1. No category

PDF hosted at the Radboud Repository of the Radboud University
Nijmegen
This full text is a publisher's version.
For additional information about this publication click this link.
http://hdl.handle.net/2066/77585
Please be advised that this information was generated on 2014-10-21 and may be subject to
change.
Alpha 1 Antitrypsin Deficiency:
Focus on Liver Disease in
Young and Old
Karin Kok
Acknowledgem ent:
The research presented in this thesis was supported byTalecris Biotherapeutics
GmbH (Frankfurt am Main, Germany).
The cover painting "Veldweg" is made byTon Schulten, a digital copy of the
painting was kindly provided by "Chez-MoiTon Schulten Galerie International",
Marktstraat 4-6, Ootmarsum.
Printing of this thesis was financially supported by: ABBOTT Immunology,
Boston Scientific, Janssen-Cilag, Gilead Sciences, Novartis, Olympus, ScheringPlough, Talecris GmbH, Tramedico and Zambon.
ISBN: 978-90-9025317-6
Design and layout: J Ontwerp | janneke_tromp@ yahoo.com
Printed by: Ipskamp drukkers
Alpha i Antitrypsin Deficiency:
Focus on Liver Disease in
Young and Old
Een wetenschappelijke proeve op het gebied
van de medische wetenschappen
Proefschrift
Ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen
op gezag van de rector magnificus prof. mr. S.C.J.J. Kortmann, volgens het
besluit van het college van decanen in het openbaarte verdedigen op
3 juni 2010
om 13.00 precies
Door
Karin Francisca Kok
Geboren op 11 april 1976
te Geesteren (ov)
Promotor:
Prof. dr. JPH Drenth
Co-promoteres:
Dr. MGH van Oijen
Dr. RA de Vries (UMCG)
Manuscript commissie: Prof. dr. PMJ Stuyt (voorzitter)
Prof. dr. JL Willems
Dr. RH Houwen (UMCU)
Contents
Chapter i:
Chapter 2:
Introduction and outline of the thesis
Heterozygous A lph a-i Antitrypsin Deficiency as a Co-factor in the Development
7
17
of Chronic Liver Disease: a Review. Kok KF, Wahab PJ7 Houwen RH, Drenth JP,
de Man RA, van Hoek B, Meijer JW, Willekens FL, deVries RA. Neth J Med. 2007
M ay;65(5):i6o-6.
Chapter 3:
Prognosis of Neonatal Cholestasis due to A lp h a-i Antitrypsin Deficiency. Kok KF,
35
van Vlerken LG, Benninga MA, EscherJC, deVries RA, Houwen RHJ.
Chapter 4:
Vitamin K Deficiency Bleeding in Cholestatic Infants with A lp h a-i Antitrypsin
45
Deficiency, van Hasselt PM, Kok KF, Vorselaars AD, van Vlerken L, Nieuwenhuys E,
de KoningTJ, deVries RA, Houwen RH .Arch DisChild Fetal Neonatal Ed. 2009
N ov;94(6):F456-6 o .
Chapter 5:
Influence of A lph a-i Antitrypsin Heterozygosity on Treatment Efficacy of HCV
59
Combination Therapy. Kok KF, van Soest H, van Herwaarden AE, van Oijen MG,
Boland GJ, Halangk J, BergT, deVries RA, Drenth J P. Eur J Gastroenterol Hepatol.
2009 Sep 29. [Epub ahead of print]
Chapter 6:
Prevalence of Genetic Polymorphisms in the Promoter Region of the A lph a-i
73
Antitrypsin (SERPIN Ai) Gene in Chronic Liver Disease: a Case Control Study.
Kok KF, te Morsche RH, van Oijen MGH and Drenth JPH. BMC Gastroenterol 2010,
10:22.
Chapter 7:
The cut-offValue of 100 mg/dL is Insufficient to Detect Heterozygous A lph a-i
87
Antitrypsin deficient Liver Disease patients. Kok KF, Willems JL and Drenth JPH.
Liver Int. 2009 Nov 30; [Epub ahead of print]
Chapter 8:
Liver transplantation for A lpha-i antitrypsin deficiency and the impact of Incidental
95
Hepatocellular Carcinoma on Post-Transplant Survial. Kok KF, Porte RJ,Adam R, van
den Berg A, Muiesan P, Scheenstra R, Metselaar H, van Hoek B, van Oijen MGH ,
Drenth JPH Forthe European Liver and IntestineTransplant Association (ELITA)
Chapter 9:
General discussion
111
Summary and Conclusion
117
Samenvatting en Conclusie
121
List of publications
125
Dankwoord
127
Curriculum Vitae
131
Chapter 1
Introduction and outline of the thesis
Chapter 1
Past-to-present
In 1963, Laurell and Eriksson, presented in the "Scandinavian Journal of Clinical
and Laboratory Investigation" 5 patients with a new type of dysproteinemia,
characterized by very pronounced alp h a-i antitrypsin (A iA T) deficiency. This
deficiency was detected by paper electrophoresis. (Figure 1.1) A total of 3 out of
these 5 patients had pulmonary disease, 1 patient had a malabsorption syndrome
and the last patient had no obvious disease.1
Figure 1 . 1 : D iagnostic tools in 1 9 6 3 .
Paper e lectro phoretic protein p a tte rn o f sera
from tw o patients w ith a lp h a -i antitry p s in
deficiency and from a control. (Upper and lower:
cases; M iddle: c o n tro l )1
A gar gel electrophoresis o f serum from a p a tie n t
and from a norm al control. (U ppercase and lower:
control )1
In 1967 and 1970, they continued to establish an association between chronic
obstructive lung disease and A iA T deficiency.2-3 In 1972 these authors
published a key paper that described the presence of liver disease in adults
with A iA T deficiency4and in ig 74 th e y reported a high incidence of hepatocel­
lular carcinoma in patients with A iA T deficiency related cirrhosis.5 At that
time, others showed A iA T polymers in 52% of hepatocellular carcinom a6 and
furthermore, A iA T deficiency was considered as a cause of neonatal hepatitis.7
The first report on A iA T deficiency originating from the Netherlands was pub­
lished in 1980 and contained interesting epidemiologic data that demonstrated
that the frequency of A iA T deficiency was 8 out of 10,000 newborns.8 In 1983
the point mutation which is involved in A iA T deficiency was detected.9
Over the years, many reports concerning A iA T deficiency and its relation
introduction and outline
with prognosis, associations with environmental factors, (patho)physiologic
mechanisms and gene association studies have been published. In December
2009, a "Google-search" for A iA T deficiency resulted in no less than 1.3 million
hits. Nowadays, the diagnosis A iA T deficiency is not made by a paper patchy
spot as shown above, but current diagnostic tools are much more sophisticated.
(Figure 1.2 and 1.3)
A: BioRad ¡Cycler
B: Melting curve
C: Allelic discrimination figure
Figure 1 .3 : Real-Tim e PCR for
m u ta tio n d etection.
RRI for Allele 1
FAM-490
9
Chapter 1
The gene
A iA T deficiency is a genetic disorder and the AiATprotein is encoded by the
protease inhibitor (Pi) locus located on chromosome i4 q 3 2 .i (S E R P IN A igene). The A iA T protein is a member of the "SERPIN" family, similar to
alphai- antichymotrypsin, alpha2-antiplasmin, antithrombin and Ci-inhibitor.
The gene is 12.2 kb in length with 7 exons (4 coding and 3 non-coding) and
6 introns. (Figure 1.4) The protein encoded includes 394 amino acids. The
protease inhibitor locus is highly polymorphic with approxim ately 123 single
nucleotide polymorphisms (SNPs), resulting in numerous different A iA T
isotypes that can be detected by iso-electrophoresis. The wild type allele
results in a normal M-type protein. The most common mutations are E264V
and E342/C, resulting in the S-type protein (exon 3) and Z-type protein (exon 5)
respectively. Beside the SNPs in exons, other SNPs in the enhancer sequence
of the A iA T gene, involving the regulation of A iA T expression have been
described.10-11
Figure 1 . 4 : schem atic overview o f th e S E R P IN A i gene.
E264V
E342K
I
Exort 1A Exon IB 1C
Exon 2
Exon 3
[
E xo n 4
— M i -------------B i l l
12
34
5
6
Exo ns
7
8
3'«rh a n c« r
9
--------- H-----------1------------------------------- 1-----------1-------------- 1------¥
The protein
A iA T is mainly produced in the liver and reaches the lung by diffusion from
the circulation or by pulmonary production in macrophages and bronchial
epithelial cells.12The A iA T protein is a serum glycoprotein acting as an acute
phase protein. The major physiological function of A iA T is inhibition of
proteolytic enzymes as elastase and other proteinases which are secreted by
neutrophilic granulocytes during inflammatory processes. The normal A iA T
protein has a tertiary structure based on a large central il-sheet, surrounded
introduction and outline
by two other sheets and a reactive centre loop. The reactive centre loop binds
to proteinases and can move in and out the il-sheet. In case of a Z-type protein
(thus E342K mutation) the reactive centre loop is reacting with the il-sheet of
a second A iA T molecule. As a consequence, this leads to polymers that are
retained in the endoplasmic reticulum of hepatocytes.n '13 (Figure 1.5)
Figure 1 . 5 : Loop sheet p o ly m e riza tio n due to Z m u ta tio n in A iA T deficiency.
A: destabilization of the IS-sheet due to Z-type AIAT. B: insertion of the reactive loop of
another AIAT molecule leads to dimer formation and C: subsequently formation of
polymers. (From: Ned Tijdschr Geneeskd 2003;147(16):758-60, with permission)
Mechanism of liver disease
Several studies have provided evidence that the SNP, which is responsible
for the Z-type A iA T reduces the stability of the molecule in its monomeric
form and increases the likelihood that it will form polymers by a so-called
"loop-sheet" insertion mechanism .14 Z-type A iA T polymers can be identified
by electronm icroscopy as periodic-acid-Schiff stain positive and diastaseresistant inclusions within the endoplasmic reticulum of hepatocytes. (Figure
1.6 panel B) The S-type A iA T has less effect on loop sheet polymerization
and accumulation. It is suggested that the process of protein degradation
involves modification of misfolded proteins by the enzyme endoplasmic
reticulum m annosidase I. Mannosidase inhibitors delay degradation and
increase secretion of Z-type AiAT. Expression of liver disease in patients with
C h a p te r 1
A iA T deficiency correlates with a delay in intracellular degradation of Z-type
A iA T and accumulation of polymers. There is remarkable variability in the
phenotypic expression of disease in individuals with Pi ZZ hom ozygosity.12
Apart from environmental factors such as smoking and alcohol abuse, genetic
modifiers appear to be partly responsible for the variability in clinical expression
of patients with A iA T deficiency. For example, SNPs in the promoter region
and a variant of the endoplasmic reticulum mannosidase I (ERManl) gene were
associated with an early onset of end-stage liver disease in patients with
homozygous (Pi ZZ) A iA T deficiency.15-16 Others suggested that individual
variations in episodes of inflammation and hence to increased synthesis of
A iA T can explain the variability in severity of liver disease and its age at onset.17
Mechanism of pulmonary disease
Polymerization and retention of polymers in the endoplasmic reticulum of
hepatocytes, causes a decrease in the amount of serum A iATthat is available
to protect the lungs against elastolytic dam age.17 An imbalance between the
A iA T protection and elastase with consequent unchecked proteolytic activity,
can lead to emphysema. Either a mild increase in elastase (which might happen
with smoking or lung infection) or a reduction in anti-elastase defence (which
may happen in severe A iA T deficiency) can affect the elastase-anti-elastase
balance unfavourably towards accelerated lung breakdown.12 (Figure 1.6, panel
A) The decreased A iA T level is not a single issue, the Z-type A iA T is about
5 time less effective as an inhibitor of elastase than normal M-type A iA T .18
In addition, Z-type A iA T polymers might be chemoattractants for neutrophils
thereby potentially increasing lung inflam m ation.19
Outline of the thesis
Despite progress in understanding the mechanisms of disease in A iA T
deficiency, several questions with regard to the clinical consequences of A iA T
deficiency are still open.
First, we are interested in the clinical course of A iA T deficiency. What is the
course of liver disease in A iA T deficiency with respect to the pediatric and
introduction and outline
adult population? Are there specific com plicationsthat need to be considered
when treating these patients? Second, we will focus on the diagnostic strategy
for establishing A iA T deficiency. Third, we wonder whether SNPs in the
SERPIN A i gene modify the development, course, and treatment results of liv­
er diseases due to various etiologies. Finally, we addressed questions
regarding the clinical relevance of liver disease due to heterozygous A iA T
deficiency.
This thesis aims to address these issues with a primary focus on the elements
that are important for the clinical management of these patients.
Figure 1.6: Mechanisms of pulmonary and liver disease.
Lung with predominantly
basifarpanacmar
emphysema in AAT deficiency
Alveoli with panacinar destruction
and enlargement
Lung parenchyma viewed
with light microscopy
------- 17
I
V-
Enlarged
alveolar
Liver affected by AAT
Polymerized AAT
molecules in a single granule
Panel A: m echanism for th e d e v e lo p m e n t o f em physem a in p atients w ith A iA T deficiency.
Panel B: overview o f liver disease in patients w ith A iA T deficiency, (from : NEJM 20 og; 3 6 o( 2 6 ):27 Z,.9 ,
w ith permission)
13
C h a p te r 1
References
1
Laurell CB and Eriksson S. The electrophoretic alph a-i globulin pattern of serum in alpha-i
deficiency. Scan J Clin Lab Invest. i9 6 3;(i5 )i3 2-4 0 .
2
Ganrot PO, Laurell CB and Eriksson S. Obstructive lung disease and trypsin inhibitors in A iA T
deficiency. Scan J Clin Lab Invest. i967;i9(3):205-8.
3
Eriksson S. Antitrypsin deficiency in chronic obstructive lung disease. Lancet i9 7o ;(i):8 9 i-2 .
4)
Berg NO and Eriksson S. Liver disease in adults with alpha-i-antitrypsin deficiency. N Eng J Med.
i972;287(25):i264-7.
5
Erisson S and Hägerstrandt I. Cirrhosis and malignant hepatoma in alpha i-antitrypsin deficiency.
Acta Med Scand. i974 ;i95(6):45i-8 .
6
Palmer PE and Wolfe HJ. a-Antitrypsin deposition in primary hepatic carcinomas. Arch Pathol
Lab Med. 1976;100:232-6.
7
Cottral K, Cook PJL and Mowat AP. Neonatal hepatitis syndrome and alpha-i-antitryspin defi­
ciency: an epidemiological study in south-east England. Postgrad Med J. i974;5o(584): 376-80.
8
Dijkman JH, PenderTJ, Kramps JA, Sonderkamp HJA, van den Broek WGM and ter Haar BGA.
Epidemiology of Alphai-antitrypsin deficiency in the Netherlands. Hum Genet.1980;53:409-13.
9
Kidd VJ, Wallace RB, Itakura K, Woo SLC. Alpha-i-antitrypsin deficiency detection by analysis
of mutation of the gene. Nature 1983;304:230-4.
10
DeMeo DL and Silverman EK. A i- Antitrypsin deficiency 2: genetics aspects of ai-antitrypsin
deficiency: phenotypes and genetic modifiers of emphysema risk. Thorax 2004;59:259-264.
11
Hiemstra PS and Stolk J. Van gen naar ziekte; ai-antitrypsine deficiëntie. NedTijdschr Geneeskd
2003; i 47( i 6): 758- 60.
12
Stoller JK and Aboussouan LS. Ai-antitrypsin deficiency. Lancet 2005;365:2225-36.
13
Lomas DA. Loop sheet polymerization: the mechanism of alphai-antitrypsin deficiency.
Respir Med. 2000;94(suppC):S3-S6
14
Teckman JH, Qu D and Perlmutter DH. Molecular pathogenesis of liver disease in ai-antitrypsin
deficiency. Hepatology I9 9 6 ;2 4(6 ):i504-i6 .
15
Chappell S, Hadzic N, Stockley R, Guetta-Baranes T, Morgan K, Kalsheker N. A polymorphism
of the alphai-antitrypsin gene represents a risk factor for liver disease. Hepatology 2008
Jan;47(i):i27-32.
16
Pan S, Huang L, McPherson J, Muzny D, Rouhani, F, Brantly M et al. Single Nucleotide Polymorphism-MediatedTranslational Suppression of Endoplasmatic Reticulum Mannosidase I Modifeisthe
Onset of End-Stage Liver Disease in Alphai-Antitrypsin Deficiency. Hepatology 2009;50:275-281.
17
Lomas DA, Evans DL, Finch JT, Carrell RW. The mechanism of Z ai-antitrypsin accumulation in
the liver. Nature 1992;357:605-7.
14
I n t r o d u c t io n a n d o u t lin e
18
Ogushi F, Fells GA, Hubbard RC, Straus SD, Crystal RG. Z-type alpha i-antitrypsin is less
competent than M i-type alpha i-antitrypsin as an inhibitor of neutrophil elastase. J Clin
Invest 1987;80:1366-74.
19
Mulgrew AT, Taggart CC, Lawless MW et al. Z alphai-antitrypsin polymerizes in the lung and
acts as a neutrophil chemoattractant. Chest 2004;125:1952-57.
Chapter 2
Heterozygous Alpha-i Antitrypsin
Deficiency as a co-factor in the
Development of Chronic Liver
Disease; a Review
Netherlands Journal of Medicine 2007 M ay;65(5):i6o-6
Karin F Kok
Peter J Wahab
Roderick HJ Houwen
Joost PH Drenth
Rob A de Man
Bart van Hoek
Jos WR Meijer
Frans LA Willekens
Richard A deVries
Abstract
A lp h a-i antitrypsin (A iA T) is an acute phase protein that
is produced in liver cells. A iA T deficiency is a hereditary
disease that is defined by the hepatic production of an
abnormal protein that cannot be released into the plasma.
This leads to deficiency of plasma A iA T and subsequently
to an impaired protection against proteases, resulting in
pulmonary disease. Accumulation of the abnormal protein in
hepatocytes can lead to liver damage. Serum level mea­
surement, phenotyping and liver biopsy can be used for
establishing the diagnosis.
Homozygous A iA T deficiency can cause neonatal hepatitis;
in adults end stage liver disease, cirrhosis and hepatocellular
carcinoma may develop. There are strong arguments to
consider heterozygous A iA T deficiency as an important
co-factor in the etiology of chronic liver disease. Studies
have shown that A iA T heterozygosity can be considered
a modifier for HCV, end stage live disease, cirrhosis and
hepatocellular carcinoma. The accumulation of A iA T in
the hepatocytes occurs more profoundly in a sick liver, and
as a consequence it affects the natural course of the liver
disease.Therapeutic options include augmentation therapy
(infusion of purified human plasma A iA T ) in pulmonary
disease; in end stage liver disease livertransplantation is an
option. In the future other interventions such as gene ther­
apy or strategies to inhibit polym erization are promising.
A 1 A T d e f ic ie n c y a n d c h r o n ic liv e r d is e a s e : r e v ie w
Background
A lp h a-i antitrypsin (A iA T) is an acute phase protein that is produced in liver
cells. It is released into the plasma in response to an inflammatory stimulus.
A iA T deficiency is a hereditary disease that is defined by the hepatic production
of an abnormal protein that can not completely be released into the plasma.
This leads to deficiency of plasma A iA T and subsequently to an impaired pro­
tection of the lungs against proteases. This results in pulmonary emphysema,
the hepatic accumulation of the abnormal protein can lead to chronic liver
disease.This review gives an update of the present knowledge of partial A iA T
deficiency in relation to various liver diseases.
Genetics and (patho) physiology
The A iA T molecule is a serum glycoprotein acting as an acute phase protein.
It is released during inflammatory processesfrom the hepatocyte which results
in increased plasma concentrations. The major physiological function of this
protein is the inhibition of destructive neutrophil elastase, thus protecting against
pulmonary dam age.1 The A iA T protein is encoded by the protease inhibitor
(PI) locus located on chromosome i4 q 3 2 .i.T h e PI locus is highly polymorphic,
resulting in different A iA T isotypes that can be detected by electrophoresis.
The most common allele is the M allele that results in a functionally normal
protein with normal serum A iA T levels. The normal A iA T protein has a tertiary
structure based on a large central il-sheet, surrounded by two other sheets and
a reactive centre loop.The reactive centre loop can move in and out of the large
il-sheet. At highertem peratures polymerization can occur between molecules
due to insertion of the loop of one molecule into the large il-sheet of the other.
The point mutation, found in the Z variant destabilizes the loop-sheet poly­
merization of the A iA T molecule, resulting in chains of polymers that are
retained in the hepatocytes. These polymers accumulate in the endoplasmic
reticulum of the hepatocytes and may be recognized as PAS(+) inclusion
bodies. (Figure 2.1) Only 15% of the Z-variant of A iA T can be secreted into
the plasma, the other 85 % accumulates in the liver. In Pi ZZ homozygous
subjects, this results in a severe deficiency of serum A iA T and in accumulation
of the abnormal protein in the endoplasmatic reticulum of the hepatocyt
which can lead to chronic liver disease.
19
Chapter 2
The S variant of the A iA T molecule, has less effect on the loop sheet poly­
merization. Formation of S-polymers is slower, resulting in less retention
of protein in the hepatocytes compared to the Z variant. There is a mild
reduction in serum A iA T levels. When S and Z variants are co-inherited, the
two interact with the formation of polymers within the hepatocytes; this
can lead to reduction in the serum A iA T level, inclusion of the polymers,
accumulation and subsequently development of cirrhosis.
Less frequently found variants are null alleles resulting in undetectable A iA T
levels due to intracellular degradation or intracellular accumulation of the
protein; this is usually associated with severe pulmonary disease, but not with
liver disease.2-7 (Table 2.1)
Table 2.1: Characteristics of the different A iA T alleles
Allele
Mutation
Cellular defect
M
-
-
S
Glu264Val
Intermediate
z
Glu342Lys
Intracellular
degradation
Intracellular
accumulation
Null
Different
mutations
Most mutations no
mRNA
Nili
Enzymatic
activity
Normal
Low
A iA T deficiency is characterized by an imbalance between the protease
neutrophil elastase and the protease inhibitor AiAT. It has been suggested
that neutrophil elastase might promote the developm ent of cancer.8 The
exact carcinogenic mechanism or sequence is not known.
In a recent paper, the authors hypothesized, that in A iA T deficiency, the hepa­
tocytes in which A iA T is accumulated are inhibited in their growth, but still do
express regenerative signals. Relatively normal cells, without A iA T deposits
are thereby stimulated and this chronic stimulation of regeneration may lead
to the formation of neoplasmata.9
20
A1AT d eficiency and chronic liver d isease: review
Diagnostic aspects
To establish the diagnosis different methods are available. The A iA T serum
level can be determined by clinical chemistry. In homozygous A iA T deficiency
the level is very low. However in the heterozygous variant the A iA T level
can be within the normal range, especially during an acute phase reaction.
Therefore electrophoresis should be performed in any case of suspected
A iA T deficiency in order to determine the phenotype. Liver biopsy is the
gold standard for establishing A iA T accumulation and PAS positive, diastase
resistant inclusions can be found. A specific immuno-histochemical staining
can confirm the diagnosis. (Figure 2.1) It is also possible to determ ine the
phenotype on paraffin embedded liver slides.
Figure 2 .1 Liver biopsy: cirrhosis due to a lp h a -i antitry p s in deficiency.
A: H E -stain ing , 2 ox: micro nodular liver cirrhosis B: D-PAS stainin g, 4 . 0 X : Diastase resistant, PeriodicSchiff-Acid positive globu li, m atching w ith A iA T accum ulation C: A iA T antib odies, 4 . 0 X : co nfirm ation
o f A iA T accum ulation
21
Chapter 2
Epidemiology
Despite the fact that A iA T deficiency is a frequent disorder, it is poorly recog­
nised in clinical practice. There are probably two main reasons, first in patients
with liver disease, the diagnosis is not always considered and second not all
subjects with a deficient phenotype develop liver disease, i.e. the penetrance
is low.10
Given this well known underdiagnosis, the real prevalence of A iA T deficiency
has mainly been determined by epidemiological methods, either using a control
cohort from an epidemiological study, or through neonatal screening. The Zallele is especially prevalent in Northern Europe, while the S allele is prevalent
in Southern Europe (Table 2.2)
Ta b le 2 . 2 : E p id e m io lo g y *10 15*
Country
Sweden
The Netherlands
France
Spain
Italy
Portugal
Australia
USA
New Zealand
Canada
PiZZ
0,06%
Frequency
PiMZ
Z allele
0,03%
0,01%
0,01%
0,02%
0,02%
0,02%
0,02%
0,05%
0,02%
-
_
2%
2,4%
2,3%
2,5%
2,7%
4,3%
2,7%
PiSZ
0,03%
S allele
0,04%
0,15%
0,2%
0,06%
0,3%
0,12%
0,09%
0,17%
0,11%
Method
Year
Neonates, phenotyping if serum AIAT < 40%
1976
Neonates, phenotyping when
AlAT/transferrine < 1,2
Control cohort epidemiological
Control cohort epidemiological
Control cohort epidemiological
Control cohort epidemiological
Control cohort epidemiological
Control cohort epidemiological
Control cohort epidemiological
Control cohort epidemiological
1980
study
study
study
study
study
study
study
study
2003
2003
2003
2003
2003
2003
2003
2003
Clinical aspects
Pulm onary disease
A iA T deficiency is associated with less inhibition of elastase resulting in pulmo­
nary disease. A iA T homozygosity (Pi ZZ) results in a pulmonary phenotype with
early onset of emphysema, asthma and bronchi ectasia. A iA T deficiency results
in panacinar pathology and disproportionate emphysematous involvement
A 1 A T d e f ic ie n c y a n d c h r o n ic liv e r d is e a s e : r e v ie w
of the lung bases. Tobacco smoking is the most important additional riskfactor for the development of pulmonary disease. Also subjects with the
Pi MZ phenotype have an increased risk for the development of pulmonary
disease.7
Neonatal / paediatric liver disease
AIAT deficiency is the most common genetic cause of liver disease in early
childhood. The most common presentation is by prolonged jaundice. The
stools generally contain no yellow or green pigment, indicating cholestasis
and mimicking biliary atresia. All patients have hepatomegaly and about 50%
also have splenomegaly. Approximately 5% of the patients present with an
increased bleeding tendency. This is due to vitamin K deficiency caused by the
cholestasis induced malabsorption. Less commonly children present later in
childhood with hepato-splenomegaly or with cirrhosis.16-19
In Sweden, in 1972-1974 200,000 neonates were screened for A iA T deficiency:
120 Pi ZZ (0,06%), 2 Pi Z-, 54 Pi SZand 1 Pi S- children were found. Only 14 of the
Pi ZZ children had prolonged jaundice, 9 of those had severe liver disease. All
infants appeared healthy at six months of age. Infants with a Pi SZ phenotype
had no signs of liver disease.
At the age of 16 years, in 17% of Pi ZZ adolescents and in 8% of Pi SZ adolescents,
elevated liver enzymes were found.The adults with liver disease in infancy were
clinically healthy. At the age of 26 years, the Pi ZZ subjects were compared to
Pi MM individuals.The Pi ZZ subjects had normal lung function; 4-9% of them
had mild livertest abnormalities.10-20-21 In the Province of Bozen in Northern Italy,
Pi phenotyping in umbilical cord blood was performed as a routine neonatal
screening. About 5% of Pi SZ children were affected by liver involvement with
elevated liver enzymes and 7% of 833 Pi MZ heterozygotes had elevated liver
enzymes in early childhood. At age of 5 and 10 years, none had liver disease.
The serum levels of A iA T were similar in the groups with and without liver
test abnorm alities; however these values had a wide range.22-23 Although
these studies suggest a good prognosis for neonatal cholestasis due to
AiAT-deficiency, other studies described children who developed severe liver
disease.16-19
23
C h a p te r 2
Liver disease in homozygous A iA T deficient adults
Liver disease due to A iA T deficiency generally presents at adult age. One
study reviewed adult patients with liver disease and A iA T deficiency; the
mean age of the patients when liver disease became symtomatic was 58 years
for the ZZ phenotype, 66 years for the SZ phenotype and 73 years for the MZ
phenotype. If the liver disease was advanced at the time of diagnosis, 42% of
these patients died within two years.24 A review of autopsy data on 94 Pi ZZ
homozygous A iA T carriers showed that cirrhotic patients survived longer
compared to non-cirrhotic patients. The non-cirrhotic patients had more
severe lung disease and died earlier.25 A cohort of patients who are registered
in the A lp h a-i foundation Registry (a USA foundation providing increased
research and improved health for A iA T deficiency), and who had reported liver
disease or jaundice (165 of the 2175 participants in the registry) completed
a questionnaire. Of these patients 71% was Pi ZZ and 18% was Pi MZ, in the
rem ainder the phenotype was unknown. Mean age at diagnosis of liver
disease was 31 years (range 0-68 years), 30% had had livertransplant or were
on the waiting list. Male gender and obesity were risk factors for advanced
liver disease, while white race, Pi phenotype, infant jaundice, diabetes or
hypercholesterolemia were not.26Although this survey is the largest cohort of
A iA T deficiency and liver disease in the literature, the self-selected cohort
runs a risk of inclusion bias. The natural history of the disease is not completely
known. The risk of cirrhosis in adults is difficult to estim ate because most
available data are retrospective and derived from patients known to have
A iA T deficient lung disease or cirrhosis.27-28
Heterozygous A iA T deficiency and liver disease
Although the role of homozygous A iA T in liver disease is established, the
association between heterozygous A iA T deficiency and chronic liver disease
is still subject to ongoing investigation. Several studies however showed an
association between heterozygous A iA T deficiency and chronic liver disease.
In 1981 a study showed the association of Pi MZ and liver disease. A total of
1055 liver biopsies was screened for A iA T depositions in hepatocytes and 34
patients with this inclusions were found. These patients were phenotyped, the
24
A 1 A T d e f ic ie n c y a n d c h r o n ic liv e r d is e a s e : r e v ie w
prevalence of phenotype Pi MZ in the whole biopsy group was 2.4%. In
the subgroup of patients with liver cirrhosis, 9% had a Pi MZ phenotype.
A percentage of 21 % of patients with cryptogenic cirrhosis and chronic
hepatitis had a Pi MZ phenotype. This was significantly increased compared to
other causes of cirrhosis. The prognosis of the Pi MZ cirrhotic patients was
poor, most patients died within one year.29 More recently patients with end
stage liver disease, in work-up for liver transplantation were investigated.
Pi MZ was found in 7.3- 8.2%, compared to 2.8% in the control population.
A heterozygous phenotype was more prevalent in patients with hepatitis C,
alcoholic liver disease, cryptogenic cirrhosis and hepatocellular carcinom a.30-31
In one study consecutive liver biopsies and autopsies were screened for Pi Z
deposits. In the biopsy group 3.4% of cases were Pi MZ phenotyped, whereas
in the autopsy group this was 1.8%. In biopsies from older people heterozygous
for AiAT, more fibrosis and more Pi Z deposits were found; the liver involvement
seems to be age dependent. In case of coexisting liver disease, patients pre­
sented with more inflammation, more fibrosis and more Pi Z deposits than
the biopsies without concomitant liver disease.32 We described 3 patients with
alcoholic liver disease and a rapidly deteriorating clinical course, resulting in
the patients' death. All 3 patients were found to be heterozygous for A1AT.33
To summarize: These studies showed that various liver diseases influence the
A iA T accumulation and that the A iA T accumulation influences the course of
the liver disease. The risk of developing liver cirrhosis is increased in patients
with heterozygous A iA T deficiency, also without coexisting liver disease. The
exact impact and involvement of liver disease by heterozygous A iA T deficiency
are unknown. Further research is needed to provide these data.
Heterozygous A iA T deficiency and coexisting hepatitis C virus infection
The role of A iA T deficiency in the severity and the course of liver disease
in chronic hepatitis C virus (HCV) infection is not clear, despite the fact that
several studies analyzed the association of HCV-induced liver disease and
A iA T deficiency. In Austria 1865 patients referred for the evaluation of chronic
liver disease were analyzed, 9% had a deficient phenotype. From these
patients with cirrhosis, 62% was HCV positive, 33% had evidence for HBV
25
C h a p te r 2
infection, 41% abuse of alcohol and 12% had features of autoimmune liver
disease. Out of 53 cirrhotic A iA T deficient patients, only 5 had no coexisting
liver disease. The authors concluded that the risk for chronic liver disease is
increased in patients with the Pi Z gene, because they may have increased
susceptibility to viral infection or additional factors, necessary to induce
chronic liver disease.34 The same authors investigated the prognosis of
patients with A iA T deficiency. Some 54 patients with A iA T deficiency had
evidence of chronic liver disease, 78% showed positive viral markers (hepatitis
B or hepatitis C); this was compared to 106 patients with A iA T deficiency
without chronic liver disease, without signs of additional viral infection. Life
expectancy in A iA T deficient patients was significantly lower in patients with
chronic liver disease in comparison to patients without chronic liver disease.35
Patients with end stage liver disease, in work-up for livertransplantation were
also investigated. In the HCV patients Pi MZ was found in 10-13%, compared
to 2,8% in the control population. This suggests that an abnormal heterozygous
phenotype is a co-factor in the development of chronic liver disease in HCV.30-31
In contrast, other studies showed no association of hepatitis C infection and
A iA T deficiency.36-38
To conclude: the results of these studies are controversial. Some studies
showed a higher incidence of A iA T deficiency in HCV infection and an
increased susceptibility to viral infections in A iA T deficiency and other studies
do not. Different methods to determine the A iA T state were used. Further
research on the influence of A iA T deficiency in the course of HCV infection
and vice versa is necessary.
Heterozygous A iA T deficiency and hepatocellular carcinoma
Established risk factors for hepatocellular carcinoma (HCC) include amongst
others chronic hepatitis B, HCV infection and alcoholic liver cirrhosis. Several
studies investigated the correlation between A iA T deficiency and HCC.39-40 In
1986 it was suggested fo rthe first time that men with A iA T deficiency may be
at risk for cirrhosis and HCC. Autopsy was performed in 16 adult patients with
A iA T deficiency. In 5 out of these 16 patients, a HCC was found.40-41 In 317
HCC, Pi Z deposits were found in 6% compared to 1.8% in the control group.
26
A 1 A T d e f ic ie n c y a n d c h r o n ic liv e r d is e a s e : r e v ie w
In heterozygous A iA T deficiency, HCC had also developed in non-cirrhotic
livers and were frequently characterized by cholangiocellular differentiation.
In patients with A iA T deficiency bile duct lesions were frequently found.
This might reflect a predisposition for the liver tissue for developing tumours
with cholangiolar differentiation in A iA T deficiency.43-44 In contrast to these
studies, others did not show an association between A iA T deficiency and
hepatocellular carcinoma, although carcinomas in non-cirrhotic livers were
Pi MZ associated.45-48
To summarize: several studies are performed to investigate the relation between
A iA T deficiency and hepatocellular carcinoma. The outcomes are not uniform.
In our opinion studies with large cohorts of patients with hepatocellular carci­
noma are the most reliable; these studies did find an association.
A iA T deficiency and associations with other diseases
A iA T deficiency is not only associated with liver and lung disease. Associations
with panniculitis, nephrotic syndrom e, intracranial aneurysm, hereditary
haemochromatosis and celiac disease have been described.49-62
Therapy
Most therapeutic strategies in the treatm ent of A iA T deficiency are directed
towards pulmonary disease. Infusion of purified pooled human plasma A iA T is
known as augmentation therapy. The goal of this treatm ent is to raise and
maintain serum A iA T concentrations above the protective threshold. Data from
different studies suggest that intravenous augmentation therapy has a positive
biochemical and clinical effect.The therapy is expensive (US $ 28,075-65,973 per
year).7 Different concepts have been studied to prevent the polymerization
and accumulation of the A iA T protein. New peptides that block the poly­
merization of the Z protein have been developed.63-64 Gene therapy by injecting
adeno-associated virus carrying the human A iA T gene is another promising
concept.65 Livertransplantation is used in end stage liver disease and results in
acquisition of the donor phenotype, a rise in serum levels of A iA T and prevention
of associated diseases.66
27
C h a p te r 2
Conclusion
A lp h a-i antitrypsin deficieny is an autosomal recessive disorder that can
lead to chronic pulmonary disease and liver disease.The liver disease is caused
by accumulation of an abnormal, polymerized protein. Deficient phenotypes
are present worldwide. Homozygous A iA T deficiency in children can cause
neonatal hepatitis. In adults homozygous patients are at risk for developing
end stage liver disease, cirrhosis and hepatocellular carcinoma. Heterozygous
A iA T deficiency is probably an important co-factor in the etiology of chronic
liver disease, as several studies showed associations with HCV, end stage
liver disease, cirrhosis and HCC. Screening on A iA T deficiency should be
done routinely, A iA T serum levels may be used, but phenotyping is crucial,
as serum levels may be not reflect true deficiencies (due to inflam m ation
serum levels can be false normal). Especially in cryptogenic chronic liver dis­
ease and in liver disease that deteriorates faster than may be expected, A iA T
deficiency may be of clinical significance as a (co)-factor. Clinical research is
needed in AiAT-related liver disease to investigate the association between
heterozygous A iA T deficiency and the presentation and course of liver diseases.67
We believe that current data are insufficient to decide on the pro's and con's of
screening on hepatocellular carcinoma in A iA T deficiency.Therapeutic options
in A iA T deficiency include augm entation therapy in pulm onary disease; in
end stage liver disease liver transplantation is an option. Gene therapy or
strategies to inhibit polymerization are promising.
References
1
Teckmann JH, Qu D, Perlmutter DH. Molecular pathogenesis of liver disease in a-antitrypsin
deficiency. Hepatology 1996;24:1504-1516.
2
Lomas DA. Loop-sheet polymerization: the mechanism of alphai-antitrypsin deficiency.
Respir Med 2000;94SuppS3-S6.
3
DaffornTR, Mahadeva R, Elliott PR, Sivasothy P, Lomas DA. A kinetic mechanism forthe
polymerization of ai-antitrypsin. J Biol Chemi999;274:9548-9555.
4
Lomas DA, Evans DL, Finch JT, Carrell RW. The mechanism of Z ai-antitrypsin accumulation
in the liver. Nature 1992;357:605-607.
28
A 1 A T d e f ic ie n c y a n d c h r o n ic liv e r d is e a s e : r e v ie w
5
DeMeo DL, Silverman EK. ai-Antitrypsin deficiency 2: genetic aspects of ai-antitrypsin
deficiency: phenotypes and genetic modifiers of emphysema risk. Thorax 2004;59:259-264.
6
Carrell RW, Lomas DA. Alphai-antitrypsin deficiency-a model for conformational diseases.
N Eng J Med 2002;346:45-53.
7
Stoller JK, Aboussouan LS. ai-antitrypsin deficiency. Lancet 2005;365:2225-2236.
8
Sun Z, Yang P. Role of imbalance between neutrophil elastase and ai-antitrypsin in cancer
development and progression. Lancet oncol 2004;5:182-190.
9
Perlmutter DH. Pathogenesis of chronic liver injury and hepatocellular carcinoma in alpha-iantitrypsin deficiency. Resp Res 2006;60:233-238.
io
Luisetti M, Seersholm N. ai-Antitrypsin deficiency 1: epidem iology of ai-antitrypsin deficiency.
Thorax 2004;59:164-169.
n
Sveger T. Liver disease in ai-antitrypsin deficiency detected by screening of 200,000 infants.
N Eng J Med 1976;294:1316-1321.
12
Dijkman JH, PendersTJ, Kramps JA, Sonderkamp HJA, Broek van den WGM, Haar ter BGA.
Epidemiology of Alphai-antitrypsin deficiency in the Netherlands. Hum Genet 1980;53:409-413.
13
Serres de FJ7 Blanco I, Fernandez-Bustillo E. Genetic epidemiology of alpha-i antitrypsin
deficiency in southern Europe: France, Italy, Portugal and Spain. Clin Genet 2003;63:490-509.
14
Serres de FJ, Blanco I, Fernandez-Bustillo E. Genetic epidemiology of alpha-i antitrypsin
deficiency in North America and Australia/New Zealand: Australia, Canada, New Zealand,
and the United States of America. Clin Genet 2003;64:382-397.
15
Serres de FJ. Worldwide racial and ethnic distribution of ai-antitrypsin deficiency: summary
of an analysis of published genetic epidemiologic surveys. Chest 2002;122:1818-1829.
16
Mowat AP. Alphai-antitrypsin deficiency (PiZZ): features of liver involvement in childhood.
Acta Paediatr 1994;393:13-17.
17
O'Brien ML, Buist NRM, Murphey WH. Neonatal screening for alphai-antitrypsin deficiency.
J Pediatr 1978;921:1006-1010.
18
Cottrall K, Cook PJL. Neonatal hepatitis syndrome and alpha-i-antitrypsin deficiency:
an epidemiological study in south-east England. Postgrad Med J 1974;50:376-380.
19
Grishan FK, Greene HL. Liver disease in children with PiZZ ai-antitrypsin deficiency.
Hepatology 1988;8:307-310.
20
Sveger T, Eriksson S. The liver in adolescents with ai-antitrypsin deficiency. Hepatology
1995;22:514-517.
21
Pittulainen E, Carlson J, Ohlsson K, Sveger T. ai-Antitrypsin deficiency in 26-year-old subjects:
lung, liver, and protease/protease inhibitor studies. Chest 2005;128:2076-2081.
22
Pittschieler K. Liver disease and heterozygous alpha-i-antitrypsin deficiency. Acta Paediatr Scan
1991;80:323-327.
23
Pittschieler K. Liver involvement in alphai-antitrypsin deficient phenotypes PiSZ and PiMZ.
Acta Paediatr 2002;91:239-240.
29
C h a p te r 2
24
Rakela J, Goldschm iedt M, Ludwig J. Late manifestation of chronic liver disease in adults with
alp h a-i antitrypsin deficiency. Dig dis Sci 1987;32:1358-1362.
25
Eriksson S. ai-Antitrypsin deficiency and liver cirrhosis in adults. Acta Med Scan 1987;221:461-467.
26
Bowlus CL, Willner I, Zern MA et al. Factors associated with advanced liver disease in adults with
alphai-antitrypsin deficiency. Clin gastroenterol hepatol 2005;3:390-396.
27
Schonfeld v. J, Breuer N, Zotz R et al. Liverfuntion in patients with pulmonary emphysema due
to severe alpha-i-antitrypsin deficiency (Pi ZZ). Digestion 1996;57:165-169.
28
Larsson C, Eriksson S. Liver function in asymptomatic adult individuals with severe a iantitrypsin deficiency (Pi Z). Scand J Gastroent 1977;12:543-546.
29
Hodges R , Millward-Sadler GH, Path MRC, Barbatis C, Wright R, Phil D. Heterozygous MZ
alphai-antitrypsin deficiency in adults with chronic active hepatitis and cryptogenic cirrhosis.
N Eng J Med 1981;304:557-560.
30
Eigenbrodt ML, McCashland TM, Dy RM, Clark J, Galati J. Heterozygous ai-antitrypsin
phenotypes in patients with end stage liver disease. Am J Gastroenterol 1997;92:602-607.
31
Graziadei IW, Joseph J J, W iesner RH, Therneau TM, Batts KP, Porayko MK. Increased risk
of chronic liver failure in adults with heterozygous ai-antitrypsin deficiency. Hepatology
1998;28:1058-1063.
32
Fischer HP, Ortiz-Pallardo ME, KoY, Esch C, Zhou H. Chronic liver disease in heterozygous
ai-antitrypsin deficiency PiZ. J Hepatol 2000;33:883-892.
33
Kok KF, Wahab PJ, Vries RA. Heterozygosity for alphai-antitrypsin deficiency as a cofactor in
the development of chronic liver disease. Ned Tijdschr Geneeskd 2005;149:2057-61.
34
Propst T, Propst A, Dietze O, Judmaier G, Braunsteiner H, Vogel W. High prevalence of viral
infection in adults with homozygous and heterozygous alphai-antitrypsin deficiency and
chronic liver disease. Ann Int Med 1992;117:641-645.
35
Propst A, Propst T, Ofner D, Feichtinger H, Judmaier G, Vogel W. Prognosis and life expactancy
in alpha-i-antitrypsin deficiency and chronic liver disease. Scan J Gastroenterol i9 9 5 ;3o :iio 8 1112.
36
Serfaty L, Chazouilleres O, Poujol-Robert A et al. Risk factors for cirrhosis in patients with chronic
hepatitis C virus infection: results of a case-control study. Hepatology:i997;26:776-779.
37
Elzouki AN, Verbaan H, Lindgren S, Widell A, Carlson J, Erikssonn S. Serine protease inhibitors
in patients with chronic viral hepatitis. J Hepatol 1997;27:42-48.
38
Scott BB, EgnerW and on behalf of theTrent hepatitisC study group. Does ai-antitrypsin
phenotype PiMZ increase the risk of fibrosis in liver disease due to hepatitis C virus infection?
Eur J Gastroenterol Hepatol 2006;18:521-523.
39
Palmer PE, Wolfe HJ. ai-antitrypsin deposition in primary hepatic carcinoma. Arch Pathol Lab
Med 1976;100:232-236.
40
Reintoft I, Hagerstrand IE. Does the Z gene variant of alpha-i-antitrypsin predispose to hepatic
cancer? Hum Pathol 1979;10:419-424.
41
Eriksson S, Carlson J, Velez R. Risk of cirrhosis and primary liver cancer in alphai-antitrypsin
deficiency. N Eng J Med 1986;314:736-739.
30
A 1 A T d e f ic ie n c y a n d c h r o n ic liv e r d is e a s e : r e v ie w
42
Elzouki AN, Eriksson S. Risk of hepatobiliary disease in adults with severe ai-antitrypsin
deficiency (PiZZ): is chronic viral hepatitis B or C an additional risk factor for cirrhosis and
hepatocellular carcinoma? Eur J Gastroenterol Hepatol 1996;8:989-994.
43
Zhou H, Ortiz-Pallardo ME, KoY, Fischer HP. Is heterozygous alpha-i-antitrypsin deficiency
type PiZ a risk factor for primary liver cell carcinoma? Cancer 2000;88:2668-2676.
44
Zhou H, Fischer HP. Liver carcinoma in PiZ alpha-i-antitrypsin deficiency. Am J Pathol
1998;22:742-748.
45
Propst T, Propst A, Dietze O, Judmaier G, Braunsteiner H, Vogel W. Prevalence of hepatocellular
carcinoma in alpha-i-antitrypsin deficiency. J Hepatol 1994;21:1006-1011.
46
Rabinovitz M, Gavaler JS, Kelly RH, Prieto M, Thiel van D. Lack of increase in heterozygous a iantitrypsin deficiency phenotypes among patients with hepatocellular and bile duct carcinoma.
Hepatology 1992;15:407-410.
47
Govindarajan S, Ashcaval M, Peters RL. a-i-Antitrypsin deficiency phenotypes in hepatocellular
carcinoma. Hepatology i9 8 i;i:6 2 8 :6 3 i.
48
Sparos L, TountasY, Chapuis-Cellier C, Thedoropoulos G, Trichopoulos D. Alphai-antitrypsin
levels and phenotypes and hepatitis B serology in liver cancer. Br J Cancer 1984;49:567-570.
49
Edmonds BK, Hodge JA, Rietschel RL. Alphai-antitrypsin deficiency-associated panniculitis:
case report and review of the literature. Pediatr Dermatol 1991;8:296-299.
50
O'Riordan K, Blei A, Rao MS, Abecassis M. ai-Antitrypsin deficiency-associated panniculitis:
resolution with intravenous ai-antitrypsin administration and liver transplantation. Transplantation
1997;63:480-482.
51
Phelps RG, ShojiT. Update on panniculitis. Mt Sinai J Med 2001;68:262-267.
52
McBean J, Sable A, Maude J, Robinson-Bostom L. ai-Antitrypsin deficiency panniculitis. Cutis
2003;71:205-209.
53
Chowdhury MMU, Williams EJ, Morris JS al. Severe panniculitis caused by homozygous ZZ
ai-antitrypsin deficiency treated successfully with human purified enzyme (Prolastin®). BrJ
Dermatol 2002;147:1258-1261.
54
Elzouki AN, Lindgren S, Nilsson S, Veress B, Eriksson S. Severe ai-antitrypsin deficiency
(PiZ homozygosity) with membranoproleferative glomerulonephritis and nephrotic syndrome,
reversible after orthotopic livertransplantation. J Hepatol 1997;26:1403-1407.
55
Elzouki AN and Eriksson S. Severe ai-antitrypsin deficiency and intracranial aneurysms.
Lancet ig94;343:i037.
56
Elzouki AN, Hultcrantz R, Stal P, Befrits R, Eriksson S. Increased PiZ gene frequency for a iantitrypsin in patients with genetic haemochromatosis. Gut 1995;36:922-926.
57
Rabinovitz M, Gavaler JS, Kelly RH, Thiel van DH. Association between heterozygous a iantitrypsin deficiency and genetic hemochromatosis. Hepatology 1992;16:145-148.
58
Fargion S, Bissoli F, Francanzant AL, Suigo E, Sergi C, Taioli E. No association between genetic
hemochromatosis and ai-antitrypsin deficiency. Hepatology 1996;24:1161-11164.
59.
Eriksson S, Lindmark B, Olsson S. Lack of association between hemochromatosis and
a-antitrypsin deficiency. Acta Med Scand 1986;219:291-294.
31
C h a p te r 2
60
W alker-Sm ithJ, Andrews J. Alpha-i-antitrypsin, autism, and coeliac disease. Lancet 1972:
883-884.
61
Pons Romero F, Casafont F, Rodriquez de Lope C, San Miguel G, Artinano E Cagigas J. Could
alphai-antitrypsin deficiency have any role in the development of celiac sprue after gastric
operations? J Clin Gastroenterol 1986;8:559-561.
62
Klasen EC, Polanco I, Biemond I, Vazquez C, Pena AS. ai-Antitrypsin deficiency and celiac
disease in Spain. Gut 1980;21:948-950.
63
Burrows JAJ, Willis LK, Perlmutter DH. Chemical chaperones mediate increased secretion of
mutant ai-antitrypsin (ai-A T ) Z: a potential pharmacological strategy for prevention of liver
injury and emphysema in a i-A T deficiency. Proc Natl Acad Sei USA 2000;97:1796-1801.
64
Lomas DA, Mahadeva R. ai-antitrypsin polymerization and the serpinopathies: pathobiology
and prospects fortherapy. J Clin Invest 2002;110:1585-1590.
65
Stecenko AA, Brigham KL. Gene therapy progress and prospects: alph a-i antitrypsin. Gene
Therapy 2003;10:95-99.
66
Vennarecci G, Gunson BK, Ismail T et a I. Transplantation for end stage liver disease related to
alpha 1 antitrypsin. Transplantation 1996;61:1488-1495.
67
American Thoracic Society/ European Respiratory Society Statement: Standards for the
diagnosis and management of individuals with alph a-i antitrypsin deficiency. Am J Respir
CritCare Med 2003;168:818-900.
32
Chapter 3
Prognosis of neonatal cholestasis
due to Alpha-i Antitrypsin Deficiency
Karin F Kok
Lotte G van Vlerken
Marc A Benninga
Johanna C Escher
Richard A deVries
Roderick HJ Houwen
Abstract
Background: A lp h a-i antitrypsin (A iA T) deficiency is the
most common genetic cause of liver disease in early child­
hood. Several studies, mostly from transplantation centers,
have reported a poor outcome of patients with neonatal
cholestasis due to A iA T deficiency. We retrospectively
studied the natural course of liver disease in a cohort of
patients with neonatal cholestasis due to A iA T deficiency.
Methods: We studied neonates who were diagnosed with a
ZZ or SZ phenotype within the first 3 months of life and who
had neonatal cholestasis. Patients born in the Netherlands
between January 1991 and December 2006 were included.
Clinical and biochemical parameters at presentation were
recorded. Outcome was reported at end of follow-up and
in 5- and 10- years follow-up cohorts.
Results: In this 16 years time frame 50 patients were
identified. A total of 15 neonates presented with vitamin K
deficiency bleeding. During follow-up 5 (10%) patients
died and 2 (4%) patients received liver transplantation. At
the end of follow-up 12% had cirrhosis, 62% had abnormal
liver enzymes and/or hepato-splenomegaly and 12% had
no biochemical or clinical abnormalities. Survival without
liver transplantation was 87% in the 5-years follow-up
cohort and 73% in the 10-years follow-up cohort.
Conclusions: In our study, the natural outcome of patients
with neonatal cholestasis due to A iA T deficiency is better
than was suggested in studies originating from transplan­
tation centers, but similar to other cohort studies.
P ro g n o s is o f n e o n a t a l c h o le s ta s is d u e t o A 1 A T d e f ic ie n c y
Background
A iA T deficiency is a hereditary disease characterized by the hepatic production
of an abnormal A iA T protein that can not completely be released into the
plasma. 1This leads to deficiency of serum A iA T and subsequently to an impaired
protection of the lungs against proteases resulting in pulmonary emphysema;
the hepatic accumulation of the abnormal protein can lead to liver disease.The
A iA T protein isencoded bythe protease inhibitor(Pi) locus located on the long
arm of chromosome i4q3i-32.3.This locus is highly polymorphic, and so far more
than 100 different A iA T isotypes have been identified, which can be detected by
iso-electrophoresis of the protein or mutational analysis. The most common M
allele results in a functional normal protein with normal serum A iA T levels. Two
frequent variants p.glu342lys (denoted as Z allele) and the p.glu264val (com­
monly referred to asthe S allele) are encountered in Western Europe.2The point
mutation found in the Z variant destabilizes the loop-sheet polym erization3
of the A iA T molecule, resulting in chains of polymers that are retained in the
liver cells.4 In Pi ZZ homozygous subjects, this results in a severe deficiency of
serum A iA T and in accumulation of the abnormal protein in the endoplasmatic
reticulum of the hepatocytes, which may lead to symptomatic liver disease.5
The S variant of the A iA T molecule has less deleterious effect on the protein.
Consequently liver disease is uncommon, and only seen in SZ patients.6-7
A iA T deficiency type Pi ZZ isthe most common genetic cause of liver disease in
early childhood. Symptomatic patients generally present with neonatal cholesta­
sis, or less commonly later in childhood with hepato- and/or splenomegaly or
cirrhosis.6-8-9 In Sweden 200,000 neonates were screened for A iA T deficiency;
120 patients with Pi ZZ were identified but only 22 of these children presented
with neonatal cholestasis.10 After 4 years follow-up 4 patients had died (18%);
3 patients due to their liver disease and 1 patient of aplastic anemia. This patient
had cirrhosis at autopsy.11-13 However in other studies up to 50% of the patients
with A iA T deficiency related neonatal cholestasis either died or had to be
transplanted.14-18 Because these studies were mainly done in transplantation
centres, they may not be representative forthe natural course of A iA T deficiency
related neonatal cholestasis.
37
C h a p te r 3
Therefore we set out to study the natural course of liver disease due to A iA T
deficiency in all children presenting with neonatal cholestasis in the Netherlands
born between January 1991 and December 2006.
Methods
Patients were included when they were born between January i st 1991
and December 31st 2006 and had presented with neonatal cholestasis due to
A iA T deficiency. Patients were identified by searching the databases of the
5 centers in the Netherlands where phenotyping or genotyping for A iA T
deficiency is routinely performed. Files of all patients, that had genotyping
and/or phenotyping done within the first three months of life were reviewed
by contacting the referring physician. Subsequently, a clinical database was
established that contained the following data: date of birth, sex, age at
presentation, birth weight, presenting symptoms, AiAT-phenotype, duration
of follow-up with biochemical markers and clinical outcome.
Results
Patients
A total of 56 children who were diagnosed with a Pi ZZ of Pi SZ phenotype
were analyzed; 4 children were sibs of known patients and had no evidence of
neonatal cholestasis; they were excluded from further analysis. Therefore 52
patients met the criteria for inclusion in this study. Two of these patients, for
whom clinical and biochemical data at presentation and follow-up were
unavailable, had to be excluded. Our cohort therefore consists of 50 patients
(32 boys, 18 girls). Pi ZZ phenotype was found in 49 children; one child had a
Pi SZ phenotype.
Presentation
The median age at presentation was 5 weeks (range 0-12 weeks). Median birth
weight was 3205 grams (n=43, range: 1675-4595 grams).The initial prestentation
within the group of 50 patients, was jaundice in 26 patients, 9 children presented
38
Prognosis of neonatal cholestasis due to A1AT d eficiency
initially with failure to thrive, which could subsequently be attributed to
malabsorption due to cholestasis, while 15 patients presented with clinical
signs of vitamin K dependent hemorrhagic disease. Intracranial hemorrhage
w asthe presenting symptom in 7/15 children.19 Base-line patient and laboratory
characteristics are outlined in table 3.1.
Table 3 . 1 : Baseline characteristics in 50 neonates w ith a lp h a -i antitry p s in deficiency
Total number of patients
Male/female
Median age at presentation in weeks
(range)
Median AST (range)
50
32/18
5 (0-12)
Median ALT (range)
54 (14-1490)
Median GGT (range)
4 5 7 (80-2526)
89 (17-2106)
(n<45 IU/1)
(n<45 IU/1)
(n<45 U /l)
Median total bilirubin (range)
106 (18-490)
______________________________________ (n<17 fjm o l/l)_____
Baseline characteristics and biochemical results of 50 patients with neonatal cholestasis due to A1AT deficiency.
Overall outcome
Median follow-up time was 6,1 years (range: 1 month - 1 5 years). In this group of
50 patients 5 neonates died (10%). Two boys died as a result of decompensated
liver failure at an age of 26 weeks and 49 weeks respectively. One boy died at
an age of 26 months after screening for living related transplantation which
was refused by his parents. Furthermore, 1 boy, born in 2005, died at an age of
6 weeks as a result of massive intracranial hemorrhage. Lastly, 1 girl, born in
1991, died at an age of 7 months after reconstruction of an atrio-ventricular
septum defect; she developed extensive cardiac and respiratory complications.
Given the unrelated cause of death this patient was excluded from further
outcome analysis. Two girls (4%) underwent orthotopic liver transplantation
(OLT) at age 1 respectively 7 years. After transplantation these children had no
evidence for liver disease. At last follow-up six patients had clinical or histo­
logical signs of cirrhosis, but were clinically stable; laboratory data showed
normal synthetic capacity of the liver. Of the remaining 37 children, 6 patients
had elevated liver enzymes with hepatomegaly without splenomegaly and
39
Chapter 3
25 patients had persistent elevated liver enzymes, but no hepato- and/or
splenomegaly. Lastly, 6 patients had normal liver enzym es and no clinical
evidence of hepato- and/or splenomegaly.
5-years and 10-years follow-up cohorts
Five-years outcome could be analyzed in 32 children born before 2001. In this
group3 children died and 1 child underwent livertransplantation; consequently,
28/32 (87%) survived without liver transplantation. A total of 22 out of these
28 patients had elevated liver enzymes, of whom 4 had clinical signs of cirrhosis;
the remaining 6 children had normal liver enzymes. Ten-years outcome could
be analyzed in 15 children born before 1996. In this group 2 children died and
2 children underwent liver transplantation, thus 11/15 (73%) survived without
liver transplantation. A total of 6 out of these 11 children patients had elevated
liver enzymes and 5 children had normal liver function tests. Our data are
compared with the data as could be retrieved from the Sveger studies.10-12
(Table 3.2)
Ta b le 3 . 2 : Fo llow -up in n e o n ata l cholestasis due to a lp h a -i antitry p s in deficiency. C urrent data
com pared w ith studies fro m Sveger *10 12*
Normal liver
enzymes
Elevated ALT
Cirrhosis
OLT or death
Sveger 4- y e a r fo llo w -u p
4 /1 6 * (25% )
Present study 5-year follow-up
6/28 (21%)
1 2 /1 6 * (75%)
2 /1 8 (1 1 % )
4 /22(18% )
22/28 (79%)
4/28 (14%)
Sveger 8- y e a r fo llo w -u p
4/32 (13%)
4 /1 1 * (37% )
7/11* (63%)
0 /1 1 (0%)
Present study 10-year follow-up
6/21(29% )
5/11 (45%)
6/11 (55%)
0/11 (0%)
4/15 (27%)
* Irrespective of the presence of cirrhosis; ALT is considered elevated when > 45 IU/l.
+assessment of liver enzymes not performed in all patients
Discussion
Several studies have reported the outcome of liver disease in children
with A iA T deficiency. Most of these studies were done in transplantation
centers, suggesting that up to 50% of this group has to be transplanted, with
a substantial proportion of cirrhosis in the rem ainder.15-18
40
Prognosis of neonatal cholestasis due to A1AT d eficiency
Based on ourclinical experience in children with symptomatic A iA T deficiency,
we had the impression that the majority of patients have a better prognosis.
As neonatal cholestasis is by far the most common presentation of A iA T
deficiency we set out to identify all children with this presenting symptom and
born in the Netherlands from 1991-2006. Survival without livertransplantation
was 87% in the 5-years follow-up cohort and 73% in the 10-years follow-up
cohort.The studies initiated by Sveger more than 30 years ago, showed a similar
outcome. (Table 3.2) The data of these two studies, both without obvious
selection bias, are in contrast with data from transplantation centers which
generally show a worse prognosis.15-18 An overview of these studies and a
summary of our overall results is given. (Table3-3)
Table 3.3: Prognosis of neonatal cholestasis due to A iA T deficiency. Results of several
studies coming from transplantation centers ^15 l8 ^and an overview of the results of our
present study
Psacharopolous
G rishan
(1988)
Ibarguen
(1990)
67
17
19 (28%)
19 (28%)
14 (21%)
15
851
3,7
4 ,8
5 ( 33 %)
3 (20% )
43 (51 %)
7 (8 %)
7 (47 %)
20 (24%)
Francavill
a (2000)
82
10
21 (26%)
6 (7%)
30 (37%)
15 (22%)
0 (0%)
15 (18%)
25 (30%)
(1983)
N um ber of patients
M e d ia n fo llo w -up (years)
Death or transplantation (%)
Cirrhosis (%)
Elevated liver enzym es and/or
hepato-/splenom egaly (%)
N orm al liver functio n (%)
Present
study (2009)
5°
6,1
7 (14%)
6 (12%)
31 (62%)
6 (12%)
1 Number of children presenting with liver disease due to AIAT deficiency; 50 children presented with neonatal
cholestasis; the others presented later in childhood.
In our cohort 15 patients presented with clinical signs of vitamin K deficiency,
with 7 having an intracranial hem orrhage19. Although it is well known that
neonatal cholestasis due to A iA T deficiency is a risk factor for hemorrhagic
disease of the newborn6-19, it was thought that this problem could be effec­
tively prevented when giving vitamin K supplements to breastfed infants20.
However recent evidence suggests that the daily oral vitamin K prophylaxis
for breast fed children, as recommended inthe Netherlands, hasa sub-optimal
efficacy, especially in patients with cholestasis.21-22 Therefore this vitamin K
prophylaxis schedule is now being revised. Only a minority of patients with
41
C h a p te r 3
A iA T deficiency presents with neonatal cholestasis. Similarly, within this
group, only a subgroup will progress to severe liver disease. It is tempting to
speculate that, apart from environmental factors, modifier genes might be
responsible for this difference. In fact a recent study showed an association
between a functional SNP in the A iA T gene and liver disease.23
We conclude that the majority of patients presenting with neonatal cholestasis
due to A iA T deficiency remains symptom free or only has elevated serum
transaminases. Nevertheless, a significant proportion of patients develops
cirrhosis, sometimes with end-stage liverdisease; inthese patients livertransplantation may be necessary.
References
1
DaffornTR, Mahadeva R, Elliott PR, Sivasothy P, Lomas DA. A kinetic mechanism forthe
polymerization of alphai-antitrypsin. J Biol Chem 1999 Apr 2;274(i4):9548-55.
2
DeMeo DL, Silverman EK. Alphai-antitrypsin deficiency. 2: genetic aspects of a lp h a iantitrypsin deficiency: phenotypes and genetic modifiers of emphysema risk. Thorax 2004
Mar;59(3):259-64.
3
Lomas DA. Loop-sheet polymerization: the mechanism of alphai-antitrypsin deficiency.
Respir Med 2000 A ug;94 Suppl C:S3-S6.
4
Lomas DA, Evans DL, Finch JT, Carrell RW. The mechanism of Z alpha l-antitrypsin accumula­
tion in the liver. Nature 1992 Jun i 8;357(6379):6 o 5-7.
5
Stoller JK, Aboussouan LS. Alphai-antitrypsin deficiency. Lancet 2005 Jun 25;365(9478):2225-36.
6
Prim ha k RA, Tanner MS. A lp h a-i antitrypsin deficiency. Arch Dis Child 2001 Jul;85(i):2-5.
7
VandenplasY, Franckx J, Liebaers I, Ketelbant P, Sacre L. Neonatal hepatitis with obstructive
jaundice in an SZ heterozygous alpha l-antitrypsin-deficient boy and destructive lung disease
in his SZ mother. A review of the literature. Eur J Pediatr 1985 N ov;i44(4):39i-4.
8
Hussain M, Mieli-Vergani G, Mowat AP. Alpha l-antitrypsin deficiency and liver disease: clinical
presentation, diagnosis and treatment. J Inherit Metab Dis i9 9 i;i4 (4 ):4 9 7 -5 ii.
9
Mowat AP. Alpha l-antitrypsin deficiency (PiZZ): features of liver involvement in childhood.
Acta Paediatr Suppl 1994 Feb;393:i3-7.
10
Sveger T. Liver disease in alphai-antitrypsin deficiency detected by screening of 200,000
infants. N Engl J Med 1976 Jun io ;2 9 4 (2 4 ):i3 i6 -2 i.
11
Sveger T. Prospective study of children with alpha l-antitrypsin deficiency: eight-year-old
follow-up. J Pediatr 1984 Ja n ;i0 4 (i):9 i-4 .
42
P rognosis o f n e o n a ta l ch o le sta sis d u e to A1AT d e fic ie n c y
12
SvegerT, Thelin T. Four-year-old children with alpha l-an titryp sin deficiency. Clinical follow -up
and parental attitudes tow ards neonatal screening. Acta Paediatr Scand 1981 M a r;7o (2 ):i7i-7.
13
SvegerT. The natural history of liver disease in alpha l-a ntitryp sin deficient children. Acta
Paediatr Scand 1988 N ov;77(6):847-5i.
14
Dick MC, M owat AP. Hepatitis syndrom e in infancy--an epidem iological survey with 10 year
follow up. Arch Dis Child 1985 Jun;6o(6):5i2-6.
15
Francavilla R, Castellaneta SP, Hadzic N, Cham bers SM, Portmann B, T ungJ, et al. Prognosis of
alph a-i-antitrypsin deficiency-related liver disease in the era of paediatric liver transplantation.
J Hepatol 2000 Jun;32(6):986-92.
16
Ghishan FK, Greene HL. Liver disease in children with PiZZ alpha l-an titryp sin deficiency.
Hepatology 1988 M ar;8(2):307-i0.
17
Ibarguen E, Gross CR, Savik SK, Sharp HL. Liver disease in alph a-i-antitrypsin deficiency:
prognostic indicators. J Pediatr 1990 D e c;ii7(6 ):8 6 4 -70 .
18
Psacharopoulos HT, M owat AP, Cook PJ, Carlile PA, Portmann B, Rodeck CH. Outcom e of liver
disease associated with alpha 1 antitrypsin deficiency (PiZ). Im plications for genetic counselling
and antenatal diagnosis. Arch Dis Child 1983 N ov;58( i i ):882-7.
19
Hasselt van PM Kok KF, Vorselaars ADM, van Vlerken L, de Vries R, Nieuwenhuys E et al. Vitamin
K deficiency bleeding in cholestatic infants with a lp h a -i antitrypsin deficiency. Arch Dis Child
Fetal Neonatal Ed. 2009 Nov;94(6):F456-6o
20
Cornelissen EA, Hirasing RA, M onnens LA. [Prevalence of hem orrhages due to vitam in K
deficiency in The Netherlands, 1992-1994]. Ned Tijdschr Geneeskd 1996 A pr 2 7;i4 o (i7):9 35-7.
21
Ijland MM, Pereira RR, Cornelissen EA. Incidence of late vitam in K deficiency bleeding in
new borns in the Netherlands in 2005: evaluation of the current guideline. Eur J Pediatr 2008
Feb;i67(2):i65-9.
22
van Hasselt PM, de KoningTJ, Kvist N, de VE, Lundin CR, Berger R, et al. Prevention of vitam in K
deficiency bleeding in breastfed infants: lessons from the Dutch and Danish biliary atresia
registries. Pediatrics 2008 A p r;i2 i(4 ):e 8 57-e 8 6 3.
23
Chappell S, Hadzic N, Stockley R, G uetta-Baranes T, Morgan K, Kalsheker N. A polym orphism
of the alph ai-an titrypsin gene represents a risk factor for liver disease. Hepatology 2008
Ja n ;4 7 (i):i2 7 -3 2 .
43
Chapter 4
Vitamin K deficiency bleeding in
Cholestatic Infants with Alpha-i
Antitrypsin Deficiency
Archives of Disease in Childhood- Fetal and Neonatal Edition 200g
Nov; 94(6)^456-60.
Peter M van Hasselt
Karin F Kok
Adriane DM Vorselaars
Lotte van Vlerken
Ed Nieuwenhuys
Tom J de Koning
Richard A deVries
Roderick HJ Houwen
Abstract
Objective: Exclusively breast-fed infants with unrecognized
cholestatic jaundice are at high risk of a vitamin K deficiency
(VKD) bleeding. It is presently unknown whether this risk
depends on the degree of cholestasis. Since alpha-i-antitrypsin deficiency (A iA T deficiency) induces a variable
degree of cholestasis, we assessed the risk ofVKD bleeding
in infants with cholestatic jaundice due to A iA T deficiency.
Patients and methods: Infants with a ZZ or SZ phenotype
born in the Netherlands between January 1991 and
December 2006 were identified from the databases of the
5 Dutch diagnostic centres for alpha-i-antitrypsin phenotyping and/or genotyping. We determined the risk ofVKD
bleeding upon diagnosis of A 1 AT deficiency in breastfed
and formula-fed infants and searched for correlations
between serum levels of conjugated bilirubin and the risk
of bleeding.Results: A total of 40 infants with A iA TD were
studied. VKD bleeding was noted in 15/20 (75%) of breast­
fed infants, compared with 0/20 of formula-fed infants
with A iA T deficiency. The relative risk for VKD bleeding in
breast-fed versus formula-fed infants was at least 15.8
(95% confidence interval 2.3 to 108). Conjugated bilirubin
levels at diagnosis did not correlate with the risk ofV K D
bleeding.
Conclusions:The risk ofVKD bleeding in breast-fed infants
with A iA T deficiency was high and did not correlate with
serum level of conjugated bilirubin at diagnosis. A similar
absolute risk was previously reported in breast-fed infants
with biliary atresia under the same prophylactic regimen.
This confirms that - without adequate prophylaxis - the
risk ofVKD bleeding is uniformly high in exclusively breast­
fed infants with cholestatic jaundice.
B le e d in g in in fa n ts w ith A1AT d e fic ie n c y
Introduction
Vitamin K prophylaxis aims to protect infants against the potentially life
threatening coagulation disorder that results from severe vitamin K deficiency
(VKD). In breastfed infants VKD bleeding may occur within days of birth dueto
the low vitamin K stores at birth1, the short half life of vitamin K2-3 and the low
vitamin K content of human milk4. An adequate dose of vitamin K given orally or
by injection at birth effectively prevents VKD bleeding in the first week of life.
However, ongoing vitamin K supplementation is necessary to prevent VKD bleed­
ing after the first week of life (late VKD bleeding) in breastfed babies having oral
prophylaxis. Even with these regimens some infants do develop late VKD bleed­
ing, usual ly due to unrecognized cholestasis causing vita mi n K m alabsorption.5-8
As late VKD bleeding is associated with intracranial haem orrhage (ICH)
- resulting in significant mortality and long term m o rb id ity- in approximately
50% of cases,9-11 it is especially important to prevent this complication. We
recently reported that - despite routine vitamin K supplementation - t h e risk
ofVKD bleeding in Dutch breastfed infants with biliary atresia (BA) is approxi­
mately 80 percent. VKD bleeding has also been described in infants with various
etiologies of neonatal cholestasis,5-8 but it is unknown w hetherthe risk ofVKD
bleeding in these diseases is equally high. In these other causes of cholestasis
(and contrary to BA) obstruction of bile flow is generally incomplete, allowing
some bile to enter the duodenal lumen. Partial obstruction of bile flow is associ­
ated with lower degrees of conjugated bilirubin.12Theoretically, residual bile flow
could aid vitamin K absorption, and reduce the risk ofVKD bleeding. A lpha-iantitrypsin (AiAT) deficiency - the next most frequent cause of cholestatic
jaundice in infancy after BA - is associated with varying degrees of liver
involvem ent.13-14 Cholestatic jaundice is present in only approxim ately 1 in 10
infants with a ZZ phenotype13 and is even less frequent in patients with an SZ
phenotype. Moreover, the degree of cholestasis in symptomatic infants can vary
from mild to severe.13Thus, A iA T deficiency induced cholestatic liver disease
represents an attractive model to study the influence of varying degrees of
cholestasis on the development ofV K D bleeding at initial presentation. The
objectives of this study were 1) to determine the absolute and relative risk ofVKD
bleeding at initial presentation of cholestaticjaundice induced byA iA T deficiency
in exclusively breastfed versus formula fed infants and 2) to investigate whether
varying degrees of jaundice are associated with a different risk of bleeding.
47
C h a p te r 4
Methods
Study population
Dutch patients born from January 1991 until December 2006 presenting with
neonatal cholestasis due to A iA T deficiency were identified from the databases
of the 5 centres in the Netherlands involved in A iA T genotyping or phenotyping.
Patients were provisionally included if a ZZ or SZ genotype or phenotype was
generated within the first 6 months after birth. Subsequently, patient files
were reviewed at the hospital where the patients originated to obtain relevant
demographic data and clinical characteristics. The same exclusion criteria
were used as described previously for BA.11 Briefly, exclusion criteria were:
1) gestational age less than 37 weeks 2) birth weight below 2000 grams,
3) absence of cholestasis (with cholestasis defined as a total serum bilirubin
concentration above 50 |_imol/L, with a conjugated fraction of at least 20
percent). External Quality Assessment is in place throughout the Netherlands
fo rth is measurement, 4) a diagnosis of cholestasis before age 8 days. An addi­
tional exclusion criterion was the presence of an older sib with cholestatic
jaundice due to A iA T defiiciency, as this should affect the advised prophylactic
regimen. Infants were categorized as "breast-fed" if they had received exclu­
sively breastfeeding from birth onwards. All other infants were categorized as
"form ula-fed".
Vitamin K prophylaxis
Since 1990 all infants born in the Netherlands receive an oral dose of 1 mg
vitamin K directly after birth. Upon breastfeeding, parents are advised to give
their child a daily oral dose of 25 |_ig vitamin K from the second week of life
until the end of the 13th week. For daily dosing of vitamin K, a dietary supplement
is used in which vitamin K is solved in arachid oil. The vitamin K prophylaxis
can be stopped earlier if breastfeeding accounts for less than fifty percent of the
daily intake.15 Formula-fed infants only receive vitamin K prophylaxis directly
after birth.Thereafter, they are expected to receive sufficient amounts of vitamin
K, since the formula feedings commercially available in the Netherlands contain
approxim ately 50 |_ig vitamin K per litre.16
48
B le e d in g in in fa n ts w ith A1AT d e fic ie n c y
Vitamin K deficiency
We calculated the prothrombin ratio (PR) at initial diagnosis to be able to com ­
pare coagulation parameters from different hospitals, and used it to assess the
presence ofVKD. In case of a prothrombin time (PT) in seconds the PR was deter­
mined asfollows: PR = PTpatient / PTcontrol. If a PTcontrol was not determined
by the laboratory, it was defined as the mean of the provided reference range.
If not available, the International Normalized Ratio (INR) was used as PR. VKD
was defined as a PR above 1.5 in line with our previous study.11 VKD bleeding
was defined as bruising, bleeding or ICH in combination with a PR of more
than 4 in any infant between eight days and six months of age and normalizing
after administration of vitamin K.9 A diagnosis of ICH was made only if there
was radiological confirmation. Clinical signs suggesting ICH (high pitched cry,
irritable, convulsions) but without radiological confirmation were also listed, but
alone were not taken as diagnostic. Life threatening VKD bleeding was defined
as either an ICH or isolated respiratory and /or circulatory insufficiency.
Statistical analysis
We performed a student t-test for clinical and biochemical parameters with a
normal distribution pattern to test for statistical differences between groups.
Kruskal-Wallis analysis was used for parameters with a non-normal distribution.
A Chi Square test was performed to determine statistical significance between
groups in case of dichotomous parameters. The relative risks and 95% confi­
dence intervals (Cl) for VKD bleeding and biochem ical levels o fV K D were
calculated. Since computation of a RR requires at least one positive case in
each quadrant a dummy case was added if necessary to be able to compute a
(minimal) relative risk. Approval for the study was obtained from the University
Medical Center Utrecht ethics committee.
Results
Between January 1991 and December 2006, 56 patients with A iA T deficiency
were added to the Dutch A iA T deficiency registry. Fifty-five infants were ZZ,
one was SZ. Sixteen infants did not meet the inclusion criteria for this study,
49
C h a p te r 4
as described in detail in table 4.1. Twenty (50%) of the remaining 40 infants
were exclusively breast-fed. Infants were categorized according to the type of
prophylaxis they received. Table 4.2 summarizes the clinical characteristics of
breast-fed infants on daily oral prophylaxis and of formula-fed infants.
Table 4.1: Patients and population
Live births
No. of documented cases
Incidence of documented cases
Excluded
Positive family history, no cholestasis
Positive family history, cholestasis
Gestational age<37 weeks
Diagnosed prior to 1st week
Unknown feeding type
Included
Breast-fed
Formula- fed
3,530,583
56
1.6/100,000
5
2
3
4
2
40
20
20
*derived from the central bureau o f statistics (http://statline.cbs.nl)
Table 4.2: Patient characteristics
Total
Male sex-no. {%)*
Birthweight-g*
Age at diagnosis-days§
Weight at diagnosis-g*
Bilirubin total-umol/l*
Bilirubin direct-umol/l*
ASAT-U/I§
ALAT-U/I§
GGT-U/I
28 ( 70)
32851548
35 ( 28^ 7)
3907±660
107±41
69±32
98 ( 78- 141)
57 ( 39-83 )
556 (315-874)
Breast-fed + 25ug vitamin K
daily oral n=20
16 (80)
Formula-fed
n=20
12 ( 60)
3388±523
31 (27-42 )
4134±736
104±34
63±25
95 (83- 138)
56 ( 31- 72)
567 ( 230-875)
3193±554
39 ( 30- 60)
3660±483
177±55
74±38
100 (82- 141)
6 1 (43- 108)
584 ( 335-918)
P-value
0.30
0.27
0.10
0.09
0.36
0.35
0.70
0.22
0.81
* Plus-minus values are means ±SD. p value was determined w ith independent sample t-test. § Values are median
(interquartile range), p value was determined w ith Mann Whitney U. *P-value was determined w ith Fisher Exact.
Vitamin K deficiency in AiATdeficient infants at initial presentation
VKD was evident in 16 of 20(80%) exclusively breast-fed infants with A iA T
deficiency and was associated with VKD bleeding in 15/20 (75% ).(table 4.3)
The PR in patients presenting with VKD bleeding was above the upper limit of
quantification in 13/15 cases.(table 4.4) Life threatening consequences were
found in 8/20(40%) infants. One infant presented with respiratory insufficiency
50
B le e d in g in in fa n ts w ith A1AT d e fic ie n c y
without documented ICH. In seven infants an ICH was documented and one of
these infants died as a consequence. Additionally, two infants had symptoms
suggestive of an ICH (convulsions in 1, irritability in 1), without radiological
confirmation.
Table 4 .3 : Risk of vitamin K deficiency in breast-fed and formula-fed infants with alpha -1
antitrypsin deficiency.
Vitamin K deficiency*
Vitamin K deficiency
bleeding
Intracranial
haemorrhage
Breastfed + 25 ug vit
K daily oral
Formula
fed
Pvalue
16/20 (80%)
15/20 ( 75%)
1/20 (5%)
0/20 (%)
<0.001
<0.001
Breastfed + 25ug vit K daily oral
vs formula fed
RR
(95%CI)
16.0
( 2.3- 109)
15.8 *
( 2.3- 108)*
7/20 (35%)
0/20 (0%)
0.008
7.4 *
( 1.0-54 .5)*
Data are num ber/total number (%). The p value was determined w ith Fisher exact, ^defined by PR>1.5 fo r VKD and PR>4.0
fo r severe VKD and normal thrombocyte count. * a positive dummy was added in the formula group to compute a
(minimal) RR.
The risk of (severe) VKD was significantly lower in infants receiving formula
feeding (p<o.ooi). None of the infants receiving formula presented with a
VKD bleeding, and only one had a (mild) VKD. The minimal relative risk for
VKD bleeding in breast-fed versus formula-fed infants was 15.8 (95% Cl 2.3108).This calculation was performed after adding a positive dummy, based on
the assumption that the next included patient would have received formula
and presented with a VKD bleeding. Breast-fed infants presented earlier than
formula-fed infants, although this was not statistically significant (table 4.2)
This is in line with our previous study of BA infants.11 This indicates that the
high rate ofVKD bleeding in breast-fed infants with A iA T defidnecy accelerated
recognition of cholestasis, and argues against delayed recognition as a cause
for the higher risk of bleeding.
Cholestatic jaundice and the risk ofVitam in K deficiency bleeding
Mean total and conjugated bilirubin levels in the group of 20 breast-fed
infants, as shown in table 4.2, were 104 (± 34) |_imol/L and 63 (± 25) |_imol/L,
respectively and did not correlate with the risk of VKD or VKD bleeding.
51
C h a p te r 4
Instead, total bilirubin levels in breast-fed infants presenting with an ICH (n=7)
were significantly lower than in infants without ICH (n=i3), (80 vs 116 |_imol/l,
p = 0 .0 2 ).
Table 4.4 Infants with A iA T deficiency presenting with biochemical vitam in K deficiency.
gender
Age
(days)
bilirubin
(conjugated/
total)
PR
APTT
(sec)
M
29
52/96
6.2
V
M
41
35
56/73
86/106
M
20
M
other
bleeding sites
Neurological
outcome
86
Normalisation ICH
after vit. K
(PR if
available)
+
1.1
-
>10
>10
>240
>300
1.4
1.0
+
+
54/79
>10
>100
+
+
melena,
hematemesis
umbilical,
melena
31
45/81
>10
>100
0.9
+
normal up to
ly r
epilepsy
epilepsy and
PMR
aggression,
development
al delay
coma, died
M
21
52/70
>10
>240
0.9
+
V
37
30/54
>10
173
1.5
+
-
pyramidal
tract signs,
normal
development
hemianopsia
M
28
69/112
>10
>240
1.1
+/- irritable
hematomas
normal
M
37
90/100
>10
>200
+
±r
convulsion
normal
M
27
30/59
>10
>150
1.1
M
13
36/84
6.4
170
1.5
-
hematemesis,
hematomas
bleeding after
vena
puncture
umbilical
M
47
31/110
>10
>150
1.1
-
normal
M
47
132/174
>10
181
1.3
M
42
85/141
>10
>180
1.0
M
50
61/127
>10
171
+
bleeding after
vena
puncture$
oozing
scratchmark
on hand
hemotomas,
hematemesis
hematomas
V
28
75/80
4.8
132
1.0
v *
35
70/96
1.8
49
#
-
bleeding after
vena
puncture
bleeding after
vena
puncture
* Formula fed; # isolated reduction o f vitamin K dependent coagulation factors;$ respiratory insufficiency.
52
normal
normal
normal
normal
normal
B le e d in g in in fa n ts w ith A1AT d e fic ie n c y
Discussion
The results of this study indicate that exclusively breast-fed infants with A iA T
deficiency have a high risk of developing a VKD bleeding before recognition of
cholestatic jaundice underthe present Dutch vitamin K prophylaxis. Of breast­
fed infants, 75% presented with a VKD bleeding and 35% presented with an
ICH. No VKD bleeding was documented in (partially) formula-fed infants.
These findings extend our previous observations in infants with BA11 to a
disease entity with a milder and more variable degree of conjugated hyper­
bilirubinemia. To our knowledge this is the first nation-wide study examining
the risk of VKD bleeding in infants with A iA T deficiency. Based on the premise
that A iA T phenotyping is a regular part ofthe workup of cholestatic jaundice
in infancy we investigated the databases from the diagnostic centers performing
A iA T phenotyping in the Netherlands. Out of a birth cohort of 3,138,576 infants
55 cases of Pi ZZ (and one with Pi SZ) were identified. Fifty ofthe Pi ZZ infants
had documented cholestatic jaundice, amounting to an incidence of 1.6/100,000.
This incidence is very similar to the expected incidence of 1.3/100,000, which
is based on the estimated incidence of Pi ZZ in the Netherlands (i:g 5 36 )17 and
the previously documented risk of developing cholestatic jaundice in infancy
in case of Pi ZZ phenotype of i2 % 13.
Since we aimed to estimate the risk of late VKD bleeding under the current
Dutch prophylaxis we carefully excluded infants who may have received addi­
tional vitamin K, such as premature infants. Another source of misclassification
may originate from the assumption that all breast-fed infants indeed received
the prophylaxis according to national guidelines, unless otherwise stated in
patient files. However, data from other European countries indicate that adher­
ence to a prophylactic regimen executed by parents is approximately 85%.8'18'19
In the group of breast-fed infants with A iA T deficiency, total and conjugated
bilirubin levels (104/63 |_imol/L) were significantly lowerthan those in breast-fed
babies with BA in the Netherlands (147/98 |_imol/L, p <o .ooi).n Nevertheless,
the risk of developing VKD bleeding was similar in the two groups (75 and 83%,
respectively, p=o.4g), as was the age at presentation (35 and 36 days, respec­
tively). No association was found in the present study between the degree of
(conjugated) hyperbilirubinemia and the risk ofVKD bleeding. These findings
53
C h a p te r 4
suggest that infants with cholestatic jaundice are at a similarly high risk
of VKD bleeding, regardless of its degree. Although counter-intuitive, this
notion is supported by other lines of evidence. First, VKD bleeding has been
repeatedly reported in infants with minimal cholestatic jaundice (total bilirubin
21-62 |_imol/L).20-23 Second, in a Japanese study investigating the incidence of
subclinical VKD, highly elevated PIVKA-II levels were found in 8/22,000 infants
at age 4 weeks, all of whom were breast-fed and had moderately elevated
conjugated bilirubin levels (range 17-45 |-imol/l).24The incidence of cholestatic
jaundice found in this study was 1/2750, similar to the reported incidence of
(transitory) cholestasis in infancy25, suggesting that nearly all breast-fed infants
with some degree of cholestatic jaundice developed coagulation abnormalities.
It could be argued that infants with underlying cholestatic liver disease
should be protected against VKD bleeding by early diagnosis and appropriate
treatment, rather than by routine prophylaxis. Indeed, 2/7 cases of ICH in
this report could have been prevented if doctors had rightly regarded the
harmless-looking (warning) bleeds as a medical emergency and reacted with
prompt parenteral administration of vitamin K. Moreover, all infants with
A iA T deficiency that developed VKD bleeding had bilirubin levels that are
recognisable on physical examination, suggesting that earlier recognition of
jaundice could help prevent VKD bleeding. However, various attempts aimed
at increasing awareness and early recognition of cholestatic jaundice by
parents26 and health professionals27-28 had only a modest effect on the age
at diagnosis.
It has been recently suggested that the decreasing incidence of VKD bleeding
in the UK is due to increased awareness of the importance of jaundice beyond
two weeks of life29, although the effect of this program on the age at presen­
tation of cholestatic jaundice has not (yet) been published. Our data in infants
with A iA T deficiency and BA11 indicate that the age at diagnosis would have
to be reduced by approximately 3 w eeksto prevent all bleedings.Theoretically,
newborn screening would be an attractive option, but quantification of serum
bile acids failed to separate infants with cholestatic jaundice from healthy
infants.30Therefore, underthe present circumstances adequate routine vitamin
K prophylaxis seems to be the most reliable way to protect all infants with
54
B le e d in g in in fa n ts w ith A1AT d e fic ie n c y
unrecognized cholestasis. In previous studies vitamin K prophylaxis was shown
not only to reduce the risk of VKD bleeding but also to postpone its
occurrence.5-10-29
Interestingly, the age at presentation in the cohort described here is similar
to the age reported in infants who received only a single oral dose at birth.
Therefore, it is doubtful whetherthe daily oral administration of 25 microgram
vitamin K in the first 3 months of life, as is customary in the Netherlands, had any
effect at all in infants at the highest risk of VKD bleeding. If oral prophylaxis
is to be continued in the Netherlands it seems prudent to choose a dose and
formulation with proven efficacy. However, the formulation which significantly11
protected cholestatic infants is no longer commercially available. As surveillance
data indicate that a mixed micellarformulation is equally effective,5-7 we suggest
using this.
In conclusion, the absolute risk ofVKD bleeding in exclusively breast-fed infants
with cholestatic jaundice due to A iA T deficiency is similar to that in infants
with BA and does not correlate with the degree of conjugated hyperbili­
rubinemia. These findings suggest that infants with cholestatic jaundice,
regardless of etiology, are at a high risk o fV K D bleeding under the current
Dutch regimen of vitamin K prophylaxis and change is needed. Meanwhile,
doctors should be aware that early recognition of cholestatic jaundice followed
by prompt treatment with vitamin K can reduce the risk of life threatening
VKD bleeding in breast-fed infant.
References
1.
W iddershoven J, Lam bert W, Motohara K, Monnens L, de Leenheer A,M atsuda I, et al. Plasma
concentrations of vitam in K i and PIVKA-II in bottlefedand breast-fed infants with and w ithout
vitam in K prophylaxis at birth. Eur J Pediatr 1988;148:139-42.
2.
Shearer MJ, Barkhan P, W ebster GR. A bsorption and excretion of an oral dose o f tritiated
vitam in K i in man. Br J Haem atol 1970;18:297-308.
3.
McNinch AW, Upton C, Sam uels M, Shearer MJ, M cCarthy P, Tripp JH, et a I. Plasma concen­
trations after oral or intram uscular vitam in K i in neonates. ArchD is Child 1985;60:814-8.
4.
Haroon Y, Shearer MJ, Rahim S, Gunn WG, M cEnery G, Barkhan P. Thecontent of phylloquinone
(vitamin K i) in human milk, cows' m ilk and infantfformula foods determined by high-perform ance
liquid chrom atography. J Nutr 1982;112:1105-17.
55
C h a p te r 4
5.
von Kries R, Hachm eister A, Gobel U. Oral mixed m icellar vitam in K for prevention of late
vitam in K deficiency bleeding. Arch Dis Child Fetal Neonatal Ed 2 0 0 3 ;8 8 :F i0 9 -F ii2 .
6.
von Kries R, Hachm eister A, Gobel U. Can 3 oral 2 mg doses of vitam in K effectively prevent
late vitam in K deficiency bleeding? Eur J Pediatr 1 9 9 9 ^ 58 Suppl 3:8183-6.
7.
Schubiger G, Berger TM, W eber R, Banziger O, Laubscher B. Prevention of vitam in K deficiency
bleeding with oral mixed m icellar phylloquinone: results of a 6-year surveillance in Sw itzerland.
E u rJ Pediatr 2003;162:885-8.
8.
W ariyar U, Hilton S, Pagan J, Tin W, Hey E. Six years' experience of prophylactic oral vitam in K.
A rch Dis Child Fetal Neonatal Ed 2000;82:F64-F68.
9.
Sutor AH, von Kries R, Cornelissen EA, McNinch AW, A ndrew M. Vitam in K deficiency bleeding
(VKDB) in infancy. ISTH Pediatric/Perinatal Subcom m ittee. International Society on Throm bosis
and Haem ostasis. Throm b Haem ost 1999;81:456-61.
10.
McNinch AW, Tripp JH. H aem orrhagic disease of the newborn in the British Isles: two year
prospective study. BMJ 1991;303:1105-9.
11.
van Hasselt PM, de Koning TJ, Kvist N, de VE, Lundin CR, Berger R, et al. Prevention of vitam in
K deficiency bleeding in breastfed infants: lessons from the Dutch and Danish biliary atresia
registries. Pediatrics 2 008 ;i2 i:e 8 57-e 8 6 3.
12.
Spivak W, Sarkar S, W inter D, Glassm an M, Donlon E, Tucker KJ. D iagnosticutility of hepa­
tobiliary scintigraphy with ggm Tc-DISIDA in neonatal cholestasis.J Pediatr 1987;110:855-61.
13.
SvegerT. Liver disease in a lph ai-an titrypsin deficiency detected by screening of 200,000
infants. N E n glJ Med 1976;294:1316-21.
14.
Odievre M, Martin JP, Hadchouel M, A lagille D. A lp hai-an titrypsin deficiency and liver disease
in children: phenotypes, m anifestations, and prognosis. Pediatrics 1976;57:226-31.
15.
Uitentuis J. [Adm inistration of vitam in K to neonates and infants]. NedTijdschr Geneeskd
1990;134:1642-6.
16.
G reer FR, Marshall SP, Foley AL, Suttie JW. Im proving the vitam in K status of breastfeeding
infants with m aternal vitam in K supplem ents. Pediatrics 1997;99:88-92.
17.
Blanco I, de Serres FJ, Fernandez-Bustillo E, Lara B, M iravitlles M. Estim ated num bers and
prevalence of PI*S and PI*Z alleles of alph ai-an titrypsin deficiency in European countries.
Eur Respir J 2006;27:77-84.
18.
Hansen KN, M inousis M, Ebbesen F. W eekly oral vitam in K prophylaxis in Denmark. Acta
Paediatr 2003;92:802-5.
19.
C roucher C, Azzopardi D. Com pliance with recom m endations for giving vitam in K to newborn
infants. BMJ 1994;308:894-5.
20.
Sutor AH, Dagres N, N iederhoff H. Late form of vitam in K deficiency bleeding in Germany.
Klin Padiatr 1995;207:89-97.
21.
Ono S, Tokiwa K, A oi S, Iwai N, Nakanoin H. A bleeding tendency as the first sym ptom of a
choledochal cyst. Pediatr Surg Int 2000;16:111-2.
22.
von Kries R, Kordass U, Shearer M, Gobel U. Idiopathic late hem orrhagic disease of newborn
and conjugated hyperbilirubinem ia. J Pediatr G astroenterol Nutr 1993;16:328-30.
56
B le e d in g in in fa n ts w ith A1AT d e fic ie n c y
23.
Humpl T, Bruhl K, Brzezinska R, Hafner G, Coerdt W, Shearer MJ. Fatal late vitam in K-deficiency
bleeding after oral vitam in K prophylaxis secondary to unrecognized bile duct paucity. J Pediatr
G astroenterol Nutr 1999;29:594-7.
24.
M atsuda I, Nishiyam a S, M otohara K, Endo F, Ogata T, Futagoishi Y. Late neonatal vitam in K
deficiency associated with subclinical liver dysfunction in human milk-fed infants. J Pediatr
1989;114:602-5.
25.
Dick MC, M owat AP. Hepatitis syndrom e in infancy--an epidem iological survey with 10 year
follow up. Arch Dis Child 1985;60:512-6.
26.
M atsui A, Ishikawa T. Identification of infants with biliary atresia in Japan. Lancet i9 94;343:9 25.
27.
Hussein M, Howard ER, Mieli-Vergani G, M owat AP. Jaundice at 14 days of age: exclude biliary
atresia. Arch Dis Child 1991;66:1177-9.
28.
Hsiao CH, Chang MH, Chen HL, Lee HC, Wu TC, Lin CC, et al. Universal screening for biliary
atresia using an infant tool color card in Taiwan. Hepatology 2008;47:1233-40.
29.
M cNinch A, Busfield A, Tripp J. Vitam in K deficiency bleeding in Great Britain and Ireland: British
Paediatric Surveillance Unit Surveys, 1993 94 and 2001-02. Arch Dis Child 2007;92:759-66.
30.
Mushtaq I, Logan S, Morris M, Johnson AW, W ade AM, Kelly D, et al. Screening of newborn in­
fants for cholestatic hepatobiliary disease with tandem mass spectrom etry. BMJ 1999;319:471-7.
57
Chapter 5
Influence of Alpha-i Antitrypsin
Heterozygosity on Treatment
Efficacy of HCV Combination Therapy
European Journal of Gastroenterology and Hepatology 200g Sep 29.
[Epub ahead of print]
Karin F Kok
Hanneke van Soest
Teun (A) E van Herwaarden
Martijn GH van Oijen
Greet J Boland
Juliane Halangk
Thomas Berg
Richard A deVries
Joost PH Drenth
Abstract
Background: The role of heterozygosity for alp h a-i anti­
trypsin (AiAT) alleles in patients with chronic hepatitis C
virus (HCV) is unclear.There is limited evidence to suggest
that there is an increased prevalence of heterozygous A iA T
carriers in HCV, but it is unclear how this affects treatment
success.
Aim: To investigate the (1) prevalence of A iA T heterozy­
gosity among 2 HCV cohorts and (2) its effect on treatment
outcome.
Methods: We performed a retrospective cohort study using
2 different cohorts. Cohort 1 consists of 678 German HCV
patients, 507 of them were treated for HCV with standard
therapy. Cohort 2 consists of 370 Dutch HCV patients of
which 252 HCV patients were part of a clinical trial (treat­
ment with amantadine or placebo, in combination with
PEG-interferon alfa-2b and ribavirin) while 37 HCV patients
received standard therapy. We analyzed A iA T status using
direct sequencing of the A iA T gene (cohort 1) or iso-electro
focusing of serum (cohort 2). In addition we measured
A iA T serum levels (cohort 2).
Results: In total we included 1048 HCV patients; 986 (94%)
were wildtype (Pi MM), while 61 (6%) were heterozygous
for a mutant A iA T allele (41 Pi MS, 20 Pi MZ). Mean A iA T
serum levels (370 patients) were lower in A iA T heterozygous
patients (1.68 g/l vs. 1.36 g/l), (p<o.o5) compared to wildtypes.
Sustained viral response (SVR) after treatment was equal
between the wildtypes and heterozygotes. (54% vs 56%)
Conclusion: We found a heterozygosity rate of 0.06, in line
with healthy controls in other studies. Serum A iA T levels
from A iA T heterozygous HCV patients are significantly lower
compared to wild type patients, although they do not dis­
criminate on an individual level. Finally, SVR in A iA T w ild­
types was not different from SVR in A iA T heterozygotes.
A1AT h e te ro z y g o s ity in HCV p a tie n ts
Introduction
A iA T deficiency is a hereditary disease that is characterized by the hepatic
synthesis of an abnormal A iA T protein that cannot be released into the plasma
completely. Accumulation of mutant A iA T protein in hepatocytes causes tissue
damage and eventually ensues in liverand pulmonary disease.The A iA T protein
is encoded by the protease inhibitor (Pi) locus located on the long arm of
chrom osom e i4 q 3 i.2 . This locus is highly polymorphic, and so far more
than 100 different A iA T isotypes have been identified. They can be detected
by iso-electrophoresis of the protein or by mutational analysis of the allele.1
The wild type allele (Pi MM) results in a functionally normal protein with
normal serum A iA T levels. In W estern Europe there are 2 frequent variants
P.G342K (denoted as Z allele) and the P.G264V (commonly referred to as the
S allele).2 For example, in the general Dutch population the allele frequency
for these A iA T variants is estimated to be 0.02 (Z allele) and 0.04 (S allele).3-5
The Z allele variant results in polymers that are retained in hepatocytes while
the S variant of the A iA T molecule has less deleterious effect on the protein.6
Hom ozygosity for A iA T deficiency is associated with liver and pulm onary
disease, but the picture is less clearfor A iA T heterozygosity.7-11
Hepatitis C virus (HCV) is mainly transmitted through contact with blood and
blood products. More than 80% of all HCV-infected patients will develop
chronic hepatitis, and in 20% this will lead to liver cirrhosis.12 Treatment
of HCV aims to eliminate the virus and current standard treatm ent regimen
consist of PEG-interferon and ribavirin.13 Treatment success greatly depends
on viral characteristics such as HCV genotype, and viral load but also host
related factors play a role. For example, presence of the ACC IL10 promoter
diplotype increasesthe likelihood of reaching a sustained viral response (SVR)
by more than 3-fold.14
If we focus on the role of A iA T in HCV, studies have merely investigated the
prevalence of A iA T heterozygosity in HCV induced liver disease. Studies in
pre-transplantation HCV patients demonstrated presence of Pi MZ in 10-13%
of the patients, compared to 2.8% in controls; suggesting that A iA T hetero­
zygosity influenced the development and/or course of the liver disease in
HCV.10-11 A case-control study in Egyptian HCV patientsfound a relatively high
frequency of Pi M S.15 In contrast, other studies failed to conform th is.16-18
61
C h a p te r 5
In vitro studies showed that A iA T plays a central role in inflammation, both as
a regulator of proteinase activity and as a signaling molecule for the expression
of pro-inflammatory m olecules.19 Also, A iA T has a role in viral clearance, as it
inhibits human im munodeficiency virus type 1 replication.20 We hypothesize
that A iA T heterozygosity puts the host at a higher risk for HCV and decreases
viral elimination after treatment because of the decreased anti-inflam matory
response. As a consequence, A iA T genotype status might be associated with
the infection rate and treatm ent outcome.
Therefore the aim of our study was (1) to investigate the prevalence of A iA T
heterozygosity in a large cohort of Dutch and German HCV patients and to
assess (2) its consequence on treatm ent outcome.
Methods and materials
Patients
We selected two different retrospective cohorts of patients. Cohort 1 consists
of 678 German HCV patients; 507 of them were treated for HCV and completed
therapy. Cohort 2 consists of 370 Dutch HCV patients; 289 patients of cohort 2
were treated and completed therapy. Patients were included regardless of
their degree of liver fibrosis. We did exclude patients with a hepatitis B or HIV
co-infection. Our study was approved by the ethical committee.
HCV Treatment
Patients from cohort 1 were treated with interferon monotherapy, a com bina­
tion of interferon and ribavirin or with PEG-interferon in combination with
ribavirin. A total of 252 patients of cohort 2 were part of a nation-wide, double
blind, placebo-controlled, randomised multicentre study (CIRA-study).21 Pa­
tients were treated with amantadine or placebo, in all cases combined with
weight-based ribavirin and high-dose interferon induction therapy followed
by PEG-interferon alfa-2b. The remaining 37 patients (cohort 2) were treated
with standard combination therapy. Treatment was given for 24-52 weeks de­
pending on genotype. SVR was defined as a negative qualitative HCV RNAtest
at 1 year after end of treatment.
62
A1AT h e te ro z y g o s ity in HCV p a tie n ts
Assays
A iA T status was analyzed in all patients. We performed mutational analysis
(cohort 1) and iso-electric focusing and nephelometric measurement of A iA T
serum levels (cohort 2).
Mutational analysis (Cohort 1)
DNA was extracted from peripheral blood leukocytes. Primer sequences
used for polymerase chain reaction (PCR) were as follows: exon 5 (PiS), 5'GATGAGGGGAAACTACAGCACCTCG-3'and5'-GGGCCTCAGTCCCAACATGG-3'
and for exon 7 (PiZ), 5'-GCATAAGGCTGTGCTGACCATCGTC-3' and s'-AGGGTTTGTTGAACTCGACCTC-3'. An automated thermal cycler (Biometra, Gottingen,
Germany) was used. A iA T heterozygosity was determined using restriction
fragment length polymorphism (RFLP). Amplification of the desired products
was confirmed by direct DNA sequencing. DNA sequences were analyzed by
sequencing both strands.The reaction products were loaded into an ABI 373A
fluorescence sequencer (Applied Biosystems, Weiterstadt, Germ any).22
Iso-electric focusing (Cohort 2)
Iso-electric focusing was performed in serum by iso-electric focusing (Phastsystem, GE Healthcare, München, Germany) and subsequent protein silver
staining. Patterns were compared to MM, MZ, MS, SZ and ZZ controls. MS
and MZ designated patterns were subsequently validated by an iso-electric
focusing with an immuno-fixation method in an independent laboratory.23-24
Nephelometric plasma measurement (Cohort 2)
Determination of the serum A iA T level was performed by nephelometry. The
A iA T reagent was used from the Immage Immunochemistry system (Immage
Beckmann-Coulter, W oerden, The Netherlands), and a calibrator with an
assigned A iA T value and a human quality control sample (Biorad, Liquichek,
Hercules, CA, USA) were used.25
HCV testing
HCV RNA testing was performed qualitatively (Cobas amplicor HCV MonitorTest,
version 2.0, detection limit 600 lU/ml, Roche Diagnostics) and quantitatively at
63
C h a p te r 5
weeks 24, 52 and 104 (Cobas Am plicor HCVtest, version 2.0; detection limit 50
lU/ml, Roche Diagnostics). Genotyping was performed by sequence analysis
of the 5' untranslated region of the HCV genome by the Visible Genetics
TrueGene Hepatitis C Assay.
Statistical Analyses
The HCV characteristics (genotype) of the wild type patients and heterozygous
patients were analyzed using Pearson chi-squared test and Fisher's exact test
where appropriate. The difference in age at start of treatm ent was analyzed
using student t-test. Differences between serum levels of the wild type patients
(Pi MM) and heterozygous patients (Pi MS, Pi MZ and Pi SZ) were analyzed using
student t-test. We used Pearson chi-squared test for analysis of the differences
in SVR between wild types and heterozygous patients and for analysis of the
degree of fibrosis in wild types and heterozygotes. All statistical analyses were
performed using GraphPad Prism 4 (Version 4.02, GraphPad Software Inc.,
San Diego, CA, USA). Atwo-sided p-value of < 0.05 was considered statistically
significant.
Results
Genotyping
Genotyping was performed in all 1048 patients. Due to technical difficulties,
iso-electro-focussing failed in a single patient, resulting in 1047 complete
analyzed patients. We found 986 Pi MM (94%), 41 Pi MS (4%) and 20 Pi MZ
(2%) patients. (Figure 5.1) Genotype results were within the in Hardy-Weinberg
equilibrium.
Serum levels
A iA T levels were measured in serum from 370 HCV patients from cohort 2
drawn before treatm ent. A iA T serum levels were higher in Pi MM carriers (340
patients, mean 1.69 g/l, 95%CI 1.66-1.71 g/l) compared to Pi MS carriers (22
patients, mean 1.43 g/l, 95%CI 1.34-1.53 g/l) or Pi MZ carriers (7 patients, mean
64
A1AT h e te ro z y g o s ity in HCV p a tie n ts
Figure 5.1: Prevalence of different A iA T carrier rates in 1047 patients w ith HCV.
Pi MZ
,1.9%
Pi SZ
. 0 . 1%
1.14 g/l, 95%CI 0.80-1.49 g/l) respectively. The differences between the wild
types and heterozygous patients were statistically significant (p<o.oi); this also
holds for the differences between the heterozygous Pi M Sand Pi MZ patients
(p<o.oi). (Figure 5.2) We detected A iA T serum levels above 1.0 g/l in all hetero­
zygous Pi MS patients and in 5/ 7 (71%) of the heterozygous Pi MZ patients.
Figure 5.2: A iA T serum levels in 370 HCV patients arranged to the different A iA T phenotypes.
(p<o.o5 for all com parisons)
3.0«
£
aj
to
1 . 0*
O.fr
MM
MS
MZ
n=341
n=22
n=7
65
C h a p te r 5
Degree of fibrosis
We assessed the stage of fibrosis using the Metavir score in liver biopsies
obtained from 409 patients.There was no statistically significant difference in
the degree of fibrosis between wildtypes and heterozygous patients, (table 5.1)
Table 5.1: Degree of fibrosis in the different A i AT groups (n = 409)
Metavir score *
F0 - F2
F3 - F4
Pi MM
n= 383 (%)
269 ( 70)
114 ( 30)
*Fo = no fibrosis; F i = minimal fibrosis,
F4 = cirrhosis or advanced fibrosis
Pi MZ and Pi MS
n=26 (%)
19 (73)
7 ( 27)
F2 = periportal fibrosis; F3 = bridging fibrosis;
HCV characteristics and treatm ent response
A total of 796 patients completed HCV treatm ent. A iA T carrier status and
qualitative HCV RNA 1 year after end of treatm ent were not available for 10
patients (n = i cohort i;n=g cohort 2). These patients were excluded from fur­
ther analysis; and data from 786 HCV patients remained available for analysis.
The distribution of the HCV genotypes at baseline did not differ among the
groups with different A iA T genotypes. (Table 5.2)
Table 5.2: Distribution of the HCV genotypes and age in the different A iA T states (11=795)
Pi MM
n=745 (%)
Pi MZ
n =17 (%)
Pi MS
n =32 (%)
Pi SZ
n = l (%)
3
4
450 ( 60,4 )
5 6 ( 7,5 )
170 ( 22,8 )
34 (4 , 6)
10 (58,8 )
1 (5 ,9 )
5 (29,4 )
1 (5 ,9 )
20 ( 62,5 )
4 ( 12,5 )
4 ( 12,5 )
2 ( 6, 3)
6
unknown
Mean age (yrs)
1 (0,1)
34 (0,46 )
46,6
0(0 )
0(0 )
0(0 )
2 ( 6, 3)
0(0)
0 (0 )
0 (0 )
0 (0 )
0 (0 )
1 (100)
47,9
45,5
59
HCV genotype
1
2
* Pi MM vs Pi MZ, Pi MS and Pi SZ
66
p value *
p=0,89
p=0,93
A1AT h e te ro z y g o s ity in HCV p a tie n ts
SVR was reached in 395/736 (54%) wild type patients, in 8/17 (47%) A iA T Pi MZ
patients and in 20/32 (62%) of A iA T Pi MS patients. One Pi SZ patients did not
achieve SVR. These differences were not statistically significant. (Figure 5.3)
The addition of am antadine to the com bination of interferon and ribavirin in
patients from cohort 2 did not affect treatm ent response.26
Figure 5.3: HCV treatm ent response in 736 A iA T w ildtype (Pi MM) patients, 17 A iA T heterozygous Pi MZ
patients, 32 A iA T heterozygous Pi MS patients and 1 Pi SZ patient.
Pi MM
Pi MZ
Pi MS
A1AT Phenotype
Pi SZ
Discussion
This study demonstrates that the efficacy of HCV combination treatm ent with
(PEG)-interferon and ribavirin is independent of carrier state of AiAT, as
heterozygous (Pi MZ / Pi MS / Pi SZ) HCV patients had a similar treatment
response compared to wild type patients (Pi MM). In addition, we did not
detect an association between A iA T heterozygosity and the degree of liver
fibrosis. We found a high rate of SVR in our study population, this can be
explained by the fact that we included only patients who finished treatment
with (PEG)-interferon and ribavirin; we did not include patients who stopped
treatm ent for any reason or who were lost to follow-up.
67
C h a p te r 5
There are no previous studies documenting HCV treatm ent response in A iA T
heterozygosity. Most of the studies in this field have focused on apparent
differences in A iA T allele frequency between those with and without a certain
liver disease.10-11 The common denominator is that a higher prevalence of
heterozygotes is found in different causes of liver disease; this is particularly
true for HCV and alcoholic liver disease in pre-transplantation patients.10-11
We could not confirm the finding of higher A iA T heterozygosity rates in HCV
patients, and our data indicate that the A iA T carrier distribution in HCV
is in line with the prevalence of control populations in the Netherlands
and Germany.3-5 Our results are at odds with an Egyptian study that found
higher prevalence of Pi MS in HCV cirrhotic patients compared with normal
controls.15 This particular A iA T allele is rare in Germany and the Netherlands.27
It is tem pting to speculate why other case control studies in this field
documented differences in A iA T carrier rate between patient and controls.
We surmise that there are 3 possible explanations. First, the variation might
depend on real population differences. On the other hand, it cannot be
excluded that bias in selecting controls and patients might have affected
the results. In addition, the low prevalence of patients with heterozygote
genotypes might have introduced a type II error giving rise to spurious results.
Serum A iA T levels were statistically significantly lower in heterozygous carriers
compared to A iA T wild type patients. There was no cut-off point that discrimi­
nated between heterozygotes and wild types. In some laboratories in the
Netherlands, genotyping is only performed if the serum level is below 1,0 g/l.
This strategy will miss obvious cases; in our data set, all Pi MS heterozygotes
and 71% of the Pi MZ heterozygotes. Previous studies also indicated the
variability of A iA T serum levels in heterozygous carriers.28-29 Some case
reports document A iA T heterozygotes with concomitant HCV30 or alcoholic
liver disease31 who had consistent normal A iA T serum levels but A iA T globules
in their liver biopsy specimen. It is possible that the liver inflammation is a driver
for A iA T production, hence giving rise to 'falsely' elevated A iA T serum levels.
Our study was adequately powered to detect differences in treatment outcome
and to calculate the prevalence of A iA T heterozygosity in HCV patients in
the Netherlands and Germany. Nevertheless, our study comes with several
68
A1AT h e te ro z y g o s ity in HCV p a tie n ts
limitations. The lack of a control population does not allow a firm estimate
of the allele frequency in relation to the background population.
While A iA T does not a play a role in the HCV treatm ent response, it is possible
that there are other important genetic factors. For example, hereditary
hemochromatosis is a common genetic liver disease that appears to be over­
represented in HCV patients15. Currently, it is not known whether possession
of the mutant hereditary hemochromatosis allele affects treatment response.
In conclusion: We found no evidence for an increased prevalence of A iA T
heterozygosity in HCV patients. Serum A iA T levels from A iA T heterozygous
HCV patients are significantly lower compared to wild type patients, however
they do not discriminate on an individual level. Finally, we could not confirm
our hypothesis that HCV patients with A iA T heterozygosity deficiency had
worse outcomes on HCV combination therapy.
References
1
Kok KF, W ahab PJ, Houwen RH, Drenth JP, de Man RA, van HB, et al. Heterozygous alpha -1
antitrypsin deficiency as a co-factor in the developm ent of chronic liver disease: a review.
Neth J Med 2007 M ay;65(5):i6o-6.
2
Stoller JK, Aboussouan LS. A lphai-antitrypsin deficiency. Lancet 2005 Jun 25;365(9478):2225-36.
3
Blanco I, de Serres FJ7 Fernandez-Bustillo E, Lara B, M iravitlles M. Estim ated num bers and
prevalence o f PI*S and PI*Z alleles of alph ai-an titrypsin deficiency in European countries.
Eur Respir J 2006 Jan ;27(i):77-84.
4
Hoffmann JJ7 van den Broek WG. Distribution of alpha-i-antitrypsin phenotypes in two Dutch
population groups. Hum Genet 1976 A pr 15;32( i ):43-8.
5
Klasen EC, Biemond I, W eterm an IT. alpha l-A n titryp sin -leve ls and phenotypes in Crohn's
disease in the Netherlands. Gut 1980 O c t;2 i(io ):8 4 0 -2 .
6
Teckm an JH, Qu D, Perlm utter DH. M olecular pathogenesis of liver disease in alphai-antitrypsin
deficiency. Hepatology 1996 D e c;2 4 (6 ):i50 4 -i6 .
7
Hodges JR, M illw ard-Sadler GH, Barbatis C, W right R. Heterozygous MZ alpha l-antitrypsin
deficiency in adults with chronic active hepatitis and cryptogenic cirrhosis. N Engl J Med 1981
Mar 5;304(i o ):557-6 o .
8
Fischer HP, Ortiz-Pallardo ME, KoY, Esch C, Zhou H. Chronic liver disease in heterozygous
alph ai-an titrypsin deficiency PiZ. J Hepatol 2000 Dec;33(6):883-92.
9
Zhou H, Ortiz-Pallardo ME, KoY, Fischer HP. Is heterozygous alph a-i-antitrypsin deficiency
type PIZ a risk facto r for prim ary liver carcinom a? Cancer 2000 Jun i5 ;8 8 (i2 ):2 6 6 8 -76 .
69
C h a p te r 5
10
Eigenbrodt ML, M cCashlandTM , Dy RM, Clark J, Galati J. Heterozygous alpha i-antitrypsin
phenotypes in patients with end stage liver disease. Am J G astroenterol 1997 A pr;92(4):6o2-7.
11
Graziadei IW, Joseph J J, W iesner RH, Therneau TM, Batts KP, Porayko MK. Increased risk
of chronic liverfailure in adults with heterozygous alphai-antitrypsin deficiency. Hepatology
1998 O ct;28(4):i058-63.
12
PoynardT,Yuen MF, RatziuV, Lai C L.V iral hepatitis C. Lancet 2003 Dec 2 0;36 2(94 0i):2 095-i00.
13
Manns MP, W edem eyer H, Cornberg M. Treating viral hepatitis C: efficacy, side effects, and
com plications. Gut 2006 Sep;55(9):i350-9.
14
Morgan TR, Lam brecht RW, Bonkovsky HL, Chung RT, Naishadham D, Sterling RK, et al. DNA
polym orphism s and response to treatm ent in patients with chronic hepatitis C: Results from the
HALT-C trial. J Hepatol 2008 O ct;49(4):548-56.
15
Settin A, El-Bendary M, bo-AI-Kassem R, El BR. M olecular analysis of A iA T (S and Z) and HFE
(C282Y and H63D) gene m utations in Egyptian cases with HCV liver cirrhosis. J Gastrointestin
Liver Dis 2006 Ju n ;i5 (2 ):i3 i-5 .
16
Scott BB, Egner W. Does alph ai-an titrypsin phenotype PiMZ increase the risk of fibrosis in liver
disease due to hepatitis C virus infection? Eur J G astroenterol Hepatol 2006 M a y ;i8 (5 ):5 2 i-3 .
17
Elzouki AN, Verbaan H, Lindgren S, W idell A, Carlson J, Eriksson S. Serine protease inhibitors in
patients with chronic viral hepatitis. J Hepatol 1997 Ju l;27(i):4 2-8.
18
Serfaty L, Chazouilleres O, Poujol-Robert A, M orand-Joubert L, Dubois C, Chretien Y, et al. Risk
factors for cirrhosis in patients w ith chronic hepatitis C virus infection: results of a case-control
study. Hepatology 1997 Sep;26(3):776-9.
19
Aldonyte R, Jansson L, Janciauskiene S. Concentration-dependent effects of native and
polym erised alph ai-an titrypsin on prim ary human m onocytes, in vitro. BMC Cell Biol 2004
Mar 2 9 ;5 :ii.
20
Shapiro L, PottGB, Ralston AH. A lpha-i-antitrypsin inhibits human im m unodeficiency virus type
1. FASEB J 2001 J a n ;i5 (i):ii5 - 2 2 .
21
van SH, Boland GJ, van Erpecum KJ. Hepatitis C: changing genotype distribution with im portant
im plications for patient m anagem ent. Neth J Med 2006 A pr;64(4):96-9.
22
W itt H, Kage A, Luck W, Becker M. A lp hai-an titrypsin genotypes in patients with chronic
pancreatitis. Scand J G astroenterol 2002 M ar;37(3):356-9.
23
Jeppsson JO, Franzen B. Typing of genetic variants of alpha i-an titryp sin by electrofocusing.
Clin Chem 1982 Ja n ;2 8 (i):2 i9 -2 5 .
24
Zerim ech F, Hennache G, Bellon F, Barouh G, Jacques LJ, Porchet N, et al. Evaluation of a
new Sebia isoelectrofocusing kit for alpha i-an titryp sin phenotyping with the Hydrasys System.
Clin Chem Lab Med 20o8;46(2):26o-3.
25
Ferrarotti I, Scabini R, Campo I, O ttaviani S, Zorzetto M, G orrini M, et al. Laboratory diagnosis
of alph a(i)-an titryp sin deficiency. Transl Res 2007 N ov;i5o(5):267-74.
26
van Soest H, van der Schaar PJ, K oekG H , deV ries RA, van Ootegehem NAM, van HB, et al. No
beneficial effect of adding am antadine to standard PEG-interferon alfa-2b and ribavirin therapy
in naive chronic hepatitis C patients. 2008. Oral presentation Dutch Society for Gastroenterology
O ctobre 2nd 2008.
70
A1AT h e te ro z y g o s ity in HCV p a tie n ts
27
de Serres FJ. W orldw ide racial and ethnic distribution of alphai-antitrypsin deficiency: sum m ary
of an analysis of published genetic epidem iologic surveys. Chest 2002 N o v ;i2 2 (5 ):i8 i8 -2 9 .
28
Pittschieler K. Liver disease and heterozygous alph a-i-antitrypsin deficiency. Acta Paediatr
Scand 1991 Mar;8o(3):323-7.
29
W ard AM, W hite PA, W ild G. Reference ranges for serum alpha 1 antitrypsin. Arch Dis Child 1985
M ar;6o(3):26i-2.
30
Banner BF, Karam itsios N, Sm ith L, Bonkovsky HL. Enhanced phenotypic expression o f a lp h a-iantitrypsin deficiency in an MZ heterozygote with chronic hepatitis C. Am J G astroenterol 1998
Sep ;93(9):i54i-5-
31
Kok KF, W ahab PJ, de Vries RA. [H eterozygosity for a lph ai-an titrypsin deficiency as a cofactor in
the developm ent o f chronic liver disease]. Ned Tijdschr Geneeskd 2005 Sep i0 ;i4 9 (3 7 ):2 0 5 7 -6 i.
71
Chapter 6
Prevalence of Genetic
Polymorphisms in the Promoter
Region of the Alpha-i Antitrypsin
(SERPlNAi) Gene in Chronic Liver
Disease: a Case Control Study
BMC Gastroenterology 2010,10:22
Karin F Kok
René H te Morsche
Martijn GH van Oijen
Joost PH Drenth
Abstract
Backg round: Al p h a-i ant it rypsin(A iA T) deficiency, caused
by the Z allele (P.E342K) and S allele (P.E264V) in the
SERPINAi gene, can induce liver and pulmonary disease.
Different mechanisms appear to be responsible for the
pathogenesis of these divergent disease expressions.
The c.-ig73T>C polymorphism located in the SERPINAi
promoter region is found more frequent in A iA T deficiency
patients with liver disease compared to patients with
pulm onary disease, but data are lacking regarding con­
tribution to the development of liver diseases caused by
other etiologies.
A im :T o study the prevalence of c.-igj3T>C, Z allele and S
allele in a cohort of patients with liver disease of various
etiologies compared with healthy controls and to evaluate
its effect on disease progression.
Methods: A total of 297 patients with liver disease from
various etiologies and 297 age and gender matched healthy
controls were included.The c.-igj3T>C polymorphism and
Z and S alleles of the SERPINAi gene were analyzed by
real-time PCR.
Results: c.-igj3T> C was similarly distributed between
patients with liver disease of various origins and healthy
controls. Furthermore, the distribution of c.-igj3T> C was
independent from etiology subgroup. In patients with liver
disease mean ages at of onset of liver disease were
44.4, 42.3 and 40.7 years for the c.-1973 T/T, T/C and C/C
genotype respectively (NS). S allele heterozygosity was
increased in patients with drug induced liver injury (DILI).
(OR 4.3; 95%CI 1.1-17.2)
Conclusion: In our study, c.-igj3T> C polymorphism was
not a risk factor for liver disease of various etiologies. In
addition, S allele heterozygosity might contribute to the
development of DILI.
P o ly m o rp h is m s o f SERPINA1 gene in c h ro n ic liv e r disease
Background
A lp h a-i antitrypsin (A iAT) deficiency is a hereditary disease and can induce
end-organ damage caused by defective A iA T protein processing. Liver disease
in A iA T deficiency is caused by hepatic accumulation of the A iA T protein and
pulmonary disease is induced by an impaired protection against neutrophil
elastase due to decreased serum A iA T .^ T h e A iA T protein is encoded by the
protease inhibitor (Pi) locus located on chromosome i4q32.i(SERP/A /A igene).
In Western Europe, A iA T deficiency most commonly results from presence of
2 genetic variants P.E342K (denoted as Z allele) and the P.E264V (commonly
referred to as the S allele).3 Liver disease in A iA T deficiency has a bimodal
presentation affecting children in neonatal life4 and, less commonly, adults in
late middle life.5
It is unclear why A iA T deficiency leads to liver disease in some patients and
lung disease in others. It appears that environm ental factors are in part
responsible for this difference. Pulmonary disease develops preferably in
homozygous Pi ZZ persons who are tobacco smokers or are exposed to airway
irritants.6 Indeed, smoking is an established risk factor for lung disease as
A iA T deficient smokers will develop emphysema at considerable younger
age, while non-smokers are at risk for liver disease developing later in life.7
Some 32-37% of A iA T deficient non-smoking patients will die as a result of
A iA T deficiency induced liver disease.8
Apart from environmental factors there is some evidence that genetic factors
modify the risk for A iA T deficiency related end-organ damage. For example,
72% of siblings of probands with A iA T deficiency related liver disease suffered
from liver disease, which was concordant for severity in 29%, while 28% had no
liver involvement.9This suggests presence of genetic modifiers. Indeed a recent
study identified a novel single nucleotide polymorphism (SNP) g.i26oj6T>C
( c.-igj3T>C ; ^8004738) in the promoter region of SERPINAi. It appeared that
the SNP was enriched (15.5%) in a cohort of A iA T Pi ZZ homozygotes with liver
disease relative to those with pulmonary disease (6.5% ).10
As a result of the above-m entioned observations we hypothesized that
c.-ig73T>C polymorphism affects susceptibility fo rthe progression of liver dis­
ease in patients with liver disease of various etiologies. Therefore we investigated
75
C h a p te r 6
the association of c.-igyjT>C, P.E342K (Z allele) and p.E26^V(S allele) polymor­
phisms in a cohort of patients with liver disease of various etiologies compared
with healthy controls and evaluated its consequence on course of disease.
Methods
Patients
We recruited patients with various liver disorders, visiting the outpatient clinic
of the Department of Gastroenterology and Hepatology of the Radboud
University Nijmegen Medical Center. In addition, age and gender matched
persons who were unrelated to our patients served as healthy controls. In the
patient population, clinical and demographic data including age, sex, age at
first presentation of liver disease, etiology of liver disease and presence or
absence of cirrhosis were obtained. The absence of liver disease in ourcontrol
population was established on the basis of self-reporting and none of the
patients used any medication. Whole blood samples were stored at -20°C.
Altogetherthe study population comprised 297 patients with various etiologies
of liver disease and 297 controls. Clinical and demographic data are given. (Table
6.1) The study was approved by the local ethical committee. (Medical Ethical
Committee of the Radboud University Nijmegen Medical Center.)
Table 6.1: Baseline characteristics of patients and controls
Male (%)
Mean oge (range)
Mean oge at onset liver disease (range)
Cirrhosis (%)
Cause o f liver disease (%)
-HCV
-AIH/PBC/PCS
-HBV
-Alcoholic liver disease
-NASH/metabolic
-Cryptogenic
-Drug induced liver injury
-Vascular
Patients
(n= 297 )
168 (57)
Controls
(n= 297 )
163 (55 )
p-value
51.4 yrs ( 19-85 )
43 yrs ( 7-82)
69 (23)
51.5 yrs ( 20-85 )
0.94
0.74
129 (43 )
53 ( 18)
50 ( 17)
2 1 (7)
16 (5 )
1 3 (4 )
11 (4 )
4 ( 1)
HCV= hepatitis C virus, AIH/PSC/PBC= autoimmune hepatitis/ primary biliary cirrhosis/ primary sclerosing cholangitis,
HBV= hepatitis B virus, NASH= non-alcoholic steato hepatitis
P o ly m o rp h is m s o f SERPINA1 gene in c h ro n ic liv e r disease
Laboratory
DNA was isolated from peripheral blood using the High Pure PCR Template
preparation kit (Roche, Mannheim, Germany). The c.-igj3T>C (^8004738),
P.E342K (Z allele; c.i024G >A; ^28929474) and P.E264V (S allele; c.7giA>T;
rsi758o) polymorphisms of the SERPINAi gene were analyzed by real-time
polymerase chain reaction (PCR) using a dual-color, allele-specific discrimination
assay with fluorescent labelled probes on the ¡Cycler iQ Multicolour real-time
detection system (Bio-Rad Laboratories Inc, Hercules, CA, USA). Primer and
probe sequences (Sigma, St Louis, MO, USA) used for PCR and real time
detection are listed in the supplem entary table. All of the genotyping results
in the control population were in the Hardy-Weinberg equilibrium. Haplotypes
and diplotypes were determined using the Partition-Ligation-ExpectationMaximization (PLEM) 1.0 softw are.11
Statistical analysis
Baseline characteristics, differences in allele frequency, diplotypes and
haplotypes were analyzed using student t-tests, Pearson chi-squared test
and Fisher's exact test where appropriate. Odds Ratio's (OR) were calculated
for the association between the studied polymorphisms and the presence of
liver disease. In addition we calculated OR's for the association between the 3
polymorphisms and all different subgroup etiologies. We analyzed differences
in age at onset of liver disease between the 3 different C.-1973 genotypes
using ANOVA. All statistical analyses were performed using GraphPad Prism
Version 4.02 (GraphPad Software Inc., San Diego, CA, USA). A two-sided p-value
of < 0.05 was considered statistically significant. Post-hoc power calculation
showed the study being adequate powered (88%) to negate our hypothesis
that c.-igj3T>C is associated with liver disease of various etiologies (a= 0.05,
difference in group proportions 10% (20% and 30%)). Power calculations were
performed using nQuery Advisor 4.0 software (Statistical Solutions Ltd, Cork,
Ireland). Linkage disequilibrium (LD) values were performed with Haploview
4.0 software.
77
C h a p te r 6
Results
c .- ig j3 T > C polymorphism
The c.-igj3T>C polymorphism distribution in 297 patients with liver disease
due to various etiologies was in line with 297 healthy controls. ic.-1gy3-.JfT 32%,
T/C 44% and C/C 24% in patients andT/T30% ,T/C 50% and C/C 20% in controls).
(Figure 6.1) We found no association of the c.-igj3T>C polymorphism with any
of the distinct liver diseases investigated. Next, we observed that mean age at
onset of liver disease was non-significantly lower in patients with the c.-1973
C/C allele, as mean ages of onset of liver disease were 44.4 (T/T), 42.3 (T/C) and
40.7 (C/C) years. (Figure 6.2) Further, possession of c.-igj3T>C polymorphism
had no influence on the presence of cirrhosis.
Figure 6.1: Distribution of the c ,-ig /3 T > C genotypes in 297 patients with liver disease o f various
etiologies and 297 controls.
100%
1--------------------------------------------
90%
j------------------------------------------------------------♦
70% \------------------------------------------------------------60% 4—
80%
Patients
Controls
Figure 6.2: Age at onset of liver disease in 297 patients arranged to the different c .-ig /3 T > C
genotypes.
T
T/T
1-------
T/C
c.-1973T>C
78
“ i—
C/C
P o ly m o rp h is m s o f SERPINA1 gen e in c h ro n ic liv e r disease
Z (p.£342/0 allele and S (P.E264V) allele heterozygosity
We observed a similar Z allele and S allele heterozygosity rate in patients and
controls (Z allele: 3.0% and 4.7%; S allele: 6.7% and 8.0%). The distribution of
Z allele heterozygosity was similar among all liver diseases of various etiologies.
In contrast, S allele heterozygosity was more frequently present in drug induced
liver injury (DILI) compared with healthy controls (11 patients, 27% vs 8%; OR
4.27; 95%CI 1.06-17.15). A total of 8 wildtype patients developed DILI likely due to
clavulanic acid (n=3), azathioprine, sulfasalazine, pantoprazole, methotrexate
and quetiapine; addionally, the S allele heterozygotes had DILI caused by an­
aesthetic compounds (isoflurane/nestonal), clavulanic acid and celecoxib. Lastly,
age at onset of liver disease was independent of Z or S allele heterozygosity as
mean ages were 48.3 (Z allele), 43.2 (S allele) and 42.4 (wildtypes) years.
Haplotyping / diplotyping
Based on the 3 polymorphisms tested, haplotype and diplotype analysis
were performed. A total of 11 diplotypes could be distinguished. We could not
detect differences in diplotypes between patients with liver disease of various
etiologies and controls. Only the CTG haplotype was more frequent found in
patients with DILI, according to the above-described association between S
allele heterozygosity and DILI. (Table 6.2) Finally, linkage disequilibrium (LD)
between the 3 SERPINAi alleles was absent. This was true for patients as well
as controls. (Figure 6.3)
Table 6.2: Diplotypes in patients with liver disease of various etiology and healthy controls
Diplotype
CAG/TAG
TAG/TAG
CAG/CAG
CTG/TAG
CAG/CTG
TAG/TAA
CAG/TAA
TAG/TTG
CTG/TAA
CTG/CTG
CAG/CAA
Controls (%)
n =297
130 (43.8 )
78 (26.3)
51 ( 17.2)
11 ( 3.7 )
8 ( 2. 7 )
8 ( 2. 7 )
5 ( 1.7 )
4 ( 1.3 )
1 (0 .3 )
1 (0 .3 )
Patients (%)
n =297
112 ( 37. 7)
9 3 ( 31.3)
63 ( 21.2)
12 (4 .0 )
7 (2.4 )
2 (0 .7 )
6 (2.0)
1 (0 .3 )
1 (0 .3 )
Order of alleles : c.-1973T>C- c.791A>T - c.1024G>A/ c.-1973T>C- c.791A>T - c.1024G>A (mutations in bold)
79
C h a p te r 6
Figure 6.3: Linkage disequilibrium (LD) plot across the S ER P IN A i gene.
Patients
Controls
The box at the top indicates the S ERP IN A i gene with the 3 investigated SNP's (i=p.£342/C, 2= P.E264.V
and 3= c .-ig /3 T > C ). The LD plot is based on the measurement of R2 (values x 0.01). Each diamond
indicates the pair w ise magnitude of LD. (LD: linkage disequilibrium is the non-random association of
alleles at two or more loci. LD describes a situation in which some com binations of alleles o r genetic
markers occur more or less frequently in a population than would be expected from a random formation
of haplotypes from alleles based on their frequencies.)
Discussion
We show that the distribution of SERPINAi c.-igj3T>C in patients with liver
disease from various etiologies is similar compared to healthy controls. The
genotype frequencies in the patient and control groups were in line with those
from the HapMap database (T/T 30%, T/C 45% and C/C 25%)12. Moreover, we
found that the distribution of c.-igj3T>C is independent from etiology of the
liverdisease. We also examined whetherthe c.-igj3T>C polymorphism affected
age at onset of liver disease. We found a non-significant lower age at onset in
patients with the C.-1973 C/C genotype compared to other C.-1973 genotypes.
In addition, we could not demonstrate an association of c.-igj3T>C on severity
of liver disease, e.g. cirrhosis. We also investigated other SERPINAi variants and
80
P o ly m o rp h is m s o f SERPINA1 gene in c h ro n ic liv e r disease
found that there was no increased prevalence of Z and S alleles in patients with
liver disease of various etiologies compared with healthy controls, even though
we observed an enrichment of S allele heterozygosity in patients with DILI.
Other investigators have studied the presence of c.-igj3T>C and A iA T deficiency
and chronic obstructive pulmonary disease (COPD). Chappell et al. reported an
enrichment of c.-igyjT>C \n homozygous Pi ZZ neonates with hepatitis (15.5%)
compared to homozygous Pi ZZ controls (adults with COPD and unaffected
subjects) (6.5% ).10 Since our study population consisted for a large extent of
patients and controls with the Pi MM genotype (wildtypes) and hardly any
subjects with Z and S allele heterozygosity, we cannot compare our data with
the results of the above mentioned study. Another study showed a decreased
prevalence of c.-igyjT>C \n patients with COPD (48.7%) compared to controls
(52.2%), suggesting a protective effect against COPD.13 It might be possible
that c.-ig73T>C influences the genesis of liver disease in childhood, in line with a
recently published report implicating that a variant of the endoplasmic reticulum
mannosidase I (ERManl) gene is associated with an early onset of end-stage
liver disease in patients with homozygous (Pi ZZ) A iA T deficiency.14
Our data were in contrast with other reports as Z and S allele heterozygosity
were previously associated with (end-stage) liver disease due to HCV, alcoholic
liver disease and cryptogenic cirrhosis15-19 and not with DILI. We found a higher
frequency of S allele heterozygosity in DILI patients. Indeed, experimental
evidence supports a relation between A iA T deficiency and DILI as adm inistra­
tion of indomethacin in a homozygous Pi ZZ mouse model leads to increased
hepatic injury20 and a case report described prochlorperazine induced liver
injury in a homozygous (Pi ZZ) A iA T deficient patient21.
We could not demonstrate an association between c.-igj3T>C and the presence
of cirrhosis. There have been several genetic case control studies that have
attempted to detect associations between genetic variations and liver
fibrosis. For example, the combination of angiotensinogen (ATG) gene variant
c.1-44 and transforming growth factor beta (TGF|3 i) P.R25P is associated with
advanced hepatic fibrosis in obese patients with non alcoholic fatty liver
disease22 but not in patients with other chronic liver diseases.23 Lastly, matrix
metalloproteinase (M M P)-7(Asp-i37) confers risk of liver cirrhosis.24
81
C h a p te r 6
Our study comes with limitations. Our cohort could have suffered from selection
bias due to patient recruitment in a tertiary referral centre.The strength of our
study is the sufficiently power to negate a 10% difference in prevalence of the
c.-ig73T>C polymorphism between groups. However, the study lacks power
to demonstrate smaller differences and to perform a thorough subgroup
analysis. Further research regarding c.-igj3T>C should include homozygous
Pi ZZ adults with liver disease to evaluate whether c.-igj3T>C is a risk factor
for a hepatic expression of A iA T deficiency.
In conclusion: We demonstrated that, in our study, c.-igj3T>C polymorphism
was not associated with liver disease of various etiologies. In addition, S allele
heterozygosity might be a risk factor for the genesis of DILI.
References
1
Stoller JK, Aboussouan LS. A lp hai-antitrypsin deficiency. Lancet 2005 Jun 25;365(9478):2225-36.
2
Kok KF, W ahab PJ, Houwen RH, Drenth JP, de Man RA, van Hoek B, M eijer JW, W illekens FL,
de Vries RA. Heterozygous alpha -1 antitrypsin deficiency as a co-factor in the developm ent of
chronic liver disease: a review. Neth J Med. 2007 M ay;65(5):i6o-6.
3
de Serres FJ. W orldw ide racial and ethnic distribution of alph ai-an titrypsin deficiency: sum m ary
of an analysis of published genetic epidem iologic surveys. Chest 2002 N o v ;i2 2 (5 ):i8 i8 -2 9 .
4
SvegerT. Liver disease in a lph ai-an titrypsin deficiency detected by screening of 200,000 in­
fants. N Engl J Med 1976 Jun io ;2 9 4 (2 4 ):i3 i6 -2 i.
5
Rakela J, G oldschm iedt M, Ludwig J. Late m anifestation of chronic liver disease in adults with
alph a-i-antitrypsin deficiency. Dig Dis Sci 1987 D e c;32(i2):i358-62 .
6
Piitulainen E, Tornling G, Eriksson S. Effect of age and occupational exposure to airw ay irritants
on lung function in non-sm oking individuals with alpha l-an titryp sin deficiency (PiZZ). Thorax
1997 M ar;52(3):244-8.
7
Eriksson S. Alpha l-a ntitryp sin deficiency and liver cirrhosis in adults. An analysis of 35 Swedish
autopsied cases. A cta Med Scand i9 8 7 ;2 2 i(5 ):4 6 i-7 .
8
Tanash HA, Nilsson PM, Nilsson JA, Piitulainen E. Clinical course and prognosis of never-smokers
with severe alph a-i-antitrypsin deficiency (PiZZ). Thorax 2008 D e c ;6 3 (i2 ):io 9 i-5 .
9
Hinds R, Hadchouel A, Shanm ugham NP, A l-H ussaini A, Cham bers S, Cheesem an P, et a I.
Variable degree of liver involvem ent in siblings with PiZZ alpha-i-antitrypsin deficiency-related
liver disease. J Pediatr G astroenterol Nutr 2006 Ju l;4 3 (i):i3 6 -8 .
10
Chappell S, Hadzic N, Stockley R, G uetta-Baranes T, Morgan K, Kalsheker N. A polym orphism
of the alph ai-an titrypsin gene represents a risk factor for liver disease. Hepatology 2008 Jan;47
82
P o ly m o rp h is m s o f SERPINA1 gene in c h ro n ic liv e r disease
11
Qin ZS, Niu T, Liu JS. Partition-ligation-expectation-m axim ization algorithm fo r haplotype
inference with single-nucleotide polym orphism s. Am J Hum Genet. 2002 N o v ;7i(5 ):i2 4 2 -7.
(i):i2 7 -3 2 .
12
http://w w w .ncbi. nlm. nih.gov/projects/SN P/snp_ref. cgi?rs=8oo4738
13
Chappell S, Daly L, Morgan K, Guetta BT, Roca J, Rabinovich R, et al. Cryptic haplotypes of
S E R P IN A i confer susceptibility to chronic obstructive pulm onary disease. Hum M utat 2006
Ja n ;2 7 (i):i0 3 -9 .
14
Pan S, Huang L, McPherson J, Muzny D, Rouhani, F, Brantly M et al. Single Nucleotide
Polym orphism -M ediated Translational Suppression of Endoplasm atic Reticulum M annosidase
I M odifeis the Onset of End-Stage Liver Disease in A lp h ai-A n titryp sin Deficiency. Hepatology
2009; 50:275-281.
15
Eigenbrodt ML, M cCashland TM, Dy RM, Clark J, Galati J. Heterozygous alpha l-antitrypsin
phenotypes in patients with end stage liver disease. Am J G astroenterol 1997 A pr;92(4):6o2-7.
16
Graziadei IW, Joseph JJ, W iesner RH, Therneau TM, Batts KP, Porayko MK. Increased risk
of chronic liver failure in adults with heterozygous alph ai-an titrypsin deficiency. Hepatology
1998 O ct;28(4):i058-63.
17
Halangk J, W itt H, Puhl H, Gabelein G, Pascu M, Mu Her T, W iedenm ann B et al. Heterozygous
a lp h a -i antitrypsin deficiency as an inherited risk fa c to r in the developm ent of chronic liver
disease. Journal o f H epatology 50 suppl 1. 2009.
18
Kok KF, van Soest H, van Herwaarden AE, van Oijen MG, Boland GJ, Halangk J, BergT, deVries
RA, Drenth JP. Influence of a lp h a -i antitrypsin heterozygosity on treatm ent efficacy of HCV
com bination therapy. Eur J G astroenterol Hepatol. 2009 Sep 29. [Epub ahead of print]
19
Kok KF, W illem s JL and Drenth JPH.The cut-offV alue of 100 m g/dL is insufficient to detect
heterozygous A lp h a -i A ntitrypsin deficient Liver Disease patients. Liver Int [Epub ahead of
print]
20
Rudnick DA, Shikapw ashya O, Blom enkam p K ,Teckm an JH. Indom ethacin increases liver
dam age in a murine model of liver injury from alpha-i-antitrypsin deficiency. Hepatology 2006
Oct;44(4 ):976-82.
21
Mindikoglu AL, Anantharaju A, Hartman GG, Li SD, Villanueva J and van Thiel DH. Prochlorperazineinduced cholestasis in a patients with a lp h a -i antitrypsin deficiency. Hepatogastroenterology
2003 S e p -0 ct;50(53):i338-40.
22
Dixon JB, Bhathal PS, Jonsson JR, Dixon AF, Powell EE, O'Brien PE. Pro-fibrotic polym orphism s
predictive of advanced liver fibrosis in the severely obese. J Hepatol 2003 D ec;39(6):967-7i.
23
Halangk J, Berg T, Neumann K, Sarrazin C, Hinrichsen H, Fitz C, et al. Evaluation of angiotensinogen c.i-4 4 G > A and P.M 268T variants as risk factors for fibrosis progression in chronic
hepatitis C and liver diseases of various etiologies. Genet Test Mol Biom arkers 2009
Jun;i3(3):407-14.
24
Hung TM, Chang SC, Yu WH, W ang YW, Huang C, Lu SC, et al. A novel nonsynonym ous variant
of m atrix m etalloproteinase-7 confers risk of liver cirrhosis. H epatology 2009 O ct;5o (4 ):ii8 4 -9 3.
83
Supplem entary: Overview of primers and probes used for real time PCR reactions
Z allele
(p.E 3 4 2 K)
S allele
(p. E2 6 4 V)
Forward primer
Reverse primer
Probe wildtype
Probe mutation
Forward primer
Reverse primer
Probe wildtype
Probe mutation
c.- 1 9 7 3 T>C
84
Forward primer
Reverse primer
Probe C
Probe T
AAAACATGGCCCCAGCAG
AG CCTTACAAGTGTCTCTG
Fam-TCAGTCCCTTTCTCGTCGATGGTCAG-B HQ1
Hex-TCAGTCCCTTTCTTGTCGATGGTCAG-BHQl
AACATGGCTAAGAGGTGTGG
ATCTTCTTCCTG CCTG ATG AG
Fam-CTACAGCACCTGGAAAATGAACTCACCCBHQ1
Hex-CTACAGCACCTGGTAAATGAACTCACCCBHQ1
AGGCTGGCAGGAGGTTGC
GAG CTG AACCAAG AAG GAG G AG
Fam-TGGCAGCAGCAGCGGCTAGGCC-BHQl
Hex-TG G CAG CAG CAG CAG CTAG G CC- BHQ1
Chapter 7
The cut-off Value of 100 mg/dL is
Insufficient to Detect Heterozygous
Alpha-i Antitrypsin deficient Liver
Disease Patients
Liver International 200g Nov 30; [Epub ahead of print]
Karin F Kok
Hans (J) L Willems
Joost PH Drenth
Abstract
Background: A lp ha-i antitrypsin(AiAT) deficiency is a com­
mon disorder and leads to pulmonary and/or liver disease.
Some authors indicate that testing for A iA T deficiency
should start with quantification of the serum level, followed
by iso-electrofocussing and/or polymerase chain reaction
( PCR) for the Z and S alleles if serum levels are below a cut­
off point of 100 mg/dL.
We wondered w hetherthis diagnostic approach is suitable
for the detection of A iA T deficiency in chronic liver disease.
Patients and methods: We studied A iA T serum levels in 3
sets of Pi MZ persons. First, 12 Pi MZ patients with liver
disease due to various etiologies, second a cohort of 6 Pi
MZ patients with end-stage liver cirrhosis caused by A iA T
deficiency and lastly 15 Pi MZ controls.
Results: Mean serum levels were 106 mg/dL in patients
with liver disease of various etiology, 146 mg/dL in patients
with cirrhosis and 86 mg/dL in controls; p<o.oi. A iA T serum
levels were above the suggested threshold of 100 mg/dL in
50%, 83% and 6.7% respectively.
Conclusion: We showed that A iA T serum level quantifica­
tion is not suitable for the detection of Pi MZ heterozygosity
in patients with liver disease. We might have missed Pi MZ
heterozygosity in the majority of our patients if genotyping
only was performed when the A iA T serum level was below
100 mg/dL. Starting the diagnostic path with quantification
of the A iA T serum level runs the risk of missing subjects
with A iA T deficiency related liver disease. To prevent
underdiagnosis, we propose using iso- electrofocussing
or PCR ratherthan quantification of serum levels.
D iffic u ltie s in d e te c tio n o f h e te ro z y g o u s A IA T d e fic ie n c y
Background
A lp h a-i antitrypsin (A iA T) deficiency is common and affecting 3.4 million
subjects worldwide and leads to pulmonary and/or liver disease. Most A iA T
deficient patients are Pi ZZ, Pi MZ or Pi SZ genotype carriers, and A iA T
deficiency is frequently not recognized.1 Some authors indicate that testing
for A iA T deficiency should start with quantification of the serum level,
followed by iso-electrofocussing and/or polymerase chain reaction (PCR) for
the most common deficient alleles (Z and S) if serum levels are below a cut-off
point of 100 mg/dL2-3.
We wondered whether the diagnostic approach starting with serum level
quantification is suitable for the detection of A iA T deficiency in chronic liver
disease.
Methods
We studied A iA T serum levels in 3 sets of Pi MZ persons. First, 12 Pi MZ
patients with liver disease due to various etiologies (8 HCV, 2 alcoholic liver
disease, 1 AIH and 1 PSC), second a cohort of 6 Pi MZ patients with end-stage
liver cirrhosis caused by histological proven A iA T deficiency and lastly Pi MZ
controls (11 unaffected Pi MZ sibs of known patients and 4 Pi MZ controls
found through a genetic association study). Median ages were 56, 52 and 46
years respectively (p=o.o5) and gender distribution was similar in the 3 groups
(male: 56%, 83% and 53% respectively). A iA T serum levels were performed
using routine nephelometry, whereas A iA T carrier state was determined by
iso-electrofocussing or PCR.
Results
Mean serum levels were 106 mg/dL in patients with liver disease of various
etiology, 146 mg/dL in patients with A iA T deficiency related cirrhosis and
86 mg/dL in controls; p<o.oi. (Figure 7.1) A iA T serum levels were above the
suggested threshold of 100 mg/dL in 6 out of 12 (50%) patients with liver
C h a p te r 7
disease of various etiologies and in 5 out of 6 (83%) patients with A iA T
deficiency induced cirrhosis. In contrast, in healthy Pi MZ carriers we measured
an A iA T serum level above 100 mg/dl_ in only 1 out of 15 (6.7%).
Figure 7.1: A lp h a -i antitrypsin (A iA T ) serum levels of 15 Pi MZ controls, 12 Pi MZ patients w ith liver
disease of various etiologies and 6 Pi MZ patients w ith A iA T related cirrhosis.
250-1
"O
0) 2001
150-
E
3
100“
a>
A
••
■
■■ ■■■■■
a)
c/)
H
<
50 -
<
0
1
controls
.............. *:*
•
•
■*
***
1
various liver diseases
1
A1AT cirrhosis
In order to exclude that A iA T levels were falsely elevated due to a general
acute phase response we measured C-reactive protein (CRP) levels in a subset
of 11 subjects (3 controls, 5 patients with various liver disease and 3 cirrhotic
patients). We failed to detect a correlation between A iA T levels and CRP.
(r2= 0.13, p=o.28). This suggests that the acute phase response is not the
major driver of the A iA T levels in our study population.
Discussion
We might have missed Pi MZ heterozygosity in the majority of our patients
if genotyping only was performed when the A iA T serum level was below or
sim ilar to the cut-off point of 100 mg/dl_. Our finding is corroborated by a
report of a normal A iA T serum level in a Pi MZ heterozygous patient with
A iA T containing globules in liver biopsy specimen and concomitant HCV4.
In addition, in 11 out of 20 (55%) patients with end stage liver disease due
to A iA T deficiency (Pi MZ and Pi ZZ), A iA T serum level quantification prior to
90
D iffic u ltie s in d e te c tio n o f h e te ro z y g o u s A1AT d e fic ie n c y
liver transplantation was above the suggested cut-off point of 100 mg/dL.5
This suggests mere serum quantification will miss the significant number of
patients with bonafide A iA T deficiency related liver disease. Lastly, another
study reported the results of 512 samples referred for A iA T phenotyping.
These sam ples were analyzed by quantification of the serum level, isoelectrofocussing and PCR. In 2% of the samples, results from the 3 tests
were discordant, frequently due to presence of'norm al' serum levels.6
We can speculate about the pathophysiologic mechanism of the normal
serum levels in Pi MZ heterozygotes with liver disease. As A iA T is an acute
phase protein and its production and secretion increases with inflammation7-8
serum levels might be "falsely elevated" and are not reflecting the genotype.
This is confirmed by a recent study which found that Pi MZ individuals with a
CRP level >o .8 m g /d Lhad higher A iA T concentrations than Pi MZ individuals
with low erCRP levels.9 It is possible that the absence of a correlation between
CRP and A iA T serum levels in our population can be explained by the hepatic
origin of CRP. Our dataset does not allow a firm conclusion in either direction.
Conclusion
We showed that A iA T serum level quantification is not suitable for the detection
of Pi MZ heterozygosity in patients with liver disease. Starting the diagnostic
path with quantification of the A iA T serum level runs the risk of missing
subjects with A iA T deficiency related liver disease. To prevent underdiagnosis,
we propose starting the diagnostic route with iso- electrofocussing or PCR
(dependent on the local expertise) ratherthan quantification of serum levels.
References
1
Stoller JK, Aboussouan LS. A lp hai-an titrypsin deficiency. Lancet 2005 Jun 25;365(9478):2225-36.
2
Stoller JK, Aboussouan LS. Myths and m isconceptions about {alph a}i-an titrypsin deficiency.
Arch Intern Med 2009 Mar 23;i69(6):546-50.
3
Ferrarotti I, Scabini R, Campo I, Ottaviani S, Zorzetto M, Gorrini M, et al. Laboratory diagnosis of
alph ai-an titrypsin deficiency. Transl Res 2007 N ov;i5o(5):267-74.
91
C h a p te r 7
4
Banner BF, Karam itsios N, Sm ith L, Bonkovsky HL. Enhanced phenotypic expression of a lp h a-iantitrypsin deficiency in an MZ heterozygote with chronic hepatitis C. Am J G astroenterol 1998
Sep ;93(9):i54i-5-
5
Vennarecci G, Gunson BK, Ismail T, Hubscher SG, Kelly DA, M cM aster P, et al. Transplantation
for end stage liver disease related to alpha 1 antitrypsin. Transplantation 1996 May
27; 6 i(io ):i4 8 8 -9 5 .
6
Snyder MR, Katzmann JA, Butz ML, W iley C, Yang P, Dawson DB, et al. Diagnosis of a lp h a-iantitrypsin deficiency: An algorithm of quantification, genotyping, and phenotyping. Clin Chem
2006 D ec;52(i2):2236-42.
7
Lomas DA, Evans DL, Finch JT, Carrell RW. The mechanism of Z alpha i-antitrypsin accumulation
in the liver. Nature 1992 Jun i 8;357(6379):6 o 5-7.
8
Kalsheker NA. alph ai-A n titrypsin deficiency: best clinical practice. J Clin Pathol 2009
O ct;62(io):865-9.
9
Zorzetto M, Russi E, Senn O, Imboden M, Ferrarotti I, Tinelli C, et al. S E R P IN A i gene variants
in individuals from the general population with reduced alphai-antitrypsin concentrations. Clin
Chem 2008 A u g ;5 4 (8 ):i33 i-8 .
92
Chapter 8
LiverTransplantation for
Alpha-i Antitrypsin Deficiency
and the Impact of Incidental
Hepatocellular Carcinoma on
Post-Transplant Survival
Karin F Kok
Robert J Porte
Rene Adam
Paolo Muiesan
Aad P van de Berg
Rene Scheenstra
Herald J Metselaar
Bart van Hoek
Martijn GH van Oijen
Joost PH Drenth
Forthe European Liverand IntestineTransplant Association (ELITA)
Abstract
Background: A lp h a-i antitrypsin (A iA T) deficiency is a
genetic disorder that can result in lung and liver injury.
Both homozygous (Pi ZZ) and heterozygous (Pi MZ and Pi
SZ) disease appear to cause liver disease, however it is
unknown whether heterozygous disease induces end stage
liver disease and hepatocellular carcinoma (HCC). Liver
transplantation (LT) is the only curative approach at this
stage of liver disease.
Aim: To study (1) A iA T deficiency characteristics with
respect to serum level, phenotype and concomitant liver
disease (particularly HCC) and (2) patient and graft survival
in LT for A iA T deficiency.
Methods: We studied 2 LT cohorts: First, LT recipients with
A iA T deficiency in the Netherlands and second, a replicative
cohort with data from A iA T deficiency patients collected
in the European LiverTransplantation Registry (ELTR).
Results: We analyzed the medical records of 24 adult and 8
pediatric LT recipients from the Netherlands. HCC was found
in 7/24 (29%) adult LT recipients; the time between i st pre­
sentation of liver disease and LT was longer in patients with
HCC. (p<o.oi).The majority of HCC (4/7) was not recognized
prior to LT. HCC was absent in the pediatric LT recipients.
To replicate our findings we retrieved data from 317 adult
and 185 pediatric European A iA T deficient LT recipients.
HCC was present in 24/317 adults (8%) and presence of
HCC was associated with a poor 5 years survival (69% vs.
83%, HCC patients vs non-HCC patients, p=o.oi).
The 15 years patient survival rates were 58% for adult LT
recipients and 86% for pediatric LT recipients (p=o.ooi).
The 15 years graft survival rates were 57% for adult LT
recipients and 76% for pediatric LT recipients (p=o.o2).
Conclusion: Patients with A iA T deficiency induced cirrhosis
are at high risk fo rthe development of HCC of which some
were not recognized before LT. Presence of HCC predisposes
for an inferior survival after LT.
Liver tra n s p la n ta tio n and HCC in A1AT d e fic ie n c y
Background
A lp h a-i antitrypsin (AiAT) deficiency is a genetic disorderthat most frequently
presents as emphysema and less commonly as liver disease.1-2 In childhood,
A iA T deficiency liver disease presents as neonatal cholestasis which can lead
to cirrhosis.The clinical phenotype of A iA T deficiency in adults isthat of mild
liver enzyme disturbances with or without advanced disease such as cirrhosis
and hepatocellular carcinoma (HCC).1 Once diagnozed, most patients with liver
cirrhosis in A iA T deficiency will develop rapid progressive liverdisease.3 Atthat
stage, the only curative approach is livertransplantation (LT).Theoretically, LT
has a dual benefit, as it not only cures the liver disease but also restores A iA T
serum levels by adopting the donor phenotype and thus offering protection
against emphysem a.3 Most LT series are small and stem from the beginning of
the livertransplantation era.They often have been performed in the pediatric
population4-7 and originate from single centers. A recent series from the United
Network for Organ Sharing (UNOS) database reported an excellent survival in
A iA T LT recipients.8
We investigated the clinical characteristics and prognosis after LT in both adult
and pediatric patients with A iA T deficiency. We addressed the following ques­
tions. First, we were interested in clinical characteristics of patients receiving
LT for A iA T deficiency. We hypothesized that the proportion of HCC cases would
be low in A iA T deficient LT recipients and we tested this hypothesis in 2 different
cohorts. Secondly, benefiting from the comprehensive European LiverTransplantation Registry (ELTR) we examined patient and graft survival after LT.
Methods
Our study included 2 different cohorts. First, we studied the clinical charac­
teristics in a cohort of A iA T deficient LT recipients in the Netherlands and
secondly, we studied a cohort of LT patients with A iA T deficiency that was
submitted to the ELTR.
The Netherlands cohort
We analyzed the medical records of all adult and pediatric LT recipients with
A iA T deficiency who were transplanted from January 1985 onwards in all 3 LT
centers in the Netherlands (University Medical Center Groningen, Erasmus
97
C h a p te r 8
Medical Center Rotterdam and Leiden University Medical Center). We used the
local databases to identify LT recipients with A iA T deposits on explant histology,
and were able to include 32 patients. We assessed patient demographics, date
of i st presentation of liver disease, date of LT, date of last outpatient clinic visit,
results of A iA T iso-electrofocussing and A iA T serum level quantification,
spirometry: Forced Expiratory Volume in lse c/V ita l Capacity (FEVi/VC), histo­
logical report of the explant specimen, routine laboratory parameters, life style
characteristics such as smoking and alcohol use, concomitant liver disease
and causes and dates of re-transplantation or death. When HCC was present:
size, number and alpha feto protein level.
ELTR cohort
We retrieved data collected in the ELTR to study patient and graft survival after
LT for A iA T deficiency in a large number of patients. The European Liver and
Intestine Transplant Association (ELITA) board approved the study request
and provided the data. We included patients who underwent LT from January
i st 1993 to December 31st 2007 and who had A iA T deficiency registered as
primary disease or as associated disease. We excluded Dutch patients because
these were studied in more detail in the first part of our study.
Statistics
Descriptive statistics were performed on baseline characteristics and presented
as median and range. We used Mann-Whitney-U tests, Chi-square tests and
Fisher exact tests where appropriate to analyze baseline characteristics. Odds
Ratio's (OR) were calculated for gender distribution, presence of HCC, diabetes,
smoking and alcohol history between patients with homozygous and hetero­
zygous disease in the Dutch cohort. In addition, we used OR's for calculating
differences in baseline characteristics between adult and pediatric LT recipients
in the ELTR cohort. For time-to-death or re-transplantation we applied the
Kaplan-Meier method and compared groups using the log-rank test. All statis­
tical analyses were performed using GraphPad Prism Version 4.02 (GraphPad
Software Inc., San Diego, CA, USA). Atwo-sided p-value of <0.05 was considered
statistically significant.
98
Liver tra n s p la n ta tio n and HCC in A1AT d e fic ie n c y
Results
The Netherlands cohort
We included 32 patients (24 adults and 8 pediatric LT recipients) who underwent
LT from January i st 1985 until May 26th 2009. Pre-LT characteristics are given
in table 8.1. One adult woman received auxiliary LT in 1987, the remaining 31
patients were treated with standard orthotopic LT.
We detected HCC in 7 out of 24 (29%) adult LT recipients (71% males), and
mean age (±SD) at diagnosis HCC was 50.6 (±10.3) years). The time between
i st presentation of liver disease and LT was longer in adult LT recipients with
HCC compared to adult LT recipients without HCC (p<o.01).(Figure 8.1)
Table 8.1: Pre-liver transplant characteristics of 24 adult and 8 pediatric LT recipients w ith
a lp h a -i antitrypsin deficiency in the Netherlands
Median age at LT (yrs) (range)
Male (%)
Median ALT (lU/ml) (range)
Median bilirubin (umol/l) (range)
Median albumin (g/l) (range)
Prothrombin time (s) (range)
FEV1/VC < 70% (%) (n= 23)
Diabetes
Time between 1st presentation of liver disease
and LT (yrs) (range)
Adult n=24
50 .6 ( 18.2- 66. 7)
19 ( 79%)
5 3 ( 26- 337)
44 (9-800)
30 ( 16-44 )
1 7 ( 11.8-25.3)
6 (26%)
3 ( 13%)
7 .0 (0 .5- 27.5)
Pediatric n=8
7. 1 ( 1. 1- 14.3 )
5 ( 63%)
124 (47 - 610)
53 ( 10-266)
32 ( 18-41)
1 8 ( 12.2- 35.7 )
NA
0 (0%)
7. 1 (0.9 - 14. 14)
LT=Liver Transplantation; AlAT=alpha-l antitrypsin deficiency; FEV1/VC= Forced Expiratory Volume in 1 sec/Vital Capacity;
NA=not available
Figure 8.1: Time interval between 1st presentation of liver disease and liver transplantation.
99
C h a p te r 8
Age at LT and age at first presentation of liver disease were similar in patients
with and without HCC. In addition, other baseline characteristics (in particular
presence of homozygous or heterozygous disease) were comparable for adult
LT recipients with and without HCC. In all HCC cases the diagnosis was con­
firmed by histological examination of the surgical explant specimen. A total of
4 out of 7 HCCs was not diagnozed prior to LT. These had an incidental HCC
and had normal alpha feto protein levels (< io ug/l) prior to LT. Two patients
with incidental HCC had small single tum ors of 8 and 15 mm, 1 patient had
multifocal H CCw iththe largerone m easuring30 mm and ip a tie n t had atum or
measuring 80 mm. This tum or of 80 mm exceeded by far the Milan Criteria,
though the patient was misclassified as having focal nodular hyperplasia prior
to LT. All 3 patients with HCC who were recognized priorto LThad elevated alpha
feto protein levels. We found no HCC in pediatric LT recipients, although one
girl, transplanted at age 3.5 had a dysplastic node of 6 mm in her liver explant.
A iA T phenotyping was performed using iso-electrofocussing.The entire pop­
ulation of pediatric LT recipients was A iA T homozygous and median serum
level was 0.4 g/l (range: o.2-0.5 9 / 0 - A iA T phenotypes in 21 adult LT recipients
were: 14 Pi ZZ (homozygous) and 5 Pi MZ, 1 Pi SZ and 1 Pi Mnull (heterozygous).
Baseline characteristics were similar in homozygous and heterozygous adult
LT recipients. A iA T serum levels were lower in homozygous adults compared
to heterozygous adults (median 0.3 g/l vs. 1.4 g/l, p=o.ooi). (Figure 8.2)
Figure 8.2: A lp h a -i antitrypsin serum levels in adult liver transplantation recipients with homozygous
(Pi ZZ) and heterozygous (MZ or SZ) A iA T deficiency.
100
Liver tra n s p la n ta tio n and HCC in A1AT d e fic ie n c y
Spirometry results were available in 23 adult LT recipients but not for pediatric
LT recipients. A total of 6/23 (26%) adult LT recipients had signs of obstructive
pulmonary disease with FEVi/VC <0.7 prior to LT. One of these patients was a
recipient of a simultaneous LT and unilateral lung transplant; this patient had
increased spirometry results after transplantation.
Median time of follow-up was 5.3 years (range 0.1-24) ¡n adult and 3.4 years
(range 0.7-16) in pediatric LT recipients. Five adults died after LT: due to cardiac
disease (n=2, 21 and 2.5 years after LT), multi-organ-failure and in-hospital
death (n=2, 0.1 and 0.4 years after LT) and metastatic melanoma (n= i, 6.4
years after LT). Re-transplantation was performed in 2 adult LT recipients
because of ischemic bile duct lesions 0.2 years after LT and liver failure of
unknown origin at 3.4 years after LT. Three pediatric LT recipients, all trans­
planted in 1989, died. Reasons of death were sepsis (n=2, 1.7 and 2.6 years
after LT) and pulmonary embolism (n=i, 16 years after LT).
ELTR cohort
As we observed an unexpected high frequency of HCC in our initial cohort
we sought confirmation of this observation in an independent LT cohort. This
replicative cohort comes from the ELTR database and we included 317 adult
and 185 pediatric LT recipients, originating from 78 centers in 20 European
countries. (See for complete list: addendum) A iA T deficiency was registered
as the leading indication for LT in 231 (73%) adult and 179 (97%) pediatric LT
recipients and was registered as an associated disease in the remainder. The
distribution in the absolute number of LT was bimodal, as we found a peak in
early childhood and a second peak in middle-aged adults. (Figure 8.3)
We observed a male predominance in adult and pediatric LT recipients, but in
adults this unbalanced gender distribution was more apparent (OR 1.8, 95%CI
1.2-2.7). When corrected fo rth e presence of alcoholic liver disease, the differ­
ence was not significant (OR 1.4, 95%CI 0.95-2.2). In adult LT recipients the
registered associated diseases were different compared to pediatric LT recipi­
ents as concomitant alcoholic liver disease, HCC and cryptogenic cirrhosis
were more frequently reported in adults and biliary atresia was unique to pe­
diatric LT recipients. (Table 8.2)
C h a p te r 8
F ig u re 8 .3 : A g e d is tr ib u tio n a m o n g 502 p a tie n ts w h o u n d e rw e n t liv e r tr a n s p la n ta tio n d u e t o a lp h a - i
a n titry p s in d e fic ie n c y b e tw e e n 1 9 9 3 and 2007 a c c o rd in g to th e ELTR.
O
Age (yrs)
Table 8.2: Baseline characteristics of 502 patients, registered in the ELTR database, w ho underwent
liver transplantation because of a lp h a -i antitrypsin deficiency between 1993 and 2007
Adult n=317
Median age (range)
Male (%)
Associated diseases t(%)
HCC
Alcoholic liver disease
Cryptogenic cirrhosis
Viral hepatitis
(Semi) Acute liver failure
AIH/PSC/PBC
Hemochromatosis
Polycystic liver disease
Hemangioma
Tyrosemia°°
Biliary atresia
Cystic Fibrosis
52.5 ( 18.7- 70.4 )
244 ( 77)
24 (8)
65 (21)
1 7 (5 )
1 1 ( 3)
2 (1)
6(2)
6(2)
2 (1 )
1(0 )
1(0)
0(0)
0(0)
Pediatric n=185
5.6 (0 .3- 17. 1)
120 ( 65)
1 (1 )
0(0)
2 (1)
1 (1 )
1 (1 )
0(0)
0(0)
0(0)
0(0)
0(0 )
6 (3 )
1 (1 )
OR (95 % Cl)
1.8 (1.2-2.7) t
15 ( 2.0 - 112) +
NA
5.2 ( 1.2- 22.7)t
1.8 (0 . 18- 17.0 )
1.2 (0 . 11- 13. 1)
NA
NA
NA
NA
NA
NA
NA
^associated disease when A1AT deficiency was registered as indicator fo r LT and indicator fo r LT when A1AT deficiency
was registered as associated disease. NA: not available, tstatistic significant. °°This patient was transplanted at age 20.
The 1-, 5-,io - and 15- year patient survival rates were 89%, 82%, 63% and 58%
for adult LT recipients and 93%, 89%, 86% and 86% for pediatric LT recipients
(p=o.ooi adult vs pediatric cohort).The 1-, 5-,10- and 15- year graft survival rates
were 87%, 79%, 62% and 57% for adult LT recipients and 88%, 84%, 76% and
76% for pediatric LT recipients (p=o.o2 adult vs pediatric cohort). (Figure 8.4)
1 02
Live r tra n sp la n ta tio n an d H CC in A 1 A T d e fic ie n c y
F ig u re 8 .4 : P a tie n t su rv iv a l (p ane l A ) and g ra ft su rv iv a l (p anel B) a fte r live r tra n s p la n ta tio n in 502
p a tie n ts w ith a lp h a - i a n titry p s in d e fic ie n c y re giste re d in th e E L T R .
----- Pediatric
—
Adult
Time (yrs)
-----Pediatric
—
Adult
Time (yrs)
The 1- and 5-year patients survival rates were 86% and 69% respectively in
adults with HCC (n=24, 92% male) and 89% and 83% respectively in adults
without HCC (n=293, 76% male). Therefore survival in patients without HCC
was superior compared to patients with HCC (p=o.oi). (Figure 8.5) A total of
9 adult LT recipients with HCC died after LT, the reason of death was tumor
recurrence in 5 out of 9 (56%) patients.
103
C h a p te r 8
F ig u re 8 .5 : P a tie n t su rv iv a l in 3 1 7 a d u lt live r tra n s p la n t re cip ie n ts w ith A i A T d e ficie n cy : p a tie n ts w ith
H C C co m p are d to p a tie n ts w ith o u t H CC .
— no HCC (n=293)
■J- HCC (n=24)
Time (yrs)
Discussion
This study shows a high rate of HCC in adult LT recipients due to A iA T
deficiency as we found HCC in 29% of liver explants from an extensively
phenotyped cohort of 32 patients. Some ofthese were unrecognized priorto
LT (so-called incidental HCC). We confirmed these data in a larger replicative
cohort and found that 24/317 (8%) had HCC in their explants. This series differs
from other studies in the field as it adds several elements of novelty. First,
we analyzed the frequency of HCC in a large number of LT patients with A iA T
deficiency and second, we compared survival after LT in A iA T deficient
patients with and without HCC.
Literature is equivocal whether A iA T deficiency represents a risk factor for
HCC. Some studies indicate that HCC in patients with A iA T deficiency is related
to concom itant viral hepatitis9-11, others reported an association between
HCC and the Z allele whereas others do not.12-14 The research strategies
that these researchers have used to associate HCC with A iA T deficiency
are diverse. There have been a number of pathology studies that detected
a higher incidence of A iA T containing globules in HCC livers compared to
control livers and epidem iology studies have studied whether A iA T was a
covariate factor for HCC on top of underlying disorders such as HBV or HCV.
104
Liver tra n sp la n ta tio n and HCC in A1A T d e ficie n cy
Both strategies come with disadvantages as they do not accurately estimate
the prevalence of HCC in patients with A iA T deficiency. Our study provides
data that helps to address this issue. The high rate of HCC in our study is similar
to that reported for genetic hemochromatosis (7/22, 32% ).15 We also show
that survival rates after LT in A iA T deficient adults with HCC are clearly inferior
to those without HCC. Although these results are intuitive, the formal proof
for this concept was absent from literature. LT recipients with HCC died due to
tum or recurrence in 56% of cases. Earlier published guidelines did not recom­
mend for or against surveillance for HCC in A iA T deficiency because a lack of
data precluded an assessment of whether surveillance would be beneficial.16
Our data have possible clinical implications as the risk for HCC in A iA T deficiency
appears to be real, although this cohort does not allow to provide an accurate
estimate.
We found that the clinical course in adult LT recipients is similar in homozygous
and heterozygous patients. Separate analysis did not yield different results
with respect to gender, age at first presentation of liver disease, age at LT and
concomitant alcohol abuse (data not shown). This was reason for us to lump
these data together. All heterozygous A iA T deficiency patients had serum
levels > 1.0 g/l (thus in the normal range) and as a consequence, these patients
could have been missed if only serum level quantification was performed. We
previously showed that A iA T serum levels were not suitable for the detection of
heterozygous A iA T deficiency in patients with chronic liver disease and sug­
gested to start the diagnostic path with iso-electrofocussing or genotyping.17
The ELTR based part of our study shows that LT is a successful treatm ent for
liver disease in A iA T deficiency with excellent graft and patients' survival in
adult and pediatric LT recipients. Both graft and patients' survival are superior
in pediatric LT recipients compared to adult LT recipients.These results can be
compared with the study that used the UNOS database to determine survival
after LT in A iA T deficiency. This study included 406 adult and 161 pediatric LT
recipients who were transplanted between 1995 and 2004. The 1- and 5-year
graft and patients survival in adult and pediatric LT recipients are sim ilarto our
data.8 An earlier published study showed inferior results and reported a l-ye ar
patient survival of 73% in adult and 88 % in pediatric LT recipients.3The better
105
C h a p te r 8
survival in our cohort may be due to improvement in clinical care with tim e.18
The superior LT results in pediatric patients relative to adults, is known to
be true for other indications such as cholestatic disease and other metabolic
diseases.18
The strength of our study is the replication of data from (1) a thorough analysis
of clinical characteristics in patients with A iA T deficiency who received LTinthe
Netherlands by a second data set from (2) a large number of European patients
with a long follow-up.The limitations of our study are asfollows: selection bias
might be introduced as we only selected patients who received LT and therefore
we cannot extrapolate our findings to all patients with A iA T deficiency related
liver disease. In addition, because of the retrospective design, there were
missing data.
In conclusion: first, adult patients with end-stage liver disease due A iA T
deficiency, requiring LT, are at risk for the development of HCC and some are
not recognized prior to LT. Patients with HCC had inferior survival outcomes.
Second, baseline characteristics were similar in patients with homozygous
compared to heterozygous A iA T deficiency. Third, this study showed that LT
is an effective treatm ent of liver disease in A iA T deficiency with excellent
graft and patients' survival in adult and pediatric LT recipients.
References
1
Stoller JK, Aboussouan LS. A lp hai-antitrypsin deficiency. Lancet 2005 Jun 25;305(9478):2225-36.
2
Kok KF, W ahab PJ, Houwen RH, Drenth JP, de Man RA, van Hoek B, M eijer JW, W illekens FL,
de Vries RA. Heterozygous alpha-1 antitrypsin deficiency as a co-factor in the developm ent of
chronic liver disease: a review. Neth J Med. 2007 M ay;65(5):i6o-6.
3
Vennarecci G, Gunson BK, Ismail T, Hubscher SG, Kelly DA, M cM aster P, et al. Transplantation
for end stage liver disease related to alpha 1 antitrypsin. Transplantation 1996 May
27; 6 i(io ):i4 8 8 -9 5 .
4
Francavilla R, Castellaneta SP, Hadzic N, Cham bers SM, Portmann B, Tung J, et al. Prognosis of
alph a-i-antitrypsin deficiency-related liver disease in the era of paediatric liver transplantation.
J Hepatol 2000 Jun;32(6):986-92.
5
Prachalias AA, Kalife M, Francavilla R, Muiesan P, Dhawan A, Baker A, et al. Livertransplantation
for alph a-i-antitrypsin deficiency in children. Transpl Int 2 000;i3(3):2 0 7-i0 .
106
Live r tra n sp la n ta tio n and HCC in A1A T d e fic ie n c y
6
Baku la A, Socha P, Pawlowska J, Teisseyre M, Jankowska I, Ka licinski P. Good and bad prognosis
of alph a-i-antitrypsin deficiency in children: when to list for liver transplantation. Transplant
Proc 2007 D e c;39 (io ):3i8 6 -8 .
7
Nemeth A. L ivertransplantation in alph a(i)-an titryp sin deficiency. Eur J P e d ia trig g g D e c;i58
Suppl2:S 85-8.:S 85-S 88.
8
Kem m er N, KaiserT, Zacharias V, Neff GW. A lp ha-i-an titryp sin deficiency: outcom es after liver
transplantation. Transplant Proc 2008 Jun ;4o (5):i492-4.
9
Fattovich G, Stroffolini T, Zagni I, Donato F. Hepatocellular carcinom a in cirrhosis: incidence and
risk factors. G astroenterology 2004 N o v;i27(5 Suppl i):S 35-S50.
îo
PropstT, Propst A, Dietze O, Judm aier G, Braunsteiner H, Vogel W. Prevalence of hepatocellular
carcinom a in alph a-i-antitrypsin deficiency. J Hepatol 1994 D e c ;2 i(6 ):io o 6 -n .
il
Rabinovitz M, G avaler JS, Kelly RH, Prieto M, Van Thiel DH. Lack of increase in heterozygous
alpha i-a ntitryp sin deficiency phenotypes among patients with hepatocellular and bile duct
carcinom a. Hepatology 1992 M a r;i5 (3 ):4 0 7-i0 .
12
Reintoft I, Hagerstrand IE. Does the Z gene variant of alph a-i-antitrypsin predispose to hepatic
carcinom a? Hum Pathol 1979 Ju l;io (4 ):4 i9 -2 4 .
13
Zhou H, Fischer HP. Liver carcinom a in PiZ alpha-i-antitrypsin deficiency. Am J Surg Pathol 1998
Jun;22(6):742-8.
14
Zhou H, O rtiz-Pallardo ME, KoY, Fischer HP. Is heterozygous alpha-i-antitrypsin deficiency type
PIZ a ris k fa c to rfo r prim ary liver carcinom a? C ancer 2000 Jun i5 ;8 8 (i2 ):2 6 6 8 -76 .
15
Dar FS, Faraj W, Zaman MB, Bartlett A, Bom ford A, O'Sullivan A, et al. Outcom e of liver
transplantation in hereditary hem ochrom atosis. Transpl Int 2009 Jul;22(7):7i7-24.
16
Bruix J, Sherm an M Practice Guidelines Com m ittee, Am erican Association fo rth e Study of Liver
Diseases. M anagem ent o f hepatocellular carcinom a. Hepatology. 2005 N ov;42(5):i2o8-36.)
17
Kok KF, W illem s JL DJ. "The cut-off Value o f 100 m g/dL is insufficient to detect heterozygous
A lp h a -i A ntitrypsin deficient Liver Disease patients ". Liver International. (Epub ahead o f print)
18
Adam R, Hoti E. Livertransplantation: the current situation. Semin Liver Dis 2009 F eb ;29 (i):3-i8.
107
C h a p te r 8
Addendum
Centres participating in the ELITA cohort study:
AUSTRIA:
Transplantationszentrum A llegem eines Krankenhaus W ien, Prof. F Muhl bâcher;
University H ospital-Dept of General & Transplant Surgery Innsbruck, Prof. R.
M argreiter
BELGIUM:
C atholic University o f Louvain, Prof. J. de Ville de Goyet and Prof. J.Lerut; Hôpital
Erasme, UL Bruxelles, Prof. V. Donkier; U niversity Hospital Ghent, Prof. B. de
Hem ptinne; U niversity Hospital Leuvin, Dr. J. Pirenne; Domaine U niversitairer du
SartT ilm an, Prof. M. Meurisse
CZECH REPUBLIC: Transplancenter Ikem Prague, Prof. S. Vitko
DENM ARK:
University Hospital Copenhagen, Prof. P. Kirkegaard
FINLAND:
U.C.J. Helsingfors, Helsinki, Prof. K. Hockerstedt
FRANCE :
CHU M arseille Hopital de la Conception, Prof. Y. Letreut ; Hopital Paul Brousse
Villejuif, Prof. D. Castaing ; CHU, Necker Paris Cedex, ProfY. Revillon ; CHRU de
Strasbourg, Prof. D. Jaeck and Prof. P. Wolff; CHU Rennes, Prof. K. Boudjema;
CHU de Besançon Hopital Jean Minjoz, Prof. G. Mantion; Hopital Beaujon, Prof.
J. Belghiti; Hopital de la Pitie Salpétriére Paris, Prof. L. Hannoun; Hopital E.
Herriot Lyon, Prof. O. Bouillot; Hopital Henri M ondor Creteil, Prof. D. Cherqui;
CHU Hopiteaux de Bordeaux, Prof. J. Saric; Hopital Pasteur Nice, Prof. A. Bourgeon
GERMANY:
Medizinische Hochschule Hannover, Prof. J. Klempnauer; Klinikum der Universität
M ünchen-G rosshadern, Prof. R. Schauer; Charite- C am pus-Virchow Klinikum
Berlin, Prof. P. Neuhaus; Chirurgische Klinik und P o lik lin ik -K lin ik rechts der Isar
München, Prof. M. Stängel; U niversitatskrankenhaus Eppendorf Hamburg, Prof.
X. Rogiers; Universitatsklinikum Heidelberg, Prof.T. Kraus; Universitätsklinikum
Schlesw ig-Holstein Kiel, Prof. B. Kremer; Klinikum der Universität zu Köln, Dr. D.
Stippel; Universität Rostock-M edizinische Fakultät, Prof. W. Schareck; Universitätsklin ikTübingen, Prof. A. Königsrainer and Prof. W. Steuer; CUK GHs Essen,
Prof. C. Broelch; Universitätsklinikum Münster, Prof. N. Senninger; U niversity of
Bonn, Prof. M. Wolff; Johannes Gutenberg, Univ. Mainz Klinikum , Prof. G. Otto
GREAT BRITAIN:
The Queen Elizabeth Hospital Birmingham, Prof. J. Buckels; Adenbrooke's Hospital
Cam bridge, Mr. N. Jam ieson; Royal Infirm ary of Edinburgh, Prof. J.Garden; St
Jame's & Seacroft University Hospital Leeds, Dr. S. Pollard; King's College Hospital
London, Prof. J. O'Grady; Royal Free Hospital London, Dr. K. Rolles; The Freeman
Hospital Newcastle, Mr. D. Manas
IRELAND:
St Vincent Hospital Dublin, Prof. O. Traynor
ITALY:
Centro Trapianti di Fegato Bari, Prof. V. Memeo; Ospedali Riuniti di Bergam o, Dr.
M. Colledan; Ospedale San Martino e Cliniche Universitarie Convenzionate Genova,
Prof. U. Valente; Ospedale M aggiore Policlinico Milano, Prof. G. Rossi; Istituto di
Chirurgia G enerale Padova, Prof. A. Maffei Faccioli; Instituto M editerraneo per i
Trapianti e Terapie ad Alta Specializzazione Palermo, Prof D. d'Amico; Università
Degli Studi di Roma, Prof. M. Rossi; University Hospital Udine, Prof. F. Bresadola
NORWAY:
Rikshopitalet Oslo, Prof. A. Foss
108
Live r tra n sp la n ta tio n and HCC in A1A T d e fic ie n c y
POLAND:
Children's M emorial Health Institute W arsaw, Dr. P. Kalicinski; Medical University
of W arsaw, Prof. W. Rowinski
PORTUGAL:
Hospitais da Univesidade de Coim bra, Prof. F. de Oliviera
ROM ANIA:
University of Medicine "Carol Davila" Bucharest, Prof. I. Popescu
SLOVENIA:
University Medical Centre Lubljana, Prof. S. M arkovic
SPAIN:
Hospital Central de A sturias Oviedo, Dr. D. Gonzalez Pinto: Hospital Clinic I
Provincial de Barcelona, Prof. J. G arcia-Valdecasas: Hospital Infantil Barcelona,
Prof. V. M artinez Ibanez; Hospital Universitario "Reina Sofia" Cordoba, Prof. C.
Pera; Hospital Juan Canalejo la Coruna, Dr. C. Fernandez Sellez; Hospital 12 de
Octubre Madrid, Prof E. Moreno Gonzales; Hospital G regoria M aranon Madrid,
Prof. C. Kempin; Hospital Infantil La Paz Madrid, Dr. J. Larrucea; Hospital Regional
"Carlos Haya" Malaga, Prof. A. de la Fuente Perucho; Hospital Universitario Puerta
de Hierro Madrid, Prof. V. SanchezTurrion; Clinica Universitaria Pamplona, Dr. F.
Sanchez; Hospital Universitario "M arqués de Va Idecilla" Santander, Prof. L. Herrera
Norena: Hospital Ntra Sra de Candelaria Santa Cruz deTenerife, Dr. A. Soriano;
Hospital Universitario La Fe Valencia, Dr. J. Berenguer; Hospital de Cruces Varacaldo-Vizcaya, Dr. J. Ortiz de Urbina; Hospital Clinico Universitario de Zaragoza,
Dr. T. Serrano
SW EDEN:
Sahlgrenska University Hospital G oteborg, Dr. S. Friman; Huddinge Hospital,
Prof. B. Ericzon
SW ISS:
U niversitatsspital Bern, Prof. D. Candinas; Hôpital Cantonal Universitaire de
Geneve, Prof. P. Morel and Prof. G. Mentha; U niversitatspital Zurich, Prof. P.
Clavien
109
Chapter 9
General discussion
G ene ral d iscu ssio n
In 2005 a total of 3 patients presented with disturbed liver enzymes. They
deteriorated rapidly and all died as a result of their liver disease. All 3 patients
had normal serum levels of a lp h a -i antitrypsin (A iA T ) and the diagnosis
A iA T deficiency was only made by examination of a liver biopsy specimen. We
discovered that all 3 patients were A iA T deficient and heterozygous carriers
ofthe Pi MZ allele.This was eventually confirmed with iso-electrofocussing. In
hindsight, we relied too much on the "pseudo"- normal A iA T levels and did
not proceed to iso-electrofocussing or genotyping.1
These 3 cases spurred our interest in A iA T deficiency as a cause of liver disease.
We performed an extensive literature search that aimed to address the following
questions.
First, we were interested in the clinical course ofthe disease. Do patients with
A iA T deficiency develop end-stage liver disease? What are the complications
of A iA T deficiency related liver disease? And what about the characteristics
and prognosis of neonatal cholestasis due to A iA T deficiency? We tried to find
these answers and wrote a review ofthe current literature.2 Subsequently, we
studied the clinical characteristics of neonates with cholestasis due to A iA T
deficiency. We focused on prognosis and on bleeding complications due to
vitamin K deficiency. Vitamin K deficiency is caused by cholestasis induced
malabsorption. We detected a high proportion of vitamin K deficiency related
bleeding, particularly in exclusively breast-fed infants.3 In addition, we studied
clinical characteristics and prognosis after liver transplantation (LT) in adult
and pediatric L.T recipients with A iA T deficiency. We detected a high proportion
of HCC in adult L.T recipients and the majority was diagnosed upon examination
o fth e explanted liver specimen. Indeed, A iA T L.T recipients with HCC had an
inferior survival compared to patients without HCC.
Second, we focused on the diagnostic approach for detecting A iA T deficiency
in liver disease. Guidelines and flow charts regarding the evaluation of A iA T
deficiency proposed to start the initial evaluation with A iA T serum level
quantification and performing iso-electrofocussing or genotyping only in case
serum level is decreased.4These flow charts are merely initiated by pulmonary
centers. As serum levels >0.5 g/l represent a protective threshold above which
the risk of emphysema is not increased, the cutoff point fits with the detection
G en eral d iscu ssio n
of A iA T deficiency in pulmonary disease. Given that aberrant A iA T phenotypes
with serum levels above this protective threshold will not induce emphysema,
the detection of an abnormal phenotype with a normal serum level is clinically
less important for pulmonologists. In contrast, in the evaluation for hepatic
disease the protective threshold is not the issue but abnormal phenotypes are
relevant. We showed that the majority of liver patients with a heterozygous
(Pi MZ) phenotype had serum levels above the cutoff value of 1.0 g/l. Addition­
ally, heterozygous (Pi MZ) A iA T deficiency LT recipients had serum levels in
the normal range. We concluded that relying on A iA T serum levels alone, runs
the risk of missing the right diagnosis.5
Third, other reports focused on associations between the Z allele and different
liver diseases.The first reports describing associations between heterozygous
A iA t deficiency and end-stage liver disease due to various etiologies were
published in 1997 and 1998.6-7 In recent years a plethora of studies reported
the results of the search for single nucleotide polym orphism s (SNPs) and
associations with A iA T deficiency.The topics which were studied can be divided
into 2 groups. First, studies involving the relation between SNPs and the pheno­
type of homozygous (Pi ZZ) A iA T deficiency. For example, the c.-igj3T>C
polymorphism located in the 5 ERPINAi promoter region (c.-igj3T>C ), appeared
to be associated with neonatal liver disease compared to pulmonary disease
in homozygous A iA T deficiency patients.8 Second, studies reporting SNPs in the
SERPINAi gene and their association with different liver diseases. For example,
a report showed an association between the Z allele in the SERPINAi gene
and the presence of liver disease in cystic fibrosis patients.9
We studied the association between the presence of heterozygous A iA T
deficiency and treatm ent outcomes in hepatitis C virus infection.10 Further­
more, we performed a case-control study to detect an association between
c.-ig73T>C and liver disease of various etiologies. We could not show an
association between treatm ent response among wild types and patients
with heterozygous A iA T deficiency and neither c.-igj3T> C was associated
with liver disease of various etiologies.
All in all, these case- control genetic association studies require many patients
and controls (mostly >500) to achieve statistical significant differences. When
113
G ene ral d iscu ssio n
this large sample sizes are required for the detection of statistically significant
differences, we question whether these findings are of major clinical relevance.
Therefore, we addressed our fourth question. What is the clinical relevance
of liver disease due to heterozygous A iA T deficiency? We showed that the
development of chronic liver disease due to heterozygous liver disease is a
serious issue. Heterozygous disease can progress to end-stage liver disease and
HCC requiring liver transplantation as well as homozygous A iA T deficiency.11
The studies described in this thesis might have clinical consequences as we
hope that clinicians would be more aware of A iA T deficiency as cause of liver
enzyme disturbances and cirrhosis. In case of unexplained liver enzyme
elevations or cryptogenic cirrhosis, diagnostic tests for A iA T deficiency should
be performed. To prevent under-diagnosis, they should start the diagnostic
path with search for genetic mutations rather than quantify serum levels.
In addition, the clinical picture of liver disease due to heterozygous A iA T
deficiency is similar to homozygous A iA T deficiency. Both heterozygous and
homozygous A iA T deficiency related liver disease can induce end-stage liver
disease requiring livertransplantation. HCC isa common complication of liver
disease in A iA T deficiency; however the majority is so-called incidental. Our
findings support surveillance for HCC in A iA T deficient patients with liver
disease, however the effect on survival still needs to be established.
Further research should aim at the detection of modifiers for the clinical
phenotype of A iA T deficiency. In other words: why do some patients develop
liver disease, others develop pulmonary disease and do some A iA T deficient
subjects have no complications? Future investigation might follow research
examples in cystic fibrosis, as the clinical expression of this hereditary disease
is influenced by different genetic modifiers. For example, a recent study
showed an association of CF transmembrane conductance regulator (CFTR)
mutations and the presence of diabetes in cystic fibrosis.12
114
G en eral d iscu ssio n
References
1
Kok KF, W ahab PJ7 de Vries RA. Heterozygosity for alph ai-an titrypsin deficiency as a cofactor in
the developm ent o f chronic liver disease. Ned Tijdschr Geneeskd. 2005 Sep i0 ;i4 9 (3 7 ):2 0 5 7 -6 i.
2
Kok KF, W ahab PJ, Houwen RH, Drenth JP, de Man RA, van Hoek B, M eijer JW, W illekens FL,
de Vries RA. Heterozygous alpha-1 antitrypsin deficiency as a co-factor in the developm ent of
chronic liver disease: a review. Neth J Med. 2007 M ay;65(5):i6o-6.
3
van Hasselt PM, Kok KF, Vorselaars AD, van Vlerken L, Nieuw enhuys E, de KoningTJ, deVries
RA, Houwen RH. Vitam in K deficiency bleeding in cholestatic infants with A lp ha-i-A n titrypsin
deficiency. Arch. Dis. Child. Fetal Neonatal Ed. 2009;94;F456-F460.
4
Ferrarotti I, Scabini R, Campo I, Ottaviani S, Zorzetto M, G orrini M, et al. Laboratory diagnosis
of alph ai-an titrypsin deficiency. Transl Res 2007 N ov;i5o(5):267-74.
5
Kok KF, W illem s JL and Drenth JPH. The cut-offValue of 100 m g/dL is insufficient to detect
heterozygous a lp h a -i antitrypsin deficient liver disease patients. Liver International 2009
(In press).
6
Eigenbrodt ML, M cCashland TM, Dy RM, Clark J, Galati J. Heterozygous alpha l-antitrypsin
phenotypes in patients with end stage liver disease. Am J G astroenterol 1997 A pr;92(4):6o2-7.
7
Graziadei IW, Joseph J J, W iesner RH, Therneau TM, Batts KP, Porayko MK. Increased risk of
chronic liver failure in adults with heterozygous alphai-antitrypsin deficiency. Hepatology 1998
O ct; 28(4 ) :i 058- 63.
8
Chappell S, Hadzic N, Stockley R, G uetta-Baranes T, Morgan K, Kalsheker N. A polym orphism
of the alph ai-an titrypsin gene represents a risk factor for liver disease. Hepatology 2008
Ja n ;4 7 (i):i2 7 -3 2 .
9
Bartlett JR, Friedm an KJ, Ling SC, Pace RG, Bell SC, Bourke B et al. G enetic modifiers of liver
disease in cystic fibrosis. JA M A 2009 Sep 9 ;30 2 (i0 ):i0 76 -8 3 .
10
Kok KF, van Soest H, van Herwaarden AE, van Oijen MG, Boland GJ, H alangkJ, BergT, deVries
RA, Drenth JP. Influence of a lp h a -i antitrypsin heterozygosity on treatm ent efficacy of HCV
com bination therapy. Eur J G astroenterol Hepatol. 2009 Sep 29. [Epub ahead of print]
11
Kok KF, Porte RJ,Adam R, van den Berg A, Scheenstra R, M etselaar H, van Hoek B ,van Oijen
MGH , Drenth JPH For the European Liver and Intestine Transplant Association (ELITA)Liver
transplantation for A lp h a -i antitrypsin deficiency and the im pact of incidental hepatocellular
carcinom a on Post-Transplant Survial. (subm itted)
12
Scott M. Blackm an, Stephanie Hsu, Lori L. Vanscoy, J. M ichael Collaco, Sarah E. Ritter, Kathleen
Naughton, and Garry R. Cutting genetic modifiers play a substantial role in diabetes com plicating
cystic fibrosis. J Clin Endocrinol Metab. A pril 2009, 9 4 (4 ):i3 0 2 -i3 0 9
115
Summary and
Conclusion
Su m m a ry and C o n clu sio n
A lp h a-i antitrypsin (A iAT) deficiency is a genetic disorder which may lead to
pulmonary and/or liver disease. In this thesis, the clinical consequences of A iA T
deficiency are addressed. First, we described the clinical course of the disease;
second the diagnostic pitfalls; third we focused on genetic associations and
finally on the role of heterozygous A iA T deficiency.
Chapter 1 is an introduction to A iA T deficiency. The history of the detection of
A iA T deficiency and also the SERPINAi gene and the mutations involved in
A iA T deficiency are recorded. Furthermore, we provided an outline of the
physiology of the normal M-type A iA T and the pathophysiology of Z-type
AiAT, which can lead to A iA T deficiency. In addition we described the
pathophysiologic mechanisms of liver and pulmonary disease. Chapter 2 is an
overview of the literature regarding clinical consequences of A iA T deficiency.
The clinical presentation, diagnostic strategy, prognosis and therapy of A iA T
deficiency related liver disease in children and adults are reviewed.
In chapter 3 and 4 studies in neonates with A iA T deficiency are reported.
In chapter 3, we studied the prognosis of neonatal cholestasis due to A iA T
deficiency. We included all neonates who were born in the Netherlands
between 1991 and 2006 and who were diagnosed with A iA T deficiency in the
first 3 months of life. We studied 50 neonates and 15 out of them presented
with vitamin K deficiency bleeding. During follow-up, 5 patients (10%) died
and 2 patients (4%) had to be transplanted. At the end of follow-up 12% had
cirrhosis, 62% had liver enzyme abnormalities and/or hepato-splenomegaly
and 12% had no signs of liver disease. Next, in chapter 4 we assessed the risk
of vitamin K deficiency bleeding in neonates with cholestatic jaundice due
to A iA T deficiency. A total of 40 neonates were studied, 20 of them were
breast-fed and received the recommended daily vitamin K prophylaxis and 20
neonates were form ula-fed. Bleeding tendency was increased in breast-fed
neonates compared to formula-fed neonates.
We performed 2 genetic association studies which are reported in chapter 5
and 6. In chapter 5 we described the prevalence of the Z mutation in 2 cohorts
of patients with hepatitis C virus infection (together 1,048 patients). We found
a heterozygosity rate (Pi MS and Pi MZ) of 0.06, in line with healthy controls
in other studies. Furthermore, we could not detect an association between S
ll8
S u m m a ry an d C o n clu sio n
and Z allele mutations and sustained viral response after treatm ent with
(PEG-)interferon and ribavirin. Additionally, in chapter 6 we reported the results
of a study to investigatethe association betweenthe c.-igj3T>C mutation inthe
promoter region of the SERPINAi gene and liver disease of various etiologies.
We included 297 patients and 297 controls, and performed real-time PCR for
the c.-ig73T>C mutation and Z and S allele. We could not demonstrate that
c.-ig73T>C polym orphism was a risk factor for liver disease due to various
etiologies. However, S allele heterozygosity was increased in patients with
drug induced liver injury and therefore, might contribute to its development.
In chapter 7 we explored the diagnostic strategy for detection of A iA T
deficiency. Testing for A iA T deficiency in current practice starts with serum
level quantification then followed by iso-electrofocussing or genotyping when
the serum level is below a cut-off point of 1.0 g/l. We had the clinical impression
that we missed liver disease patients with heterozygous A iA T deficiency.
Therefore, we studied A iA T serum levels in 3 sets of heterozygous (Pi MZ)
persons. We included 12 patients with liver disease due to various etiologies, 6
patients with cirrhosis due to A iA T deficiency and 15 controls. Serum levels
were higher in patients with cirrhosis and liver disease of various etiologies
compared to unaffected controls. The majority of heterozygous Pi MZ patients
might be missed when we relied on serum level quantification alone.To prevent
under diagnosis, we proposed starting the diagnostic path with iso-electrofocussing or PCR rat her than quantification of serum levels. In chapter 8, we assessed
liver transplantation, the only curative treatment option in end-stage liver dis­
ease. We started with an analysis of the medical records of all patients who had
received liver transplantation in the Netherlands. We detected a higher than
expected prevalence of hepatocellular carcinoma in liver explant specimen:
29% of adult liver transplantation recipients. We sought confirmation of this
finding and studied data of livertransplantation recipients with A iA T deficiency,
collected in the European LiverTransplantation Registry. Similarly, we detected
a high frequency of hepatocellular carcinoma. Furthermore we could dem on­
strate that hepatocellular carcinoma negatively affects outcome after liver
transplantation. Moreover we showed that both homozygous and heterozygous
A iA T deficiency can be associated with end-stage liver disease.
119
Su m m a ry and C o n clu sio n
In conclusion: First, A iA T deficiency is a disorder with a low penetrance; however,
in affected adults and children it is a serious issue and it can be associated with
life-threatening complications. Second, to prevent under diagnosis we strongly
suggest to start the diagnostic path with iso-electrofocussing or genotyping
rather than serum level quantification. Finally, both heterozygous and homo­
zygous A iA T deficiency can be associated with end-stage liver disease.
120
Samenvatting en
Conclusie
Sa m e n va ttin g en Co n clu sie
A lfa -i antitrypsine (AiAT) deficiëntie is een erfelijke aandoening die kan leiden
tot zowel long- als leverziekte. In dit proefschrift proberen we de klinische
consequenties van A iA T deficiëntie te onderzoeken. Ten eerste hadden we
vragen over het klinisch beloop van de ziekte, ten tweede over de diagnostische
valkuilen, ten derde hebben we ons gericht op genetische associaties en ten
slotte bogen we ons over de rol van heterozygote A iA T deficiëntie.
Hoofdstuk i is een inleiding in A iA T deficiëntie. We beschreven de geschiedenis
van het herkennen van A iA T deficiëntie en tevens het SERPINAi gen en de
mutaties die betrokken zijn bij A iA T deficiëntie. We verklaren de fysiologie
van het normale M-type A iA T en de patho-fysiologie van het Z-type AiAT,
welke kan leiden tot A iA T deficiëntie. Daarnaast beschreven we de pathofysiologische mechanismen van lever- en longaandoeningen. Hoofdstuk 2 is
een overzicht van de literatuur met betrekking tot de klinische consequenties
van A iA T deficiëntie. De klinische presentatie, het diagnostisch traject, de
prognose en de therapie van leverziekten door A iA T deficiëntie bij kinderen
en volwassenen komen in dit hoofdstuk aan bod.
In hoofdstuk 3 en 4 beschreven we de studies die we deden bij pasgeborenen met
A iA T deficiëntie. In hoofdstuk 3 rapporteerden we de prognose van neonatale
cholestase door A iA T deficiëntie. Wij includeerden alle pasgeborenen die zijn
geboren in Nederland tussen 1991 en 2006 en bij wie A iA T deficiëntie is vast­
gesteld in de eerste 3 maanden van het leven. We vonden 50 pasgeborenen
met A iA T deficiëntie en 15 van hen hebben zich initieel gepresenteerd met
vitamine K deficiëntie afhankelijke bloedingen. Tijdens de follow-up zijn 5
patiënten (10%) overleden en 2 patiënten (4%) moesten worden getrans­
planteerd. Aan het einde van de follow-up had 12% cirrose, 62% had afwijkende
leverenzym en en/of hepato-splenom egalie en 12 % had geen tekenen van
leverziekte. Vervolgens werd in hoofdstuk 4 het risico op vitamine K deficiëntie
afhankelijke bloedingen bij pasgeborenen met cholestatische geelzucht als
gevolg van A iA T deficiëntie onderzocht. Er werden 40 pasgeborenen bestu­
deerd, 20 van hen kregen borstvoeding en ontvingen dagelijks de aanbevolen
vitamine K profylaxe en 20 pasgeborenen kregen flesvoeding. Het risico op
bloedingen was verhoogd bij pasgeborenen die borstvoeding kregen in verge­
lijking met pasgeborenen die flesvoeding kregen.
Sa m e n va ttin g en C o n clu sie
Wij hebben 2 genetische associatie studies uitgevoerd die worden beschreven in
hoofdstuk 5 en 6. In hoofdstuk 5 beschreven we een studie naar de prevalentie
van de Z-mutatie in 2 cohorten van patiënten met hepatitis C virus infectie
(samen 1048 patiënten). We vonden een heterozygotie prevalentie (Pi MS
en Pi MZ) van 0.06, in overeenstemm ing met gezonde controles in andere
studies. Bovendien konden we geen associatie ontdekken tussen S en Z alIel
mutaties en therapiesucces 1 jaar na behandeling met (PEG-) interferon en
ribavirine. Tevens onderzochten we in hoofdstuk 6 de associatie tussen de
c.-i973T> C mutatie in de promotor regio van de SERPINAi gen en leverziekten
door verschillende oorzaken. Wij bestudeerden 297 patiënten en 297 controles
en we hebben real-time PCR verricht voor het vaststellen van de c.-igj3T> C
mutatie en het Z en S-allel. We konden niet aantonen dat c.-igj3T> C polymorfisme een risicofactor is voor leverziekte door verschillende oorzaken. Echter,
S allel heterozygositeit bleek vaker voor te komen bij patiënten met een
geneesmiddel geïnduceerde leverschade, dit suggereert dat S allel hetero­
zygositeit zou kunnen bijdragen aan de ontwikkeling van door geneesmiddelen
geïnduceerde leverziekten.
In hoofdstuk 7 gaan we nader in op het diagnostisch traject van A iA T deficiëntie.
Sommige auteurs beargumenteren dat het testen op A iA T deficiëntie zou
moeten beginnen met het bepalen van een serumspiegel, gevolgd door isoelectrofocussing of genotypering als de serumspiegel kleiner is dan 1,0 g/l. We
hadden klinisch de indruk dat we heterozygote leverpatiënten misten bij het
volgen van die strategie. Daarom bestudeerden we A iA T serum spiegels in 3
groepen van heterozygote (Pi MZ) personen. Wij onderzochten 12 patiënten
met een leverziekte door meerdere oorzaken, 6 patiënten met cirrose als
gevolg van A iA T deficiëntie en 15 controles. Serumspiegels waren hoger bij
patiënten met cirrose en leverziekten door verschillende oorzaken vergeleken
met controles. De meerderheid van de heterozygote Pi MZ patiënten zouden
zijn gemist als we alleen serumspiegels hadden bepaald. Om het missen van
de diagnose te voorkomen, stelden wij voor om het diagnostische traject te
starten met iso-electrofocussing of PCR in plaats van het bepalen van serum ­
spiegels. In hoofdstuk 8 onderzochten we levertransplantatie bij patiënten
met A iA T deficiëntie, dit is de enige curatieve behandeling in het eindstadium
123
Sa m e n va ttin g en Co n clu sie
van deze leverziekte. We begonnen met een analyse van de medische dossiers
van alle patiënten die een levertransplantatie hadden ondergaan in Nederland.
Tegen alle verwachtingen in, ontdekten we hepatocellulair carcinoom in 29%
van de leverexplantaten bij volwassen. We zochten bevestiging van deze
bevinding en bestudeerden vervolgens de gegevens van patiënten die een
levertransplantatie moesten ondergaan wegens A iA T deficiëntie, zoals deze
verzameld waren in het Europese levertransplantatie register. Wij vonden ook
in deze patiëntengroep een groot aantal hepatocellulair carcinomen. Daarnaast
konden we aantonen dat patiënten met een hepatocellulair carcinoom een
slechtere overleving hadden na levertransplantatie in vergelijking met patiënten
zonder hepatocellulair carcinoom. Verder hebben we laten zien dat zowel
homozygote als heterozygote A iA T deficiëntie kan leiden tot eindstadium
leverziekte.
Concluderend: Ten eerste, A iA T deficiëntie is een aandoening met een lage
penetrantie, echter bij aangedane kinderen en volwassenen is het een serieus
probleem en het kan leiden tot levensbedreigende complicaties. Ten tweede,
om het missen van de diagnose te voorkomen adviseren wij om het diagnostisch
traject te starten met iso-electrofocussing of genotypering in plaats van het
bepalen van serumspiegels. Ten slotte, zowel heterozygote als homozygote
A iA T deficiëntie kan leiden tot eindstadium leverziekte.
124
List of publications
Prevalence of Genetic Polymorphisms in the Promoter Region of the A lp h a-i
Antitrypsin (SERPIN A i) Gene in Chronic Liver Disease: a Case Control Study.
BMC Gastroenterology 2010, 10:22 Kok KF, te Morsche RH, van Oijen MGH
and Drenth JPH.
The cut-off Value of 100 mg/dL is insufficient to detect heterozygous A lp h a-i
Antitrypsin deficient Liver Disease patients. Kok KF, W illems JL and Drenth
JPH. Liver Int. 2009 Nov 30. [Epub ahead of print]
Influence of alp h a-i antitrypsin heterozygosity on treatm ent efficacy of HCV
combination therapy.Kok KF, van Soest H, van Herwaarden AE, van Oijen MG,
Boland GJ, Halangk J, Berg T, de Vries RA and Drenth JP. Eur J Gastroenterol
Hepatol. 2009 Sep 29. [Epub ahead of print]
Vitamin K deficiency bleeding in cholestatic infants with Alpha-i-Antitrypsin
deficiency, van Hasselt, PM, Kok KF, Vorselaars R, van Vlerken L, de Vries RA,
Nieuwenhuis E, de KoningTJ and Houwen RHJ. Arch Dis Child Fetal Neonatal
Ed. 2009 Nov;94(6):F456-6o.
Value of molecular analysis of W ilson's disease in absence of tissue copper
deposits: diagnosis at adulthood and novel ATPB7 mutation. Kok KF, Hoevenaars
B, W aanders E and Drenth JPH. Neth J Med. 2008 Sep;66(8):348-50.
Heterozygous alpha -1 antitrypsin deficiency as a co-factor in the development
of chronic liverdisease: a review. Kok KF, W ahab PJ, Houwen RHJ, Drenth JPH,
de Man RA, van Hoek B, Meijer JWR, Willekens FLA and de Vries RA. Neth J
Med. 2007 M ay;65(5):i6o-6.
Pu b licatio n s
Acute pancreatitis bij het gebruik van peg-interferon en ribavirine bij een
hepatitis C positieve patiënt. Kok KF and de Vries RA. Ned Tijdschr Geneesk
2006;150:681-3.
Heterozygotie voor alpha-i antitrypsine deficiëntie als co-factor bij de ontwik­
keling van een chronische leveraandoening. Kok KF, W ahab PJ and de Vries
RA. NedTijdschrG eneesk 2005;149:2057-61.
Dankwoord
Het project dat in dit proefschrift wordt beschreven is in meerdere fases ver­
richt. Tijdens de opleiding tot MDL-arts leek het op een dieseltreintje dat op
elk stationnetje stopt en hortend en stotend steeds weer verder gaat. Na het
afronden van de opleiding, mocht ik in de "kelder", in alle rust, ver weg van de
kliniek, plaatsnemen tussen de onderzoekers. Daar kreeg het project het tempo
van een Intercity, vervolgens ging de trein steeds harder rijden en werd in de
laatste maanden een echte HSL. Daardoor kon het eindstation worden bereikt
voordat ik in het Canisius Wilhelmina Ziekenhuis ging starten als MDL-arts.
Het afronden van het promotietraject was niet mogelijk geweest zonder de
hulp van velen.
Beste professor dr. JPH Drenth, beste Joost,
Hoeveel dankwoorden kan ik op papier zetten? Eén ding is zeker, zonder jou
was het dieseltreintje geen intercity geworden en was het eindstation nimmer
bereikt. Ik heb genoten van onze wekelijkse bijeenkomsten die inspirerend en
leerzaam waren. Je opmerkingen bij teksten waren sporadisch ontm oedi­
gend, soms grappig ("dame, wat doet deze zin hier?"), meestal helder en altijd
leerzaam.
Beste dr. MGH van Oijen, beste Martijn,
In de fase dat dit project als dieseltreintje reed, was jij degene die de motor
soms iets extra brandstof gaf zodat het bleef rijden. Nadat ik bij jou in de
"kelder" kwam wonen, heb je je over het project ontfermd en ben je een steeds
grotere rol gaan spelen. Ik ben er trots op dat ik je eerste (co)promovendus
ben, velen zullen volgen.
127
D ankw oord
Beste dr. RA de Vries, beste Richard,
Ik wil je danken voor je wijze, klinische lessen. Dank voor het feit dat je me
hebt gestimuleerd om dit promotietraject te starten. Je hebt het onderzoek
dat in dit proefschrift staat beschreven op de rails gezet.
Beste professor dr. JBMJ Jansen, beste Jan,
Dank voor het "warme nest" datje als afdelingshoofd wist te bieden aan artsen
en onderzoekers. Jij komt echt op voor je mensen. Ik heb het gewaardeerd dat
ik na het beëindigen van de opleiding tot MDL-arts het onderzoeksproject kon
afronden.
Daarnaast hebben vele anderen een rol gespeeld bij het tot stand komen van
dit proefschrift.
Beste Roderick, Peter en Lotte (Wilhelmina Kinderziekenhuis Utrecht), ik wil
jullie bedanken voor de prettige samenwerking bij de studies bij kinderen met
alp h a -i antitrypsine deficiëntie.
Greet en Hanneke (ook Utrecht) bedankt voor jullie bijdrage aan de studie
bij hepatitis C patiënten. De phenotyperingen zijn verricht in het Rijnstate
Ziekenhuis te Arnhem doorTeun, Frans en Yvonne: dank voor het analyseren
van de vele monsters.
Ik heb met veel plezier gewerkt op het "Lab Gastro" in het Radboud: René,
Hennie, Jannes en W ilbert bedankt. René, dank voor het ontwerpen van
de primers en probes; ik heb vol bewondering toegekeken hoe jij mijn
Excel-bestanden onderhanden nam waarna er allerlei mogelijke variaties en
associaties uit de computer kwamen rollen.
Dank aan de collega's afkomstig uit Groningen, Rotterdam en Leiden die
gegevens beschikbaar stelden zodat de studie naar de uitkomst van levertransplantatie bij ap h a-i antitrypsine deficiëntie verricht kon worden.
D ankw oord
Petra, dank voor je hulp en voor de gezelligheid onderweg bij het doorkruizen
van het land op zoek naar statussen. Ook de zoektocht naar diepgevroren
monsters op een donkere zolder zal ik niet snel vergeten.
Janneke, wat heb je er een mooi boekje van gemaakt, bedankt.
Mijn keldergenoten Serena, Mieke en Merel, jullie waren prettige collega's,
maar wat voelde ik me soms oud als we op maandagochtend het weekend
doornamen.
Daarnaast wil ik dank uitspreken aan alle klinische collega's uit het UMC St
Radboud: Nynke, Eric,Tessa, Aura, Roland, Manon,Tanya, Joris, Marije, Bertram,
Marriëtte, Dirk, Geert en Fokko, ik heb veel van jullie geleerd.
Mijn collega's in de maatschap internisten / MDL van het CWZ en in het bijzonder
Adriaan, Ellen en Nieves: dank voor de prettige samenwerking.
Promoveren is vaak organiseren, ook thuis moest het (en moet het nog steeds)
vlotjes lopen. Dank aan iedereen die hulp biedt bij het oppassen en huishouden.
Dan mijn paranimfen, Annemieke en Maureen. Onze vriendschap is ontstaan
in de m entorgroeptijdens de introductieweek in 1994; nu 16 jaar later hebben
we alle 3 onze plek in het medische veld gevonden. Annemieke en Maureen
dank voor jullie steun tijdens de lastige momenten in het promotietraject.
Mijn ouders, jullie hebben ons (Dennis, Ivo en mij) altijd gesteund om het beste
uit onszelf te halen. Jullie lieten ons zien dat er meer bestond dan het prachtige
Twente en stimuleerden ons om ons eigen pad te kiezen.
Familie en vrienden, dank voor de gezelligheid en voor de belangstelling voor
het onderzoek.
En dan als laatste, Martin en Lisa, jullie zijn m ijnthuis. Lieve, kleine Lisa, mama
houdt van jou en zal nu niet meer gaan computeren op jouw "mama-dagen".
129
Curriculum Vitae
Karin Francisca Kok werd geboren op 11 april 1976 te Geesteren (Ov). Ze volgde
het Voorbereidend Wetenschappelijk Onderwijs (VWO) aan het PiusX College
te Almelo. In 1994 startte ze met de opleiding geneeskunde aan de Katholieke
Universiteit Nijmegen (thans: Radboud Universiteit). In 2000 behaalde ze cum
laude het artsexamen. Na haar afstuderen startte ze als arts-assistent-nietin-opleiding op de afdeling interne geneeskunde in Ziekenhuis Rijnstate te
Arnhem. Begin 2002 begon ze daar met de opleiding tot internist (opleiders:
dr. R van Leusen en dr. L Verschoor). In 2005 is de overstap gemaakt naar de
opleiding tot MDL-arts in hetzelfde ziekenhuis (opleider dr. PJ Wahab). In die
periode werd gestart met het onderzoek naar leverziekten bij alp h a-i antitrypsine deficiëntie. In 2007 werd de opleiding voortgezet in het Universitair
Medisch Centrum St Radboud Nijmegen (opleider: prof. dr. JBMJ Jansen). Onder
leiding van prof. dr. JPH Drenth kreeg het onderzoek verder vorm en werd de
eALTA award verkregen. Na haar opleiding tot MDL-arts (2009) heeft ze de
mogelijkheid gekregen om zich bijna volledig op het onderzoek te richten. Ze
is begin 2010 gestart als MDL-arts in de maatschap internisten/MDL in het
Canisius Wilhelmina Ziekenhuis te Nijmegen. Ze is getrouwd met Martin
Tassche, ze kregen in 2008 een dochter: Lisa.