA synthesis. ACV was first discovered in

1974 and became available for clinical use in 1982. Despite its rather poor

in vitro activity, it has been found effective compared to placebo for pro-

phylaxis in SCT patients [83], renal transplant patients [84], and AIDS

patients [85].

• Valacyclovir (VACV) – a valine ester of ACV with increased bioavail-

ability that has been in use since the early 1990s. VACV has been shown

to reduce both CMV disease and acute rejection in renal transplant recipi-

ents when administered in a high (8g daily) or low (3g daily) dose regi-

men [86, 87]. It is also more effective than ACV as a prophylactic agent

in SCT patients [88]. However, neurotoxic adverse effects such as confu-

sion and hallucinations have been seen with high-dose VACV[86, 89].

• Ganciclovir (GCV) – a cyclic nucleoside homologue of guanosine that

also is phosphorylated by viral thymidine kinase (UL97) and cellular ki-

nases to a triphosphate form that inhibits viral DNA polymerase and act

as a chain terminator of CMV DNA synthesis. Used since the 1980s, its

effect on CMV is superior compared to that of ACV, and despite the poor

bioavailability (3-5%) of the oral form of GCV, studies have confirmed

its benefits when used as a prophylactic agent in liver transplant [90] and

high-risk kidney transplant recipients [91]. In populations with the highest

risk of CMV disease, such as those receiving lung or heart-lung transplan-

tats, intravenous (iv) GCV is used initially. The most common side effects

are renal toxicity and neutropenia, which often cause problem in SCT pa-

tients.

24

• Valganciclovir (VGCV) – a valine ester of GCV with improved bio-

availability that has been used since 2000. The efficacy of VGCV as a

prophylactic agent in various transplant populations is comparable to that

of GCV [92]. VGCV prophylaxis is as economical as sequential GCV

prophylaxis in high-risk D+/R� SOT patients [93]. The spectrum of side

effects is the same as GCV.

• CMV immunoglobulin (CMVIG) – a pool of anti-CMV IgG antibodies

that has been used since two trials demonstrated a decrease in ”severe”

CMV-associated disease in kidney and liver transplant recipients [94, 95].

The trials did not, however, demonstrate any reduction in the overall risk

of CMV disease, and CMVIG is mostly used in combination with GCV in

high-risk heart or lung transplant recipients in some centers.

Pre-emptive therapy

The alternative approach to minimizing CMV disease is pre-emptive ther-

apy, which is initiated when a laboratory marker (e.g., antigenemia, DNAe-

mia) indicative of a high-level of replication and/or disease becomes posi-

tive. The aim is to reduce the viral burden and to prevent progression from

asymptomatic infection to CMV disease. A successful strategy is built on

several factors, including:

• Selection of the appropriate patient population

• The use of rapid, sensitive and reliable diagnostic methods such as PCR

or pp65Ag assays

• Regular monitoring, with excellent patient and doctor compliance

• Initiation of treatment quickly after detection of viral replication

• Choosing the appropriate type, dose, and duration of an antiviral agent

If these factors cannot be achieved, it is likely that the prophylactic strategy

might be the more suitable option. The advantages of pre-emptive therapy

include a reduced drug exposure and decreased risk of toxicity and resis-

tance. It is also thought that pre-emptive therapy enables limited replication

of CMV to occur, thus facilitating immune priming, which may be of impor-

tance for future control of CMV replication [96]. The disadvantages of this

strategy include the logistic demands of laboratory testing and uncertainty

about the impact on the indirect effects of CMV disease. It has also been

reported that up to 33% of patients receiving pre-emptive treatment later

develop CMV DNAemia that requires treatment.

25

The drugs most commonly used for antiviral pre-emptive therapy are:

• Ganciclovir • Valganciclovir • Foscarnet – an inorganic pyrophosphate analogue that directly inhibits

the CMV DNA polymerase. Foscarnet has been used since the 1980s,

most often in the event of GCV resistance or in cases of SCT-associated

neutropenia. It can only be administered iv, and the most common side ef-

fects are renal toxicity and hypocalcemia.

To summarize the preventive strategies: There is currently more or less

agreement that both prophylactic and pre-emptive strategies reduce the im-

pact of CMV, and it is likely that a mixed approach is needed to optimize the

well-being of patients. Evidence-based guidelines from the International

Herpes Management Forum (IHMF) [43] and the Canadian Society of

Transplantation (CST) [97] recommend universal prophylaxis for patients at

the highest risk for CMV disease whereas preemptive therapy may be most

appropriate for those at a moderate or low risk of CMV. The choice of anti-

viral drug regimens used for universal prophylaxis depends on the type of

organ transplanted and the CMV serostatus of the donor and recipient. The

optimal pre-emptive drug regimen and laboratory monitoring strategy are

still unknown.

Treatment

Despite the use of preventive strategies, some patients ultimately develop

CMV disease, and the treatment options available for these patients are simi-

lar to those used in the pre-emptive strategies. In recent years there has been

a gradual shift from iv GCV toward the use of VGCV [98], and in the case

of neutropenia and/or GCV resistance, foscarnet is used/added. Treatment

with GCV in neonates has been proven to be effective in reducing hearing

loss [99]. Additional drugs used in treating established CMV disease are:

• Cidofovir – a nucleotide monophosphate that is triphosphorylated by

cellular kinases to triphosphate that directly inhibits CMV DNA poly-

merase. It has a long half-life that allows once-weekly therapy. The most

common side effect is a significant renal toxicity.

• Maribavir – a novel benzimidazole drug in ongoing phase 3 trials that

directly inhibits the UL97 kinase and CMV DNA synthesis. It has also

been shown to be effective against GCV-resistant CMV strains. It is not

associated with nephrotoxicity or hematologic toxicities.

26

• Leflunomide – A pyrimidine synthesis inhibitor that inferes with virion

assembly. It has a good bioavailability (80%) and is already approved for

autoimmune conditions because of its immunosuppressive abilities. It has

been used in SOT recipients with good effect, whereas the results in SCT

recipients have been less positive [100, 101].

Adoptive immunotherapy

Since the early 1990s [24], infusing CMV-specific CD8 and CD4 T cells

into patients to stop viral replication has been a subject of research. Together

with the progress associated with new immunological methods such as MHC

class I tetramers, the introduction of this approach has enabled several

groups to use this method as a treatment option in SCT patients who are

unresponsive to GCV [102, 103]. This method is not yet in routine use, but it

could be used as a final option in patients in whom all other treatment op-

tions have been explored.

CMV resistance

The emergence of resistance to anti-CMV drugs has become an increasing

problem in recent decades. Already in the late 1980s, shortly after the intro-

duction of GCV, sporadic cases of GCV-resistant CMV were reported in

severely immunocompromised patients [104], and during the AIDS epi-

demic, the GCV resistance increased. In recent years, the widespread use of

more intensive immunosuppressive strategies and prophylaxis with GCV in

the SOT population has resulted in an increasing level of GCV resistance.

Risk factors for the development of GCV resistance are: D+/R- serological

status, prolonged exposure to GCV, potent immunosuppression, suboptimal

GCV levels, and high viral load [105-107]. The highest incidence of GCV-

resistance (20-30%) is seen in the D+/R- lung transplant and kidney-

pancreas transplant populations, whereas the incidence in D+/R- kidney

transplant patients ranges from 2 to 5%. In the R+ transplant population,

however, the incidence is virtually zero, except for lung transplant patients,

resulting in an overall incidence of GCV resistance of 0-13% [108].

The main mechanisms of GCV resistance involves mutations in UL97

(viral phosphotransferase), UL54 (DNA polymerase), or both [109-111].

Mutations in UL97 ultimately result in decreased levels of active GCV,

whereas mutations in UL54 result in a virus that is less susceptible to GCV,

with cross-resistance to cidofovir and foscarnet. Mutations in UL97 are far

more common, and a majority of the mutations occur in one specific region.

It has been demonstrated that the mutation in UL97 can be detected at up to

1% in the original CMV population in patients, and this resistant virus is

27

selected by GCV treatment [18]. The frequency of GCV-resistant virus has

been found to increase from 8% after 3 months of GCV treatment to 64%

after 15 months [112, 113]. In patients who develop CMV disease because

of drug-resistant virus, the clinical outcome is generally poor, with a mortal-

ity rate of up to 20% [108, 114, 115]. The reason for this high rate may be

that the second-line therapy options are limited; thus, the need for additional

drugs with good anti-CMV effect is evident.

Vaccine development

Because of the high risk in neonates and the immunocompromised individu-

als of developing CMV infection and disease, intensive research has been

carried out during the past 30 years to develop a vaccine against CMV infec-

tion. Despite these efforts, no CMV vaccine has yet been licensed. The de-

velopment of a vaccine to prevent CMV infection or disease has therefore

been assigned the highest priority by medical authorities such as the U.S.

Institute of Medicine, but there are still many problems to be solved.

What viral proteins should be included?

In order to be effective, viral vaccines must induce antibodies that neutralize

viral entry into the host cells. Earlier studies of laboratory CMV strains have

focused on viral entry into fibroblasts and have revealed that the major neutraliz-

ing epitopes reside in glycoproteins gB, gH, and gM/gN [116-118]. It has also

been suggested that epitopes within gB can represent up to half the neutralizing

activity of human immune sera [116]. Recent findings, however, have indicated

that CMV enters different cell types via various distinct mechanisms. Fibroblast

entry occurs at the cell membrane and is pH-independent, while entry into endo-

thelial and epithelial cells requires a complex of gH and gL, along with three

additional viral proteins that are unnecessary for fibroblast entry, UL128,

UL130, and UL131 [119, 120]. This discovery suggests that previously unrec-

ognized neutralizing epitopes may exist. Recent findings also imply that in natu-

rally infected individuals, the titers of neutralizing antibodies to these epitopes

far exceed those seen in fibroblast-neutralizing responses [121].

At present, two vaccines are in phase II studies, one based on the gB complex

and one based on a live attenuated Towne strain. The gB vaccine has been

shown to induce high levels of monoclonal antibodies that neutralize both wild-

type virus and laboratory strains. It is safe and immunogenic and is probably the

most likely candidate for a subunit vaccine. While the gB vaccine would not be

expected to elicit antibodies to UL128–131, antibodies to certain epitopes within

gB could selectively inhibit epithelial or endothelial cell entry. The Towne vac-

cine has been used extensively in clinical trials; it is safe and stimulates neutral-

izing antibodies. This vaccine, however, contains a mutation in UL 130 but

28

could potentially elicit antibodies to epitopes within UL128 or UL131. These

antibodies can specifically neutralize epithelial or endothelial cell entry [120],

but the levels of antibody produced have been demonstrated to be much lower

than those found in natural sera [122]. These findings may indicate that the en-

docytic entry mediators UL128, UL130, and UL131 are strong potential candi-

dates for antigens containing epithelial neutralizing epitopes. It is therefore pos-

sible that the induction of epithelial-neutralizing antibodies comparable to those

induced by natural infection may be necessary for an effective CMV vaccine.

What population should a clinical vaccine trial focus on?

Another problem is how to design a vaccine trial: Should it be aimed at im-

munocompromised patients or at mothers who are at risk for transmitting

congenital CMV? If women who are not yet pregnant are the target popula-

tion, one problem would be the great number of subjects needed to demon-

strate efficacy. Given an estimated efficacy of the vaccine of between 50 and

80% with congenital infection as an endpoint, somewhere between 3,000-

10,000 mothers and their newborns would have to be enrolled [123].

A vaccine would have the greatest impact if it were given to infants be-

fore the age of two. In this population, much smaller number of subjects

would be needed (around 300). In children however, it has been demon-

strated that those who shed CMV for a prolonged period have greatly dimin-

ished CMV-specific CD4 responses [25]. Therefore, it may be necessary to

develop a vaccine that induces CMV-specific CD4-responses in order for it

to be effective.

Renal transplantation and allograft rejection

Since the onset of transplantation procedures in the 1950s, patient and graft

survival has improved, mostly because of the availability of better immuno-

suppressive drugs, but also as a result of better surgical techniques and infec-

tion control. Patient and graft survival after 1 year is now approaching 100%

in living donor transplants (98% patient and 95% graft survival) and >90%

in deceased donor renal transplants (95% patient and 89% graft survival).

Five-year survival, however, is less encouraging with 90% (patient) and 80%

(graft) survival in living donors and 80% (patient) and 67% (graft) survival

in deceased donor transplants [124]. Today, the major causes of late allograft

loss are death with a functioning graft and chronic renal allograft nephropa-

thy (CAN, see below), each of which accounts for about 50% of all allograft

loss. The major identified causes of mortality among renal transplant recipi-

ents are cardiovascular disease (22%), infection (16%), malignancy (7%),

cerebrovascular disease (6%), and graft failure (1%) (Fig.4).

29

Figure 4. Causes of death in deceased donor renal transplant recipients, 1993 to 2004, based on OPTN data as of September 2008 (www.optn.org).

Rejection

There are four major clinical types of rejection in human renal allografts:

• Hyperacute

• Humoral/antibody-mediated

• Acute cellular/vascular

• Chronic

Hyperacute rejection Hyperacute rejection occurs very rapidly upon reperfusion of the graft:

within minutes to 24 hours post-transplantation. It requires pre-formed anti-

bodies of either the HLA or ABO type against the donor tissue. These anti-

bodies are produced by plasma cells and have been induced by blood trans-

fusions, previous transplantation, or pregnancies. The antibodies attack the

endothelium and cause activation of both the coagulation and complement

cascades, which in turn leads to vascular coagulation and immediate necrosis

and graft failure. These rejections can usually be avoided by a thorough

HLA matching of the donor and recipient to avoid an HLA mismatch that

increases the risk.

30

Humoral/antibody-mediated rejection Humoral- or antibody-mediated rejection often occurs within 30 to 90 days

after transplantation. Clinical signs of rejection are a rapid increase in serum

creatinine, hypertension, and reduced urine output. Neutrophilic infiltration

and C4d deposition in the glomeruli can be seen on biopsy. The pathogenesis

is based on the post-transplantation development of anti-HLA antibodies.

This condition requires intensive treatment with monoclonal or polyclonal

antibodies, possible plasmapheresis and/or immune globulin, and a change in

the maintenance immunosuppressive therapy.

Acute rejection In recent years, the incidence of acute rejection has decreased dramatically,

from >50% in the pre-cyclosporine era to about 10% with the newer immu-

nosuppressive protocols. Acute rejection occurs mainly within the first 12

months after transplantation. Clinically, an increase in the serum creatinine

and blood pressure are typically noted. The acute rejection, further divided

into cellular and vascular rejection, is characterized by an infiltration of both

CD4+ and CD8+ T cells, together with histiocytes, macrophages, and granu-

locytes. In cellular rejection, the infiltration is seen in the renal tubules, and

in the event of a vascular rejection in the intima of small arteries. Acute re-

jection is still classified according to the BANFF criteria [125, 126], which

combines the topographical categorization of the immune injury and the

severity of the injury. These scores are then summarized into a scheme with

different severity grades. Primary treatment of the acute rejection is based on

intravenous methylprednisolone and/or a change in the maintenance immu-

nosuppressive therapy.

Chronic rejection Chronic rejection can occur months to years after transplantation, with clini-

cal signs of a gradual increase in serum creatinine, proteinuria and hyperten-

sion. The pathology of this condition is diverse and features interstitial fibro-

sis, tubular atrophy, and glomerulosclerosis. The pathogenesis involves fac-

tors such as HLA-antibody formation, inadequate immunosuppression, drug

toxicity, and previous rejections and infections. Another condition that re-

sembles chronic rejection is the multi-etiological, non-specific and chroni-

cally progressive parenchymal scarring previously [127] known as chronic

allograft nephropathy (CAN), the commonest cause of graft failure in the

first decade after transplantation. This allograft dysfunction may have an

unpredictable onset but occurs after the first 3 months post-transplantation,

with a slow deterioration of renal function. Biopsies display a nonspecific

histology with tubular atrophy and thickening of arteries. Depending on the

cause (which is often unknown), therapy may be beneficial, but in most

cases there is no cure.

31

Immunosuppressive treatment in renal transplantation

The central issue in organ transplantation remains suppression of allograft

rejection. Immunosuppressive drugs are used for induction (intense immuno-

suppression during the initial days after transplantation), maintenance ther-

apy, and reversal of established rejection. Until recently, the strategy was

quite simple: After induction with depleting antibodies, maintenance therapy

consisted of three drugs (a calcineurin inhibitor, azathioprin, and corticoster-

oids) that were continuously reduced. Rejection was treated with high-dose

steroid or depleting antibodies. Today, hundreds of combinations exist, a

situation that has improved both graft outcome and patient morbidity and

mortality, and the combinations are now being more and more individual-

ized, partly as a result of our increasing knowledge in the field of pharmaco-

genomics [128, 129].

Immunosuppression can be achieved by depleting lymphocytes, diverting

lymphocyte traffic, or blocking lymphocyte response pathways (Fig.5). As a

result of the immunosuppression, rejection is suppressed, but this situation

also leads to undesired consequences such as infection, cancer, post-

transplant lymphoproliferative disease (PTLD), and nonimmune toxicity.

Immunosuppressive drugs can be classified into the following categories:

• Small-molecule drugs (e.g., cyclosporine)

• Depleting protein drugs (e.g., ATG)

• Nondepleting protein drugs (e.g., basiliximab)

• Fusion proteins (e.g., belatacept)

• Glucocorticoids

• Intravenous immune globulin

Small-molecule drugs

Most small-molecule immunosuppressive agents are derived from microbial

products and target proteins that have been highly conserved in evolution.

These drugs probably do not saturate their targets at clinically tolerated con-

centrations, and without target saturation, the drug's effects are proportional

to the concentration of the drug, which makes dosing and monitoring critical.

Azathioprine was the first immunosuppressive drug widely used in organ

transplantation, but it is now considered a second-line drug because of the

introduction of cyclosporine. Its effect is mainly on DNA synthesis, with

serious side effects that include infections, neoplasias, and hematologic con-

dition (leukopenia and/or anemia).

32

Calcineurin inhibitors Calcineurin inhibitors can be considered the cornerstone of immunosuppres-

sion and can be used as a single drug, even though they are mainly used to-

gether with other immunosuppressive drugs.

Cyclosporine, the most widely used immunosuppressive drug, is a fungal

peptide composed of 11 aminoacids (aa). It is a prodrug that forms a com-

plex with cyclophilin, an intracellular protein that inhibits calcineurin and

reduces the transcription of IL-2, IL-4, and IFNγ. The side effects of cyc-

losporine are related to its concentration and, in addition to an elevated risk

of opportunistic and other infections, include nephrotoxicity, hypertension,

hyperlipidemia, gingival hyperplasia, hirsutism, and tremor. This drug is also

associated with the hemolytic-uremic syndrome and post-transplantation

diabetes mellitus. Drug concentrations must be monitored, but there is no

consensus on the optimal time for sampling [130].

Tacrolimus (FK506), a macrolide antibiotic licensed since 1994, uses an-

other intracellular protein, FKBP12, to create a complex similar to that in-

volving cyclosporine but with much greater potency. The effect on rejection

frequency is superior to that of cyclosporine [131]. The choice between cyc-

losporine and tacrolimus is based on the differences in their side effects:

Tacrolimus use poses the same risk of renal toxicity and the hemolytic-

uremic syndrome, but it is less likely to cause hypertension, hyperlipidemia,

gingivahyperplasia, or hirsutism. On the other hand, an elevated risk of de-

veloping post-transplantation diabetes [132] and BK-virus-related nephropa-

thy has been demonstrated with this drug [133]. The concentration needs to

be monitored, and trough levels are generally measured.

Purine synthesis inhibitors Mycophenolic acid (Myfortic) and the prodrug mycophenolate mofetil (MMF) block purine synthesis and prevent the proliferation of T and B cells. They

are widely used in combination with calcineurin inhibitors and have replaced

azathioprine because of their superior effect on graft rejection. Addition of

MMF in the event of rejection could improve graft outcome. Other advan-

tages are that here is no absolute need for concentration monitoring, and the

risk of cardiovascular events and nephrotoxicity is low. The drug has been

demonstrated to inhibit pneumocystic activity in vitro, but the risk of CMV

disease may be elevated [134].

Target-of-Rapamycin inhibitors Sirolimus (rapamycin) and everolimus are other macrolide antibiotics that

form a complex together with FKBP12, but their effect is inhibition of the

rapamycin target instead of calcineurin. This inhibition prevents cytokine

33

(IL-2) receptors from activating the cell cycle. These drugs are not nephro-

toxic by themselves, but they increase the nephrotoxicity of calcineurin in-

hibitors [135]. Other important side effects are hyperlipidemia, pneumonitis,

and thrombocytopenia, and recently an association between sirolimus and

new-onset diabetes has been demonstrated [136]. Sirolimus has also been

demonstrated to have both antineoplastic and arterial protective effects [137,

138], and the available evidence indicates that with respect to tumor growth,

rapamycin’s anti-cancer activity outweighs its immunosuppressant effects

[139]. Its arterial protective effects could be due to an impaired wound heal-

ing and/or an antiangiogenetic effect and have led to the development of

drug-eluting stents used in coronary arteries.

Figure 5. Immunosuppressive drugs and sites of action. CsA-cyclosporine. MPA-mycophenolic acid. See text for details. (Figure reprinted by permission from Macmillan Pub-

lishers Ltd: Nature Reviews Immunology 3, 831-838, copyright 2003)

34

Depleting antibodies

Depleting protein immunosuppressive agents are antibodies targeted against

T cells, B cells, or both. When T-cell depletion treatment is initiated, large

amounts of cytokines are released, causing severe systemic symptoms. De-

pleting antibodies are used to reduce early rejection but also increase the risk

of infection and PTLD. When the immune system recovers, after months to

years, late rejection may occur. The depletion of antibody-producing cells is

better tolerated than is T-cell depletion because of the absence of cytokine

release and the maintainance of immunoglobulin levels.

Antithymocyte globulin (ATG) is produced by immunizing horses or rab-

bits (more potent) with human lymphoid cells; IgG without toxic antibodies

is then harvested. ATG blocks T-cell membrane proteins and causes a pro-

found and durable T-cell functional depletion as a result of both altered func-

tion and lysis. Polyclonal ATG can be used as an induction agent for 3 to 10

days or in the event of steroidresistant rejection. Unwanted side effects are,

most notably, the cytokine release syndrome (fever, chills, and hypotension),

neutropenia, thrombocytopenia, and serum sickness.

Muromonab-CD3 (OKT3) is a mouse monoclonal antibody against CD3

that is used to treat rejection and for induction. It binds to T-cell receptor-

associated CD3-complexes and triggers a massive cytokine-release syn-

drome before depleting T cells and altering T-cell function. Because of the

severe immunosuppression and elevated risk for CMV disease that are asso-

ciated with this drug, prophylaxis against CMV is given when muromonab-

CD3 is used as rejection treatment. Unfortunately, humans can make neutral-

izing antibodies to the drug, limiting its reuse; also, prolonged use increases

the risk of PTLD. Its other main adverse effects are gastrointestinal and neu-

rotoxic. In general, the use of OKT-3 has decreased since the incidence of

rejection has declined.

Alemtuzumab is a humanized monoclonal antibody against CD52 that

massively depletes lymphocyte populations. It was originally approved for

treating refractory B-cell chronic lymphocytic leukemia but has recently

been used as an induction agent in renal transplantation. The risks of immu-

nodeficiency complications with alemtuzumab, such as infections and can-

cer, are being investigated, but some studies have demonstrated a similar or

reduced incidence of infectious complications [140, 141]. Toxicity is gener-

ally related to neutropenia, anemia, and a mild cytokine-release syndrome.

35

Rituximab is a monoclonal antibody that targets anti-CD20, eliminating B

cells for 6-9 months; it is considered a very safe drug. It was originally ap-

proved for treating refractory non-Hodgkin's B-cell lymphomas and PTLD in

organ-transplant recipients, but has recently been used in both autoimmune

disorders and transplant settings. It is used to diminish the levels of alloreac-

tive antibodies in highly sensitized patients [142], to manage ABO-

incompatible transplants [143], and to treat rejection associated with B cells

and antibodies [144, 145]. The use of rituximab in the transplant setting

needs further studies in randomized controlled trials before it can be recom-

mended as a standard treatment.

Non-depleting antibodies

Non-depleting protein drugs are monoclonal antibodies or fusion proteins

(still in clinical trials) that target proteins expressed only in activated im-

mune cells; this strategy effectively reduces the immune response without

depleting entire lymphocyte populations. The somewhat limited efficacy of

non-depleting antibodies is balanced by an absence of immunodeficiency

complications and low levels of non-immune toxicity and cytokine release.

Basiliximab and daclizumab are anti-CD25 monoclonal antibodies that

block the IL-2 receptor α-chain. They are widely used for induction, with a

low-to-moderate risk of rejection, but they require T-cell activation and

therefore only deplete activated T-cells. They are considered moderately

effective, since they only reduce rejection by about one-third in combination

with calcineurin inhibitors. On the other hand, they have minimal toxic ef-

fects and are associated with a reduced risk of CMV infection [146].

Fusion proteins

Belatacept (LEA29Y) is a second-generation cytotoxic T lymphocyte–

associated antigen 4 (CTLA-4) immune globulin. It is a fusion protein that

combines CTLA-4 (which engages CD80 and CD86) with the Fc portion of

IgG, thereby inhibiting a T-cell costimulatory pathway. Belatacept, now in

phase III trials, is used as a long-term iv-administered drug to reduce reli-

ance on toxic small-molecule immunosuppressive drugs and has been dem-

onstrated to have a superior effect on graft function when compared to

treatment with cyclosporine.

36

Glucocorticoids

Glucocorticoids have played a critical role in the evolution of organ trans-

plantation. Their mechanism of action is yet not fully understood, but they

are known to affect T-cell activation both directly and indirectly and display

different effects in low- and high-dose regimens. Higher doses are used dur-

ing the first days after transplantation or in the case of rejection to reduce

non-specific inflammation; the lower doses used during maintenance therapy

act via receptor-mediated effects to target transcription factors. Long-term

glucocorticoid use is associated with numerous adverse effects, including an

increased risk of infection, new-onset diabetes, hyperlipidemia, hyperten-

sion, and accelerated bone loss. An early steroid withdrawal in renal trans-

plant recipients would reduce many of these steroid-related side effects, and

it has recently been demonstrated that this can be accomplished, at 2 to 6

days post-transplantation, without compromising graft function [147, 148].

Intravenous immune globulin

The mode of action of this therapy depends largely on antigen binding and

modification of effector functions. Antigen binding is mediated via immune

antibodies and a wide spectrum of autoantibodies. The effector functions

include modulation of expression and function of Fc receptors, complement

activation, complement binding, anti-inflammatory effects due to interfer-

ence with the cytokine network, and modulation of T- and B-cell activation.

Organ transplantation is an important field for this treatment, primarily for

patients with high titers of anti-HLA antibodies.

37

AIMS

General aim

CMV infections are still a source of great concern, especially in transplant

populations. A better understanding of the immunologic response in primary

and reactivated CMV infection as well as better diagnostic methods and

optimal strategies for preventing and treating CMV disease in transplant

populations are still needed. In these papers, we wanted to evaluate a modi-

fied prevention strategy in high-risk renal transplant patients, and to charac-

terize the dynamics of the CMV-specific immune response in both renal

transplant patients and healthy controls.

Specific aims

Paper I

The aim of this retrospective study of CMV-mismatched (D+/R-) renal

transplant patients was to determine whether a comparatively low dose of

prophylactic valacyclovir (3g/day) given for 90 days post-transplantation

could prevent CMV disease with fewer CNS adverse effects than were re-

ported by participants in an earlier study in which 8g/day was used. We also

wanted to compare the incidence of rejection to that of historical controls.

Paper II

The aim of this study was to characterize the CMV-specific CD8+ immune

response in healthy individuals with primary or latent CMV infection. We

wanted to see whether, in latently infected individuals, the frequencies of

different CMV epitopes within the same donor were of equal magnitude or

whether any of the epitopes displayed immunodominance. The frequencies

of the CMV-specific T cells were then related to the function of the CMV-

specific T cells.

38

Paper III

The aim of this longitudinal, prospective study of CMV-seropositive renal

transplant recipients was to evaluate the dynamics of CMV-specific T cells,

in terms of both function and absolute number, as well as the viral load up to

1 year after transplantation. We also wanted to assess possible correlations

between CMV-specific T-cell activity and clinical symptoms of CMV infec-

tion.

Paper IV

The aim of this retrospective study of CMV-mismatched (D+/R-) renal

transplant patients given low-dose valacyclovir (3g/day) for 90 days post-

transplantation was to evaluate patient and graft survival up to 5 years post-

transplantation and to compare these survival rates to international data from

the CTS database. We also wanted to analyze the incidence of CMV disease,

CMV resistance, biopsy-proven acute graft rejection, and major neurotoxic

adverse effects up to 1 year after transplantation.

39

METHODS

Ethics

The ethics committees at Uppsala University Hospital (papers I-IV, when

applicable) and Rikshospitalet, Oslo (paper III) approved the studies. When

applicable, information about the study was provided to the participants by

research nurses and physicians, and informed consent was given by all par-

ticipants. To protect the integrity and privacy of the participating individuals,

all participants were assigned study codes, when applicable, which were

used throughout our analyses and in correspondence.

Subject inclusion (I-IV)

Renal transplant patients (I, III, IV)

The patients in the study described in papers I and IV were studied retro-

spectively on the basis of medical charts and nursing notes. All CMV-

seronegative patients at Uppsala University Hospital who consecutively re-

ceived a renal transplant from a CMV-seropositive donor between Septem-

ber 1998 and November 2000 (I) or between September 1998 and June 2007

(IV) and also received low-dose prophylaxis with valacyclovir 3g/day for 90

days post-transplantation were included.

In paper I, 25 patients (D+/R-) were followed for 6 months after trans-

plantation. Nine patients who received basiliximab induction therapy were

analyzed as a subgroup.

In paper III, all consecutive CMV-seropositive renal transplant recipients

(n=17) with haplotype HLA A*0201 and/or B*0702, from two transplant

centers, Uppsala University Hospital (n=13) and Rikshospitalet – Radium-

hospitalet Medical Centre, Oslo (n=4), between March 2002 and September

2003 were recruited. All donors were CMV-seropositive, resulting in a ho-

mogenous D+/R+ group.

In paper IV, 102 patients (D+/R-) were followed for up to 5 years after

transplantation. The results from these patients were compared to analogous

results from approximately 4,000 patients (see paper IV for details) from the

international Collaborative Transplant Study (CTS) database, to which Upp-

sala is one of the almost 300 contributing renal transplant centers.

40

Patients with primary CMV disease (II)

Five otherwise healthy persons with a symptomatic primary CMV infection

diagnosed at the Department of Infectious Diseases, Uppsala University

Hospital were included. They displayed classic symptoms of CMV infection

in combination with positive CMV-specific IgM and low IgG levels. They

had to express HLA allele A1 or A2, since these tetramers were the only

ones available at the start of the study.

Healthy controls with latent CMV infection (II, III)

CMV-seropositive and serologically HLA typed blood donors (II) and staff

from the Departments of Clinical Immunology (II, III) and Infectious Dis-

eases (III) were included in these studies. In paper II, 53 subjects (30 male

and 23 female) aged 27-80 were recruited, and in paper III, 11 subjects (1

male and 10 female) aged 29-55 were studied. In paper II, all subjects had to

express at least two of the following HLA alleles: HLA A1, A2, A3, A24,

B7, B8, and B35; in paper III, they had to express the HLA A*0201 and/or

B0702 haplotypes (corresponding to the only tetramers available at the start

of the study). All of the subjects stated that, as far as they knew, they were

perfectly healthy and had not experienced any signs of infection during the

past 2 weeks during the sampling period.

Blood sampling (II, III)

In paper II, symptomatic patients were sampled as soon as possible after

diagnosis, and subsequent samples were then obtained on weeks 4, 6, and 8,

and then every month up to 6 months after the onset of symptoms. Of the 53

latently infected subjects, 41 were tested twice, the second occasion occuring

at a median of 58 days (range 14-269) after the first sample.

The patients in paper III were repeatedly tested on day 0 (the day of

transplantation) and days 14 and 28, then at least once a month up to 1 year

after transplantation. Blood samples were collected for routine laboratory

tests and for quantification of CMV DNA, tetraCD8, IFNγCD8, and

IFNγCD4. Samples were drawn by nurses at hospitals and other health care

facilities and sent to the Department of Clinical Immunology for analysis

within 6 hours (functional assays) or 24 hours (tetramer assays) of venipunc-

ture. Samples for CMV PCR analysis were frozen for storage and analyzed

later (III). The latently infected subjects were repeatedly tested at the same

intervals as the patients.

41

Determination of CMV load (III, IV)

In paper III, CMV-DNA was quantified by a real-time PCR assay (TaqMan),

and the sequences of the PCR primers and probes were selected from the

UL65 gene encoding pp67 [149]. In paper IV, the commercial COBAS Am-

plicor CMV Monitor (Roche) was used from 1998 to 2000. From 2001 to

2008, we used an in-house PCR from our local laboratory [150].

MHC I tetramer analysis (II, III)

The numbers of CMV-specific CD8+ T cells were determined using MHC I

CMV tetramers (II,III) as previously described [38]. Titrated amounts of

phycoerythrin (PE)-labeled HLA-matched MHC I tetramers (Table 1) and an

antibody mixture containing αCD8-FITC and αCD3-APC were added to

EDTA-blood and incubated for 30 min at 4°C. FACS Lysing Buffer was

added, and the cells were repeatedly washed with PBS containing 1% FCS

and 0.1% sodium azide before FACS analysis was performed. In patients

with ongoing rejection and/or CMV disease, excessive nonspecific binding

was sometimes seen. In these cases, the experiment was repeated, with the

order of the lysing and staining reversed. A total of 20,000 to 30,000 CD3+

lymphocyte events were collected, and CMV-specific cells were expressed

as the percentage of CD8+ cells and considered positive if the value was

>0.10%. The number of CMV-specific cells/μl was recalculated by using the

absolute number of CD8+ cells, which was simultaneously determined.

Figure 6. A schematic drawing of a CD8 T cell binding a MHC I tetramer. To the right, a FACS plot showing tetramer-binding cells in the upper right quadrant. (Reproduced with the permission of Anna-Karin Lidehäll.)

42

Table 1. MHC I tetramers used in the studies

HLA A or B allele Peptide sequence Peptide Paper #

restriction element origin

A*0101 YSEHPTFTSQY pp65 II

A*0201 NLVPMVATV pp65 II,III

A*0301 TTVYPPSSTAK pp150 II

A*2402 QYDPVAALF pp65 II

B*0702 TPRVTGGGAM pp65 II,III

B*0801 ELRRKMMYM IE-1 II

B*3501 IPSINVHHY pp65 II

CMV-specific T-cell stimulation and intracellular cytokine staining (II, III)

The function of CMV-specific CD4+ and CD8+ T cells was measured by

assessing their ability to secrete IFNγ in response to specific stimulation; no

cytotoxic assays or proliferation assays were used. After blood sampling, T

cells were stimulated by adding CMV antigen (see below) to the sodium-

heparinized whole blood and incubating the samples at 37oC for 6 hours.

Stimulation of CD4+ T cells

CMV-specific CD4+T cells were stimulated by adding CMV-infected lysed

fibroblasts. Viral proteins were processed and presented by antigen-

presenting cells in the blood. The CD4+ T cells with the right specificity

became activated upon recognition of the combination of MHC II and viral

peptides [151].

Stimulation of CD8+ T cells

CMV-specific CD8+ T cells were stimulated by two different methods: In

paper II, a pool of overlapping peptides spanning the CMV protein pp65 was

used. The peptide pool consisted of 15 amino acid-long peptides covering

the whole protein and overlapping by 9 amino acids. Peptide pools mainly

activate CD8+ T cells but also activate CD4+ T cells to some extent [152].

In paper III, stimulation was achieved by adding the pp65-derived immu-

nodominant peptides NLVPMVATV (A*0201) or TPRVTGGGAM

(B*0702).

43

Intracellular cytokine staining

After the first 2 hours of a 6-hour of incubation with CMV antigen, brefeldin

A was added to inhibit the Golgi apparatus, thus containing the cytokines

intracellularly. For practical reasons, the last 4 hours of incubation took

place in a programmable waterbath that automatically cooled the samples

after 4 hours and kept the tubes cold overnight. On day 2, the erythrocytes in

the sample were lysed (with FACS Lysing Solution) and the T cells were

then permeabilized (FACS Permeabilizing Solution 2). Antibody staining

was performed with the addition of αCD3-APC, αCD8-PerCP, αCD38-, or

69-PE and αIFN�-FITC antibodies. The tubes were then incubated for 30

minutes at 4°C and then washed once with PBS containing 1% BSA before

FACS analysis was performed.

Figure 7. Intracellular cytokine staining after stimulation with CMV antigen. (a) and (b) - gating of CD3+ lymphocytes in FACS plots. (c) - IFNγ-production in an unstimulated sample. (d) - IFNγ-production in CD8 T cells after in vitro stimulation with a mixture of overlapping peptides covering the entire CMV protein pp65. (e) - IFNγ-production in CD4 T cells after in vitro stimulation with a CMV-infected and lysed fibroblast cell line. (Reproduced with the permission of Anna-Karin Lidehäll)

44

Absolute counting of lymphocyte subsets (II, III)

Enumeration of CMV-specific CD8+ and CD4+ T cells

The numbers of CMV-specific CD8+ T cells were determined using MHC I

CMV tetramers. IFNγ-production was examined by intracellular cytokine

staining after stimulation with a peptide pool spanning the entire pp65 (II) or

with a HLA-specific peptide (III). Because of the greater difficulty experi-

enced in synthesizing MHC class II tetramers, we used cytokine responses

against CMV lysate (III) to determine the number of CMV-specific CD4+ T

cells.

Flow cytometry and FACS analysis

Flow cytometry-based absolute cell counting was performed essentially as

described by Gratama et al. [153]. An antibody mixture containing titrated

amounts of �CD45-FITC, �CD8-PE, �CD4-PerCP, and �CD3-APC was

incubated with exactly 100 �l of EDTA-anticoagulated blood. After lysis of

the erythrocytes, 100 �l of Flow-Count Beads was added. To add exact

amounts of blood and beads, a reverse pipetting technique was used. Gating

and data analysis were performed as described by Gratama et al. [153]. All

samples were analyzed in duplicate, and mean values were calculated.

For all flow cytometric analyses, a four-color FACSCalibur instrument

with CellQuest or CellQuest Pro Software was used. FACS analyses were

performed within a few hours after staining. A total of 20,000 to 30,000

CD3+ lymphocytes were collected, and the percentage of IFN�-producing

CD8+ cells, as determined by cell counting, was calculated and expressed as

IFN�-producing CD8+ cells/μl. To correct for background cytokine release,

the IFN� result from a control tube without antigen was subtracted from the

experimental results (Fig. 7c).

Statistics

In paper I, the differences in the incidence of CMV disease/graft rejection

between the various groups with different immunosuppressive protocols

were analyzed with Fisher’s exact test.

In paper II and III, paired data obtained at various time points in the same

subject were compared using the Wilcoxon signed rank test. Differences

between groups were assessed by Mann-Whitney U-test and correlations

were calculated using Spearman´s nonparametric correlation test. Median

(minimum-maximum) values were used unless stated otherwise. In paper III,

45

Fisher´s exact test (two-tailed) was used to determine sensitivity and speci-

ficity.

In paper IV, univariate analysis of quantitative variables was conducted

using Student’s t-test, Mann-Whitney U-test or one-way ANOVA, whereas

binary categorical variables were analyzed using the chi-square test or

Fisher’s exact test according to sample sizes. The Kaplan-Meier method was

used to estimate patient and graft survival, and the results are indicated as

mean percentages. All statistical tests were two-tailed, and p-values <0.05

were considered significant. Median values with ranges (minimum-

maximum) are reported unless stated otherwise. All statistical calculations

were made with Statistica software (StatSoft, Inc., Tulsa, USA).

46

RESULTS

CMV disease in renal transplant patients (I, III, IV)

In Paper I, we retrospectively studied 25 CMV-mismatched (D+/R-) renal

transplant patients who received low-dose prophylaxis with VACV (3g/day).

The primary endpoint was CMV disease and time to onset of symptoms.

CMV disease was considered to be present if a patient had CMV DNAemia,

fever (≥38.0°C) for >2 consecutive days, and one or more of the following:

leukopenia, thrombocytopenia, pneumonitis, gastrointestinal disease, hepati-

tis, or ophthalmologically detected retinitis. Alternatively, patients were

diagnosed with CMV disease if they presented with CMV DNAemia and

any kind of illness responsive solely to treatment with ganciclovir or foscar-

net. Six of the 25 (24%) patients developed CMV disease within 6 months

after transplantation, a number that was about half of that recorded for our

historical controls [154]. The time to onset of symptoms was a median 156

days (92-191). None of the patients developed CMV disease during the pe-

riod of VACV prophylaxis. The symptoms were mild or moderate in five of

the six cases, and these patients recovered completely with GCV therapy for

3 weeks. None of the nine patients treated with basiliximab induction and

VACV developed CMV disease, whereas one patient outside the protocol,

who received basiliximab but not VACV developed a symptomatic CMV-

infection.

In paper III, 17 CMV-seropositive renal transplant recipients (D+/R+)

were prospectively followed for 1 year post-transplantation. CMV disease in

these patients was defined as CMV DNAemia in the plasma, together with

fever (>38.0° C) for >2 consecutive days and any clinical symptoms com-

patible with CMV-infection: leukopenia (below reference values); elevated

liver transferases (above reference values); or organ-specific disease, such as

colitis or pneumonia, with no other plausible explanation. One patient (6%)

developed CMV disease.

In paper IV, we retrospectively studied 102 CMV-mismatched (D+/R-)

renal transplant patients who received low-dose prophylaxis with VACV

(3g/day). Of these, 26 patients (25%) were diagnosed with CMV disease

during the first year post-transplantation (Fig.8). CMV disease was classified

into three different categories: CMV syndrome, tissue-invasive CMV dis-

ease, and clinical CMV disease. CMV syndrome required a fever (>38.0°C)

for >2 consecutive days without other possible cause and detection of CMV

47

in blood or serum and any of the following symptoms: fatigue, malaise, ar-

thralgias, decreased renal function, leukopenia, or elevated transaminases.

Tissue-invasive disease required any of the above symptoms and detection

of CMV in biopsies. Clinical CMV disease was any condition that did not

fulfil the criteria listed above but clinical judgment led to treatment with

anti-CMV drugs. The time to CMV disease was a median of 124 days (26-

191). Seven patients were diagnosed with CMV disease during prophylaxis

at a median time of day 60 (26-91), but six of them did not fulfil the criteria

for CMV syndrome or tissue-invasive disease.

Figure 8. Incidence of CMV disease. Comparison of the results of paper IV (Present study, VACV 3g) to other comparable studies on D+/R- renal transplant populations with different prophylaxis protocols. The Paya studies are based on a mixed solid organ transplant population, and therefore include more than just renal transplants. The shaded portion of the bars indicates clinical or investigator-treated CMV dis-ease.

48

CMV resistance (IV)

Resistant CMV was considered to be present if, despite 4 or more weeks of

treatment with any anti-CMV drug, the antiviral therapy had to be switched

and the CMV load in the blood failed to decline or CMV could be demon-

strated in biopsies. One patient (4%) with CMV disease (CMV colitis) ful-

filled the criteria and developed a GCV-resistant CMV strain with a muta-

tion in UL97.

Rejection and graft function in renal transplant patients (I, III, IV)

In paper I, biopsy-proven acute rejection (BPAR) was seen in 32% of the

patients (8/25), as compared to 38% in our group of historical controls. At 6

months post-transplantation, impaired renal function with creatinine levels

>200 umol/l was seen in 20% (5/25), including one patient on hemodialysis.

In paper III, 53% (9/17) of the patients experienced BPAR. Renal func-

tion, as measured by serum creatinine levels, was essentially unchanged

between 1 year post-transplantation (median level, 138 μmol/l; (range 98-

419), and 3 years post-transplantation (median level, 142 μmol/l; (range 101-

388).

In paper IV, 17 patients experienced one BPAR, five were diagnosed with

two episodes of BPAR, and one was treated for a BPAR on three different

occasions during the first year after transplantation. Thus, the total rejection

frequency was 22% (Fig.9). The mean time to rejection was 50.2 +/- 58.0

days (median, 28 days; range, 5-236) and the rejection was classified as

Banff I (A+B) in 52% and Banff II (A+B) in 48% of the cases. Rejection

was generally treated with steroids (n=22), and in six cases (26%) antithy-

mocyte globulin (ATG) was given. The use of basiliximab as induction ther-

apy was correlated with an absence of rejection episodes (p=0.008). One-

year graft survival was 95%; the mean serum creatinine level at 1 year after

transplantation was 153.8 μmol/l (+/-65.4), and the median was 131 μmol/l

(74-485).

49

Figure 9. Incidence of rejection. Comparison of the results of paper IV (Present study, VACV 3g) to other comparable studies on D+/R- renal transplant populations with different prophylaxis protocols. The Paya studies are based on a mixed solid organ transplant population, and therefore include more than just renal transplants.

CMV DNAemia and graft function in renal transplant patients (III)

In paper III, 16 of the 17 patients (94%) were positive at least once for CMV

DNA in the plasma. In these patients, the first positive CMV DNA sample

was detected at a median of 40 days (range, 12-227) after transplantation.

The maximum level of CMV DNAemia (CMVmax) during follow-up in pa-

tients without pre-emptive therapy (n=13) was a median of 8,400 (820-2.9

million) copies/ml, detected at a median of 95.5 (52-258) days after trans-

plantation. A correlation was found between the initial viral load and

CMVmax, but there was no correlation between CMVmax and graft function at

either 1 or 3 years after transplantation. Two patients with CMVmax >1 mil-

lion copies/ml displayed no symptoms of CMV disease and received no an-

tiviral therapy. Creatinine levels in the three patients with CMVmax > 1 mil-

lion copies/ml were 118, 138, and 180 μmol/l at 1 year and 118, 136, and

151 μmol/l at 3 years after transplantation.

50

Long-term clinical outcome in D+/R- renal transplant patients (IV)

The 5-year patient survival in the 102 patients (IV) was 92%, comparable to

the data from the CTS database (Fig.10, top). Graft survival at 5 years after

transplantation was also comparable to the data from the CTS (Fig.10, bottom).

Figure. 10. Patient survival (top) and graft survival (bottom) rates up to 5 years after transplantation. Comparison between all kidney transplants in Uppsala and Europe (EU w/o Uppsala; data from CTS) from September 1998 – June 2007. All patients were CMV D+/R- and received CMV prophylaxis.

51

Neurotoxic adverse effects of valacyclovir prophylaxis (I, IV)

In paper I, the follow-up period was 6 months and no CNS adverse effects

such as hallucinations, confusion, or paranoia were observed.

In paper IV, major neurotoxic adverse effects could be demonstrated in

two (2%) of the examined patients (Fig.11). One patient developed confu-

sion on day 31 despite a rather low dose of VACV (500mg thrice daily with

a GFR of 30 ml/min). One patient experienced episodes of hallucinations on

day 23 when he received GCV iv as a result of ATG-treated rejection, and

on day 119 when he received VACV 1g thrice daily despite a diminshed

GFR (40 ml/min). The neurotoxic effects disappeared in both patients with-

out discontinuation of the VACV prophylaxis. We found no evidence of

other major neurotoxic side effects such as paranoia or anxiety. Insomnia,

the most common neurotoxic side effect, was observed in 7 patients (7%).

The prophylaxis was switched to oral GCV in only one patient, as a result of

nephritis with a possible connection to VACV.

Figure 11. Major neurotoxic adverse effects. Comparison of the results of paper IV (Present study, VACV 3g) to other comparable studies on D+/R- renal transplant populations with different prophylaxis protocols.

52

Frequencies of CMV-specific CD8+ T cells (II, III)

Healthy subjects with latent CMV infection

In paper II, 53 latently infected but otherwise healthy individuals were stud-

ied. The tetramers used in the study are displayed in Table 1. The individuals

were HLA-matched with at least two of these tetramers. Of the 53 individu-

als, 51 bound at least one, and 43 bound at least two of the tested tetramers.

The frequencies of the different tetramer-binding populations were added,

and the average (mean+/-SEM) frequency of CMV-specific CD8+ T cells

was 3.5+/-0.8%, or 16+/-3.5 T cells/μl blood. Interestingly, the frequencies

of CMV-specific CD8+ cells varied considerably between individuals, but

the variation within each individual over time was small. Very high persis-

tent levels of CMV-specific CD8+ T-cell binding to the HLA B*0801ERL

tetramer were seen in two subjects: 25.9% (65 cells/μl) and 28.1% (118

cells/μl), respectively (see paper II, Fig.1 H for details). No clinical signs of

reactivation or CMV DNAemia could be detected, and the reason for these

high numbers remains unknown.

In paper III, 11 healthy subjects were longitudinally sampled during one

year. These subjects were positive for HLA A*0201 or B*0701; they also

displayed a large variation in CMV-specific CD8+ T cell levels between

individuals (0.1-60.2 cells/μl), but the intra-individual variations were low

during follow-up.

Healthy subjects with primary CMV infection

In paper II, the frequencies of CMV-specific T cells were investigated in five

healthy individuals with primary CMV infection. In these patients, CMV-

specific CD8+ T cells were found in high numbers within weeks after the

first clinical symptoms. The pattern of T-cell levels during follow-up, ex-

pressed as the mean (range), resembled that seen in previous studies of pa-

tients with primary Epstein-Barr-virus [155] and immunosuppressed patients

with primary CMV [156]. At 1 month after the onset of disease, the fre-

quency HLA A*0101YSE tetramer expression was 35 cells/μl (18.8-50.5),

and that for the A*0201NLV tetramer was 83.1 cells/μl (20.5-181.7). At 3

months, the frequencies were 5.1 cells/μl (4.9-5.2) and 18.8 cells/μl (4.3-

33.3) for HLA A*0101YSE and A*0201NLV, respectively, and at 6 months,

frequencies were down to levels seen in subjects with latent CMV infection.

Immunosuppressed subjects

In paper III, we demonstrated that the levels of CMV-specific CD8+ T cells

before transplantation did not differ from those in control subjects. Immedi-

ately after transplantation, the T-cell levels decreased rapidly, and at 1 month

53

after transplantation, the tetraCD8 (both A*0201 and B*0701) counts were

lower (median, 6.1 cells/μl; range, 0.3-29.7) than those at baseline. How-

ever, at 2 months, the levels were already comparable (median, 11.1

cells/μl; range, 0.6-119.1) to baseline levels, and at 4 months after transplan-

tation, the frequency of CMV-specific T cells was significantly higher (me-

dian, 26.1 cells/μl; range, 3.7-145.9) than at baseline. The tetraCD8 levels

then remained above baseline throughout the follow-up period, but this dif-

ference from baseline values was not statistically significant (Fig.12).

CD4+ and CD8+ T-cell function (II, III)

Healthy subjects with latent CMV infection

In paper II, functional analyses of the CMV-specific CD8+ T cells from 42

subjects were analyzed by intracellular cytokine staining. The cells were

stimulated with a pp65 peptide pool, and the mean number of CMV-specific

CD8+ T cells showing IFNγ production was found to be 3.9+/-0.6/μl blood

(range, 0-17.8).

Immunosuppressed subjects

In Paper III, the levels of IFNγ−producing CD8 (peptide-stimulated) and

CD4 (lysate-stimulated) T cells before transplantation did not differ from

those in the controls. Immediately after transplantation, the T-cell levels

decreased rapidly. At 14 days after transplantation, IFNγCD8 (median, 0.9

cells/μl; range, 0-6.0) and IFNγCD4 (median, 1.8 cells/μl; range, 0-14.4)

were already significantly reduced (p<0.05) and continued to decrease there-

after. At 2 months after transplantation, the IFNγCD8 numbers reached a

nadir, with a median of <0.03 cells/μl (range, 0-10.4; p<0.05), and the

IFNγCD4 reached a plateau, with a median of 1.2 cells/μl (range, 0-10.3;

p<0.005). The IFNγCD8 values then rebounded to levels similar to baseline

(p=n.s) at 4 months after transplantation, whereas the IFNγCD4 values re-

mained low until 1 year after transplantation (Fig.12).

54

Figure 12. From the top: The number of tetramer-selected CD8+ T cells, CMV-specific interferon-γ-producing CD8+ T cells and interferon-γ producing CD4+ T cells in renal transplant patients (filled bars, �) monitored up to 1 year after trans-plantation, and in controls (unfilled bars, �) monitored for 1 year. The values are expressed as medians and 25%-75% range. The dotted line (---) refers to baseline values obtained immediately before transplantation in patients, and at the start of the study in controls. Significant changes were seen in the transplant group in all T-cell populations, whereas the variation was small in the control group. See text for de-tails.

55

R2 = 0,786

CMV specific CD8+ T cells per μl blood

20 6040 80 100 120 14020 6040 80 100 12020 6040 80 100 120 140

2

4

68

10

12

14

16

18

2

4

68

10

12

14

16

18

4

68

10

12

14

16

18

Frequency and function of CMV-specific CD8+ T cells (II)

When we correlated the frequency of tetra+CD8 T cells with the number of

functional (IFNγ-producing CD8+ and CD4+) T cells in paper II, we saw

two patterns (Fig.13): A majority of the latently infected healthy individuals

displayed low numbers of both tetra+CD8 and functional T cells. One-third

of these subjects, however, had high levels of one of these two parameters;

none were high in both.

IFNγCD8 cells/μl

Figure 13. Relationship between the total number of CMV-specific CD8+ T cells (total tetramer-binding cells) and number of IFN�-producing CD8+ T cells in sub-jects eligible for analysis with three or four tetramers. A straight line obtained by least-squares analysis is shown for data in the upper quartile of either of the two parameters, represented by the open symbols (�).

These findings suggest that CMV remains inactive in most individuals,

whereas in a minority of CMV-infected people, an ongoing low-grade repli-

cation can occur that may be handled either by producing a small clone of

capable CD8+ effector cells or by up-regulating the size of the less effective

T-cell clone(s).

56

CMV-specific CD4+ T cells and CMV DNAemia (III)

In paper III, correlations were sought between the various T-cell populations

and CMVmax in patients not receiving pre-emptive therapy (n=13). The rela-

tive reduction in IFNγCD4 levels at 2 months after transplantation, when

compared to baseline values, was found to correlate with the CMVmax

(Fig.14). All six patients with IFNγCD4 levels <15% of baseline values at 2

months after transplantation developed high-grade DNAemia, with CMVmax

>10,000 copies/ml, either at 2 months after transplantation (n=4) or during

follow-up (n=2). None of the seven patients with IFNγCD4 levels >15% of

baseline values at 2 months after transplantation developed high-grade CMV

DNAemia. The sensitivity and specificity of this IFNγCD4 cut-off level in

this group of patients were 100% (p<0.001).

0 20 40 60 80 100

Proportion IFNγCD4 in %

1

1000

10000

100000

1000000

CM

V max

(cop

ies/

ml)

(r= -0.81; p<0.005)

Figure 14.Correlation between the relative IFNγCD4 reduction from baseline to 2 months after transplantation and CMVmax. High-grade CMV DNAemia, >10,000 copies/ml, occurred when the percentage of IFNγCD4 cells decreased to below 15% of baseline values.

57

The absolute number of IFNγCD4 at 2 months after transplantation was di-

rectly correlated with the CMVmax, but no correlation was found at baseline

or during the first month after transplantation. However, both patients with

undetectable levels of IFNγCD4 at baseline developed high-grade DNAe-

mia. No correlation was found between the CMVmax and the absolute values

for IFNγCD8 or tetraCD8, or the absolute reduction in IFNγCD4, IFNγCD8,

or tetraCD8 during the first 2 months after transplantation.

Immunodominance (II)

The T-cell response in persons with latent CMV infection is directed against

several CMV eiptopes. In paper II, we wanted to see whether the frequencies

of different CMV epitopes within the same donor were of equal magnitude

or whether any of the epitopes displayed immunodominance. In 18 of 37

individuals (49%) who were able to bind at least two tetramers, the fre-

quency of CD8+ T-cell binding to one of the tetramers was at least 10 times

higher than that seen for the tetramer with the second-highest binding fre-

quncy. The immunodominant epitope was identified by comparing T-cell

frequencies for different tetramers within the same donor. Using these fre-

quencies, we calculated the immunodominace value for each tetramer (Table

2, details in paper II, Fig.2). For example, the A2 tetramer bound with higher

frequency in 27 of 57 tetramer pairs examined, and therefore had an immun-

dominance value of 27/57=0.47. We found that the B*0702TPR epitope (0.82)

was the most immunodominant of the epitopes available to us.

Table 2. Immunodominance values based on observations from 53 healthy donors

HLA HLA HLA HLA HLA HLA HLA

B’0702TPR B*0801ERL A*0201NLV A*0101YSE B*3501IPS A*0301TVY A*2402QYD

0.82 > 0.72 > 0.47 > 0.21 > 0.19 > 0.03 > 0

_____________________________________________________________________________________

58

DISCUSSION

CMV, termed the “troll of transplantation” by Balfour as long ago as 1979

[157], remains an important pathogen in transplant patients. Both prophylac-

tic and pre-emptive strategies have been developed during the past 20 years

to decrease the risk of CMV disease. In 1989, prophylaxis with ACV was

demonstrated to be successful in reducing CMV disease [84]. In 1999,

VACV 8g/day was shown to yield a significant reduction in CMV disease

[86], and in 2004, VGCV was introduced as prophylaxis at many transplant

centers after a publication by Paya et al. [92]. On the other hand, pre-

emptive strategies involving both GCV and VGCV have also been demon-

strated to reduce the incidence of CMV disease [158, 159]. There are pros

and cons to both strategies and there is still ongoing debate among different

transplant populations and centers about which approach to choose [80, 96,

114, 160]. Recently published evidence-based guidelines [43, 97] recom-

mend universal prophylaxis for patients at the highest risk for CMV disease,

whereas preemptive therapy may be most appropriate for those with a mod-

erate or low risk of developing CMV disease. There are, however, still ques-

tions to be answered: Issues such as what drug to use, how to respond to

adverse reactions and the emergence of resistance, what constitutes an opti-

mal dose, when to start prophylaxis, and how long therapy should be contin-

ued still need to be defined.

VACV prophylaxis in the high-risk (D+/R-) population

In the high-risk D+/R- renal transplant population, the risk of developing

CMV disease when high-dose VACV prophylaxis is used is comparable to

oral GCV [161], whereas the risk with oral GCV is comparable to VGCV

[92]. The use of high-dose VACV has been limited because the potential

neurotoxic adverse effects associated with this protocol can be profound and

have lead to discontinuation of the drug in as much as 6 to 24% of patients in

some studies [80, 161], however, in one study high-dose VACV was used

without neurotoxic adverse effects [162]. These effects are presumably re-

lated to the use of the high VACV dose [163]; we hypothesised that a lower

dose of VACV would still be able to prevent CMV disease without causing

neurotoxic adverse effects. This prediction was confirmed in paper I and IV,

in which we demonstrated that prophylaxis with low-dose VACV, 3g/day,

59

decreased the incidence of CMV disease without any drug-related neurotoxic

adverse effects. These findings have also been supported by other studies

[164]. The incidences of CMV disease in paper I and IV was 24% and 25%

respectively, and these rates are at the high end of the range reported for

similar prophylaxis studies [86, 92]. These high rates may be attributed to

our wide-ranging definition of CMV disease; if a stricter definition had been

used in paper IV, the incidence would have been 10%.

In general, CMV morbidity was a minor problem, since most infections

were easily treated, and in only a few cases was treatment given repeatedly.

One observation that deserves comment is the fact that seven patients devel-

oped CMV disease during prophylaxis (IV). This frequency is higher than

that observed in other prohylaxis studies [86, 165] and could indicate that a

low-dose VACV protocol is insufficient in some cases. These patients who

developed CMV were, however, easily treated without sequelae, with the

exception of the patient with GCV-resistant CMV who developed a severe

colitis.

Concerns have been raised about prophylaxis because of the possible in-

crease in late-onset CMV disease and the higher proportion of tissue-

invasive disease that has been demonstrated after VGCV prophylaxis [165,

166]. These findings could potentially be explained by the profound viral

suppression that is caused by VGCV, which could prevent an effective anti-

gen-induced priming of the host defence. As a result, when anti-viral pro-

phylaxis is withdrawn, the late-onset CMV infection might rapidly develop

into CMV disease because of the absence of CMV-specific T cells. If so,

VACV, which is not as highly effective as VGCV, could be a more appeal-

ing choice because low-level viral replication could more easily occur, prim-

ing their CMV-specific immunity.

Two very important endpoints after a renal transplantation, are long-term

patient and graft survival rates, and it has been shown that persistant CMV

infection can affect long-term graft survival [167, 168]. The impact of CMV

disease on long-term graft outcome, however, has been a matter of discus-

sion and studies have produced conflicting results [52, 75, 167]. In paper IV,

we demonstrated that low-dose VACV prophylaxis produced results that

were equal or better than those from other prophylaxis studies [86, 92, 169]

and the CTS database (paper IV, Fig.1 and 2), with 5-year survival rates of

>90% (patient) and >80% (graft).

VACV therefore provides an alternative choice for CMV prophylaxis, al-

lowing GCV and VGCV -the current prophylactic drugs- to be reserved for

the treatment of CMV disease. This could potentially decrease the likelihood

of creating strains that are GCV-resistant.

60

CMV resistance

The emergence of resistance is a matter of increasing concern; in patients

with CMV disease who are infected with drug-resistant virus, the clinical

outcome is generally poor, with a mortality rate of up to 20% [108, 114,

115]. It is believed that the higher bioavailability of GCV that follows the

administration of VGCV could reduce the risk of GCV resistance [170], but

in a recent article by Eid et al. [114], in which VGCV prophylaxis was used

in D+/R- solid organ transplant patients, as many as 14% of the patients with

CMV disease had drug-resistant virus. The risk of developing drug-resistant

CMV in the mismatched renal transplant population has been estimated to be

as high as 5% [108], and since the number of renal transplants is high, the

need for an optimal prophylactic strategy is clear. One way to reduce GCV-

resistant CMV might be to reserve GCV for treatment and switch to other

compounds, such as VACV, for prophylaxis. In a study of VACV prophy-

laxis by Alain et al. [171], a very low incidence of GCV-resistance was

demonstrated, and it was postulated that VACV may be less likely to select

UL97-mutated GCV-resistant strains than is GCV. This theory is supported

by the findings in the present low-dose VACV prophylaxis study (IV), in

which we found a very low incidence (1%) of drug-resistant virus.

Strategies in non-high-risk populations

In contrast to CMV mismatched (D+/R-) patients, the risk of developing

CMV disease in seropositive patients receiving organs from seropositive

donors (D+/R+) is regarded to be low to intermediate [63]. Data from the

CTS has also indicated that graft survival in the D+/R+ group is unaffected

by the use of prophylaxis (Fig.15). According to current guidelines, patients

with low or intermediate risk, based on CMV serostatus, could benefit from

pre-emptive therapy [43]. This strategy, however, is logistically demanding,

and therapy is often introduced when the patient already has become symp-

tomatic. Also, a large number of patients are monitored and treated “in

vain”. It would be better to be able to pinpoint individuals at risk of develop-

ing CMV disease, but at present we lack the diagnostic tools to identify these

individuals. Therefore, other approaches such as measuring patients’ specific

immune status, could be of interest (III).

61

w/o UppsalaAll Kidney Transplants in Europe Sep1998-Jun2007

CMV Prophylaxis for CMV R+/D+

0

100

95

90

85

80

75

70

65

% G

raft

Sur

viva

l

0 1 2 3 4 5Years

5,686 n=No Prophylaxis2,975 n=With Prophylaxis

Figure 15. Graft survival up to 5 years after transplantation in D+/R+ kidney trans-plant patients with and without CMV prophylaxis.

CMV-specific immunity

In paper II and III, we have used tetramer analysis to enumerate CMV-

specific CD8+ T cells. This T-cell population has been demonstrated to be of

importance in SCT patients, in whom the number of tetraCD8 T cells has

been reported to correlate with the risk of developing CMV disease [172]. In

paper II, we found that the inter-individual variation in CMV-specific CD8+

T cells was large, and two healthy latently infected subjects displayed un-

usually high levels. Several factors could have contributed to these findings:

One possibility is that these two patients experienced a local, low-grade re-

activation of CMV. Another is the fact that they were slightly older (47 and

70 years of age), since some studies have reported a correlation between age

and an increase in CMV-specific T cells [173, 174]. A third, and perhaps the

most plausible explanation, is that the functional capacity of this clone was

reduced and that an up-regulation of the numbers of T cells is needed to

maintain an adequate total immunologic capacity versus CMV. This expla-

nation is supported by the findings in paper II, in which we found a correla-

tion between the frequency and function of CD8+ T cells (paper II, Fig.4);

however, this concept requires further study.

62

Tetramer analysis: pros and cons

Tetramer analysis is rapid and specific and can be performed with small

amounts of blood. However, there are limitations to this technique: The pa-

tient must have HLA alleles that match the available tetramers, and even

though the number of different tetramers has increased recently, there are

still a number of alleles lacking. Another disadvantage is that the number of

peptides is limited, and even if patients have the right combination of tet-

ramer and HLA, their T-cell clones may be directed against epitopes other

than those presented by the tetramer.

These restrictions make it clear that tetramer analysis cannot be used to

demonstrate the whole repertoire of CMV-specific T cells in an individual. It

can, however, still be of value for monitoring a patient over time, since an

increase or decrease in one tetramer-binding population should reflect

changes in the total CMV-specific lymphocyte population. One other disad-

vantage of tetramer analysis is that the number of CMV-specific CD8+ T

cells does not necessarily reflect the individual’s level of CMV-specific im-

munity. This situation was seen in both latently infected subjects (paper II),

in whom high numbers of CMV-specific CD8+ T cells were found to corre-

late with low levels of functional T cells, and in immunosuppressed subjects

(paper III), in whom the dynamics of CMV-specific CD8+ T cells differed

markedly from those of their IFNγCD8 and IFNγCD4 T cells (paper III,

Fig.3). Several studies have tried to use levels of tetramer-binding cells as a

marker for clinical outcome in various transplant populations, but the results

have proved inconclusive [172, 175-177].

Dynamics of IFNγCMV-specific T cells

When it comes to monitoring CMV-specific immunity, it seems logical to

measure the levels of functional T cells, and even though functional assays

such as intracellular cytokine staining after pulsing with CMV antigens are

time-consuming and laborious, this approach may be more promising. In

paper III, we demonstrated that after an initial rapid decrease in all T cells,

IFNγCD8 regeneration occured, but with a delay of approximately one

month when compared to the tetraCD8 response. This pattern suggests that

the first CD8+ cells may have expanded rapidly but were ineffective and that

more effective IFNγCD8 cells were generated over time. Recovery of

IFNγCD8 activity has been suggested to be protective against CMV disease

in transplant patients [178, 179]. Even though we were unable to demon-

strate a direct correlation, the lack of IFNγCD8 response in one of the pa-

tients (UA-15, described in paper III, Fig.5), might explain the prolonged

DNAemia observed in one patient.

63

High IFNγCD4 levels have also been shown to correlate with protection

from CMV disease [180, 181]. In paper III, the IFNγCD4 response in our

transplant patients was severely impaired during the entire first year after

transplantation, suggesting that IFNγCD4 restoration in reactivated CMV

infection in renal transplant recipients is delayed to a greater extent than had

previously been thought [180, 182].

How should CMV-specific T cells be measured?

While previous studies have tried to correlate the percentage of IFNγ-

producing CMV-specific cells with clinical CMV complications such as

CMV DNAemia and CMV disease [180], we have focused on the absolute

number of IFNγ-producing T cells. The reason for taking this approach is

that CMV-specific T-cell number vary extensively between individuals with

latent CMV infection (paper II, [20]), indicating that there is no general level

of T-cell response required to maintain the infection in a latent state. Instead,

a balance between the immune system and the virus is apparently established

in each individual, and we suggest that the percentage of IFNγCD4 cells, as

compared to baseline level, might be a more accurate way of defining the

immunological capacity of a specific recipient. Therefore, the relative reduc-

tion in T-cell responses rather than the reduction in absolute numbers could

be important. This hypothesis is supported by the findings in paper III, in

which patients with a rapid and substantial decrease (>85%) in IFNγCD4

during the first 2 months after transplantation had a higher risk of developing

high-grade CMV DNAemia.

We suggest that the dynamics of IFNγCD4, and specifically the level of

reduction during the first 2 months after transplantation, when compared to

baseline, may serve as valuable indicators for predicting the relative risk of

CMV DNAemia. These prognostic factors could also be used for identifying

subgroups of patients that could benefit from surveillance or prophylactic

therapy. Today, prophylactic, pre-emptive, and deferred therapy are all used

in various categories of patients with different levels of risk of developing

CMV disease [96, 160]. Characterizing the CMV-specific immune responses

with baseline samples obtained before the onset of immunosuppression is

therefore of highest importance in the development of strategic approaches

for pre-emptive or prophylactic therapy, or when monitoring functional T

cells in clinical situations involving therapy failure. The dynamic changes in

the levels of CMV-specific T-cell responses during the first months after

transplantation could possibly even serve as a tool for monitoring the im-

mune status at a more general level, providing the clinician with information

for decision-making with regard to immunosuppression levels and/or pro-

phylactic or pre-emptive therapy. These applications will have to be evalu-

ated in larger cohorts of patients, using broadly targeted antigens for T-cell

stimulation.

64

CONCLUSIONS

Paper I

• Low-dose (3g/day) VACV prophylaxis for 90 days post-transplantation

reduces the incidence of CMV disease in CMV-seronegative renal trans-

plant patients from 54% to 24%. The graft rejection rate (32%) was simi-

lar to those of historical controls (38%).

• Neurotoxic adverse effects like those previously reported with a higher

VACV dose (8g/day) were not observed.

• CMV disease was not seen in the subgroup that received basiliximab as

induction therapy.

Paper II

• In healthy subjects with a primary CMV infection, high frequencies of

CMV-specific CD8+ T cells are seen initially after the infection, but after

a few months, the levels are similar to subjects with a latent CMV infec-

tion.

• In subjects with latent CMV infection, the frequencies of CMV-specific

CD8+ cells vary considerably between individuals, but the intra-individual changes over time are small.

• In subjects with latent CMV infection, the T-cell response is directed

against several CMV eiptopes. The B*0702 TPRVTGGGAM epitope was

the most immunodominant epitope of those available for testing.

65

Paper III

• CMV DNAemia is present in >90% of the CMV-seropositive renal trans-

plant patients within the first year after transplantation.

• CMV-specific T-cells decrease rapidly after transplantation. TetraCD8

and IFNγCD8 regenerate within 3 months, whereas IFNγCD4 recovery is

impaired during the entire first year after transplantation.

• The percentage of IFNγCD4 T cells at 2 months post-transplantation, as

compared to baseline, is strongly correlated with CMV DNAemia.

Paper IV

• Low-dose VACV prophylaxis (3g/day) for 90 days post-transplantation in

D+/R- renal transplant patients results in a high 5-year patient and graft

survival, comparable to that obtained with other prophylaxis strategies.

• Low-dose VACV prophylaxis reduces the incidence of CMV disease and

acute rejection to the same levels as those of other prophylaxis strategies,

such as high-dose VACV, oral GCV, or VGCV.

• Major neurotoxic adverse effects are minimal with low-dose VACV pro-

phylaxis.

66

FUTURE PERSPECTIVES

Despite years of intense research on CMV and its effects, infection with this

virus remains a multifaceted problem, and it is evident that much research

still needs to be done. The major areas for future research are outlined be-

low.

Among the most important tasks to focus on is the development of a

CMV vaccine, with the primary target populations being CMV-seronegative

women of childbearing age and high-risk transplantation populations. If a

vaccine were successfully developed, a dramatic decrease in congenital

CMV disease and transplantation-associated CMV disease would be ex-

pected, with a subsequent substantial reduction in morbidity, mortality, and

hospital-associated costs. However, recent studies have revealed that vaccine

development is very complex, and unfortunately there is no prospect of an

effective vaccine becoming available in the immediate future.

In solid organ transplant populations with low or intermediate risk of de-

veloping CMV disease, as indicated by their CMV serostatus (e.g., D+/R+),

it is of interest to target those patients who are subject to a higher risk for

reasons unrelated to CMV serostatus. One potentially useful approach would

be to explore CMV-specific immunity further and to establish certain cut-off

levels for initiating more active strategies to prevent CMV disease. If pro-

phylactic strategies are used, it is also important to decide whether and how

these patients should be monitored after the cessation of prophylaxis in order

to reduce late-onset CMV disease.

The development of antiviral drugs is another area in which increased re-

search is needed. During the past 20 years, only a few anti-CMV drugs have

been licensed, and the adverse effects of those that are currently available are

troublesome. There is an urgent need for more alternatives, both in terms of

decreasing adverse effects and as a source of possible combination therapies

for minimizing the emergence of resistance. This need is particularly acute

because the number of immunosuppressed patients continues to increase,

thanks to the availability of more powerful treatment options for patients

with both cancer and autoimmune diseases, and to the growth in the SOT

and SCT populations.

67

Finally, more resources need to be channeled into research on the indirect

and chronic effects of CMV, such as arteriosclerosis and cancer, and the

potential exhaustion of immune defenses in the elderly. The question of

whether CMV is a bystander in the chronic inflammatory process or an ini-

tiator is a matter for further exploration.

In summary, CMV is a fascinating and multi-talented virus that has

evolved for millions of years, collecting a range of powerful tools that enable

it to evade a variety of defense mechanisms and infect most of the human

population. Another remarkable feature of this virus is its ability to maintain

an equilibrium with the host’s immune defense after infection, allowing it to

establish life-long persistence in the host. This persistence achieved by co-

evolution has been viewed as a disadvantage but may also be of some sig-

nificance for humans. This alternative remains to be explored, but it is just

possible that this little microbial companion may prove vital to our survival

for millions of years to come...

68

ACKNOWLEDGMENTS

This thesis is the sum of many peoples’ hard work, and I would really like to

express my sincere gratitude to all those who have contributed and helped

me through this process, both directly and indirectly. There are, however,

some persons who have contributed a little extra, and I thank you all for that!

Britt-Marie Eriksson, my supervisor. The pace of this project has been

close to perfect, with inspiring studies that have had a clinical aspect and

relevancy. You have always been confident and believed in me, so thank you

for letting me live my kind of life during these years! It has been a long

journey but we have learned a lot along the way...

Olle Korsgren, Professor at the Department of Clinical Immunology and my

co-supervisor. He always has a smile on his lips, but don’t let that fool you

too often... Thank you for our ongoing collaboration and possible future

projects.

Jan Sjölin, Professor at the Department of Infectious Diseases. One of the

few people that I rank highly in all three areas: research, clinical medicine,

and educational skills. A tough negotiatior, while still managing to share a

smile and a good story. I hope that we can have fun together for many more

years to come!

Göran Friman, Professor emeritus at the Department of Infectious Dis-

eases, who has supported me from the start, inspiring me with his vast

knowledge of both infectious diseases and wine tasting.

Björn Olsen, our latest Professor at the Department of Infectious Diseases

who, despite the little time we have spent together, has already inspired and

supported me in scientific matters.

Göran Günther, the head of the Department of Infectious Diseases, who

has supported me in both research and management decisions.

69

Birgitta Sembrant, what can I say.... Without you this thesis would never

have made it through the formalities! You have a tremendous knowledge of

how to get through the red tape, and for you, problems don’t exist!

All co-authors and research facilitators, but especially:

Anna Karin Lidehäll, my predecessor in the CMV thesis work, who, apart

from invaluable laboratory work and knowledge in the immunological area,

has managed to give birth to two children during the time we worked to-

gether, thus giving me some extra time to catch up! You have also been the

mastermind behind the immunological figures in the thesis and sometimes

more of a co-supervisor (at least in the immunological field!).

Kerstin Claesson, a transplant surgeon who not only read but scrutinized

our common manuscript, and then, uncompelled, offered to help me with the

chapter on immunosuppression and renal rejection in my thesis.

Gunnar Tufveson, Professor at the Transplant section, with whom I just

recently started to collaborate. Another person with a smile below the mus-

tache! I really hope that there will be more mutual research projects in the

future.

Ingrid Skarp Örberg, the best transplant coordinator there is, with a fantas-

tic organizational capacity. Without your personal knowledge of all the renal

transplant patients and your connections to all the regional hospitals, little of

this research would have been done.

Selina Parvin and Tobias Lundberg at the Department of Clinical Immu-

nology, for your hard lab work when all the immunological samples had to

be analyzed.

Kåre Bondeson, Jonas Blomberg, and Björn Herrmann at the Department

of Clinical Virology, for collaboration and fruitful discussions about the

mysteries of CMV and other viruses.

Deborah ”Debbie” McClellan for the fast, humorous and excellent lan-

guage editing!

I also want to thank all my great colleagues and personnel at the Department

of Infectious Diseases who still make my ”working life” interesting and fun,

but in particular:

70

Elisabeth Löwdin, my mentor, who has taught me everything (almost!) I

need to know when it comes to practising infectious disease medicine!

Mia Furebring, my partner on the clinical board, with remarkable capabili-

ties. You always prove that it is possible to do more if you just stay focused,

not to mention getting things done administratively...

Ingrid Uhnoo, my first clinical supervisor, who incessantly demonstrated

how thoroughly you could (should?) investigate a patient.

All the amazing secretaries; especially Marie Sehlbrand, the ”spider” on

the clinical side, who has facilitated clinical work and practical matters since

I started at the clinic, Marianne Söderblom, for all your help and suppor-

tive discussions concerning the schedule and other matters... and finally,

Titti Hamström for the fantastic and invaluable help with the pictures on

the thesis cover!

All the personnel at our regular and emergency wards, especially Sissi (our

family’s own nurse!), and Eva Söderberg, with whom I have worked with

since my first scary on-call nights! I have learned a lot from you both!

A great thank you to all my friends in different groups: other colleagues,

personnel, relatives, choirmembers, neighbors, soccerteam members, old

students, pokerplayers... all you friends who makes life worth living! Among

all of these I would especially like to mention:

Magnus Kaijser and Lena Rosenberg and family, for pleasant, inspiring,

and thoughtful conversations and family events, and also for toastmastering

the two most important parties of my life, for my wedding and dissertation.

Katja Fall, who has demonstrated that even the hardest times can be dealt

with without losing a taste for the good life with wine and cheese. I look

forward to more of that, maybe in combination with some future CMV stud-

ies.

And last, some thanks to the people closest to me:

My second family: Otto Cars, also Professor at the Department of Infec-

tious Diseases, who inspired me to start at the clinic, with whom I have had

many stimulating discussions about work, research, music, nephews, practi-

cal Utö-matters and daughters... I’m also glad that we mutually decided that

it would be better for me to start my research in the viral arena, since the

fight against bacteria is almost over...

71

Karin Cars, my mother-in-law, for always stimulating the wine part of my

brain, and providing me with clothes refused by the Cars-family members...

Thomas Cars, the nephew with whom you always can have a teasing chat.

The amazing combination of a pharmacist, ”practical” wizard, and singer-

songwriter whose music have entertained the most memorable moments of

my life.

Stefan Cars, my second nephew, who has arranged for me to be able to

write much of the thesis in my livingroom.

My dearest brother Krister, with Marie and their two adorable daughters

Vera and Vanja. Thank you for being with me almost all my life and for all

your skills that have forced me (as a big brother) to develop my own and

have kept me on my toes.

My parents Robert and Margareta, who have provided me with the essen-

tial genes and environment to manage this kind of life. I especially appreci-

ate the opportunities you have given me in terms of all the musical educa-

tion, the endless support with childcare, and finally, the fact that you actually

moved into our house to take care of my family during the final 2 months

while I was finishing the thesis...

And finally: Ninna (or Katarina), for your never-ending energy, all kinds of

support, including one figure in the thesis (!), and help with the dissertation

party. You are a remarkable companion in life, and I love you!

And my three fantastic children, who with your different personalities al-

ways inspire me and remind me what life is really worth living for. You have

provided plenty of moments of distraction, reduced my sleeping hours,

helped me let go of my frustration, but most of all gived me all the love that

I’m really sure I needed!

Till mina tre älsklingar som alla har sett dagens ljus under avhandlingsarbe-

tet: Edla, Dante och Tyra - nu är virusboken klar och vi kan fortsätta upp-

täcka världen tillsammans!

The studies were conducted with financial support from the Olinder-Nielsen Foundation, Beckman-

Coulter Inc. (III) and GlaxoSmithKline (I).

72

REFERENCES

1. Mocarski ES, C.C., Cytomegaloviruses and their replication., in Fields Virology, H.P. Knipe DM, et al, Editor. 2001, Lip-

pincott Williams and Wilkins Publishers: Philadelphia. p.

2629-2673.

2. McGeoch, D.J., et al., Molecular phylogeny and evolutionary timescale for the family of mammalian herpesviruses. J Mol

Biol, 1995. 247(3): p. 443-458.

3. Taylor-Wiedeman, J., et al., Monocytes are a major site of persistence of human cytomegalovirus in peripheral blood mononuclear cells. J Gen Virol, 1991. 72 ( Pt 9): p. 2059-

2064.

4. Mendelson, M., et al., Detection of endogenous human cy-tomegalovirus in CD34+ bone marrow progenitors. J Gen Vi-

rol, 1996. 77 ( Pt 12): p. 3099-3102.

5. Slobedman, B. and E.S. Mocarski, Quantitative analysis of latent human cytomegalovirus. J Virol, 1999. 73(6): p. 4806-

4812.

6. Soderberg, C., et al., Definition of a subset of human periph-eral blood mononuclear cells that are permissive to human cy-tomegalovirus infection. J Virol, 1993. 67(6): p. 3166-3175.

7. Jarvis, M.A. and J.A. Nelson, Human cytomegalovirus persis-tence and latency in endothelial cells and macrophages. Curr

Opin Microbiol, 2002. 5(4): p. 403-407.

8. Jarvis, M.A. and J.A. Nelson, Human cytomegalovirus tropism for endothelial cells: not all endothelial cells are created equal. J Virol, 2007. 81(5): p. 2095-2101.

9. Jarvis, M.A. and J.A. Nelson, Mechanisms of human cy-tomegalovirus persistence and latency. Front Biosci, 2002. 7:

p. 1575-1582.

10. Gandhi, M.K. and R. Khanna, Human cytomegalovirus: clini-cal aspects, immune regulation, and emerging treatments. Lancet Infect Dis, 2004. 4(12): p. 725-738.

73

11. Rowshani, A.T., et al., Clinical and immunologic aspects of cytomegalovirus infection in solid organ transplant recipients. Transplantation, 2005. 79(4): p. 381-386.

12. Bergström, M., Amning av barn födda 2005: Report from The National Board of Health and Welfare, Sweden. 2007.

13. Ahlfors, K., IgG antibodies to cytomegalovirus in a normal urban Swedish population. Scand J Infect Dis, 1984. 16(4): p.

335-337.

14. M. Hecker, D.Q., K. Marquardt, G. Bein, H. Hackstein,, Con-tinuous cytomegalovirus seroconversion in a large group of healthy blood donors. Vox Sanguinis, 2004. 86(1): p. 41-44.

15. Ryckman, B.J., et al., Human cytomegalovirus entry into epithelial and endothelial cells depends on genes UL128 to UL150 and occurs by endocytosis and low-pH fusion. J Virol,

2006. 80(2): p. 710-722.

16. Soderberg-Naucler, C., K.N. Fish, and J.A. Nelson, Reactiva-tion of latent human cytomegalovirus by allogeneic stimulation of blood cells from healthy donors. Cell, 1997. 91(1): p. 119-

126.

17. Homman-Loudiyi, M., et al., Envelopment of human cy-tomegalovirus occurs by budding into Golgi-derived vacuole compartments positive for gB, Rab 3, trans-golgi network 46, and mannosidase II. J Virol, 2003. 77(5): p. 3191-3203.

18. Emery, V.C., et al., The dynamics of human cytomegalovirus replication in vivo. J Exp Med, 1999. 190(2): p. 177-182.

19. Landini, M.P. and M. La Placa, Humoral immune response to human cytomegalovirus proteins: a brief review. Comp Immu-

nol Microbiol Infect Dis, 1991. 14(2): p. 97-9105.

20. Sylwester, A.W., et al., Broadly targeted human cytomegalovi-rus-specific CD4+ and CD8+ T cells dominate the memory compartments of exposed subjects. J Exp Med, 2005. 202(5):

p. 673-685.

21. Khan, N., et al., T cell recognition patterns of immunodomi-nant cytomegalovirus antigens in primary and persistent infec-tion. J Immunol, 2007. 178(7): p. 4455-4465.

22. Kuijpers, T.W., et al., Frequencies of circulating cytolytic, CD45RA+CD27-, CD8+ T lymphocytes depend on infection with CMV. J Immunol, 2003. 170(8): p. 4342-4348.

23. Reusser, P., et al., Cytotoxic T-lymphocyte response to cy-tomegalovirus after human allogeneic bone marrow transplan-

74

tation: pattern of recovery and correlation with cytomegalovi-rus infection and disease. Blood, 1991. 78(5): p. 1373-1380.

24. Walter, E.A., et al., Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N Engl J Med, 1995.

333(16): p. 1038-1044.

25. Tu, W., et al., Persistent and selective deficiency of CD4+ T cell immunity to cytomegalovirus in immunocompetent young children. J Immunol, 2004. 172(5): p. 3260-3267.

26. Chen, S.F., et al., Antiviral CD8 T cells in the control of pri-mary human cytomegalovirus infection in early childhood. J

Infect Dis, 2004. 189(9): p. 1619-1627.

27. Bronke, C., et al., Dynamics of cytomegalovirus (CMV)-specific T cells in HIV-1-infected individuals progressing to AIDS with CMV end-organ disease. J Infect Dis, 2005. 191(6):

p. 873-880.

28. Arase, H., et al., Direct recognition of cytomegalovirus by ac-tivating and inhibitory NK cell receptors. Science, 2002.

296(5571): p. 1323-1326.

29. Basta, S. and J.R. Bennink, A survival game of hide and seek: cytomegaloviruses and MHC class I antigen presentation pathways. Viral Immunol, 2003. 16(3): p. 231-242.

30. Hegde, N.R., M.S. Chevalier, and D.C. Johnson, Viral inhibi-tion of MHC class II antigen presentation. Trends Immunol,

2003. 24(5): p. 278-285.

31. Mocarski, E.S., Immune escape and exploitation strategies of cytomegaloviruses: impact on and imitation of the major his-tocompatibility system. Cell Microbiol, 2004. 6(8): p. 707-717.

32. Odeberg, J., et al., The human cytomegalovirus protein UL16 mediates increased resistance to natural killer cell cytotoxicity through resistance to cytolytic proteins. J Virol, 2003. 77(8):

p. 4539-4545.

33. Humar, A., et al., A prospective assessment of cytomegalovirus immune evasion gene transcription profiles in transplant pa-tients with cytomegalovirus infection. Transplantation, 2007.

83(9): p. 1200-1206.

34. Hernandez-Fuentes, M.P., A.N. Warrens, and R.I. Lechler,

Immunologic monitoring. Immunol Rev, 2003. 196: p. 247-

264.

75

35. Miyahira, Y., et al., Quantification of antigen specific CD8+ T cells using an ELISPOT assay. J Immunol Methods, 1995.

181(1): p. 45-54.

36. Manz, R., et al., Analysis and sorting of live cells according to secreted molecules, relocated to a cell-surface affinity matrix. Proc Natl Acad Sci U S A, 1995. 92(6): p. 1921-1925.

37. Pala, P., T. Hussell, and P.J. Openshaw, Flow cytometric measurement of intracellular cytokines. J Immunol Methods,

2000. 243(1-2): p. 107-124.

38. Altman, J.D., et al., Phenotypic analysis of antigen-specific T lymphocytes. Science, 1996. 274(5284): p. 94-96.

39. Goodpasture E.W. and T. F.B., Concerning the Nature of "Protozoan-like" Cells in Certain Lesions of Infancy. Ameri-

can Journal of Diseases of Children, 1921: p. 21:415.

40. Kirubakaran, S., Update: cytomegalovirus infection in HIV-infected patients - a review. Clinical Microbiology Newsletter,

2004. 26(18): p. 137-144.

41. Lazzarotto, T., et al., Congenital cytomegalovirus infection: recent advances in the diagnosis of maternal infection. Hum

Immunol, 2004. 65(5): p. 410-415.

42. Goldstein, L.C., et al., Monoclonal antibodies to cytomega-lovirus: rapid identification of clinical isolates and prelimi-nary use in diagnosis of cytomegalovirus pneumonia. Infect

Immun, 1982. 38(1): p. 273-281.

43. Razonable, R.R., V.C. Emery, and Management of CMV infec-tion and disease in transplant patients. 27-29 February 2004. Herpes, 2004. 11(3): p. 77-86.

44. van der Bij, W., et al., Comparison between viremia and anti-genemia for detection of cytomegalovirus in blood. J Clin Mi-

crobiol, 1988. 26(12): p. 2531-2535.

45. Drew, W.L., Laboratory diagnosis of cytomegalovirus infec-tion and disease in immunocompromised patients. Curr Opin

Infect Dis, 2007. 20(4): p. 408-411.

46. Anonymous, Cytomegalovirus. Am J of Transplant, 2004.

4(suppl 10): p. 51-58.

47. Boivin, G., et al., Comparative evaluation of the cytomegalovi-rus DNA load in polymorphonuclear leukocytes and plasma of human immunodeficiency virus-infected subjects. J Infect Dis,

1998. 177(2): p. 355-360.

48. Piiparinen, H., et al., Comparison of two quantitative CMV PCR tests, Cobas Amplicor CMV Monitor and TaqMan assay,

76

and pp65-antigenemia assay in the determination of viral loads from peripheral blood of organ transplant patients. J

Clin Virol, 2004. 30(3): p. 258-266.

49. von Muller, L., et al., CMV monitoring using blood cells and plasma: a comparison of apples with oranges? Bone Marrow

Transplant, 2007. 39(6): p. 353-357.

50. Allice, T., et al., Evaluation of a novel real-time PCR system for cytomegalovirus DNA quantitation on whole blood and correlation with pp65-antigen test in guiding pre-emptive anti-viral treatment. J Virol Methods, 2008. 148(1-2): p. 9-16.

51. Garrigue, I., et al., Whole blood real-time quantitative PCR for cytomegalovirus infection follow-up in transplant recipients. J

Clin Virol, 2006. 36(1): p. 72-75.

52. Sagedal, S., et al., Impact of early cytomegalovirus infection and disease on long-term recipient and kidney graft survival. Kidney Int, 2004. 66(1): p. 329-337.

53. Sagedal, S., H. Rollag, and A. Hartmann, Cytomegalovirus infection in renal transplant recipients is associated with im-paired survival irrespective of expected mortality risk. Clin

Transplant, 2007. 21(3): p. 309-313.

54. Barbi, M., et al., Cytomegalovirus DNA detection in Guthrie cards: a powerful tool for diagnosing congenital infection. J

Clin Virol, 2000. 17(3): p. 159-165.

55. Lehto, J.T., et al., A prospective study comparing cytomega-lovirus antigenemia, DNAemia and RNAemia tests in guiding pre-emptive therapy in thoracic organ transplant recipients. Transpl Int, 2005. 18(12): p. 1318-1327.

56. Roback, J.D., et al., CMV DNA is rarely detected in healthy blood donors using validated PCR assays. Transfusion, 2003.

43(3): p. 314-321.

57. Ziemann, M., et al., High prevalence of cytomegalovirus DNA in plasma samples of blood donors in connection with sero-conversion. Transfusion, 2007. 47(11): p. 1972-1983.

58. Kenneson, A. and M.J. Cannon, Review and meta-analysis of the epidemiology of congenital cytomegalovirus (CMV) infec-tion. Rev Med Virol, 2007. 17(4): p. 253-276.

59. Ahlfors, K., S.A. Ivarsson, and S. Harris, Report on a long-term study of maternal and congenital cytomegalovirus infec-tion in Sweden. Review of prospective studies available in the literature. Scand J Infect Dis, 1999. 31(5): p. 443-457.

77

60. Boppana, S.B., et al., Symptomatic congenital cytomegalovirus infection: neonatal morbidity and mortality. Pediatr Infect Dis

J, 1992. 11(2): p. 93-99.

61. Paya CV and R. R, CMV infection after organ transplantation,

in Transplant Infections, Bowden RA, Ljungman P, and P.

CV, Editors. 2003, Lippincott, Williams and Wilkins. p. 298-

325.

62. Brennan, D.C., Cytomegalovirus in renal transplantation. J

Am Soc Nephrol, 2001. 12(4): p. 848-855.

63. Sagedal, S., et al., A prospective study of the natural course of cytomegalovirus infection and disease in renal allograft re-cipients. Transplantation, 2000. 70(8): p. 1166-1174.

64. Rubin, R.H., Cytomegalovirus in solid organ transplantation. Transpl Infect Dis, 2001. 3 Suppl 2: p. 1-5.

65. Legendre, C. and M. Pascual, Improving outcomes for solid-organ transplant recipients at risk from cytomegalovirus infec-tion: late-onset disease and indirect consequences. Clin Infect

Dis, 2008. 46(5): p. 732-740.

66. Paya, C.V., Indirect effects of CMV in the solid organ trans-plant patient. Transpl Infect Dis, 1999. 1 Suppl 1: p. 8-12.

67. Cobbs, C.S., et al., Human cytomegalovirus infection and ex-pression in human malignant glioma. Cancer Res, 2002.

62(12): p. 3347-3350.

68. Fishman, J.A., et al., Cytomegalovirus in transplantation - challenging the status quo. Clin Transplant, 2007. 21(2): p.

149-158.

69. Samanta, M., et al., High prevalence of human cytomegalovi-rus in prostatic intraepithelial neoplasia and prostatic carci-noma. J Urol, 2003. 170(3): p. 998-991002.

70. Koch, S., et al., Human cytomegalovirus infection and T cell immunosenescence: a mini review. Mech Ageing Dev, 2006.

127(6): p. 538-543.

71. Ouyang, Q., et al., Dysfunctional CMV-specific CD8(+) T cells accumulate in the elderly. Exp Gerontol, 2004. 39(4): p. 607-

613.

72. Phillips, A.C., et al., Cytomegalovirus is associated with de-pression and anxiety in older adults. Brain Behav Immun,

2008. 22(1): p. 52-55.

73. Winston, D.J., W.G. Ho, and R.E. Champlin, Cytomegalovirus infections after allogeneic bone marrow transplantation. Rev

Infect Dis, 1990. 12 Suppl 7: p. 776-792.

78

74. Doyle, A.M., et al., 24-week oral ganciclovir prophylaxis in kidney recipients is associated with reduced symptomatic cy-tomegalovirus disease compared to a 12-week course. Trans-

plantation, 2006. 81(8): p. 1106-1111.

75. Opelz, G., B. Dohler, and A. Ruhenstroth, Cytomegalovirus prophylaxis and graft outcome in solid organ transplantation: a collaborative transplant study report. Am J Transplant,

2004. 4(6): p. 928-936.

76. Snydman, D.R., Counterpoint: prevention of cytomegalovirus (CMV) infection and CMV disease in recipients of solid organ transplants: the case for prophylaxis. Clin Infect Dis, 2005.

40(5): p. 709-712.

77. Sagedal, S., A. Hartmann, and H. Rollag, The impact of early cytomegalovirus infection and disease in renal transplant re-cipients. Clin Microbiol Infect, 2005. 11(7): p. 518-530.

78. Hodson, E.M., et al., Antiviral medications for preventing cy-tomegalovirus disease in solid organ transplant recipients. Cochrane Database Syst Rev, 2005(4).

79. Hodson, E.M., et al., Antiviral medications to prevent cy-tomegalovirus disease and early death in recipients of solid-organ transplants: a systematic review of randomised con-trolled trials. Lancet, 2005. 365(9477): p. 2105-2115.

80. Reischig, T., et al., Valacyclovir prophylaxis versus preemp-tive valganciclovir therapy to prevent cytomegalovirus disease after renal transplantation. Am J Transplant, 2008. 8(1): p.

69-77.

81. Mauskopf, J.A., et al., Cost-effectiveness model of cytomega-lovirus management strategies in renal transplantation. Com-paring valaciclovir prophylaxis with current practice. Phar-

macoeconomics, 2000. 18(3): p. 239-251.

82. Peggs, K.S. and S. Mackinnon, Cytomegalovirus: the role of CMV post-haematopoietic stem cell transplantation. Int J Bio-

chem Cell Biol, 2004. 36(4): p. 695-701.

83. Prentice, H.G. and P. Kho, Clinical strategies for the manage-ment of cytomegalovirus infection and disease in allogeneic bone marrow transplant. Bone Marrow Transplant, 1997.

19(2): p. 135-142.

84. Balfour, H.H., et al., A randomized, placebo-controlled trial of oral acyclovir for the prevention of cytomegalovirus disease in recipients of renal allografts. N Engl J Med, 1989. 320(21): p.

1381-1387.

79

85. Ioannidis, J.P., et al., Clinical efficacy of high-dose acyclovir in patients with human immunodeficiency virus infection: a meta-analysis of randomized individual patient data. J Infect

Dis, 1998. 178(2): p. 349-359.

86. Lowance, D., et al., Valacyclovir for the prevention of cy-tomegalovirus disease after renal transplantation. Interna-tional Valacyclovir Cytomegalovirus Prophylaxis Transplanta-tion Study Group. N Engl J Med, 1999. 340(19): p. 1462-1470.

87. Sund, F., J. Wahlberg, and B.M. Eriksson, CMV disease in CMV-mismatched renal transplant recipients with prophylac-tic low dose valaciclovir. J Clin Virol, 2001. 23(1-2): p. 107-

111.

88. Ljungman, P., et al., Randomized study of valacyclovir as pro-phylaxis against cytomegalovirus reactivation in recipients of allogeneic bone marrow transplants. Blood, 2002. 99(8): p.

3050-3056.

89. Das, A., Cytomegalovirus Infection in Solid Organ Transplan-tation: Economic Implications. PharmacoEconomics, 2003.

21(7): p. 467-475.

90. Winston, D.J., et al., Randomised comparison of ganciclovir and high-dose acyclovir for long-term cytomegalovirus pro-phylaxis in liver-transplant recipients. Lancet, 1995.

346(8967): p. 69-74.

91. Flechner, S.M., et al., A randomized prospective controlled trial of oral acyclovir versus oral ganciclovir for cytomega-lovirus prophylaxis in high-risk kidney transplant recipients. Transplantation, 1998. 66(12): p. 1682-1688.

92. Paya, C., et al., Efficacy and Safety of Valganciclovir vs. Oral Ganciclovir for Prevention of Cytomegalovirus Disease in Solid Organ Transplant Recipients. American Journal of

Transplantation, 2004. 4(4): p. 611-620.

93. Oppenheimer, F., M. Gonzalez-Molina, and M. Rubio, Cost of prophylaxis in the management of cytomegalovirus infection in solid organ transplant recipients. Clin Transplant, 2007. 21(4):

p. 441-448.

94. Snydman, D.R., et al., Use of cytomegalovirus immune globu-lin to prevent cytomegalovirus disease in renal-transplant re-cipients. N Engl J Med, 1987. 317(17): p. 1049-1054.

95. Snydman, D.R., et al., Cytomegalovirus immune globulin pro-phylaxis in liver transplantation. A randomized, double-blind,

80

placebo-controlled trial. Ann Intern Med, 1993. 119(10): p.

984-991.

96. Emery, V.C., Prophylaxis for CMV should not now replace pre-emptive therapy in solid organ transplantation. Rev Med

Virol, 2001. 11(2): p. 83-86.

97. Preiksaitis, J.K., et al., Canadian society of transplantation consensus workshop on cytomegalovirus management in solid organ transplantation final report. Am J Transplant, 2005.

5(2): p. 218-227.

98. Asberg, A., et al., Oral valganciclovir is noninferior to intra-venous ganciclovir for the treatment of cytomegalovirus dis-ease in solid organ transplant recipients. Am J Transplant,

2007. 7(9): p. 2106-2113.

99. Kimberlin, D.W., et al., Effect of ganciclovir therapy on hear-ing in symptomatic congenital cytomegalovirus disease involv-ing the central nervous system: a randomized, controlled trial. J Pediatr, 2003. 143(1): p. 16-25.

100. John, G.T., et al., A Prospective Evaluation of Leflunomide Therapy for Cytomegalovirus Disease in Renal Transplant Re-cipients. Transplantation Proceedings, 2005. 37(10): p. 4303-

4305.

101. Battiwalla, M., et al., Leflunomide failure to control recurrent cytomegalovirus infection in the setting of renal failure after allogeneic stem cell transplantation. Transpl Infect Dis, 2007.

9(1): p. 28-32.

102. Einsele, H., et al., Infusion of cytomegalovirus (CMV)-specific T cells for the treatment of CMV infection not responding to antiviral chemotherapy. Blood, 2002. 99(11): p. 3916-3922.

103. Cobbold, M., et al., Adoptive transfer of cytomegalovirus-specific CTL to stem cell transplant patients after selection by HLA-peptide tetramers. J Exp Med, 2005. 202(3): p. 379-386.

104. Erice A, C.S., Biron KK, Stanat SC, Balfour HH Jr, Jordan

MC, Progressive disease due to ganciclovir-resistant cy-tomegalovirus in immunocompromised patients. N Engl J

Med, 1989(320(5)): p. 289-93.

105. Emery, V.C. and P.D. Griffiths, Prediction of cytomegalovirus load and resistance patterns after antiviral chemotherapy. Proc Natl Acad Sci U S A, 2000. 97(14): p. 8039-8044.

106. Drew, W.L., Ganciclovir resistance: a matter of time and titre. Lancet, 2000. 356(9230): p. 609-610.

81

107. Drew, W.L., C.V. Paya, and V. Emery, Cytomegalovirus (CMV) resistance to antivirals. Am J Transplant, 2001. 1(4): p.

307-312.

108. Limaye, A.P., Ganciclovir-Resistant Cytomegalovirus in Or-gan Transplant Recipients. Clinical Infectious Diseases, 2002.

35(7): p. 866-872.

109. Chou, S.W., Cytomegalovirus drug resistance and clinical implications. Transpl Infect Dis, 2001. 3 Suppl 2: p. 20-24.

110. Chou, S., et al., Cytomegalovirus UL97 phosphotransferase mutations that affect susceptibility to ganciclovir. J Infect Dis,

2002. 185(2): p. 162-169.

111. Erice, A., Resistance of human cytomegalovirus to antiviral drugs. Clin Microbiol Rev, 1999. 12(2): p. 286-297.

112. Jabs, D.A., et al., Cytomegalovirus retinitis and viral resis-tance: ganciclovir resistance. CMV Retinitis and Viral Resis-tance Study Group. J Infect Dis, 1998. 177(3): p. 770-773.

113. Drew, W.L., et al., Prevalence of resistance in patients receiv-ing ganciclovir for serious cytomegalovirus infection. J Infect

Dis, 1991. 163(4): p. 716-719.

114. Eid, A.J., et al., Emergence of drug-resistant cytomegalovirus in the era of valganciclovir prophylaxis: therapeutic implica-tions and outcomes. Clin Transplant, 2008. 22(2): p. 162-170.

115. Isada, C.M., et al., Clinical characteristics of 13 solid organ transplant recipients with ganciclovir-resistant cytomegalovi-rus infection. Transpl Infect Dis, 2002. 4(4): p. 189-194.

116. Marshall, G.S., et al., Antibodies to recombinant-derived gly-coprotein B after natural human cytomegalovirus infection correlate with neutralizing activity. J Infect Dis, 1992. 165(2):

p. 381-384.

117. Urban, M., et al., Glycoprotein H of human cytomegalovirus is a major antigen for the neutralizing humoral immune re-sponse. J Gen Virol, 1996. 77(7): p. 1537-1547.

118. Shimamura, M., M. Mach, and W.J. Britt, Human cytomega-lovirus infection elicits a glycoprotein M (gM)/gN-specific vi-rus-neutralizing antibody response. J Virol, 2006. 80(9): p.

4591-4600.

119. Wang, D. and T. Shenk, Human cytomegalovirus virion pro-tein complex required for epithelial and endothelial cell tro-pism. Proc Natl Acad Sci U S A, 2005. 102(50): p. 18153-

18158.

82

120. Hahn, G., et al., Human cytomegalovirus UL131-128 genes are indispensable for virus growth in endothelial cells and vi-rus transfer to leukocytes. J Virol, 2004. 78(18): p. 10023-

10033.

121. Gerna, G., et al., Human cytomegalovirus serum neutralizing antibodies block virus infection of endothelial/epithelial cells, but not fibroblasts, early during primary infection. J Gen Vi-

rol, 2008. 89(4): p. 853-865.

122. Cui, X., et al., Cytomegalovirus vaccines fail to induce epithe-lial entry neutralizing antibodies comparable to natural infec-tion. Vaccine. In Press, Corrected Proof.

123. Adler, S.P., Human CMV vaccine trials: what if CMV caused a rash? J Clin Virol, 2008. 41(3): p. 231-236.

124. OPTN, www.optn.org. 2008:8.

125. Solez, K., et al., International standardization of criteria for the histologic diagnosis of renal allograft rejection: the Banff working classification of kidney transplant pathology. Kidney

Int, 1993. 44(2): p. 411-422.

126. Racusen, L.C., P.F. Halloran, and K. Solez, Banff 2003 meet-ing report: new diagnostic insights and standards. Am J

Transplant, 2004. 4(10): p. 1562-1566.

127. Solez, K., et al., Banff '05 Meeting Report: differential diagno-sis of chronic allograft injury and elimination of chronic al-lograft nephropathy ('CAN'). Am J Transplant, 2007. 7(3): p.

518-526.

128. Cattaneo, D., N. Perico, and G. Remuzzi, From pharmacoki-netics to pharmacogenomics: a new approach to tailor immu-nosuppressive therapy. Am J Transplant, 2004. 4(3): p. 299-

310.

129. MacPhee, I.A., S. Fredericks, and D.W. Holt, Does pharmaco-genetics have the potential to allow the individualisation of immunosuppressive drug dosing in organ transplantation?

Expert Opinion on Pharmacotherapy, 2005. 6(15): p. 2593-

2605.

130. Knight, S.R. and P.J. Morris, The clinical benefits of cyc-losporine C2-level monitoring: a systematic review. Trans-

plantation, 2007. 83(12): p. 1525-1535.

131. Bowman, L.J. and D.C. Brennan, The role of tacrolimus in renal transplantation. Expert Opinion on Pharmacotherapy,

2008. 9(4): p. 635-643.

83

132. Heisel, O., et al., New Onset Diabetes Mellitus in Patients Re-ceiving Calcineurin Inhibitors: A Systematic Review and Meta-Analysis. American Journal of Transplantation, 2004.

4(4): p. 583-595.

133. Binet I, et al., Polyomavirus disease under new immunosup-pressive drugs: a cause of renal graft dysfunction and graft loss. Transplantation, 1999. Mar 27(67(6)): p. 918-22.

134. Song, A.T.W., et al., Does mycophenolate mofetil increase the risk of cytomegalovirus infection in solid organ transplant re-cipients?--A mini-review. Braz J Infect Dis, 2006. 10(2): p.

132-138.

135. Knight, R.J. and B.D. Kahan, The place of sirolimus in kidney transplantation: can we reduce calcineurin inhibitor renal tox-icity? Kidney Int, 2006. 70(6): p. 994-999.

136. Johnston, O., et al., Sirolimus Is Associated with New-Onset Diabetes in Kidney Transplant Recipients. J Am Soc Nephrol,

2008.

137. Guba, M., et al., Rapamycin inhibits primary and metastatic tumor growth by antiangiogenesis: involvement of vascular endothelial growth factor. Nat Med, 2002. 8(2): p. 128-135.

138. Morice, M.-C., et al., A randomized comparison of a si-rolimus-eluting stent with a standard stent for coronary revas-cularization. N Engl J Med, 2002. 346(23): p. 1773-1780.

139. Law, B.K., Rapamycin: An anti-cancer immunosuppressant?

Critical Reviews in Oncology/Hematology Immunosuppres-

sive Treatment and Induction of Cancer, 2005. 56(1): p. 47-60.

140. Malek, S.K., et al., Campath-1H induction and the incidence of infectious complications in adult renal transplantation. Transplantation, 2006. 81(1): p. 17-20.

141. Thomas, P.G., et al., Alemtuzumab (Campath 1H) induction with tacrolimus monotherapy is safe for high immunological risk renal transplantation. Transplantation, 2007. 83(11): p.

1509-1512.

142. Sidner, R.A., et al., In vivo human B-cell subset recovery after in vivo depletion with rituximab, anti-human CD20 mono-clonal antibody. Hum Antibodies, 2004. 13(3): p. 55-62.

143. Tyden, G., et al., ABO incompatible kidney transplantations without splenectomy, using antigen-specific immunoadsorption and rituximab. Am J Transplant, 2005. 5(1): p. 145-148.

84

144. Alausa, M., et al., Refractory acute kidney transplant rejection with CD20 graft infiltrates and successful therapy with rituxi-mab. Clin Transplant, 2005. 19(1): p. 137-140.

145. Becker, Y.T., et al., Rituximab as treatment for refractory kid-ney transplant rejection. Am J Transplant, 2004. 4(6): p. 996-

991001.

146. Mourad, G., et al., Sequential protocols using basiliximab ver-sus antithymocyte globulins in renal-transplant patients re-ceiving mycophenolate mofetil and steroids. Transplantation,

2004. 78(4): p. 584-590.

147. Gallon, L.G., et al., Effect of prednisone versus no prednisone as part of maintenance immunosuppression on long-term renal transplant function. Clin J Am Soc Nephrol, 2006. 1(5): p.

1029-1038.

148. Jaber, J.J., et al., Early steroid withdrawal therapy in renal transplant recipients: a steroid-free sirolimus and CellCept-based calcineurin inhibitor-minimization protocol. Clin Trans-

plant, 2007. 21(1): p. 101-109.

149. Mousavi-Jazi, M., et al., Point mutations induced by foscarnet (PFA) in the human cytomegalovirus DNA polymerase. J Clin

Virol, 2003. 26(3): p. 301-306.

150. Herrmann, B., et al., Comparison of a duplex quantitative real-time PCR assay and the COBAS Amplicor CMV Monitor test for detection of cytomegalovirus. J Clin Microbiol, 2004.

42(5): p. 1909-1914.

151. Nomura, L.E., J.M. Walker, and H.T. Maecker, Optimization of whole blood antigen-specific cytokine assays for CD4(+) T cells. Cytometry, 2000. 40(1): p. 60-68.

152. Kern, F., et al., Analysis of CD8 T cell reactivity to cytomega-lovirus using protein-spanning pools of overlapping pentade-capeptides. Eur J Immunol, 2000. 30(6): p. 1676-1682.

153. Gratama JW, Keeney M, and S. DR, Enumeration of CD34+ hematopoetic stem and progenitor cells, in Current Protocols in Cytometry, Robinson JP, Darzynkiewicz Z, and D. PN, Edi-

tors. 1999, John Wiley & Sons: New York. p. 6.4.1-6.4.22.

154. Eriksson, B.M., et al., A prospective study of rapid methods of detecting cytomegalovirus in the blood of renal transplant re-cipients in relation to patient and graft survival. Clin Trans-

plant, 1996. 10(6 Pt 1): p. 494-502.

85

155. Callan, M.F., et al., Direct visualization of antigen-specific CD8+ T cells during the primary immune response to Epstein-Barr virus In vivo. J Exp Med, 1998. 187(9): p. 1395-1402.

156. Sester, M., et al., Dominance of virus-specific CD8 T cells in human primary cytomegalovirus infection. J Am Soc Nephrol,

2002. 13(10): p. 2577-2584.

157. Balfour, H.H., Cytomegalovirus: the troll of transplantation. Arch Intern Med, 1979. 139(3): p. 279-280.

158. Paya, C.V., et al., Preemptive use of oral ganciclovir to pre-vent cytomegalovirus infection in liver transplant patients: a randomized, placebo-controlled trial. J Infect Dis, 2002.

185(7): p. 854-860.

159. Khoury, J.A., et al., Prophylactic versus preemptive oral val-ganciclovir for the management of cytomegalovirus infection in adult renal transplant recipients. Am J Transplant, 2006.

6(9): p. 2134-2143.

160. Gregory D. Hart, C.V.P., Prophylaxis for CMV should now replace pre-emptive therapy in solid organ transplantation. Reviews in Medical Virology, 2001. 11(2): p. 73-81.

161. Reischig, T., et al., Prospective comparison of valacyclovir and oral ganciclovir for prevention of cytomegalovirus disease in high-risk renal transplant recipients. Kidney Blood Press

Res, 2005. 28(4): p. 218-225.

162. I. D. Pavlopoulou, V.P.S., H. Chelioti, G. L. Daikos, D. Sta-

matiades, A. Kostakis, J. N. Boletis,, A comparative random-ised study of valacyclovir vs. oral ganciclovir for cytomega-lovirus prophylaxis in renal transplant recipients. Clinical Mi-

crobiology & Infection, 2005. 11(9): p. 736-743.

163. Hellden, A., et al., High serum concentrations of the acyclovir main metabolite 9-carboxymethoxymethylguanine in renal failure patients with acyclovir-related neuropsychiatric side effects: an observational study. Nephrol. Dial. Transplant.,

2003. 18(6): p. 1135-1141.

164. Reddy, S.P., et al., Low-dose valaciclovir prophylaxis against cytomegalovirus disease in renal transplant recipients. Transpl

Int, 2003. 16(10): p. 726-729.

165. Arthurs, S.K., et al., Delayed-onset primary cytomegalovirus disease and the risk of allograft failure and mortality after kidney transplantation. Clin Infect Dis, 2008. 46(6): p. 840-

846.

86

166. Singh, N., Antiviral drugs for cytomegalovirus in transplant recipients: advantages of preemptive therapy. Rev Med Virol,

2006. 16(5): p. 281-287.

167. Humar, A., et al., Association between cytomegalovirus dis-ease and chronic rejection in kidney transplant recipients. Transplantation, 1999. 68(12): p. 1879-1883.

168. Helantera, I., et al., Persistent cytomegalovirus infection in kidney allografts is associated with inferior graft function and survival. Transpl Int, 2006. 19(11): p. 893-900.

169. Collaborative Transplant Study, C., http://cts.med.uni-

heidelberg.de/ CTS E-11122-0808. 2008:8.

170. Boivin, G., et al., Absence of Cytomegalovirus-Resistance Mu-tations after Valganciclovir Prophylaxis, in a Prospective Mul-ticenter Study of Solid-Organ Transplant Recipients. The

Journal of Infectious Diseases, 2004. 189(9): p. 1615-1618.

171. Alain, S., et al., Detection of ganciclovir resistance after vala-cyclovir-prophylaxis in renal transplant recipients with active cytomegalovirus infection. J Med Virol, 2004. 73(4): p. 566-

573.

172. Gratama, J.W., et al., Tetramer-based quantification of cy-tomegalovirus (CMV)-specific CD8+ T lymphocytes in T-cell-depleted stem cell grafts and after transplantation may identify patients at risk for progressive CMV infection. Blood, 2001.

98(5): p. 1358-1364.

173. Ouyang, Q., et al., Age-associated accumulation of CMV-specific CD8+ T cells expressing the inhibitory killer cell lec-tin-like receptor G1 (KLRG1). Exp Gerontol, 2003. 38(8): p.

911-920.

174. Karrer, U., et al., Memory inflation: continuous accumulation of antiviral CD8+ T cells over time. J Immunol, 2003. 170(4):

p. 2022-2029.

175. Aubert, G., et al., Cytomegalovirus-specific cellular immune responses and viremia in recipients of allogeneic stem cell transplants. J Infect Dis, 2001. 184(8): p. 955-963.

176. Engstrand, M., et al., Characterization of CMVpp65-specific CD8+ T lymphocytes using MHC tetramers in kidney trans-plant patients and healthy participants. Transplantation, 2000.

69(11): p. 2243-2250.

177. Singhal, S., et al., Direct visualization and quantitation of cy-tomegalovirus-specific CD8+ cytotoxic T-lymphocytes in liver

87

transplant patients. Transplantation, 2000. 69(11): p. 2251-

2259.

178. Radha, R., et al., Cellular immune responses to cytomegalovi-rus in renal transplant recipients. Am J Transplant, 2005. 5(1):

p. 110-117.

179. Bunde, T., et al., Protection from cytomegalovirus after trans-plantation is correlated with immediate early 1-specific CD8 T cells. J Exp Med, 2005. 201(7): p. 1031-1036.

180. Sester, M., et al., Levels of virus-specific CD4 T cells correlate with cytomegalovirus control and predict virus-induced dis-ease after renal transplantation. Transplantation, 2001. 71(9):

p. 1287-1294.

181. Gamadia, L.E., et al., Primary immune responses to human CMV: a critical role for IFN-gamma -producing CD4+ T cells in protection against CMV disease. Blood, 2003. 101(7): p.

2686-2692.

182. Gerna, G., et al., Monitoring of human cytomegalovirus-specific CD4 and CD8 T-cell immunity in patients receiving solid organ transplantation. Am J Transplant, 2006. 6(10): p.

2356-2364.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 386

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-9324

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2008

cytomegalovirus infection in immunocompetent and ...172666/fulltext01.pdf · cytomegalovirus...

Documents