university of groningen genetic predisposition to ... · contains trophoblastic giant cells), yolk...
TRANSCRIPT
University of Groningen
Genetic predisposition to testicular cancerLutke Holzik, Martijn Frederik
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2007
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Lutke Holzik, M. F. (2007). Genetic predisposition to testicular cancer. s.n.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 29-06-2019
CIP-gegevens Koninklijke Bilbiotheek, Den Haag
Lutke Holzik, M.F.
Genetic predisposition to testicular cancer
Proefschrift Groningen. – Met lit. opg. – Met samenvatting in het Nederlands.
Verschijningsvorm: Digitaal
ISBN: 978-90-367-3109-6
© Copyright 2007 M.F. Lutke Holzik
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, mechanically, by photocopying, recording or otherwise
without the written permission of the author.
Cover and lay-out: Annemiek van der Kleijn
Printed by: Gildeprint, Enschede
RIJKSUNIVERSITEIT GRONINGEN
Genetic Predisposition to Testicular Cancer
Proefschrift
ter verkrijging van het doctoraat in de
Medische Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. F. Zwarts,
in het openbaar te verdedigen op
maandag 1 oktober 2007
om 16:15 uur
door
Martijn Frederik Lutke Holzik
geboren op 25 februari 1974
te Enschede
Promotores Prof. dr. H.J. Hoekstra
Prof. dr. D.Th. Sleijfer
Copromotores Dr. R.H. Sijmons
Dr. J.E.H.M. Hoekstra-Weebers
Beoordelingscommissie Prof. dr. S. Horenblas
Prof. dr. J.W. Oosterhuis
Prof. dr. P.H.B. Willemse
Paranimfen Stephan Lutke Holzik
Deepu Daryanani
All studies in this thesis were designed and carried out at the department of Surgical Oncology,
in cooperation with the departments of -Medical Oncology, -Genetics, -Medical Biology, -Pathology
and Wenckebach Institute, University Medical Center Groningen, the Netherlands and the
Comprehensive Cancer Center North-Netherlands, Groningen, the Netherlands, the department
of Clinical Genetics, University of Antwerp, Belgium and the department of Surgery, Medisch
Spectrum Twente, Enschede, the Netherlands.
This research was financially supported by the Dutch Cancer Society (Koningin Wilhelmina Fonds)
grant RUG: 94-873 and RUG 99-2130 and “de Jan Kornelis de Cock” foundation.
Financial support for the publication of this thesis kindly provided by:
Medisch Spectrum Twente, Stichting Werkgroep Interne Oncologie, Integraal Kankercentrum
Noord-Nederland, KWF kankerbestrijding, Novartis, GlaxoSmithKline, Sanofi-Aventis, Roche,
Astra Zeneca, KCI Medical, Bioprof, Janssen-Cilag, Amgen, Tyco Healthcare, Stryker, Laprolan,
Johnson & Johnson Medical, Ethicon Endo-Surgery, Nycomed, Promega, Graduate School GUIDE
and MSD.
Aan mijn ouders,
Voor Marjolijn en Louise.
Contents
Chapter 1 General introduction, aim and outline of the thesis ............................................................................11
Chapter 2 Genetic predisposition to testicular germ-cell tumours..................................................................23
Lancet Oncololgy. 2004; 5:363-71
Chapter 3 Syndromic aspects of testicular carcinoma ...................................................................................................43
Cancer. 2003; 97:984-92
Chapter 4 Do the eastern and northern parts of The Netherlands
differ in testicular cancer? ...................................................................................................................................................57
Urology. 2001; 58:636-7(letter)
Chapter 5 Testicular carcinoma and HLA Class II genes...............................................................................................61
Cancer. 2002; 95:1857-63
Chapter 6 Absence of constitutional Y chromosome AZF deletions in
patients with testicular germ cell tumors ........................................................................................................75
Urology. 2005; 65:196-201
Chapter 7 Re-analysis of the Xq27-Xq28 region suggests a weak association
of an X-linked gene with sporadic testicular germ cell tumour without
cryptorchidism...................................................................................................................................................................................85
European Journal of Cancer. 2006; 42:1869-74
Chapter 8 Interest in and motivations regarding genetic testing for testicular
germ cell tumour susceptibility....................................................................................................................................97
Submitted
Chapter 9 Summary, discussion and future perspectives........................................................................................ 113
Chapter 10 Nederlandse samenvatting, conclusies en toekomstperspectieven.............................. 123
Reference list ................................................................................................................................................................................... 135
Addendum........................................................................................................................................................................................... 157
Dankwoord ......................................................................................................................................................................................... 163
List of publications................................................................................................................................................................... 169
Curriculum Vitae........................................................................................................................................................................... 175
Genetic Predisposition to Testicular Cancer
1
11
Chapter 1
General introduction, aim and outline of the thesis
General introduction12
Genetic Predisposition to Testicular Cancer
1
13
General introduction, aim and outline of the thesis
General introductionTesticular tumours can be divided into germ cell tumours, stromal tumours and other tumours
(e.g. malignant lymphomas). Tumours of paratesticular structures form a separate group. The
research in this thesis focused solely on the testicular germ cell tumours (TGCT) seminoma and
non-seminoma. TGCT are rare, but they are the most frequently occurring tumour in men aged
between 15 and 40 years. In the Netherlands, 536 men were diagnosed with TGCT in 2003, while
in 2004, 30 men died of this malignancy. Although the incidence of TGCT has increased sharply
in recent years, survival of patients with TGCT has improved enormously. Five year survival in
the nineteen seventies was about 65% compared to more than 90% at present (Figures 1 and
2).(1) Improved survival can chiefly be attributed to the cisplatin-based polychemotherapy that
was introduced in the nineteen eighties to treat patients with metastasized TGCT. In addition,
new strategies have been developed in the surgical approach to metastasized/non-metastasized
TGCT and alterations have been made to the radiotherapy technique and radiation dose for
seminoma.(2;3) The progress in diagnosis, treatment and the subsequent treatment outcomes
in patients with TGCT are the ultimate result of multidisciplinary team work. At the University
Medical Center Groningen (UMCG), this multidisciplinary approach was started at the end of the
nineteen seventies to provide every patient with tailored treatment. These accomplishments in
the treatment of TGCT have led to the present goal of further optimising the treatment for TGCT,
����������
��������
����������
���������������
������
�����������������������������
�
�
�
�
�
��
��
��
��
��
���� ���� ���� ���� ����
Figure 1: Incidence of TGCT
General introduction14
Genetic Predisposition to Testicular Cancer
1
15
in which the research and treatment chiefly concentrate on reducing the toxic side-effects of
chemotherapy and radiotherapy. In patients with prognostically favourable factors (Table 2)(4),
the number of courses of chemotherapy can be reduced, whereas in patients with prognostically
unfavourable factors, more intensive chemotherapy is necessary to improve the chances
of survival. Nowadays the majority of TGCT patients can be cured by the multidisciplinary
treatment. Therefore, the number of TGCT survivors will continue to increase. In principle, these
TGCT survivors will be exposed to the long-term consequences of chemotherapy-related toxicity
(side-effects) cq the long-term side-effects of radiotherapy. TGCT survivors are mostly young men
who can be expected to have a long life ahead of them. This has meant that over the past few
years, scientific research has centred on studying the long-term effects of treatment(5-8) and the
quality of life of these TGCT survivors.(9) All this research has the ultimate aim of achieving further
improvement in the treatment and follow-up of TGCT patients.
The genetic background of TGCTCarcinoma in situ (CIS) (or intra-tubular germ cell neoplasia) is the precursor of TGCT and is
found in nearly all TGCT together with an invasive component. CIS cells originate from primordial
germ cells that “escaped” normal differentiation in utero. It is assumed that over the course of
time, CIS “develops” into an invasive TGCT, but the precise transformation of premalignant CIS
into a TGCT is not yet clear. It is suggested that the default pathway follows the development of
CIS into seminoma and that non-seminoma requires activation of pluripotency (reprogramming)
of a CIS or seminoma cell.(10;11) A theoretical model of TGCT development as part of testicular
dysgenesis, taking into account a range of reported TGCT risk factors, has been developed by
Skaekebaek et al(12) and is discussed in subsequent chapters. In recent years several studies
have looked into chromosomal abnormalities and more recently at gene mutations and gene
activity in TGCT to unravel the molecular pathways underlying these tumours. A detailed
overview of (non-inherited) genomic aberrations in TGCT was recently published by Von
Eyben.(13) Aneuploidy has been found in nearly all cases and triploidy is a common finding.
Seminoma have a mean hypertriploid DNA index and non-seminoma have a mean hypotriploid
DNA index (due to loss of chromosomal material during cancer progression).(11) When looking at
individual chromosomal regions, an isochromosome of the short arm of chromosome 12, i(12p),
(resulting in a duplication of the short arm of chromosome 12) is found in about 80% of TGCT.
The remainder have excess 12p genetic material in derivative chromosomes.(14) The exact relation
between these changes and TGCT is unclear but the absence of amplification of a section of
12p in intratubular germ cell neoplasia, suggests that this amplification may be related to
progression of the disease rather than initiation.(15) A recent gene expression profile study on
TGCT material identified differentiated expressed genes on 12p. Seventy-three genes on 12p were
significantly overexpressed, indicating that the p arm of chromosome 12 may play an important
role in TGCT tumorigenesis.(16) In addition to the genes located on 12p, a growing list of genes is
implicated in the various stages of TGCT development. In particular, TGCT has been shown to be
associated with a characteristic series of abnormalities in the retinoblastoma pathway including
upregulation of cyclin D2 and p27 and downregulation of RB1 and the cyclin-dependent kinase
inhibitors of p16, p18, p19 and p21.(13) A gain of activity of the KIT gene, a member of the tyrosine
kinase family, appears to play a role in the progression of CIS towards seminomas.(17) Recently,
the scope of genetic study of TGCT has been extended to include the role of naturally occurring
micro RNAs (miRNAs). Indeed, some of these miRNAs (miRNA-372 and 373) were shown to allow
tumorigenic growth. As they have also been observed to be expressed in human seminomas
and non-seminomas, but not in normal testicular tissue, it has been suggested that these miRNA
may represent a new class of oncogenes involved in TGCT.(18) Studies of hereditary genetic
changes and TGCT are widely discussed in the subsequent chapters.
HistologyTGCT can be divided into two important histological subtypes: seminoma and non-seminoma.
Pure seminoma occurs in about 50% of the cases and often (in 20%) contains trophoblastic
giant cells that can produce beta-human chorionic gonadotrophin (ß-HCG). Non-seminoma chiefly
comprises two or more cell types, for example embryonal carcinoma, choriocarcinoma (also
contains trophoblastic giant cells), yolk sac tumour, teratoma, whether or not in combination
with seminoma. Embryonal carcinoma and yolk sac tumour elements are associated with the
production of alpha-fetoprotein (AFP). Clinically, there are three important tumour markers in
the diagnosis and follow-up of TGCT: ß-HCG, alpha-fetoprotein (AFP) and lactate dehydrogenase
(LDH).(2)
������
������
������
�������
�
��
��
��
��
���
����������� � ��������� � ��������� � ����������
��������������������������
�
��
��
��
��
���
����������� � ��������� � ��������� � ����������
����������������������
�
��
��
��
��
���
����������� � ��������� � ��������� � ����������
�������������������� ����������������������������������������
����������������������������������
�������������������������������
����������������
Figure 2: Survival of TGCT
Figure 1 and 2 from: Signaleringsrapport kanker:
www.kwf.nl(1)
General introduction16
Genetic Predisposition to Testicular Cancer
1
17
Clinical presentation and initial therapyPatients who present with a painless swelling of the testicle and increased levels of AFP, ß-HCG
and/or LDH have a TGCT until proven otherwise. Clinical presentation varies widely. TGCT can also
show extra-testicular clinical manifestation (without any testicular complaints). Retroperitoneal
lymph node metastases can cause (lower) back pain, while haemoptysis can be the result of
pulmonary metastases. Ultrasound examination of the testicle is useful to establish possible
testicular abnormalities. Radical inguinal orchidectomy with high ligation of the spermatic cord,
blood vessels and lymph vessels is the surgical treatment for patients who are suspected of
having a TGCT. Originally, the testicles descend embryologically via the retroperitoneal route
and inguinal canal into the scrotum. This often results in regional metastases from a TGCT that
first arise in the retroperitoneal lymph nodes. A right-sided TGCT tends to metastasize to the
inter-aorto-caval lymph nodes, whereas a left-sided TGCT tends to metastasize to the para-aortal
lymph nodes. At a higher, supradiaphragmatic level, metastases can spread via the thoracic
duct and result in mediastinal or supraclavicular metastases. Haematogenically, TGCT chiefly
metastasize to the lungs, later to the liver, the skeleton and the cerebrum.(3;19)
StagingWhen a patient has been diagnosed with a TGCT, the malignancy must be staged. This can be
done by means of tumour markers (AFP and ß-HCG, LDH) and spiral CT (computed tomography)
scanning of the lungs, the retroperitoneum and pelvis. On indication (anamnestic complaints
of the cerebrum and/or sharply elevated ß-HCG), CT scanning of the cerebrum is conducted.
The roles and/or additional value of magnetic resonance imaging (MRI) and positron emission
tomography (PET) are currently under investigation.(20) At the UMCG, patients with TGCT are staged
according to the Royal Marsden classification (Table 1). Patients with stage I have no radiological
or biochemical evidence of metastases. Patients with stages II to IV have metastasized disease.
These patients are subsequently classified according to the prognostic factors formulated by the
International Germ Cell Cancer Collaborative Group (IGCCCG) (Table 2) and a treatment plan is
drawn up on the basis of the subgroup classification (good / intermediate / poor).(4)
Table 1: Royal Marsden staging classification of testicular germ cell tumours
Stage Criteria
Stage I No evidence of metastases
Stage IM No clinical evidence of metastases, but persistent
elevation of serum tumour markers AFP and/or hCG
Stage II Infradiaphragmatic lymph node metastases
IIA Metastases < 2 cm in diameter
IIB Metastases 2-5 cm in diameter
IIC Metastases > 5 cm in diameter
Stage III Supradiaphragmatic lymph node metastases;
Status A, B, C as for stage II
Stage IV Extra lymphatic metastases
L1 <- 3 Lung metastases
L2 > 3 Lung metastases, all <- 2 cm in diameter
L3 > 3 Lung metastases, one or more > 2 cm in diameter
H+, Br+, Bo+ Liver, brain, or bone metastases
Table 2: IGCCCG prognostic classification for germ cell cancer (4)
Non-seminoma Seminoma
Good prognosis Testis / retroperitoneal primary Any primary site
and No non-pulmonary and No non-pulmonary
visceral metastases visceral metastases
and AFP<1000 ng/ml and and Normal AFP, any hCG,
hCG<1000 ng/ml and any LDH
LDH<1.5x N*
Intermediate prognosis Testis / retroperitoneal primary Any primary site
and No non-pulmonary and Non-pulmonary
visceral metastases visceral metastases
and and Normal AFP, any hCG,
1000 <- AFP <- 10.000 ng/ml or any LDH
1000 <- hCG <- 10.000 ng/ml or
1.5x N <- LDH <- 10x N
Poor prognosis Mediastinal primary No patients classified as
or Non-pulmonary visceral poor prognosis
metastases
or AFP>10.000 ng/ml or
hCG>10.000 ng/ml or
LDH>10xN
* N=normal range
TreatmentStage I disease (=non-metastasized disease)
About half of the patients with a non-seminoma present with stage I disease. Presently, the
treatment comprises radical orchidectomy with high ligation of the spermatic cord, blood vessels
and lymph vessels, followed by regular outpatient visits (wait and see policy) or modified
unilateral nerve-sparing retroperitoneal lymph node dissection. The UMCG has been applying
the wait and see policy to patients with stage I disease since 1982.(21) In the meantime, world-
General introduction18
Genetic Predisposition to Testicular Cancer
1
19
Surgery after chemotherapy
After completion of the chemotherapy, patients with a metastasized tumour (TGCT) at the UMCG
undergo restaging with the aid of serum tumour marker analyses (as described above) and spiral
CT scanning of the lungs and retroperitoneum. When radiological investigation shows residual
disease after completion of the chemotherapy in patients with seminoma, surgical resection is not
performed, but instead the abnormality is followed radiologically. Generally, a new indication will
arise for chemotherapy or radiotherapy, which if necessary is combined with surgical resection of
the residual tumour. Presently, positron emission tomography (PET) scanning can be applied to
help identify viable cancer.(25) After patients with non-seminomatous TGCT have completed the
chemotherapy, there is no indication for surgical resection when they do not show any residual
disease, or the residual abnormality is smaller than 1 cm, the tumour markers have normalised
and mature teratoma is absent from the primary tumour. Follow-up is then conducted on the
basis of tumour markers. However, when residual (abdominal / pulmonary) disease is identified,
surgical resection must be performed on the residual retroperitoneal tumour mass (RRRTM), or
local resection in the case of residual pulmonary tumour tissue. When residual tumour persists
after completion of the chemotherapy and the tumour markers remain high, an individual
treatment plan must be drawn up: salvage chemotherapy or salvage surgery. In patients with
non-seminomatous TGCT, it is not possible to predict the histology of the residual tumour. After
surgical resection, histological examination shows that the mass consists of necrosis in 45% of
the cases, mature teratoma in 40%, viable tumour tissue in 10% and non-germ cell malignancies
in the remaining 5%.(26)
When the resected residual tumour only contains necrosis or mature teratoma, no further
treatment is necessary and the patient has an excellent prognosis. However, when viable
tumour tissue is present, the prognosis is less favourable and depending on factors such as
the initial prognostic classification (Table 2), the volume (percentage) of residual viable cancer
and the completeness of the resection, it may be necessary to administer adjuvant salvage
chemotherapy.(27)
The clinical significance of mature teratoma in the residual tumour is not yet completely clear
and it is impossible to predict the course that can be expected from mature teratoma left in
situ. In the literature, “growing teratoma” is a well-known phenomenon and it can lead to (very)
late tumour recurrence. In addition, there is a risk that mature teratoma will de-differentiate into
a non-germ cell malignancy (e.g. sarcoma). In such cases, the prognosis of the patient is far
less favourable. Radical surgery is the only curative treatment option for these non-germ cell
tumours.
When the primary TGCT contains elements of mature teratoma, there is a greater chance that the
residual tumour will also contain teratomatous elements. Therefore, at the UMCG, all patients
with elements of teratoma in the primary tumour undergo laparotomy and partial retroperitoneal
wide consensus has been reached about the treatment for stage I non-seminoma patients in
the low risk group (i.e. histological examination does not show any vascular invasion of the
tumour). These patients have 15% risk of TGCT relapse, thus the wait and see policy is justified
and comprises regular outpatient visits for 5 years after orchidectomy with physical examination,
tumour marker analysis and frequent radiological investigation. There is a great deal of
discussion about the current treatment policy for patients with stage I non-seminoma in the
high risk group (i.e. histo-pathological investigation shows vascular invasion). Unilateral nerve-
sparing retroperitoneal lymph node dissection (RPLND) or adjuvant chemotherapy leads to a
considerable reduction in the risk of relapse of about 5%. However, the disadvantages of RPLND
(loss of ejaculatory function) or chemotherapy (toxicity) must be taken into consideration. The
UMCG applies still the wait and see policy to these high risk patients. In the UK and many other
European countries, the preferred approach for high risk patients is often to administer two
courses of adjuvant chemotherapy, whereas in the USA it is the trend to perform nerve-sparing
unilateral RPLND.(2;3) Which treatment policy is the best for these so called ‘high-risk’ patients
is unknow.
About 75% of the patients with seminoma have stage I disease. The standard treatment after
orchidectomy is radiotherapy of the retroperitoneal para-aortal lymph nodes with a total dose of
20 Gy. In this way the relapse rate is reduced to 1-3%.(2) The advantage of para-aortal radiotherapy
(while excluding the ipsilateral lymph nodes) is less gastrointestinal and gonadal toxicity. As an
alternative for radiotherapy, adjuvant chemotherapy can be administered. The latter approach
results in the same relapse rates.(22) A wait and see policy alone would lead to relapse rates of
15% to 20% and is less suitable, because there are no sensitive tumour markers for seminoma
and the patient would therefore have to undergo frequent radiological investigation.(2;3)
Metastasized disease
Patients with stage IIa and IIb seminoma receive radiotherapy with a total dose of 30 Gy and
36 Gy, respectively. In contrast with stage I seminoma, the ipsilateral iliac lymph nodes are
also included in the treatment volume (dog-leg field) in patients with stage II seminoma. This
approach achieves 6-year relapse-free survival of 95% in stage IIa patients and 89% in stage
IIb patients. Alternative treatments for radiotherapy can be considered in stage IIb patients, for
example 3 courses of BEP (bleomycin, etoposide and cisplatin) or 4 courses of EP (etoposide
and cisplatin).(2;3)
Treatment for patients with stage IIc - IV metastasized seminoma and with stage II – IV
metastasized non-seminoma comprises chemotherapy in accordance with the prognostic factor
classification. In the good prognosis group, patients with non-seminoma / seminoma receive
chemotherapy in the form of 3 courses of BEP or 4 courses of EP. Patients in the intermediate
and poor prognosis groups are treated with 4 courses of BEP.(2;23;24)
General introduction20
Genetic Predisposition to Testicular Cancer
1
21
dissection at the “original” tumour site after chemotherapy, even when there are no radiological
signs of residual disease.(26)
Aim and outline of this thesisOver the past few years, there has been growing interest in the genetic aspects of testicular
cancer for a variety of reasons. In general terms, there is a scientific need to unravel the
oncogenetic steps of all types of cancer in the hope that understanding the genetic changes will
lead to better (early) diagnosis and treatment options. Some of these genetic changes have a
hereditary nature, i.e. they can be passed on from one generation to the next and they must
therefore be distinguished from the non-hereditary, somatic mutations that occur in the course
of life. Studies on the hereditary predisposition of testicular cancer may lead to the identification
of men with a strongly increased risk of developing testicular cancer who might benefit from
preventive measures. Cure rates for testicular cancer are now high, so there is increasing interest
in fertility aspects and the possible consequences of passing on the predisposition of testicular
cancer to the offspring. Nowadays all patients with testicular cancer who undergo chemotherapy
are in principle offered the opportunity to freeze their semen (cryopreservation). If a patient
is less fertile or infertile after chemotherapy, it is possible to achieve pregnancy by means of
artificial insemination. In this way, there is a risk of transmitting the hereditary predisposition.
Another reason to study testicular cancer predisposition is that as a rule, finding hereditary
cancer-predisposing mutations in genes leads to increased insight into the development of non-
hereditary (so-called sporadic) tumours. In many cases, it has been found that genes involved
in hereditary tumours also play a role in non-hereditary types of cancer.
Now that the genomes of humans and other species are being systematically mapped and the
scope of molecular and statistical analyses in genetic studies is expanding, there has been rapid
progress in tumour genetic research. In contrast with studies on the hereditary aspects of breast
cancer, colorectal cancer and a series of relatively rare tumour syndromes (such as multiple
endocrine neoplasia (MEN1 and MEN2), Von Hippel Lindau, neuro-fibromatosis (NF1 and NF2)
and Li-Fraumeni’s syndrome), studies on the hereditary aspects of testicular cancer have been
unable to identify any germline mutations in genes that could be associated with a high risk of
developing testicular cancer.
The research described in this thesis aimed to contribute to knowledge on hereditary
predisposition to TGCT and focuses on the following aspects:
• The literature on the hereditary and syndromal aspects of TGCT
• Epidemiological aspects of TGCT
• The HLA and Xq27 regions on the genome that might harbour genes contributing to the
development of TGCT
• Occurrence of constitutional deletions of the Yq11 region in men with TGCT
• Interest of men in genetic testing for TGCT
Genetic Predisposition to Testicular Cancer
2
23
Chapter 2
Genetic Predisposition to Testicular Germ-Cell Tumours
MF Lutke Holzik1, EA Rapley2, HJ Hoekstra1, DTh Sleijfer3, IM Nolte 4, RH Sijmons5
Departments of: 1Surgical oncology,3Medical oncology, 4Medical biology, 5Clinical genetics. University Medical Center Groningen, the Netherlands.2Institute of Cancer Research, Section of Cancer Genetics, London, United Kingdom
Lancet Oncology 2004; 5:363-371
Genetic Predisposition to Testicular Germ-Cell Tumours24 Genetic Predisposition to Testicular Cancer
2
25
Genetic Predisposition to Testicular Germ-Cell Tumours
IntroductionThe term testicular cancer encompasses a group of neoplasms that occur from childhood through
to old age. This review focuses on germ-cell tumours of adolescents and adults, specifically
seminomas and non-seminomas, which are thought to arise from carcinoma-in-situ.(28-30) We do
not discuss paediatric germ-cell tumours or spermatocytic seminomas which do not arise from
carcinoma-in-situ and probably have a different cause from germ cell tumours in adolescents
and adults.(28) We focus mainly on testicular germ-cell tumours (TGCT). However extragonadal
germ-cell tumours of adolescents and adults have a very similar cause and arise from carcinoma-
in-situ; they can therefore be thought of as a part of the same disorder.
TGCT is the most common type of malignant disorder in men aged 15 - 40 years. The
yearly incidence in this age group is about 7.5 per 100,000 people, but the incidence varies
substantially between countries. (19;28;31) TGCT can be classified histologically as seminoma (about
55%), which is most commonly diagnosed during the fourth decade of life, and non-seminoma
(about 45%), which is generally diagnosed in the third decade of life. Both subtypes probably
arise from preinvasive carcinoma-in-situ.(32;33) The strongest risk factors for TGCT are a family
history of the disorder(34;35), a previously diagnosed TGCT(36;37), and cryptorchism (undescended
testis)(38-41,81). In addition, patients with Klinefelter’s syndrome(42) or XY gonadal dysgenesis(43)
have a high risk of developing germ-cell tumours.(44) Other less strong risk factors documented
and confirmed in several studies include: testicular atrophy(45), inguinal hernia(34), infertility(46;47),
hydrocele(48), previously diagnosed extragonadal germ-cell tumour(44) and other disorders of male
sexual differentiation. Many of the risk factors for TGCT include disorders of male urogenital
differentiation. This feature has led to the term testicular dysgenesis syndrome (Figure 1), (12) and
both environmental and genetic factors are probably involved. In this review, we summarise the
current knowledge of genetic predisposition to TGCT.
Clues to the existence of genetic predisposition Family history
1 –3% of patients with TGCT report an affected first-degree relative, a proportion higher than
would be expected by chance (see Table 1).(35;48-57) The largest number of reported cases in
a family is five,(58) but most familial clustering concists of relative pairs such as two affected
brothers and to a smaller extent an affected father and son.(35;59) Figure 2 gives examples of
familial clustering in TGCT. Brothers of patients with TGCT have a relative risk of TGCT of 8-10,
and for father-son the relative risk is 4 – 6.(35;51;56) These relative risks for TGCT are higher than
for most other cancer types, which rarely exceed 4. The high relative risk for TGCT is difficult to
attribute only to a shared environmental component. To account for a relative risk greater than
3 without some form of genetic predisposition requires that the environmental risk factor or
factors are extremely potent. Khoury and colleagues(60) calculated that in this situation the risks
Genetic Predisposition to Testicular Germ-Cell Tumours26 Genetic Predisposition to Testicular Cancer
2
27
to exposed people are 50-100 times those of unexposed individuals. Such potent risk factors
have not yet been identified for TGCT.(61)
Racial differences and geographic clustering
Geographic clustering of TGCT(62) and racial differences in the incidence of this disease could
indicate a genetic component in the cause of the disease. The highest incidence is seen in white
people of northern European descent, whereas people of African or Asian descent seem to
have a universally low incidence.(28;59;61;63) The incidence of this disorder in African Americans is a
quarter of that among white Americans(61) , and is similar to that of native African populations;
thus the risk has not changed by much with migration to a new environment. Differences in
incidence persisting after migration argue in favour of genetic rather than exogenous risk factors.
This maintenance of risk contrasts sharply with the situation for cancers of breast, stomach,
colon and ovaries, for which the incidence in immigrant populations tends rapidly towards that
of the host population.(59;61;64;65)
Table 1: First-degree familial TGCT and estimates of relative risk in male first-degree
relatives of patients with TGCT 1985-2002
Author Year Study No. FTC / % FTC Sibs / RR
total Fath
group* -son
Tollerud (48) 1985 Multicentre hospital-based, 6/269 2.2 2/4 5.9
Retrospective
Kruse (52) 1987 Single-centre hospital 3/255 1.2 3/0 NC
based, Retrospective
Patel (54) 1990 Multicentre hospital-based, 5/500 1.0 1/4 NC
Retrospective
Forman (35) 1992 Multicentre hospital-based 12/794 1.5 8/4 9.8B
and national population
based, Retrospective
Westergaard (57) 1996 Data from Danish cancer 22/2261 1.0 10/12 12.3
registry, Retrospective /1.96A
Polednak (55) 1996 Population-based 12/1395 0,86 8/4 NC
Connecticut Tumor Registry
Heimdal (51) 1996 Multicentre hospital-based, 32/1159 2.8 Nc 7.6
Retrospective
Dieckmann (49) 1997 Multicentre hospital-based, 18/1692 1.1/1.7 9/9 3.1
Retrospective and 9/518 7/2 (retrospective
Prospective1 group)
Ondrus (53) 1997 Single-centre hospital 2/650 0.3 2/0 NC
based, Retrospective
Sonneveld (56) 1999 Single-centre hospital
based, Retrospective 17/686 2.5 11/6 8.5-12.7/1.7A
Dong et al.(50) 2001 Data from Swedish 62/4650 1.3% 38/24 8.3/3.9A
family-cancer database,
Retrospective
FTC = familial testicular cancer
NC = not calculated* = Number of familial TGCT cases compared with total group.
RR = Relative Risk for all male first-degree relativesA = RR for brothers / RR for fathersB = RR only for brothers1 = In a prospective multi centre study 18 of the 1692 patients had a first-degree relative with TGCT.
Also a selection of patients from the Berlin (Germany) hospital were investigated: 518 patients,
9 had a first degree relative with TGCT.
Figure 1: The testicular dysgenesis syndrome. The asterisk indicates the possibility that cryptorchidism
(testicular maldescent) acts as a causal risk factor. CIS: carcinoma in situ. Modified frame from:
Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an increasingly
common developmental disorder with environmental aspects. Hum Reprod. 2001;16:972–978.
© European Society of Human Reproduction and Embryology.
Figure 1: The Testicular Dysgenesis Syndrome
Enviromental factors
Testicular dysgenesis
Hereditary disorders
and constitutional
chromosomal
anomalies and
somatic genetic
defects
Reduced semen
quality
CIS −> Testicular
Cancer
Testicular
maldescent
Hypospadias
Disturbed sertoli
cell function
Impaired germ cell
differentiation
Decreased leydig
cell function
Androgen
insufficiency
?*
Genetic Predisposition to Testicular Germ-Cell Tumours28 Genetic Predisposition to Testicular Cancer
2
29
Bilateral TGCT and risk of other tumours
A consistent risk factor for the development of TGCT is having a previous testicular tumour. For
patients who have had a first primary TGCT, the relative risk of developing a second is 25.(37;66)
The prevalence of bilateral TGCT ranges between 1.0% and 5.8% and studies have even shown
an increase.(67) Bilateral involvement of paired organs (breast, retina, and kidney) has proved to
be a clinical marker of hereditary cancer. Bilateral cases are commonly associated with a positive
family history for the same tumours.(56) Heimdal and co-workers(51) found that 2.8% of patients
with TGCT who did not have a family history had bilateral disease compared with 9.8% of those
with a positive family history. Bilateral disease in paired organs could also be the result of a very
early somatic mutation (in embryogenesis), and evidence suggests that a proportion of bilateral
TGCT could indeed be accounted for such a mechanism. Although mutations in the KIT (tyrosine
kinase receptor) gene are rare in TGCT(68) and mediastinal germ-cell tumours(69) , mutations of
this gene occur in a verry high proportion (95%) of tumours from patients with bilateral disease
compared with a smaller proportion (3%) of tumours from those with unilateral disease.(70) When
both tumours from bilateral cases could be assesed, the same mutation was present in both.
Together these results suggest that somatic KIT mutations occur early in embryogenesis, before
the primordial germ cells have divided and migrated to the gonads. Primordial germ cells with
KIT mutations are therefore distributed to both testes; hence KIT mutations are associated with
bilateral disease.(70) In patients with a family history of TGCT, the frequency of KIT mutations
in unilateral TGCT was similar to that detected previously by Looijenga and colleagues.(70)
However, in patients with bilateral disease and a family history of TGCT, only 28% of tumours
had mutations of KIT.(71) Although the reason for this difference is unclear, the evidence suggests
that bilateral disease in the context of familial TGCT has a different pathogenesis from sporadic
bilateral cases and that most familial bilateral cases are explained by the high risk conferred by
the underlying inherited genes.(71)
A genetic predisposition to TGCT does not necessarily have to be site specific; in analogy with
the situation with many proven hereditary tumour predispositions, patients with TGCT and
their relatives could have an increased risk of developing non-TGCT tumours. Dong and co-
workers(50) investigated this trend in 4650 patients with TGCT and with a mean follow-up of 11
years. They calculated a significantly increased standardised incidence ratio for various second
primary tumours after testicular seminomas: TGCT 11.6 (95% CI 7.0-18.1), colorectal cancer 1.9
(1.1-3.1), pancreatic cancer 3.8 (2.1-6.4), renal cancer 2.2 (1.1-4.0), bladder cancer 2.4 (1.5-3.7),
thyroid cancer 5.4 (1.4-14.1) and malignant lymphomas 2.5 (1.3-4.1). Some of the second primary
tumours could be associated with treatment, since seminomas (mostly treated by radiotherapy)
were associated with a higher frequency of second malignant disorders than non-seminomas
(mostly treated with chemotherapy). Furthermore, an analysis of the prevalence of tumours
in relatives also showed an increased risk of TGCT in male relatives of patients with TGCT
(standarised incidence ratio 8.3 for brothers and 3.8 for fathers and sons).(50) For all relatives
together there was no excess risk of other cancer types in first-degree relatives of patients with
Figure 2:
Examples of familial clustering in TGCT.
A. Brother pair, the most common type of pedigree.
B. Father-son pair, the next most common relative pair.
C. Cousin pair.
D. Uncle-nephew pair.
E and F. Rare TGCT pedigrees with many affected cases.
Genetic Predisposition to Testicular Germ-Cell Tumours30 Genetic Predisposition to Testicular Cancer
2
31
TGCT. However, mothers of patients with TGCT had significantly increased standarised incidence
ratios for lung cancer (1.9; 95%CI 1.4-2.6), non-endometrial uterine cancer (2.6; 1.2-4.7), soft-
tissue tumours (2.5; 1.1-4.9) and malignant melanoma (1.8; 1.1-2.6). Other studies(72) have shown
similar trends (although some not significant) of other cancers in mothers. However, all studies (57;73) show a high frequency of TGCT in relatives of patients with TGCT. These findings generally
suggest that relatives of patients with TGCT have a greatly increased risk of developing TGCT
but may not be at risk of developing other types of cancers.
Syndromic characteristics
Two constitutional chromosomal abnormalities are clearly associated with an increased risk
of TGCT. Patients with Klinefelter’s syndrome (47 XXY) have a very high risk of mediastinal,
(extragonadal) germ-cell tumours.(74) About 8% of patients with such tumours have Klinefelter’s
syndrome.(44) Patients with this syndrome rarely develop TGCT, probably because they do not
have germ cells in the testis from shortly after birth; however, about a third of patients with
extragonadal germ-cell tumours present with testicular carcinoma-in-situ and these patients have
a substantial risk of developing TGCT (see later).(44;75;76) Patients with XY gonadal dysgenesis
have a greatly increase risk of germ-cell tumours.(74;77) This increased risk is seen only in patients
with Y chromosome material; those with gonadal dysgenesis and XX or 45X karyotype do not
have an increased risk of TGCT.(51;76;77) Patients with Down’s syndrome (trisomy 21) might also be
at increased risk of TGCT but the numbers are too small for firm conclusions to be drawn. (78;79)
The presence of TGCT in a hereditary syndrome might be an indication of a hereditary
predisposition to TGCT. We have described the prevalence of TGCT in patients with various
hereditary disorders.(76) Owing to the rarity of most of these disorders and the scarce reports
of their occurrence in combination with TGCT, at present there is no statistical proof of an
association of TGCT with these disorders. An important clinnical issue is that in a proportion
of these disorders there is also a range of defects in urogenital differentiation, which suggests
that TGCT in these disorders is indeed a further complication of such a differentiation defect, as
postulated in the model of Skakkebaek and colleagues (figure 1).(12)
Cryptorchism and other disorders of testicular differentiation in patients and relatives
Cryptorchism and other testicular abnormalities such as atrophy, infertility, hydrocele and
inguinal hernia are risk factors for TGCT.(34;45;46;80;81) As regards family history, Forman and co-
workers(35) showed that the frequency of cryptorchism in patients with TGCT did not differ
between those with or without a positive family history. However, very few studies have looked
at the frequencies of these risk factors in male relatives of patients with TGCT. A small study
by Tollerud and colleagues(48) showed that cryptorchism occurred in a significantly greater
proportions of first-degree male relatives of patients with a family history of TGCT (17%) than
of relatives of patients with TGCT who did not have such a family history (5.3%) or of controls
(2.7%). Their study also showed that 50% of patients who had TGCT and a family history
reported a first-degree relative with inguinal hernia, compared with 10.3% of those without a
family history and with 12.7% of controls. The high prevalence of cryptorchism, inguinal hernias,
and hydrocele among men in these families suggests that an underlying alteration in urogenital
embryogenesis could be associated with the familial predisposition to testicular neoplasia.
Studies on Twins
Studies on twins with cancer have been used to address two general issues. First, are there any
carcinogenic effects of twinning? This question can be answered by comparison of the occurrence
of a cancer in twins and in singletons. The second issue, what the heritable components are to
that cancer, can be addressed by a proband-wise analysis ( proband = the affected individual
through whom a family with a genetic disorder is ascertained) of monozygotic twins compared
with dizygotic twins or siblings. If there is a heritable component the relative risk should be
higher for monozygotic than for dizygotic twins.
In a cohort of 14326 elderly twins aged 66 – 77 years, Braun and co-workers(82) noted that a
personal history of TGCT was reported by five (0.08%) of 5951 monozygotic twins and 11 (0.16%)
of 6992 dizygotic twins. Swerdlow and colleagues(83) als found a significantly higher risk of
TGCT in dizygotic than in monozygotic twins (odds ratio 1.5; 95% CI 1.1-2.2).(83) These finding
suggest that an environmental component was acting in utero to cause TGCT. To find a possible
genetic effect, twin studies must asses the risk of TGCT in the twin brothers of affected patients;
however, the numbers are too small in many studies for any conclusions to be drawn. Swerdlow
and colleagues(83) identified six pairs of concordant (both affected) twins. The risk of TGCT was
raised in twin brothers of patients with TGCT (relative risk 37.5; 95% CI 12.3 – 115.6) and was
greater in monozygotic (76.5) than dizygotic (35.7) twins, which would be expected if there
is a heritable part to TGCT. This relative risk is several times that found in non-twin brothers
(although the confidence limits are wide because of small numbers) but it does imply that the
genetic element for risk is far larger for TGCT than for most other cancers. Other twin studies
on TGCT have not had sufficient numbers of concordant twin pairs and have been unable to
determine zygosity so could not confirm this result.
Environmental components
The worldwide incidence of TGCT has more than doubled over the past 40 years. The increase
follows a birth-cohort effect, and a probable explanation is that factors in embryogenesis or
early life are contributing to the development of TGCT.(28;84;85) The rapid increase highlights
the importance of environmental factors in this disease, because the genetic composition of a
population simply cannot change in the course of one or two generations. Why this increase is
occuring is unclear, but one theory is that an increase in endogenous oestrogens is contributing
to increase in TGCT in addition to risk factors for the disorder such as the increase in incidence
of cryptorchism and the decrease in fertility and sperm quality.(86;87) Another hypothesis is that
abnormally high oestrogen concentrations in pregnancy predispose the developing gonad to
Genetic Predisposition to Testicular Germ-Cell Tumours32 Genetic Predisposition to Testicular Cancer
2
33
TGCT in adulthood. Direct investigation of this idea is difficult, and many studies have looked at
surrogate features that could reflect high oestrogen concentrations in pregnancy.(87) According to
this hypothesis, a high rate of TGCT in dizygotic twins (because uterine oestrogen concentration
during pregnancy are higher for dizygotic than for monozygotic twins), as well as older ages of
mothers and a higher rate of oestrogen related cancers (eg, breast cancer) in mothers and sisters
would be expected.(88) Overall published reports(87;88) provide conflicting results on the analysis
of these variables. These features are merely weak indicators of high oestrogens concentrations,
so the sample size and statistical power of many of these case-control studies might have
been too low to show any significant association. Although hormonal factors could be causally
involved in the development of TGCT, mothers of patients with TGCT do not have an increased
risk of oestrogen related cancer and the risks of breast cancers does not seem to be increased
in the sisters of these patients.(50;72) Even though many studies have showed similar results, this
hypothesis remains unconfirmed.
Several studies have confirmed a higher frequency of TGCT in dizygotic twins than in singletons
and monozygotic twins. This difference suggests that the concentration of circulating oestrogen,
which is significantly higher in pregnancies with dizygotic than in those with monozygotic twins
or singletons, is a contributory causative factor.(82;83;88) However, more research is needed into
the pathogenetic mechanisms that could cause the increase risk of TGCT in dizygotic twins.
Twinning itself is to a sustantial extent genetically determined, and coinherited genetic factors
could conceivably be contributing to this increase rather than circulating oestrogens.
Modelling studies
Two studies so far have tried to identify the mode of inheritance of TGCT susceptibility genes.
One was based on the frequency of bilateral TGCT,(89) and the other was a segregation analysis
on a group of Norwegian and Swedish families.(90) In a segregation analysis, a disease model
is sought for familial aggregation of a disease by fitting the patients and non-affected family
members in pedigrees to, for example, a dominant or recessive disease model. Heimdal and
colleagues(90) did such an analysis on all available patients with TGCT treated at a Norwegian
hospital and a Swedish hospital between 1981 and 1991 (n=978). For 30 patients, a first-degree
relative also had TGCT; there were no families with more than two affected members. The
investigators concluded that the familial clustering was best accounted for by a major gene for
TGCT with a recessive mode of inheritance. Their analysis took into account that the incidence
of TGCT has increased substantially over the past few years and that the treatment for TGCT
has improved greatly and led to better maintenance of fertility and longer survival. The study
showed that time trends in TGCT make little difference to the evidence for the recessive mode of
inheritance.(90) Under this recessive model, the estimated gene frequency was 3.8%. Thus 7.3%
(according to the Hardy-Weinberg equilibrium) of men in the population carry the mutant allele
and that 0.1% are homozygous. According to the calculations by Heimdal and colleagues, the
life-time risk of development of TGCT in homozygous men is 43%.(90)
Nicholson and Harland(89) aimed to define the incidence of genetically determined TGCT in
the general population. They analysed published data on the age of onset of TGCT and the
prevalence of bilateral TGCT in familial and unselected general cases according to the arguments
used by Knudson’s two-hit model for tumorigenesis in patients with a familial predisposition
to a certain cancer.(91) In the general population bilateral TGCT occurs much more frequently
than could be attributed to chance. On the basis of the comparison of the distribution of age
at diagnosis between patients with bilateral TGCT and familial cases (ie, those likely to be
genetically determined), the investigators assumed in their model that the increased risk was
due solely to genetic susceptibility. Thus patients with bilateral TGCT probably carry the same
susceptibility genotype as that causing familial TGCT. Nicholson and Harland(89) estimated that
about a third of the general patients with TGCT carry the susceptibility genotype and that the
penetrance is 0.45. Calculations with these values showed that a recessive disease model
showed a better fit with the observed risks for brothers (2.2%) and fathers (0.5%) of patients
with TGCT than a dominant disease model. The frequency of the susceptibility allele in the
recessive model was estimated to be 5%. Although this analysis, based on the simplest set of
assumptions, fitted the data reasonably well, Nicholson and Harland conceded that it was based
on data obtained under diverse conditions and could be subject to several unknown biases.
Therefore, the assumptions that underlie the model might be incorrect. They also conceded
that some unilateral tumours may be predisposed to a contralateral tumour by some as yet
unknown biological mechanism (indeed, such a mechanism was later identified as early somatic
KIT mutations, as iscussed above) or that there is more than one predisposition gene and that
the mode of inheritance is more complex.
X-linkage was not specifically addressed by the two studies mentioned. The higher relative risk
for brothers than for father-son pairs is compatible with a recessive mode of inheritance, but
since the early 1990s the incidence of TGCT has increased sharply, maintenance of fertility after
treatment for TGCT has improved greatly, and the introduction of cisplatin-based chemotherapy
has substantially lowered the number of deaths from TGCT.(92-94) These changes might lead to a
situation in which the relative risk for father-son is higher than was believed initially. A further
possibility is that because of shorter survival (ie, a lower chance that a TGCT patient will father
a child) and abolition of fertility in the past (before the 1990s), less transmission of TGCT from
father to son was seen. Although the study by Heimdal and colleagues(90) tried to take this
factor into account, the investigators admitted that to do so is very difficult and that more
complex assumptions are involved. The study by Nicholson and Harland(89) made allowance
for the time trends, but partly because only two studies have attempted to define the mode of
inheritance, there is a risk that the conclusion of inheritance according to a recessive model is
incorrect. A dominant form of inheritance should therefore not be totally excluded. Because a
proportion of bilateral cases are caused by early somatic mutations in KIT and because there is
clearly more than one TGCT susceptibility gene, susceptibility to TGCT is probably more complex
than suggested by either of these models.
Genetic Predisposition to Testicular Germ-Cell Tumours34 Genetic Predisposition to Testicular Cancer
2
35
Association studies and haplotype analyses
Association studies try to find statistical evidence of an association between polymorphic loci
on, or closely linked to, candidate genes (eg, genes involved in the metabolism of mutagenic
agents, known tumour-suppressor genes or oncogenes) and a particular phenotype (eg, TGCT)
by comparing frequencies between cases and controls. Because TGCT is a rare disease and
mapping of the causal genes is difficult, TGCT patients and families from founder populations
are very suitable people in whom to detect TGCT predisposition genes with the aid of
association or haplotype analyses.(95;96) Patients in founder populations are expected to share a
high number of mutations predisposing to TGCT from recent common ancestors.(62) They will also
share segments of DNA surrounding the disease mutations. Haplotypes consisting of conserved,
ancestral alleles at genetic markers covering the region of a disease mutation will therefore be
more frequent among patients than among controls.
Several association studies have been done on TGCT. In Table 2, an overview is given of these
studies. Much attention has been paid to HLA genes and TGCT. The HLA region is thought to be
interesting for such studies, because differences in immune response based on HLA variation
might have a role in the development of cancer and metastases. Hodgkin’s lymphoma, Kaposi’s
sarcoma, colorectal cancer and breast cancer are associated with this genomic region.(97;98) The
effects (based on HLA variation) of the immune response to carcinogenic factors such as viruses
that might be associated with TGCT, could contribute to the development of the disorder.(99;100)
Previous studies have shown that HLA factors might be associated with the development of
TGCT. In particular, consistent associations were found with the HLA class II antigens. No
association was found with the HLA A or C regions, and inconsistent associations were found
with the HLA B region(96) (Table 2). The method used in most studies was serotyping, which is
much less accurate and less efficient than the more recent genotyping studies. DNA typing is
far more sensitive and can identify larger numbers of alleles.(101) The first HLA genotyping study
by Özdemir and colleagues(102) in 55 Japanese patients, showed two associations: one HLA
susceptibility allele and one HLA protective allele (relative risk 3.26 and 0.26 respectively). A
much larger genotyping study on the HLA class II region of chromosome 6p21 in 151 patients
from the northern part of the Netherlands could not replicate this association.(96) The association
between TGCT and HLA class II alleles either does not exist or is much weaker than the earlier
report suggested.
The other association studies in Table 2 give an overview of the diverse associations that
researchers have attempted to show with TGCT. In many cases, only one study has investigated
a certain region. To date, no convincing associations has been shown between TGCT and a
genetic polymorphism.
Linkage analysis and genome-wide screens
Linkage analysis relies on the fact that during the formation of gametes through meiosis
chromosomal material is exchanged between homologous chromosomes. This recombination
of genes means that were once on the same homologue of a particular chromosome pair
become separated and those once on different homologues are brought together. The process
of recombination is achieved by crossing-over of the chromosomes during meiosis. The
probability that a cross-over will occur and two loci on the same chromosome will randomly
segregate increases with the distance between the genetic loci. For two genes close together
on a chromosome, a cross-over is unlikely to occur and the two loci will be inherited together.
Linkage analysis uses this feature to map genes by means of a series of polymorphic genetic
markers (microsatelite markers or single-nucleotide polymorphisms) along the chromosomes
to follow the inheritance pattern of marker loci and the disease locus through a family. If a
particular polymorphism or marker is inherited by a higher proportion of patients than would
be expected by chance, the locus is said to be linked to the disease and is in fact probably
very closely located to the disease locus on the DNA strand. Since the position of the marker
locus is known, the location of the unknown disease locus is then found. Linkage analysis in a
set of pedigrees with many cases of a cancer has identied genes associated with breast cancer,
colon cancer, familial melanoma, and others. For many of these cancers the pedigrees used had
many affected cases spread over several generations. In many cases, a single pedigree provides
sufficient numbers of affected cases to generate a statistically significant LOD score (logarithm
of odds, a statistic that indicates whether a locus is inherited by a higher proportion of patients
than expected by chance). The search for a TGCT susceptibility gene has been hampered by
a lack of these large multigenerational pedigrees. Most families identified are relative pairs,
generally siblings. Larger pedigrees with three, four or five affected cases have been reported
but these rarely extend beyond two generations.
The International Testicular Cancer Linkage Consortium (ITCLC) has the largest collection of
TGCT pedigrees with two or more cases of TGCT in a family.(103) An analysis by this group has
shown that no single autosomal gene accounts for all TGCT pedigrees.(104) The analysis of 160
TGCT pedigrees consisted of calculations under the best autosomal dominant and autosomal
recessive models given by the segregation analysis of Heimdal and colleaguesl(90), but allowing
for the possibility of genetic heterogeneity (ie, several genes are involved separately or in
interaction with each other). An X-linked component to susceptibility was not considered. The
analysis suggests that this set of pedigrees has sufficient power to detect two TGCT genes each
contributing to 50% of the set with a statistically significant LOD score greater than 3 under
both a dominant and a recessive model. The analysis also investigated the power to detect four
TGCT susceptibility genes each contributing to a quarter of the family set. In this case, significant
LOD scores would not be possible unless the number of families in the set approached 500
pedigrees. The last report by the ITCLC was on a total of 179 pedigrees and the failure to find
LOD scores greater than 3 in any autosomal locus suggests that no single locus can explain
occurrence of TGCT in at least 50% of the families. It also suggests that there are more than
two TGCT susceptibility genes.(103) The power to detect some or all of the susceptibility genes
Genetic Predisposition to Testicular Germ-Cell Tumours36 Genetic Predisposition to Testicular Cancer
2
37
Tabl
e 2:
Tes
ticul
ar C
ance
r As
soci
atio
n St
udie
s
Auth
or
Type
of st
udy
Gen
e As
soci
atio
n Co
nclu
sion
DeW
olf et
al.
In
vest
igat
ed g
enom
ic D
NA o
f 61
TGCT
HLA
A,
Incr
ease
d an
tige
n freq
uenc
y of
DW
7
Pos
sibl
e as
soci
atio
n19
79(1
06)
patien
ts for
52
HLA
A, B, C
B,C
,D
insu
bgro
up w
ith
tera
toca
rcin
oma
betw
een
HLA
DW
7
spec
ifici
ties
. HLA
D a
ssig
nmen
ts m
ade
ant
igen
s (p
<0.0
1)
antige
ns a
nd T
GCT
by
hom
ozyg
ous
typi
ng.
Carr e
t al
.
Ass
esed
20
patien
ts w
ith
HLA
In
crea
sed
antige
n freq
uenc
y of
AW
24
Pos
sibl
e as
soci
atio
n 19
79(1
07)
tera
toca
rcin
oma
antige
ns
in p
atie
nts
with
met
asta
tic
betw
een
HLA
AW
24
dise
ase
(p=
0.00
8)
and
met
asta
tic
TGCT
Maj
sky
et a
l. Te
sted
23
HLA
ant
igen
s of
A a
nd B
HLA
A
No
sign
ifica
nt res
ult
No
asso
ciat
ion
betw
een
1979
(108
) lo
ci In
62 p
atie
nts
with
test
icul
ar
and
B
TG
CT a
nd H
LA A
or B
ge
rmin
ativ
e tu
mou
rs a
nd 3
01 h
ealthy
an
tige
ns
an
tige
ns
unre
late
d su
bjec
ts.
Pol
lack
et al
. Ass
esed
exp
ress
ion
of H
LA a
ntig
ens
HLA
A,B
, Lo
w fre
quen
cy o
f DR3
in a
ll pa
tien
ts,
Caus
al a
nd p
rogn
ostic
1982
(109
) in
145
unr
elat
ed w
hite
pat
ient
s w
ith
C
and
DR
alth
ough
no
differ
ence
was
sig
nific
ant
impo
rtan
ce o
f ov
eral
l
TGCT
an
tige
ns
afte
r co
rrec
tion
for
num
ber of
de
crea
ses
in H
LA D
R3
an
tige
ns tes
ted
rem
ains
to
be d
eter
min
edOliv
er e
t al
. 11
4 TG
CT p
atie
nts
HLA
A, B,
Incr
ease
d DR5
in s
emin
oma
patien
ts,
Pos
sibl
e as
soci
atio
n19
86(1
10)
C
and
DR
(p<0
.04)
and
inc
reas
ed H
LA-D
R7
in
betw
een
HLA
DR5
and
antige
ns
stag
e IV
dis
ease
TGCT
(p<
0.05
) DR7
antige
ns a
nd T
GCT
Aig
inge
r et
1:
inv
estiga
ted
132
TGCT
pat
ient
s HLA
B13
1:
Inc
reas
ed fre
quen
cies
of HLA
B13
in
Pos
sibl
e as
soci
atio
nal
. 19
87(1
11)
2: c
ompl
eted
joi
nt c
alcu
lation
of
and
DR
TGCT
with
hae
mat
ogen
eous
met
asta
ses
betw
een
HLA
ant
igen
s
publ
ishe
d da
ta o
n HLA
ant
igen
s in
an
tige
ns
(p<0
·01)
, of
DR2
in T
GCT
witho
ut
and
TGCT
35
1 TG
CT p
atie
nts
m
etas
tase
s (p
<0·0
01), a
nd o
f DR1
in
se
min
oma
(p<0
·035
)
2: Inc
reas
ed fre
quen
cies
of DR1
(p
<0·0
25) an
d DR5
(p<0
·015
) in
sem
inom
a; inc
reas
ed fre
quen
cy o
f DR5
(p
<0·0
5) a
nd D
R7
(p<0
·05)
in
non-
se
min
oma
with
haem
atog
eneo
us
m
etas
tase
s Au
thor
Ty
pe o
f st
udy
Gen
e As
soci
atio
n Co
nclu
sion
Die
ckm
ann
Rev
iew
ed 1
19 fam
ilies
in
HLA
In
crea
sed
freq
uenc
y of
Hap
loty
pe s
haring
in
all (fam
ilial
) et
al.
1989
(112
) pu
blis
hed
repo
rts
with
an
tige
ns
HLA
ant
igen
s HLA
A3
TG
CT c
ases
tha
t is
mor
e th
an
ag
greg
atio
n of
TGCT
and
B7
(p<0
.02)
ex
pect
ed. Add
itio
nal ev
iden
ce for
theo
ry o
f ge
netic
influ
ence
in
the
ca
use
of
TGCT
Ryb
erg
et a
l. A
sses
sed
rest
rict
ion-
frag
men
t-
HRAS
Hig
her freq
uenc
y of
rar
e
Pat
ient
s w
ith
bila
tera
l ca
ncer
or
1993
(113
) le
ngth
pol
ymor
phis
m o
f th
e
al
lele
s in
TGCT
pat
ient
s
unila
tera
l ca
ncer
you
nger
tha
n 20
VNTR
reg
ion
flank
ing
the
HRAS
(23/
696)
tha
n in
con
trol
s;
year
s of
age
had
sig
nific
antly
gene
in
geno
mic
DNA fro
m 3
48
(p=
0·00
4). Hig
her
high
er inc
iden
ce o
f ra
re H
RAS
TG
CT p
atie
nts
and
343
heal
thy
freq
uenc
y of
rar
e al
lele
s al
lele
s th
an p
atie
nts
with
cont
rols
in b
ilate
ral TG
CT (p=
0·01
) un
ilate
ral ca
ncer
old
er tha
n 20
year
sDie
ckm
ann
In
vest
igat
ed b
lood
gro
ups
of
Lew
is a
ntig
en
Mor
e freq
uent
Lew
is
Ass
ocia
tion
s of
Le(
a-b-
) w
ith
TGCT
et
al.
1993
(114
) 57
7 pa
tien
ts tre
ated
for
TGCT
an
d HLA
an
tige
n Le
(a-b
-) in
patien
ts
and
of H
LA B
w41
with
sem
inom
a
and
cont
rols
; Le
wis
ant
igen
s of
B
w41
th
an in
cont
rols
(p=
0·04
6;
supp
ort co
nten
tion
tha
t ge
netic
14
3 pa
tien
ts tre
ated
for
TGCT
,
antige
ns
rela
tive
risk
2·38
). H
LA
fact
ors
are
invo
lved
in
the
caus
e
and
cont
rols
; an
d HLA
ant
igen
s
Bw
41 a
ssoc
iate
d w
ith
and
path
ogen
esis
of TG
CT.
of
215
pat
ient
s trea
ted
for TG
CT
se
min
oma
(p=0·
0001
;
HLA
-Bw
41 c
ould
be
used
as
risk
an
d co
ntro
ls
re
lativ
e ris
k 8·
2)
mar
ker fo
r se
min
oma
Hei
mda
l et
Com
pare
d TG
CT (n
= 4
42) an
d
WT1
(W
ilm’s
A1
alle
le o
f W
T1
Find
ings
mig
ht ind
icat
e an
al. 19
94(1
15)
cont
rol (n
= 3
84) po
pula
tion
tum
or); O
ne o
f po
lym
orph
ism
ass
ocia
ted
invo
lvem
ent of
the
WT1
bot
h in
fo
r th
e al
lele
fre
quen
cies
of tw
o t
he p
olym
orph
ism
s w
ith
bila
tera
l TG
CT (p=
0.03
) su
scep
tibi
lity
to T
GCT
and
in
po
lym
orph
ic loc
i lo
cate
d at
(W
T) w
as loc
ated
an
d di
ssem
inat
ed T
GCT
pr
ogre
ssio
n of
the
dis
ease
ch
rom
osom
e ba
nd 1
1p13
w
ithi
n an
d th
e
(p=
0.03
)
ot
her (D
11S3
25)
in c
lose
pro
xim
ity
to W
T1)
Genetic Predisposition to Testicular Germ-Cell Tumours38 Genetic Predisposition to Testicular Cancer
2
39
Cont
inue
Tab
le 2
: Te
stic
ular
Can
cer
Asso
ciat
ion
Stud
ies
Auth
or
Type
of st
udy
Gen
e As
soci
atio
n Co
nclu
sion
Olie
et al
. In
vest
igat
ed p
rese
nce
of N
RAS
NRAS
or K
RAS
Mut
atio
ns fou
nd in
five
sem
inom
a NRAS
or K
RAS
mut
atio
ns19
95(1
16)
and
KRAS
mut
atio
ns, in
cod
ons
(thr
ee in
NRAS
and
two
in K
RAS,
all
are
rare
and
app
aren
tly
12
, 13
, an
d 61
, in
sna
p-froz
en
codo
n 12
), a
nd in
one
non-
sem
inom
a no
t es
sent
ial fo
r in
itia
tion
sa
mpl
es o
f 10
0 pr
imar
y TG
CT,
(K
RAS,
cod
on 1
2)
or p
rogr
essi
on o
f TG
CT
cons
isting
of 40
sem
inom
a an
d
60 n
on-s
emin
oma
tum
ours
Hei
mda
l et
Ass
esse
d al
lelic
fre
quen
cies
of Oes
trog
en-
No
differ
ence
s in
alle
lic fre
quen
cies
Dat
a do
not
ind
icat
e th
atal
. 19
95(1
17)
thre
e po
lym
orph
ism
s w
ithi
n th
e re
cept
or g
ene
betw
een
canc
er p
atie
nts
and
cont
rols
va
riat
ion
in the
5’ en
d of
oe
stro
gen-
rece
ptor
gen
e in
(p=
0.08
) th
e oe
stro
gen-
rece
ptor
te
stic
ular
can
cer (n
=45
4) a
nd
gene
con
fers
sus
cept
ibili
ty
co
ntro
l po
pula
tion
s (n
=67
2)
to tes
ticu
lar ca
ncer
Har
ries
et
Ass
ocia
tion
stu
dy w
ith
GST
P1a
G
STP1b
alle
le
6.5%
of co
ntro
ls a
nd 1
8.7%
of TG
CT
The
GST
P1b
alle
le
al. 19
97(1
18)
and
GST
P1b
in
155
TGCT
(v
aria
nt o
f pa
tien
ts h
omoz
ygou
s fo
r th
e GST
P1b
as
soci
ated
with
test
icul
ar
patien
ts
glut
athi
one
al
lele
p=
0.00
2 ca
ncer
S-tran
sfer
ase
cDNAs
)
Özd
emir e
t H
igh-
reso
lution
gen
otyp
ing
HLA
gen
es
HLA
DRB1
susc
eptibi
lity
alle
le
Susc
eptibi
lity
and
al. 19
97(1
02)
stud
y in
HLA
cla
ss II re
gion
(rel
ativ
e risk
3.2
6) a
nd D
QB1
prot
ective
alle
le for
TGCT
on 5
5 pa
tien
ts w
ith
TGCT
prot
ectiv
e al
lele
(0.
26) fo
und
in (no
n)
iden
tifie
d
sem
inom
a pa
tien
ts
Rap
ley
et
Gen
otyp
ing
stud
y on
134
Xq
27 reg
ion
Link
age
of reg
ion
Xq27
in
patien
ts
Link
age
betw
een
Xq27
and
al. 20
00(1
05)
pedi
gree
s w
ith
TGCT
with
TGCT
and
at le
ast on
e fa
mily
TG
CT (pe
rhap
s al
so
mem
ber w
ith
bila
tera
l te
stic
ular
cr
ypto
rchi
sm)
ca
ncer
p=
0.0
34
So
nnev
eld
Hig
h re
solu
tion
gen
otyp
ing
HLA
gen
es
No
asso
ciat
ion
betw
een
HLA
cla
ss
No
asso
ciat
ion
betw
een
et a
l.
stud
y in
HLA
cla
ss II re
gion
on
II
gene
s an
d (n
on) se
min
oma
HLA
cla
ss II ge
nes
and
2002
(96)
15
1 TG
CT p
atie
nts
pa
tien
ts –
log 10
(p-v
alue
)<3
TGCT
Tum
our cl
assi
ficat
ion
/ ty
pe o
f st
udy
acco
rdin
g to
the
orig
inal
artic
le. TG
CT: te
stic
ular
ger
m c
ell tu
mou
r
for TGCT will depend greatly on the degree of genetic heterogeneity of this disease. Continued
identification and recruitment of families into this linkage study will be crucial to the efforts to
identify these genes.
The first evidence for a TGCT susceptibility gene was for the X chromosome.(105) The analysis on
99 pedigrees compatible with X linkage (no male-to-male transmission) yielded a heterogeneity
LOD score of 2.01 on chromosome Xq27. The data were subsequently stratified according to
the presence of at least one bilateral case, the presence of undescended testis, histology, and
age. Families with at least one case of bilateral disease showed strong evidence of linkage to
a locus on Xq27 (heterogeneity LOD score 4.76) and were more likely to show linkage to the X
chromosome than were families without a bilateral case. This score corresponds to a genome-
wide significance of p=0.034, equivalent to a LOD score of 3.78 in a genome-wide linkage search
of affected sibling pairs without subgrouping. This proposed susceptibility gene on chromosome
Xq27 was called TGCT1.(105) In Klinefelter’s syndrome, the increased risk of extragonadal germ-cell
tumours is associated with an extra copy of the X chromosome which further suggests X-linked
involvement in TGCT.(75;105) In addition, the subset analysis provided evidence that the TGCT1
might also predispose to cryptorchism. Linkage to this locus was found in 73% of families with
cryptorchism compared with 26% of families without cryptorchism (p=0.03). The results also
suggest that about a third of the excess familial TGCT risk to brothers is from TGCT1, with little
difference in the residual risks to brothers and sons after this locus has been accounted for. This
is the first cancer gene to be mapped in a genome wide search of predominantly sibling pairs
and was the third male cancer gene mapped to the X chromosome.(105)
Conventionally, such a result would be ratified in another set of families; however no other set
of TGCT pedigrees of similar size is known. These pedigrees are rare and collection of a similar-
sized set may take several years. The ITCLC reported preliminary results on an additional 25
pedigrees compatible with X linkage but the set was to small for any firm conclusions to be
drawn.(103) The gene in this region has not been identified.
TGCT1 is unlikely to be the only TGCT susceptibility gene and it could account for as few as
25% of all TGCT pedigrees. Several genome searches have now been done on the ITCLC set,
and suggestive evidence for linkage has been obtained for several autosomal regions including
3q27, 12q12-q13, 16p13, 18q22 – qter.(103) Many more pedigrees need to be assesed before any
of these regions can be conclusively identified as including a TGCT susceptibility gene.
DiscussionThere is evidence that TGCT susceptibility genes exist and are important in this disease.
This evidence includes increased TGCT risks associated with a positive family history, the higher
frequency of bilaterality in familial cases and the ethnic and racial differences that do not
change with migration, as well as the mathematical modelling of observed familial and non-
Genetic Predisposition to Testicular Germ-Cell Tumours40 Genetic Predisposition to Testicular Cancer
2
41
familial cases, and possible associations with known hereditary syndromes and constitutional
chromosomal anomalies; the latter commonly seen within the setting of defects in urogenital
differentiation.
Human TGCT susceptibility genes have not yet been identified. The putative gene mapped to
Xq27 is postulated to confer an increased risk of TGCT as well as cryptorchism. Completion of
the human gene map, further studies on animal models, the arrival of advanced gene arrays
(chips), genome-wide single- nucleotide polymorphism technology, and applied bioinformatics
are expected to facilitate further exploration of genetic predisposition to TGCT. One point of
interest is whether such predisposition can be linked to a genetic contribution from increased
intrauterine oestrogen concentrations and susceptibility to disruption of usual urogenital
differentiation or to an environmental factor that would account for the increasing incidence
of TGCT. Insight into genetic features of TGCT not only might contribute to the identification
of individuals at increased risk of developing the disorder, but also is likely to increase our
understanding of normal urogenital differentiation and non-hereditary TGCT. This knowledge will
contribute to improving the diagnosis and treatment of TGCT in the general population. From
a practical clinical perspective, identification of men with an increased risk of TGCT depends
on the presence of known risk factors, including the family history of cancer. Clinicians should
therefore record the family history of cancer and urogenital differentiation defects as part of their
routine clinical practice in patients with TGCT or urogenital differentiation defects. Brothers of
patients with TGCT should be informed about their risk of developing the disorder and should
be encouraged to examine their testicles regularly. The US National Cancer Institute and the
ITCLC are doing a genetic and causal multidisciplinary study of familial TGCT. (http://familial-
testicular-cancer.cancer.gov/). However, unlike some other increased risk situations determined
by positive family histories (eg, those of colorectal cancer), the effects of preventive measures,
including testis self-examination, in terms of tumour risk as well as psychosocial effects remain
to be investigated.
Search strategy and selection criteria
Articles and studies included in this review were identified by searches of titles, abstracts and
keywords of reports included in PubMed and through the list of references cited by the papers
found by those PubMed searches. Searches were done with combinations of the search terms:
” testicular”, “seminoma”, “non-seminoma”, ”familial”, “risk factors”, “genetic”, “tumo(u)r”,
“gene(s)”, “linkage” and “association”. Papers published before 1970 and those published in
languages other than English, French and German were excluded.
Genetic Predisposition to Testicular Cancer
3
43
Chapter 3
Syndromic Aspects of Testicular Carcinoma
MF Lutke Holzik1, RH Sijmons2, DTh Sleijfer3, DJA Sonneveld1,
JEHM Hoekstra-Weebers4,5, J van Echten-Arends2, HJ Hoekstra1
Departments of: 1Surgical Oncology, 2Genetics, 3Medical Oncology, 4Wenckebach Institute, University Medical Center Groningen, the Netherlands.5Comprehensive Cancer Center North-Netherlands, Groningen, the Netherlands
Cancer 2003; 97:984-992
Syndromic aspects of testicular carcinoma44 Genetic Predisposition to Testicular Cancer
3
45
Syndromic Aspects of Testicular Carcinoma
IntroductionAlthough testicular carcinoma (seminoma and non-seminoma) is a rare disease, it is the most
common form of malignant disease in men between the ages of 20-40 years.(28;62) The exact
etiology of testicular carcinoma remains unknown; however, over the years, various risk
factors have been identified, including factors with an assumed or definite genetic basis (i.e.,
cryptorchidism, familial testicular carcinoma). Greater understanding of the molecular foundation
of hereditary tumor predisposition will not only facilitate the identification of men who have
an increased risk for testicular carcinoma but also will provide more insight into the origination
of nonhereditary forms of testicular carcinoma. There are various indications of a genetic
predisposition for testicular carcinoma.(61;105) Supporting arguments are the presence of familial
clustering and the racial differences in incidence of this tumor. Recently, DNA-linkage studies
have indicated the existence of a gene on the X chromosome that, when mutated in the germ
line, is associated with an increased risk for the development of testicular carcinoma. Another
argument is the presence of testicular carcinoma in patients with a hereditary abnormality or
with a constitutional chromosomal anomaly. Systematic research into the incidence of testicular
carcinoma in patients with these disorders is scarce in the literature.(119;120) The objectives of the
current study were to study disorders have been reported in combination with seminomatous
and non-seminomatous testicular carcinoma in the literature and to examine the extent to
which our current knowledge of the genetics and pathogenesis of these disorders contributes to
gaining a better understanding of the oncogenesis of testicular carcinoma.
Materials and MethodsA literature search was made for articles in the English language on seminomatous and/or
non-seminomamatous testicular carcinoma that cooccurred with hereditary disorders or
constitutional chromosomal anomalies. We also searched for articles on risk factors for testicular
carcinoma, because risk factors that have been established in the normal population might
also form part of a genetic condition and (partly) may explain testicular carcinoma in patients
with those conditions. The Literature sources were: Pubmed (www.ncbi.nlm.nih.gov/PubMed),
McKusick’s on-line catalogue of hereditary phenotypes (Online Mendelian Inheritance in Man,
www.ncbi.nlm.nih.gov/Omim) and Familial Cancer Database 1.2 (found at http://facd.uicc.org).(121)
The key words were (combinations of) testicular cancer, germ cell cancer ,seminoma, non-
seminoma, congenital anomalies, syndromes, hereditary ,inherited, mendelian , genetic and risk
factors. For an additional source, we used the references listed in the articles found using above-
described method. We excluded all articles which were not written in English and/or described
hereditary disorders or constitutional chromosomal anomalies with testicular carcinoma, other
than (non)seminoma.
Syndromic aspects of testicular carcinoma46 Genetic Predisposition to Testicular Cancer
3
47
Table 1: Hereditary Disorders and
Constitutional Chromosomal Anomalies
in Combination with Reported Testicular
Carcinoma (Alphabetical Order)
Name Syndrome a
Synonym b
OMIM # c
Androgen insensitivity syndrome XLR AR gene 1, 2, 3 X î (43;
Testicular feminization syndrome located at 152-160)
Androgen receptor deficiency Xq11-q12
Dihydrotestosterone receptor deficiency
# 300068
Adrenogenital syndrome AR CYP21 gene 1 (161;162)
Congenital adrenal hyperplasia (CAH1) located at
# 201910 6p21.3
Ataxia teleangiectasia AR ATM gene 4 11 (163)
Lois – Bar syndrome located
# 208900 at 11q22.3
Congenital total hemihypertrophy AR, (164)
Hemihyperplasia Spor,
Isolated hemihyperplasia impr
# 235000
Down’s syndrome 21 1 21 î (78;79;
Trisomie 21 165-168)
# 190685
Familial male-limited precocious AD LCGR gene 4 (169)
puberty (FMPP) located at
# 176410, 152790 2p21
Familial atypical multiple mole AD CDKN2A/p16, 12 î (170-
melanoma syndrome located at 173)
Familial dysplastic nevus syndrome 9p21, CDK4
FAMMM located at
# 155600, 155601 12q14, CMM1
located at
1p36
Hereditary persistence alpha-feto AD AFP located 4 (174-
protein (HPAFP) at 4q11-q13 177)
# 104150
Kallmann syndrome 1 XLR KAL1, located 1, 2, 4 X î (178)
# 308700 at Xp22.3
Mod
e of
In
her
itan
ce d
Gen
e /
Gen
e m
ap loc
us e
Ris
k fa
ctor
num
ber
co
rres
pond
ing
with
tab
le 2
f
Tum
orcy
togen
etic
ab
norm
ality
g
Lite
ratu
re
refe
renc
esî
î
Table 1 continued
Name Syndrome a
Synonym b
OMIM # c
Klinefelter syndrome XXY X, Y 2,4 X î (42;43;
Y 74;179-
185)
Li Fraumeni syndrome AD TP53 located (73;132)
# 151623, 114450 at 17p13.1,
CHK2 located
at 22q12.1
Marfan AD 15q21.1 (120;
# 154700 186)
Mixed gondal dysgenesis X,Y 1,2 X î (43;123;
45,X/46,XY Gonadal dysgenesis Y 187-191)
# 233420, 306100
Neurofibromatosis type I AD NF1, located (192)
Von Recklinghousen disease at 17q11.2
# 162200, 162220
Noonan’s syndrome AD PTPN11, 1,2 12 î (193;
Male turner syndrome, Pterygium located at 194)
Colli syndrome 12q24.1
# 163950
Persistent mullerian duct syndrome AR AMH located 1,2,3, 12 î (195-
# 261550 at 19p13. 4,5 200)
3-p13.2
AMHR2 gene
located at
12q13
Prader-Willi syndrome AD, cgd(15q)/UPD 1, 2, 4 (201;
Prader-Labhart-Willi syndrome impr (mat) located 202)
# 176270 at 15q11-q13
Proteus syndrome Spor, (203)
Encephalocraniocutaneous lipomatosis, AD?
ECCL
# 176920
Mod
e of
In
her
itan
ce d
Gen
e /
Gen
e m
ap loc
us e
Ris
k fa
ctor
num
ber
co
rres
pond
ing
with
tab
le 2
f
Tum
orcy
togen
etic
ab
norm
ality
g
Lite
ratu
re
refe
renc
es
î
î
Syndromic aspects of testicular carcinoma48 Genetic Predisposition to Testicular Cancer
3
49
ResultsOur literature search revealed 83 articles and a total of 25 hereditary disorders and constitutional
chromosomal anomalies in combination with testicular carcinoma (see Table 1). Table 2 presents
a list of recognized risk factors for testicular carcinoma that can occur as part of a hereditary
disorder or constitutional chromosomal anomaly (e.g., we excluded exposure to estrogens in
utero). Column 4 in Table 1 shows which risk factors for testicular carcinoma were present per
disorder in correspondence with the factors numbered in Table 2. Detection of statistically
significant correlations between the chromosomal / hereditary disorder and the testicular tumor
could not be performed due to the rarity of these disorders.
DiscussionThe literature search revealed 25 different hereditary disorders and constitutional chromosomal
anomalies that cooccurred with seminomatous or non-seminomatous testicular carcinoma.
These co-occurrences can be explained in two ways. They simply may be coincidental, or they
may result directly or indirectly, possibly interacting with other endogenous or exogenous risk
factors, from the constitutional genetic defect underlying the hereditary disorders / chromosomal
anomalies in question.
The majority of the hereditary conditions listed in Table 1 are extremely rare; only a few dozen to
100 patients with such a congenital disorder have been described. The combination of testicular
carcinoma and a hereditary condition often was described only in one or a few case reports, never as
the subject of clinical epidemiologic research into large groups of patients with a certain hereditary
abnormality. Moreover, many publications on hereditary disorders were limited to relatively young
patients, which means that complications in adulthood, such as testicular carcinoma, have received
little attention. This might have led to under-reporting of testicular tumors in patients with such
disorders. Conversely, there may be a publication bias in view of the remarkable nature of the
cooccurrence. These possible causes for bias and the small number of patients reported make it
difficult to provide statistical proof of a significant correlation between having a certain hereditary
Table 1 continued
Name Syndrome a
Synonym b
OMIM # c
Prune Belly syndrome Spor ?, 1,2 (204;
# 100100 AD?, 205)
AR?
Rubinstein-Taybi syndrome AD CREBBP (206)
Broad thumb-hallux syndrome gene located
# 180849 at 16p13.3
Russell-Silver syndrome Spor+ UPD(mat) 7 1,4,5,6 7 î (124)
Silver-Russell Dwarfism, Silver-Russell imprXL?, 17q23-q24
syndrome, SRS, Russel-Silver Dwarfism AD?,
# 180860, 312780 AR?
Supernumerary nipples, familial AD ? Multi- (207-
Polythelia, familial factorial? 209)
# 163700
Testicular germ cell tumour, familial XL Xq27 1,2,3 X î (105)
# 300228
Von Hippel-Lindau disease (VHL) AD VHL located (131)
# 193300 at 3p25-p26
X-linked ichthyosis XLR, STS located 1 X î (210;
Steroid sulfatase deficiency AD at Xp22.32 211)
# 308100 a Name: name of syndrome b Synonym = synonym of the syndrome c OMIM (#) number: McKusick’s on-line catalogue of hereditary phenotypes found at:
http://www3.ncbi.nlm.nih.gov/Omim/ d Mode of inheritance: AD = Autosomal dominant, AR = Autosomal recessive, XL = X-linked,
XLR = X-linked recessive, Spor = Sporadic, Impr = Imprinting, ? means that the mode of inheritance
is suggested in the literature, but is inconclusivee Gene/Gene map locus: name of the gene and locus, or only the locus of the gene if the gene has
only been mapped, but not cloned. UPD(mat) = uniparental (maternal) disomyf Risk factor number: risk factor corresponding with the number in Table 2g Tumorcytogenetic abnormalities: known (parts of) chromosomes that were identified as abnormal in
cytogenetic studies on (non)seminomatous tumours in general: = under-representation,
î = Over-representation
If a box is empty, the data are unknown
Mod
e of
In
her
itan
ce d
Gen
e /
Gen
e m
ap loc
us e
Ris
k fa
ctor
num
ber
co
rres
pond
ing
with
tab
le 2
f
Tum
orcy
togen
etic
ab
norm
ality
g
Lite
ratu
re
refe
renc
es
î
Table 2: Recognized Risk Factors for Testicular Carcinoma that May Be Part of Hereditary
Disorders and Constitutional Chromosomal Anomalies
Risk factor References
1. Cryptorchidism (34;212-217)
2. Subfertility (infertility) (34;46;218)
3. Inguinal hernia requiring surgery (34;39;80;214)
4. Hypogonadism (165)
5. Hypospadias (214)
6. Early age at puberty (34;218-220)
Syndromic aspects of testicular carcinoma50 Genetic Predisposition to Testicular Cancer
3
51
disorder and the development of testicular carcinoma, unless the risk of a testicular malignancy is
relatively high in certain hereditary disorders. The so-called intersex disorders, (and, in particular,
gonadal dysgenesis), are examples of such disorders. It is possible that 30% of individuals with
gonadal dysgenesis or mixed gonadal dysgenesis have an increased risk of developing gonadal
neoplasia. These patients have been known to develop a gonadoblastoma; this form of in situ germ
cell tumor has the ability to transform into an invasive germ cell tumor (e.g. seminoma). It appears
to be the presence of Y chromosome material in a dysgenetic gonad that predisposes to testicular
tumor development.(77;122;123)
Due to the fact that it is difficult to gain insight into the etiology of testicular carcinoma in patients
with a hereditary disorder through an epidemiologic-statistical approach, it is important to find other
ways to study the nature of these cooccurrences. A finding that is worthy of attention in this respect
is that many of the conditions listed in Table 1 also involve urogenital differentiation disorders,
several of which are known to be recognized risk factors for testicular carcinoma in the general
population (Table 2). An example is the Russel-Silver dwarf syndrome; over 40% of these patients
have cryptorchidism, hypospadia and an early onset of puberty.(124) Although there is no proof that
these recognized risk elements are direct causal factors or solely epidemiologic markers of an as
yet unknown causal factor, it is conceivable that, particularly in conditions that involve urogenital
differentiation disorders, testicular carcinoma develops as a further expression of such differentiation
disorders. Skakkebaek et al. developed a model that aims to explain the correlation between these
well-established epidemiologic risk factors, genetic factors and testicular carcinoma (Fig. 1).(12;125)
Those authors assume that the cause of testicular carcinoma lies in a condition they refer to as
testicular dysgenesis syndrome (TDS), which is postulated to be caused by a range of environmental
and/or genetic defects that disrupt the embryonal programming of gonadal development during fetal
life. In the Skakkebaek model, known risk factors, such as testicular maldescent (cryptorchidism),
infertility, and hypospadias, do not cause testicular carcinoma but, rather, result from TDS, as does
testicular carcinoma. The genes underlying the hereditary conditions listed in Table 1 may cause TDS
and subsequently may cause (indirectly) testicular carcinoma, together with a range of associated
urogenital anomalies. The Skakkebaek model takes into consideration a variety of urogenital
defects and their severity, including testicular carcinoma in the absence of congenital urogenital
anomalies. The type of genetic defect might influence the severity of TDS and, thus, the severity
and type of any associated urogenital anomalies. Currently, it is not clear whether cryptorchidism
results from TDS, as the model postulates or whether cryptorchidism (also) directly causes testicular
carcinoma. The decreased risk of testicular carcinoma after orchidopexia indicates a more direct role
in neoplastic development, although the data are conflicting and it remains to be seen whether
the Skakkebaek model is correct.(126-129)
From a theoretical point of view it cannot be excluded that in a number patients with testicular
carcinoma cases another pathway of tumor development has occurred, in which testicular
carcinoma develops in a normal differentiated testis, (i.e., in the absence of TDS, for instance
as a result of mutations in tumor suppressor and/or [proto] oncogenes). In the hereditary
disorders referred to as the human cancer syndromes, (130) these mutations are present in the
germline; therefore, these disorders are of special interest when looking at testicular carcinoma
predisposition. Testicular carcinoma has been reported in three of these hereditary disorders:
the Li-Fraumeni syndrome, neurofibromatosis type 1 (Recklinghausen disease) and Von Hippel-
Lindau disease.
Von Hippel-Lindau disease features (clear cell) renal cell carcinoma and epididymis cyst
adenomas as frequent urogenital anomalies.(131) However, only a single report has been published
on testicular carcinoma in Von Hippel-Lindau disease. Furthermore neurofibromatosis type 1 has
been described a number of times in combination with (bilateral) testicular carcinoma. Hartley
et al. and Heimdal et al. even suggested that testicular carcinoma may be a rare manifestation
in the Li-Fraumeni syndrome. (73;132)
Figure 1: The testicular dysgenesis syndrome. The asterisk indicates the possibility that cryptorchidism
(testicular maldescent) acts as a causal risk factor. CIS: carcinoma in situ. Modified frame from:
Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an increasingly
common developmental disorder with environmental aspects. Hum Reprod. 2001;16:972–978.
© European Society of Human Reproduction and Embryology.
Figure 1: The Testicular Dysgenesis Syndrome
Enviromental factors
Testicular dysgenesis
Hereditary disorders
and constitutional
chromosomal
anomalies and
somatic genetic
defects
Reduced semen
quality
CIS −> Testicular
Cancer
Testicular
maldescent
Hypospadias
Disturbed sertoli
cell function
Impaired germ cell
differentiation
Decreased leydig
cell function
Androgen
insufficiency
?*
Syndromic aspects of testicular carcinoma52 Genetic Predisposition to Testicular Cancer
3
53
When attempting to unravel the molecular steps that lead to testicular carcinoma, the field of tumor
cytogenetics can be very helpful. Cytogenetic studies on testicular carcinoma have demonstrated an
increase in chromosome 12p in invasive testicular carcinoma. In addition, complex rearrangements
have been found with increases and decreases of specific chromosomal material: (parts of)
chromosomes 4, 5, 11, 13, 18 and Y are under-represented, whereas (parts of) chromosomes 7, 8,
12, 21 and X are over-represented.(133-136) The search for genes in these regions that are responsible
for testicular carcinoma, including 12p, is still in its early stages and it remains unknown which of
these genes also may carry germline mutations.(137) To date, linkage studies on these regions have
isolated only Xq27 as the locus for a gene (TGCT1) that might be responsible for familial clustering
of testicular carcinoma (105). It is not clear whether the HLA regions harbor a hereditary testicular
carcinoma gene: the data are controversial.(96;102;106;110-112;138)
Table 1 shows whether the genetic defects associated with the hereditary conditions lie in
chromosomal regions that appear to be of interest in cytogenetic studies on testicular carcinoma.
If there are correlations, then this might be another clue to a causal relationship between testicular
carcinoma and the genetic defect concerned. It is striking that, in the current literature review, we
found six hereditary conditions with underlying gene defects/regions that lie on the X chromosome
(Table 1); in addition, cryptorchidism has been described in association with these six hereditary
disorders. These findings, in combination with evidence concerning the Xq27 region suggest that
the X chromosome plays a role in the etiology of cryptorchidism resulting in testicular carcinoma
predisposition whether or not according to the model of testicular dysgenesis (Fig.1).
If, on the grounds of tumor cytogenetics or other considerations (e.g., the fact that a gene is
already known to play a role in other types of tumor), a gene appears to play a candidate role
in the oncogenesis of testicular carcinoma, then further molecular studies on tumors may help to
determine whether their role is more or less probable. Such research can involve searching for
somatic gene mutations, loss of heterozygosity (LOH) (loss of the normal allele in the tumor),
changes in methylation status, and gene expression with the aid of immunohistochemistry, or on a
large scale by means of gene expression arrays. Currently, such research has only been performed
on a very limited scale on genes that are responsible for the hereditary conditions listed in Table
1. Kume et al. recently described a patient with neurofibromatosis type 1 and testicular carcinoma
in whom no LOH of the NF1 gene could be demonstrated.(139) This makes it less probable that
NF1 mutation played a role in the pathogenesis of testicular carcinoma in this patient. Studies
concerning somatic mutations of the gene for Von Hippel-Lindau disease in patients with sporadic
gonadal tumors have not yet revealed any mutations.(140) Screening for the gene of the Li-Fraumeni
syndrome (TP53), in testicular tumors failed to demonstrate any pathogenic mutations, although
some missense mutations of unknown pathogenicity have been observed.(141) Gene expression
studies have suggested that another associated gene (CHK2) may play a role in testicular
carcinoma.(142) Somatic mutations of p16 (the gene mutated in the germline in a proportion of
families with familial dysplastic nevus syndrome) have been observed in testicular carcinoma.(143)
The insulin growth factor binding protein (IGFBP)1 gene may be involved in Russell-Silver
syndrome, and immunohistochemical studies have suggested a role of this protein in testicular
carcinoma development.(144;145) Mouse models with germ line defects in the above mentioned
genes also may have provided clues to associated tumors, although the spectrum of associated
tumors may differ significantly between mice and humans. No testicular tumors were found in
knock-out mice for TP53, NF1, p16, VHL and a range of other known tumor-suppressor genes.(146)
Based on data generated by research to date, we still cannot draw any definite conclusions
regarding the role of the above-discussed germline mutations in testicular oncogenesis.
Completion of the human gene map and the availability of advanced gene arrays and
bioinformatics undoubtedly greatly will facilitate further exploration of the role of hereditary
gene defects in testicular carcinoma. The first gene expression profiling studies on testicular
carcinoma that used large-scale gene arrays (chips) have been published recently and it can be
expected that more candidate genes will be found through studies like these.(147) It would be
interesting to include the genes that are associated with hereditary disorders mentioned in Table
1 in these expression arrays, because they may prove to be associated with testicular carcinoma.
The testicular dysgenesis model has implications for interpreting the results of these expression
studies. It is conceivable that some of the genes that are important in testicular dysgenesis and
(subsequent) neoplasia are expressed normally only in a narrow time window during early gonadal
development. Therefore, those genes will not be expressed in normal adult testicular tissue;
and, if they act as a step in tumor development through loss of action, then gene expression
studies comparing adult normal and testicular carcinoma tissue will not display any differences in
expression. Gene expression studies on normal gonads or dysgenic gonads during various stages of
development (including animal models) may suggest additional genes to be included in further
screening research for mutations.(148)
The objective of this review was to summarize the current knowledge about the hereditary
predisposition of testicular carcinoma. In a last remark about the possible psychological,
social, ethical and economic implications of the identification of men who are at increased
risk of hereditary carcinoma, we note that the literature on individuals with a family history
of malignancy shows that issues such as medicalization, stigmatization, coping with disease-
related worry and anxiety, greater sense of vulnerability, difficulty understanding statistical risks
and risk perception, aspects of decision making, changes in family dynamics and planning, and
difficulties with health insurance are related to the progress of genetic science.(149-151)
In summary, the identification of a hereditary predisposition for testicular carcinoma is likely to
contribute to our understanding of the development of the nonhereditary variety and help to
identify men with an increased risk of testicular carcinoma. We present an overview of hereditary
disorders and constitutional chromosomal anomalies that have been described in the literature
in combination with testicular carcinoma. Although, from an epidemiological point of view, there
Syndromic aspects of testicular carcinoma54 Genetic Predisposition to Testicular Cancer
3
55
seems to be only a direct or indirect correlation between testicular carcinoma in mixed gonadal
dysgenesis and Xq27-linked familial testicular carcinoma, the presence of urogenital differentiation
disorders and data from tumor cytogenetic research, combined with the knowledge on gene loci of
the discussed hereditary disorders, suggest that such a relation also may exist in other syndromes.
New techniques are rapidly becoming available that will enable us to complete the human gene
map and investigate the possible role of large numbers of candidate genes in the development of
testicular carcinoma.
Genetic Predisposition to Testicular Cancer
4
57
Chapter 4
Do the Eastern and Northern Parts of The Netherlands Differ in Testicular Cancer?(letter to the Editor)
MF Lutke Holzik1, DJA Sonneveld1, HJ Hoekstra1, GJ te Meerman2, DTh Sleijfer3,
M Schaapveld4.
Departments of: 1Surgical oncology, 3Medical oncology.
University Medical Center Groningen, the Netherlands
Department of: 2Medical Genetics, University of Groningen, the Netherlands
Department of: 4Comprehensive Cancer Center North-Netherlands Groningen,
the Netherlands
Urology 2001; 58:636-637
Do the eastern and northern parts of The Netherlands differ in testicular cancer?58 Genetic Predisposition to Testicular Cancer
4
59
To the editor,
Spermon et al. (221) conclude in their article that in the Netherlands, brothers of patients with
testicular cancer have an increased risk of developing testicular cancer (TC). Recently, we also
studied familial TC in the northern part of the Netherlands and reviewed the medical records of
686 TC patients treated at the Groningen University Hospital.(56) These results compared with the
results of Spermon et al. are presented in Table I. In addition to familial TC, we studied the age-
adjusted incidence rates of TC in the Netherlands. Within a small country like the Netherlands,
there are geographic differences in incidence of TC present, with a statistically significant highest
incidence in the northern part of the Netherlands.(62) These results are in contrast to the findings
of Spermon et al., who mention that the incidence rates for cancer in the eastern parts of the
Netherlands are similar to other parts of the Netherlands. Shared environmental factors, as well
as genetic drift, might have resulted in the higher incidence of TC found in the stable founder
population in the northern part of the Netherlands.(62) The geographic clustering of TC in stable
founder populations in the northern Netherlands may lend support to a genetic susceptibility
to TC development. The lack of sufficient number of TC families with two or more affected
males has been a handicap to performing adequate linkage analysis studies.(222) Spermon et
al. recommended in their article the mapping of candidate genes for TC in TC families. We
think that not only TC families are suitable in the search for TC susceptibility genes. Our stable
founder population is very useful in tracing TC susceptibility genes because individuals in such
populations share a relatively high frequency of genes from common ancestors; that is, genes
that are identical by descent, which can be detected by association-based studies.(96)
Table 1: Familial testicular cancer: Nijmegen compared to Groningen
Item Nijmegen Groningen
Period 1986 - 1997 1977 - 1997
Patients 379 686
FTC patients 7 (1,8%) 17 (2,5%)
Seminomas 114 (30%) 153 (22%)
Nonseminomas 265 (70%) 540 (78%)
Bilateral TC en FTC 1 (14,3%) 1 (5,9%)
Prevalence of UDT in FTC’s* 0 or 1 (0 or 14,3%) 3 (17,6%)
Inguinal hernia in FTC ? 1 (5,9%)
RR father-son 0,96 1,75
RR for TC in brothers of TC patients 5,9 9 - 13
Key: TC= testicular cancer, FTC= familial testicular cancer, UDT= undescended testis, RR= relative risk.
* Spermon et al mention in their Results section that, in their population, patients with FTC had no
history of undescended testes, whereas in their Comment section (at page 751) they mention that
one of the FTC patients had an undescended testis.
Genetic Predisposition to Testicular Cancer
5
61
Chapter 5Testicular Carcinoma and HLA Class II Genes
DJA Sonneveld1, MF Lutke Holzik1, IM Nolte2, DTh Sleijfer3,
WTA van der Graaf3, M Bruinenberg4, RH Sijmons5,
HJ Hoekstra1 and GJ te Meerman2
Departments of: 1Surgical oncology, 3Medical oncology, 5Genetics.
University Medical Center Groningen, the Netherlands.
Departments of: 2Medical genetics, 4Medical biology
University of Groningen, the Netherlands
Cancer 2002; 95:1857-1863
Testicular Carcinoma and HLA Class II Genes62 Genetic Predisposition to Testicular Cancer
5
63
Testicular Carcinoma and HLA Class II Genes
IntroductionTesticular germ cell tumors (TGCT) constitute the most common malignancy in men 20-40 years
of age. The etiology of TGCT is still poorly understood. In addition to possible environmental
predisposing factors, several observations point to a genetic susceptibility to the development
of TGCT.(19;28) First, familial and bilateral testicular carcinoma cases occur more frequently than
expected by chance. The relative risk (RR) for brothers of TGCT patients ranges from 3 to
13.(49;50;56) Second, a genetic susceptibility is reflected by an increased incidence of TGCT in
persons with certain rare malformations of the urogenital system, some of which have a definite
genetic component in the etiology.(61;123) Furthermore, higher rates of urogenital developmental
anomalies have been reported in families prone to TGCT.(48) Third, the age distribution of TGCT
may suggest a genetic origin of the disease. TGCT are usually diagnosed at a young age and
the incidence declines after the age of 50 years. The young age at onset of testicular neoplasms
indicates a role of important etiologic factors operating early in life, either in utero or shortly
after birth. In addition to in utero exposure to maternal estrogens or exposure to infectious
agents in early childhood, these early operating etiologic factors may also be genetic.(61;223-225)
Finally, racial differences in the incidence of TGCT may point to a genetic component in the
etiology of the disease. The highest incidence is observed in Caucasians of Northern European
descent. People of African descent have a universally low incidence of TGCT. In the United
States, the incidence of TGCT in African Americans is only one fourth of that observed in
Caucasians.(28;61;63) Based on findings in numerous clinical and epidemiological studies, a genetic
susceptibility to TGCT is very likely. Several candidate genes have been proposed to play a role
in the etiology of the disease.(61;105;212;226-229) Recently, a high-resolution genotyping study in 55
Japanese TGCT patients showed a histocompatibility antigen (HLA)-DRB1 susceptibility allele (RR
3.26) and an HLA -DQB1 candidate protective allele (RR 0.26) for TGCT.(102) The possible role
of the HLA system in TGCT development could result from effects of HLA variation on immune
response to carcinogenic factors, for example viruses that may be etiologically associated with
the development of TGCT.(230) The importance of the HLA system in regulating susceptibility to,
and tumor development in, a growing number of neoplastic conditions is becoming increasingly
clear. An impaired immune system, genetically or acquired, favors carcinogenic factors.(97)
Linkage studies in TGCT families could be performed to map candidate genes for TGCT.
Unfortunately, to date, the lack of a sufficient number of families with two or more affected men
with TGCT has been a handicap to perform linkage studies with enough power to find effects of
frequent alleles or genes with low marginal effects.(227) An alternative approach to linkage studies
in familial cases is to search for testicular carcinoma susceptibility genes among testicular
carcinoma patients in founder populations by means of association analysis including haplotype
methods. These approaches have more power than linkage analysis when high frequency alleles
are involved in the pathogenesis. However unlike linkage analysis, they require markers that
are very close to the causative genes. Patients in founder populations are expected to share
Testicular Carcinoma and HLA Class II Genes64 Genetic Predisposition to Testicular Cancer
5
65
a relatively high number of alleles from recent common ancestors, which is also expected to
apply for mutations predisposing to testicular carcinoma. Therefore, founder populations are
particularly suitable for fi nding genes predisposing to the development of testicular neoplasms
through association-based methods. In a previous study(56) we showed the geographic clustering
of testicular carcinoma in the northern part of The Netherlands, which is indicative of the
importance of founder alleles. The current study, which includes testicular carcinoma patients
and their relatives from this founder population, is the fi rst extensive genotyping of the HLA
region on chromosome 6p21 in a large number of TGCT patients using both standard methods
and a new haplotype sharing method to examine the association between HLA class II genes
and TGCT.
Materials and MethodsPatients
A total of 151 TGCT patients treated at the University Medical Center Groningen (UMCG) in The
Netherlands during the period 1977-1998 were selected for the initial analysis. The majority
of these patients descended from three provinces (Groningen, Friesland and Drenthe) in the
northern part of The Netherlands, based on information collected about birthplace of the
patients’ great-grandparents. Patient characteristics are listed in Table 1. The difference between
the total number of nonseminomas (n=132, 87%) and pure seminomas (n=19, 13%) is due to
different referral patterns for these histologic subtypes. Histological diagnosis was established
for all patients by the Department of Pathology, UMCG. For 108 patients, DNA from both parents
was available for phase determination and for control (nontransmitted haplotype). For the
remaining 43 patients, DNA from children (older than 18 years) and spouses was available for
phase determination. In these cases, the haplotypes of the spouses were regarded as controls.
All participants gave their informed consent and the Ethical Committee of the UMCG approved
the study.
Genotyping
High molecular weight genomic DNA was extracted from peripheral lymphocytes from 20 mL
blood using standard protocols. After DNA extraction, a set of 15 polymorphic microsatellite
markers in the HLA region on 6p21 was genotyped in all patients and controls. Markers were
selected over a distance of approximately 8 cM. The marker order was determined by using
sequence data of the major histocompatibility complex (MHC).(231) Most of the 15 markers used in
the current study are located on this 3.6 megabase MHC sequence, particularly in the HLA class II
region. The position of the remaining markers on the immediate fl anking regions was determined
by using available marker information from data published in print or on the internet (see table
2 and fi gure 1 for marker information).(105;232;233) For each polymerase chain reaction (PCR): 0,25
units Taq DNA polymerase (Roche diagnostics, Mannheim, Germany) were used to amplify the
fragments. The reaction volume was 10 µl. Reaction mixtures contained 200 µM of each dNTP, 1.5
mM MgCl2 , 10 mM tris-HCL, 50 mM KCL and 0.25 µM of each primer (with one primer 5’ labeled
with a fl uorochrome 6-FAM, HEX or NED). Cycling was performed on a PTC-225 thermal cycler
(MJ research, Waltham, MA, USA). Amplifi cation consisted of an initial denaturation of 5 minutes
at 950C, 35 cycles of 30 seconds at 950C, 30 seconds at 550C and 1 minute at 720C. Post PCR
multiplexing was performed by combining 1-10 µl (based on signal strength) of PCR products.
Pooled fragments (2.3 µl)were mixed with 2.5 µl deionised formamide and 0.2 µl ET-400R size
standard (Amersham Pharmacia Biotech, Upsala, Sweden) and separated on a MegaBACE 1000
capillary sequencer (Amersham Pharmacia Biotech) according the manufacturer’s protocol.
Results were analyzed using genetic profi ler v1.1 (Amarsham Pharmacia Biotech).
Table 1: Patient Characteristics
No. of patients 151
Median age at diagnosis (range) 29.4 (15.9-63.0) yrs
Histologic type (%)
- Pure seminoma 19 (13)
- Nonseminomaa 132 (87)
Familial testicular carcinoma (%) 16 (11)
- Affected fi rst-degree relative 7 (5)
- Affected second-degree relative 9 (6)
Bilateral testicular carcinoma (%) 7 (5)
History of undescended Testis (%) 23 (15)a with or without a seminomatous component
Figure 1: Map of the HLA-region
Testicular Carcinoma and HLA Class II Genes66 Genetic Predisposition to Testicular Cancer
5
67
Statistical methods
After genotyping the 15 markers in all participants (each marker meets the Hardy-Weinberg
criteria), the set of haplotypes present in patients and the set of nontransmitted haplotypes
present in parents or spouse control haplotypes was determined. Differences between these
two haplotype sets were analyzed using Haplotype Sharing Statistic (HSS), a new method
for quantitative analysis of haplotype similarity. The validity of this method is demonstrated
elsewhere by applying it extensively to simulated and empirical data.(232;234-236) HSS assumes that
mutations that predispose to TGCT will be present more often in patients than in controls. Some
of the patients will have inherited a possible predisposing mutation to TGCT development from
a common ancestor, especially in a founder population. Due to linkage disequilibrium, a small
haplotype surrounding this mutation will also be identical by descent among these patients.
The amount of identical DNA is dependent upon the number of recombinations that have taken
place on either side of the mutation since it occurred. This number is particularly influenced by
the number of meioses that have taken place and the recombination frequency of the region.
The number of meioses between patients, selected from a population on the condition that a
specific disease mutation is present on their haplotypes, is expected to be smaller than the
number of meioses between a random sample of controls from that population. Therefore, the
length of identical or shared DNA surrounding a predisposing mutation in patients is expected
to be larger than the length of shared DNA surrounding that locus in controls. The difference
in length of haplotype sharing surrounding a specific locus between patients and controls can
be used as an indication for involvement of this locus in susceptibility to the disease. This
difference is expected to be largest at the marker locus closest to the disease mutation. HSS
defines the haplotype sharing between pairs of haplotypes as the number of intervals between
loci that shows identical alleles on a row from a locus in both directions. This can be evaluated
for all pairs of haplotypes for all marker loci.
As a test of linkage disequilibrium, the statistical significance of the overlap of the observed
haplotypes is evaluated. For this test, the observed alleles are redistributed randomly over
their loci and the haplotype sharing in this randomized set is calculated. The mean haplotype
sharing observed in the data is compared to the mean haplotype sharing in the randomized data
in which there is linkage equilibrium among all marker loci. In addition to the new haplotype
sharing test, we also performed single and two locus association tests and a transmission
disequilibrium test (TDT). For the one and two locus association test, the frequencies of the
alleles and two locus haplotypes, respectively, are compared using a chi-square test, taking
only those alleles into account that have an expected frequency of at least one copy. The TDT
test evaluates the transmission distortion of each allele versus all other alleles, using the test
proposed by Spielman et al.(237) For each locus, the result given will be the maximum distortion
at that locus.
Tabl
e 2:
Mar
ker
Dat
a
Locu
s M
arke
r na
me
Posi
tion
(bp)
* Fo
rwar
d Pr
imer
(5´
->3´
) Re
vers
e Pr
imer
(5´
->3´
)
1 D6S
1560
te
lom
eric
CT
CCAGTC
CCCA
CTGC
CCCA
AGGCC
ACA
TAGC
2 DNRNGCA
35
5440
AGGAATC
TAGTG
CTCT
CTCC
CT
CTAGCA
AAAGGAAGAGCC
3 RIN
G3C
A
3753
87
TGCT
TATA
GGGAGACT
ACC
G
GATG
GGAAGTT
TCCA
GAGTG
4 D6S
2445
46
1158
AATA
TGATG
GAAGAAGTA
ATC
CAG
GGATT
ACA
GGTA
TAAGCC
ATT
G
5 TA
P1
4986
34
GCT
TTGATC
TCCC
CCCT
C GGACA
ATA
TTTT
GCT
CCTG
AGG
6 D6S
2444
60
1183
GAGCC
AAGAACC
CAGCA
TTC
GGAAGGATT
CTAAATA
GGGGAG
7 D6S
2443
63
1213
CC
ATA
CCAAAGTA
AAACC
CAG
GAGGATG
AAGGGAAATT
AGAG
8 G51
1525
64
8774
GGTA
AAATT
CCTG
ACT
GGCC
GACA
GCT
CTTC
TTAACC
TGC
9 D6S
1666
70
0043
CT
GAGTT
GGGCA
GCA
TTTG
ACC
CAGCA
TTTT
GGAGTT
G
10
D6S
273
1577
323
GCA
ACT
TTTC
TGTC
AATC
CA
ACC
AAACT
TCAAATT
TTCG
G
11
TNFa
17
2559
2 GCC
TCTA
GATT
TCATC
CAGCC
ACA
CC
TCTC
TCCC
CTGCA
ACA
CACA
12
C3-4
-13
2839
280
GCA
TGACA
CTATA
GTG
GCT
G
CATT
GCA
CTCC
AGTC
TGGGC
13
D6S
265
3165
149
ACG
TTCG
TACC
CATT
AACC
T ATC
GAGGTA
AACA
GCA
GAAA
14
D6S
478
3376
450
CCTC
CATA
ATT
GTG
TGAGCC
CC
AATC
TTCT
AACC
CAAGCA
15
D6S
258
cent
rom
eric
GCA
AATC
AAGAATG
TAATT
CCC
CTTC
CAATC
CATA
AGCA
TGG
* Po
sitio
n (b
ase
pair)
on
sequ
ence
as
publ
ishe
d by
the
San
ger Ce
ntre
(231
)
Mar
kers
use
d in
thi
s st
udy.
The
y co
ver fu
lly the
tw
o lo
ci fou
nd b
y Özd
emir
et a
l.(102
)
Testicular Carcinoma and HLA Class II Genes68 Genetic Predisposition to Testicular Cancer
5
69
ResultsFigure 2 shows the test of linkage disequilibrium. In both patients and controls, there is
significant excess sharing (-log10 (p-value) >3) compared with random redistribution for all loci,
indicating strong linkage disequilibrium in the entire genotyped region. Haplotype sharing in
the presence of linkage disequilibrium indicates that similar haplotypes are likely to be identical
by descent, i.e., inherited from a common ancestor suggesting that HSS has the power to
detect differences between patients and controls. These differences in mean haplotype sharing
between patients and controls are plotted in Figure 3. No significant differences (-log10 (p-value)
<3) between patients and controls in haplotype sharing are observed over the entire region
between marker loci 1 and 15. In addition, single and two-locus association tests (Fig. 4), as well
as the transmission disequilibrium test (TDT) (Fig. 5), showed no significant association
(-log10 (p-value) <3) between TGCT and marker loci in the HLA region on 6p21. Subanalysis of
histological groups (i.e., seminoma, nonseminoma) demonstrated no significant differences
between patients and controls (data not shown). It must be noted, however, that the number
of seminoma patients is rather small due to the unique referral pattern of a cancer center. Our
results particularly pertain to nonseminomas.
Figure 3: HSS: difference in mean haplotype sharing between patients and controls
Figure 2: Linkage Disequilibrium
Deviation from multilocus linkage equilibrium (LE) in patients (solid line) and controls (dotted
line), evaluated by haplotype sharing. Multilocus LE is simulated by permutation of the alleles over
the haplotypes. The results are represented as the –log10 of the significance of the difference in
haplotype sharing between the observed and randomized haplotypes. Significance level is at -log10
(P value) greater than 3.
Figure 3: Difference between haplotype sharing between patient and control haplotypes as calculated
by a t test, expressed as –log10 of the significance. The standard deviation is calculated by repeated
sampling without replacement of 50% of the observed haplotypes.
Significance level is at –log10(P value) greater than 3.
Figure 4: Association of the markers with testicular germ cell tumors. Results of single-locus (solid
line) and two-locus (dotted line) association analysis. The distributions of alleles in patients and in
controls are compared using the chi-square test. The results are presented as –log10 of the significance.
Significance level is at –log10 (P value) greater than 3.6.
Figure 4: Association analysis
Testicular Carcinoma and HLA Class II Genes70 Genetic Predisposition to Testicular Cancer
5
71
DiscussionIt is likely that immune response differences based on HLA variation may play a role in
carcinoma development and metastatic patterns. Several carcinomas have been reported to be
HLA associated (e.g., Hodgkin lymphoma, Kaposi sarcoma, colorectal carcinoma, and Burkitt
lymfoma.(97) HLA variation has also been suggested to play a role in TGCT development. This may
be caused by differences in effects of HLA variation on immune response to carcinogenic factors,
for example viruses that may be etiologically associated with the development of TGCT.(230) An
impaired immune system, genetically or acquired, favors carcinogenic factors. The theory that
TGCT may arise under conditions of reduced immune capacity is supported by the observation
that patients with immune deficiencies following renal transplantation have a two to fivefold
increased risk to develop a TGCT.(99) In addition, there is evidence that the incidence of TGCT in
patients with an acquired immunodeficiency syndrome (AIDS) is higher than in the general male
population.(238-241)
Recently, a high-resolution genotyping study comprising 55 Japanese TGCT patients showed a
HLA-DRB1 susceptibility allele (RR 3.26) and a HLA -DQB1 candidate protective allele (RR 0.26) for
TGCT.(102) This study suggested that one of the genetic factors involved in TGCT development may
be associated with HLA, in particularly the HLA class II region. In addition to Özdemir et al.,(102)
Oliver(110) found an association between DR5 and seminoma and an increase of DR7 in patients
with stage IV disease (extralymphatic metastasis). Aiginger et al.(111) pooled their data with those
from Oliver et al.’s study (total = 233 patients) and found a significant increased frequency of
the HLA antigenes DR1 en DR5 in seminoma patients. The current extensive genotyping study of
a large number of Dutch TGCT patients, however, fails to confirm the associations of HLA Class
II genes with susceptibility to TGCT. A previous study showed geographic clustering of testicular
carcinoma in the northern part of The Netherlands.(62) The majority of patients participating in
this study descend from this area. In addition to possible common environmental factors, this
population is likely to share a relatively high frequency of mutations in genes involved in TGCT
from recent common ancestors. However, for the HLA class II region analyzed in this study, no
difference between patients and controls is observed using both standard methods (association
and TDT analyses) and HSS, even though HSS has the power due to the presence of strong
linkage disequilibrium (Fig. 2). Strong linkgage disequilibrium suggests that only a few different
haplotypes are present in the data. Similar haplotypes are, therefore, likely to be identical by
descent, i.e., inherited from a common ancestor. Because this similarity will be centered around
the disease locus for patients and random over the entire region among controls, differences
due to disease mutation in this region would have been be revealed by HSS. The linkage
disequilibrium observed in this study is in agreement with the results of former studies that
have reported regions within the HLA region that show very little recombination (e.g., between
HLA-DR and DQ) and regions where recombination preferentially occurs.(242)
The study by Özdemir et al.(102) was the first HLA genotyping study of TGCT patients. The HLA
alleles for which associations were reported are also prevalent in the Dutch populations.(243)
However, Özdemir et al. genotyped only 55 patients whereas the current study genotyped almost
three times that number, resulting in more power, particularly for nonseminomas.(102) Moreover
with HSS, we have additional power from phase information to detect differences due to genetic
factors in the HLA region. Systematic analysis of patient and control haplotypes using a high
density genome screen can identify identity by descent within haplotypes of unrelated patients
in a founder population. This is because haplotypes that are similar for many consecutive
marker alleles are very likely to be inherited from common ancestors. The shared segment of
the haplotypes that is most significantly overrepresented in patients compared to controls is
likely to contain a predisposing gene. The previously suggested association between HLA and
TGCT need not result from a functional role of the HLA-system itself, but may also be due to
an effect from a separate gene that is closely linked with HLA loci. The analysis of haplotypes
where most genetic variation is associated with specific haplotypes enables conclusions about
all genes present on the haplotypes. Therefore, based on the results in the current study, a role
of HLA class II genes in the development of TGCT seems much more limited than previously
suggested by a few studies.
In previous studies by Oliver et al.(110) and Aiginger et al.,(111) the association between HLA and
TGCT was found by serotyping methods that are less precise and less efficient than genotyping
Figure 5: Transmission Disequilibrium Test
Results of the transmission disequilibrium test in 108 trios represented as –log10 of the significance
of the maximal transmission distortion at each locus. Significance level is at –log10 (P value) greater
than 3.2.
Testicular Carcinoma and HLA Class II Genes72 Genetic Predisposition to Testicular Cancer
5
73
methods. DNA typing is more sensitive and identifies more alleles. The reason for this is that
the variation present in the HLA region is specific for the small number of founders. All variation
that contributes to disease can be identified with haplotype association methods. DNA based
methods are more accurate and allow higher definition HLA typing. High-resolution genotyping
is the method of choice because the polymorphism of the peptide binding domain of MHC class
II molecules is more precisely determined by genotypes than by serotypes.(101)
In conclusion, although there are proven associations between HLA and several malignancies
this is not the case for TGCT. The current genotyping did not confirm the previous reported
association between HLA class II genes and TGCT, despite a larger sample size, especially
nonseminomas. As the HLA alleles for which associations were reported are also prevalent in
the Dutch populations, these associations are likely nonexistent or much weaker than reported.
Further research focusing on other candidate loci should be performed to identify possible TGCT
susceptibility genes.
Genetic Predisposition to Testicular Cancer
6
75
Chapter 6Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors
MF Lutke Holzik1, K Storm2, RH Sijmons3, M D’Hollander2,
EGJM Arts4, ML Verstraaten1, DTh Sleijfer5, HJ Hoekstra1
Departments of : 1Surgical oncology, 3Genetics, 4Obstetrics and Gynecology, 5Medical oncology. University Medical Center Groningen, the Netherlands
Department of : 2Medical Genetics, University of Antwerp, Belgium
Urology 2005; 65:196-201
Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors76 Genetic Predisposition to Testicular Cancer
6
77
Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors
IntroductionAlthough testicular germ cell tumors (TGCT) constitute the most common malignancy in men
aged 15 to 40 years, their etiology is still poorly understood.(244) In the past few years a decrease
in fertility and an increase in TGCT has been reported.(46) This could suggest that fertility and
TGCT share a common etiological factor. A typical illustration is the East-West semen quality
gradient in the Nordic Baltic area and the incidence of TGCT. Finland and Estonia have only
one third of the TGCT incidence compared with Denmark and Norway, which is inversely related
to the lower sperm counts observed in Danish and Norwegian men compared with men from
Finland and Estonia.(245) A retrospective cohort study of more than 30,000 men from infertile
couples found an association between infertility and a subsequent risk of TGCT.(46) Men with
infertility were 1.6 times more likely to develop TGCT. The greatest risk of TGCT was in the first
2 years (standardized incidence ratio 1,8) after the first semen analysis. At 2 to 11 years after the
first semen analyses, the standardized incidence ratio was 1,5 to 1,6. This is a relatively constant
risk for TGCT, and impaired spermatogenesis may, therefore, have been present many years
before TGCT was diagnosed.(46) These results are in line with our previous results showing that
TGCT patients have already impaired spermatogenesis before orchiectomy was performed.(246)
In the model postulated by Skakkebaek et al.,(12) TGCT and poor semen quality are symptoms
of one underlying entity, the testicular dysgenesis syndrome (TDS). Endogenous, as well as
exogenous, risk factors, including inherited predisposing gene mutations may result in TDS. Past
research has traditionally focused on the range of possible exogenous risk factors for TDS and
TGCT, including prenatal exposure to maternal hormones.
Although TGCT susceptibility genes remain to be identified,(244) more light has recently been shed
on the genetics of infertility. Up to approximately 8% of infertility in the general male population
can be explained by the presence of constitutional deletions of part of the long arm of the Y
chromosome (Yq11), referred to as the azoospermia factor (AZF) region (subdivided into AZFa
to AZFd).(247-249) AZFc deletions are the most commonly found (60%). Although most of the AZF
deletions observed in infertile men are new (de novo) mutations, they have been inherited in
some cases from (apparently fertile) fathers, and 0.4% of fertile men in the general population
appear to carry an AZF deletion.(248)
Foresta et al. recently observed a particularly high percentage of 27.5% AZF (a-c) deletions
in patients with low sperm counts as well as unilateral cryptorchism.(250) Cryptorchism is an
acknowledged risk factor for TGCT and is one of the postulated other possible manifestations
of the TDS.(12;81;251) Taken together, these observations suggests that, at least from a theoretical
point of view, constitutional AZF deletions might be one of the genetic contributors to the
Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors78 Genetic Predisposition to Testicular Cancer
6
79
development of TDS and thereby of TGCT and other TDS manifestations. This possibility would
be in line with tumor cytogenetic studies that have demonstrated a nonconstitutional loss of
Y chromosome material in adult TGCT, as well as in their precursor carcinoma in situ, which
suggests that loss of Y chromosome material may indeed play a role in TGCT development.(252)
Bianchi et al.(253) have demonstrated that, noninherited mosaic AZF deletions can be observed in
tumor as well as nontumor, tissues from some patients with TGCT. Altogether, this might point
at a role for the loss of Y chromosome material, inculding AZF, in TGCT development. Given that
constitutional AZF deletions have been observed in fertile men in the general population and
TDS does not necessarily present with infertility, the possibility of constitutional AZF deletions
causing TDS and thereby TGCT in fertile men has not be ruled out. In the present study, we
investigated the frequency of Y chromosome deletions in the AZF region in a series of fertile, as
well as infertile, patients with TGCT.
Patients and MethodsPatients selection
A total of 112 patients with TGCT treated at the University Medical Center Groningen (UMCG) in
the Netherlands were randomly selected for initial analysis. Patient characteristics are listed in
Table 1. Familial TGCT was defi ned as more than one case in the family. Histologic diagnosis was
established in all patients by the Department of Pathology of the UMCG. Owing to the position
of our medical center (academic referral hospital for the northern part of The Netherlands) most
patients had undergone orchiectomy (before diagnosis) in the referring hospitals without prior
semen analysis and preservation. Therefore, in the current series of patients no data were
available on semen quality before orchiectomy; however, semen data were available for 25
patients after orchiectomy. We stratifi ed these 25 patients into four groups according to semen
concentration (Table 2). All participants gave their written informed consent and the ethical
committee of the UMCG approved the study.
Genotyping High-molecular-weight genomic DNA was extracted from peripheral blood lymphocytes according
to standard protocols.(254) After DNA extraction screening for AZF deletions was performed by
multiplex polymerase chain reaction (PCR) analysis using the Y Chromosome Deletion Detection
System, version 1.1 (Promega), and the addition of Multiplex Master Mix E (Promega). Version
1.1 has been extensively described by Aknin-Seifer et al.(255) and the addition of mix E has
improved accuracy. Currently, the system that includes mix E is known as the Y Chromosome
Deletion Detection System, version 2.0.(256) In the current study the system consisted of 24
primer pairs, of which 20 primer pairs are homologous to previously identifi ed and mapped
sequenced tag sites (STSs) within the AZF regions on the Y chromosome (locations provided in
Fig. 1 and Table 3). All the loci analyzed in this study have been recommended by the European
Quality Monitoring Network Group (EQMNG) for detection of Yq11 deletions associated with
male infertility.(256) Primers were combined into fi ve primer sets to use in fi ve parallel PCR
Table 1: Characteristics of the 112 TGCT patients
Type of Neoplasia Patients/total(%)
Patients with nonseminoma 94/112 (84%)
Patients with pure seminoma 18/112 (16%)
Other characteristics
Patients with bilateral TGCT 4/112 (3,5%)
Patients with cryptorchism 21/112 (18,8%)
Patients with familial TGCT 10/112 (9%)
TGCT = testicular germ cell tumor
Table 2: Distribution of semen concentration in 25 patients after orchiectomy
Sperm Concentration Patients (n)
Normozoospermia (>20 × 106/mL) 13*
Moderate oligozoospermia (5–20 × 106/mL) 5
Severe oligozoospermia (<5 × 106/mL) 2
Azoospermia 5** Normozoospermia and azoospermia groups both included 1 patient with cryptorchism.
Figure 1: Diagram of the Y chromosome with AZF regions, previously cloned genes
and pseudogenes, and STSs. Reprinted form Technical Manual No 248.(256) Used with
permission of Promega Corporationpermission of Promega Corporationpermission of Promega Corporation
Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors80 Genetic Predisposition to Testicular Cancer
6
81
amplifications (multiplex PCR A through E; Table 3). The slight modifications to the protocol(256)
provided by the manufacturer were: 500 ng DNA in a final volume of 25 µl multiplex Master Mix,
amplification in 35 cycles on a Perkin-Elmer GeneAMP® System 9700 thermal cycler (Applied
Biosystems), and annealing at 58°C for 1 minute 30 seconds. The control samples analyzed in
each multiplex PCR were a male genomic DNA control, a female genomic DNA control and a
blank (no-DNA) control. The separation and visualization of the PCR products were performed
by electrophoresis in 4% NuSieve 3:1 Plus agarose gels (Cambrex Bio Science), stained with
Ethidium Bromide (Fig. 2). The multiplex primer sets A through D contained a control primer pair
that amplifies a fragment of the X-linked SMCX locus.
Multiplex E contains a control primer pair that amplifies a unique region in both male and female
DNA (ZFX/ZFY). Both control primer pairs are internal controls for the amplification reaction and
the integrity of the genomic DNA sample. In addition multiplex E contains a primer pair that
amplifies a region of the SRY gene that is a control for the presence of the testis determining
factor on the short arm of the Y chromosome (Yp) and allows XX males (arising from Y to X
translocations) to be detected. The Y chromosome deletion detection system(256) is the standard
procedure in the laboratory of the Department of Medical Genetics, University of Antwerp, to
detect AZF deletions in men analyzed for infertility and subfertility. We previously found some
AZF deletions in our laboratory amongst infertile and subfertile men, who did not have a history
of TGCT (data not shown).
ResultsMicrodeletions analysis of the AZF region (Yq11) was successfully performed on genomic DNA of
112 patients with TGCT. Figure 2 includes representative examples of the electrophoresis gels,
showing amplification products for multiplex A-E. No PCR products detected in the blank (no
DNA) control. As expected the positive male control showed the appropriate number and sizes
of bands for each multiplex master mix. The positive female control only showed amplification
for the SMCX and ZFX loci. In the 112 patients with TGCT no deletions within the AZF region
were detected.
DiscussionA detailed analysis of microdeletions on the Y chromosome was performed in 112 Dutch patients
with TGCT by studying 24 STSs within the AZF regions on Yq11. These patients included bilateral
cases (n=4), cases with cryptorchism (n=21), cases with a positive family history (n=10) for TGCT
(Table 1) and patients with proven normal or low sperm counts (Table 2). No AZF deletions were
observed in any of the patients. The distribution of nonseminomatous TGCTs and seminomatous
TGCTs as shown in Table 1 is related to the referral pattern of TGCT’s to the UMCG. Our data
confirmed and extended the findings of another study recently published while our study was
in progress. Frydelund-Larsen et al.(257) screened 160 Danish TGCT patients for microdeletions
on chromosome Yq11. In 103 patients seven STSs spanning the three AZF regions (AFZa,
AZFb and AZFc) (plus SRY and ZFX/ZFY) were analyzed. In 57 patients, nine additional STSs
spanning AZFabc, (and TSPY on Yp) were studied. Four of the 16 STSs spanning AZFabc (sY84
in AZFa, sY134 in AZFb, and sY152 and sY254 in AZFc) were in common with those studied in
our patient group. No AZF deletions were observed in their study population. Because, in the
study by Foresta et al.,(250) AZF deletions were only found in a group of patients with a history
Table 3: Overview of the 24 STSs amplified
STS Locus PCR Fragment (bp) Multiplex PCR Position
sY81 DYS271 209 A distal to AZFa
sY86 DYS148 232 E AZFa
sY84 DYS273 177 E AZFa
sY182 KAL-Y 125 A proximal to AZFa
sY121 DYS212 190 C AZFb
SYPR3 SMCY 350 B AZFb
sY124 DYS215 109 D AZFb
sY127 DYS218 274 B AZFb
sY128 DYS219 228 C AZFb
sY130 DYS221 173 A AZFb
sY133 DYS223 177 D AZFb
sY134 DYS224 303 E AZFb
sY145 DYF51S1 143 C proximal to AZFc = AZFd
sY152 DYS236 285 D proximal to AZFc = AZFd
sY153 DYS237 139 D AZFd (nonpathogenic)
sY242 DAZ 233 B AZFc
sY239 DAZ 200 B AZFc
sY208 DAZ 140 B AZFc
sY254 DAZ 370 A AZFc
sY255 DAZ 126 C AZFc
sY157 DYS240 285 A distal to AZFc
sY14 SRY 400 E SRY gene
SMCX 83 A-D X chromosome (control)
ZFX/ZFY 496 E X/Y chromosome (control)
STSs = sequenced tag sites; PCR = polymerase chain reaction; AZF = azoospermia factor.
Overview of the 24 STSs amplified in five multiplex PCR amplifications
(A-E), including 20 STSs for detection of AZF deletions associated with male infertility,
one STS sY153 which seems to be polymorphic or in multiple copies and 3 control STSs
(SMCX, ZFX/ZFY and sY14).
Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors82 Genetic Predisposition to Testicular Cancer
6
83
of cryptorchism together with azoospermia or severe oligozoospermia, it is possible that AZF
deletions are only present in that minority of patients with TGCT who also have markedly
reduced fertility. In the current study, fertility status (Table 2) was known in 25 patients (22%)
and 5 of these had azoospermia after orchiectomy. Data on semen concentration before
orchiectomy were not available because the vast majority of patients were referred for treatment
after orchiectomy performed elsewhere. Furthermore, semen analyses after orchiectomy was
only offered to patients treated with adjuvant chemotherapy or radiotherapy with the intention
to father children in the near future. Frydelund-Larsen et al. (257) presented data on fertility for
70 of 160 of their patients before TGCT treatment. A total of 37 of these patients (23% of the
total group) had severe (n = 17; less than 5x106/ml) or very severe (n = 8; less than 0.2x106/ml)
oligozoospermia or azoospermia (n = 12). Although their study as well as the current study
did not detect any AZF deletions in patients with TGCT with reduced fertility, our study did not
have enough statistical power to exclude low percentages of AZF deletions in small subsets of
patients
In conclusion, the data suggest that a substantial contribution of constitutional large AZF
deletions to the development of TGCT, whether or not in the presence of reduced fertility,
cryptorchism, previous history of TGCT or positive family history, is unlikely. The present data
do not rule out the possibility of constitutional smaller deletions or other type of mutations
in genes mapped to the AZF region. These genes could therefore be the subject of additional
research. Given the complexity of urogenital differentiation and testicular tumor development,
only the ‘tip of the iceberg’ has been mapped with respect to genes involved in these processes
to date. As new tools for molecular study become available and mapping efforts advance, more
opportunities will undoubtedly arise to explore the molecular basis of the testicular dysgenesis
model and testicular tumor development and these explorations should include interactions with
environmental risk factors.
Note added in proof: A very recent study performed by Nathanson et al.(258) has revealed that
a small inherited or de novo deletion within the AZFc region, referred to as the gr/gr deletion
appears to be associated with an increased risk to develop TGCT. This gr/gr deletion on the Y
chromosome is not detected by the commonly used test for the larger AZF deletions, and was
recently found to be a risk factor for spermatogenic failure by Repping et al.(259)
Figure 2: Multiplexes A-E.
Figure 2A-2E. Electrophoresis gel (4% NuSieve 3:1 Plus agarose) showing amplifi cation products
for (A) multiplex PCR A, (B) multiplex PCR B, (C) multiplex PCR C, and (D) multiplex PCR D,
representing STSs within AZF regions and control STS (SMCX) and (E) multiplex PCR E showing
amplifi cation products for multiplex PCR E, representing STSs within AZF regions and control STSs
(ZFX/ZFY and sY14). Lanes 1 to 4 = patients with TGCT; L = 50 bp DNA Step Ladder; B = blank
(no-DNA control), M = normal male control; F = normal female control.
A B
C D
E
Genetic Predisposition to Testicular Cancer
7
85
Chapter 7Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular Germ Cell Tumour without cryptorchidism.
MF Lutke Holzik1, HJ Hoekstra1, RH Sijmons2, DJA Sonneveld1, G van der Steege3,
DTh Sleijfer4, IM Nolte3
Departments of: 1Surgical Oncology, 2Genetics, 3Medical Biology, 4Medical Oncology,
University Medical Center Groningen, the Netherlands
European Journal of Cancer 2006; 42:1869-74
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular
Germ Cell Tumour without cryptorchidism
86 Genetic Predisposition to Testicular Cancer
7
87
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular Germ Cell Tumour without cryptorchidism.
IntroductionThe incidence of testicular germ cell tumour (TGCT) is still rising.(260) Although the aetiology of
TGCT is poorly understood, it is well known that male relatives (fathers/brothers/sons) of TGCT
patients have an increased risk of developing TGCT. Currently 1-3% of TGCT patients report an
affected relative.(261) Brothers of TGCT patients have an 8-10 fold increased risk of developing
TGCT and the relative risk (RR) to fathers and sons is approximately 4-6.(104;244) Because these RRs
associated with an affected first-degree relative are considerably higher than for other cancers,
which rarely exceed four, this observation most likely points to a genetic role in the aetiology of
TGCT.(244) Efforts have been made to identify TGCT predisposing genes. Although TGCT families
have been reported in the literature, multigenerational pedigrees with several affected cases are
rare and this limits the opportunities for linkage studies. In 1994, the International Testicular
Cancer Linkage Consortium (ITCLC) was formed with the aim to collect TGCT families from all
over the world and to perform genotyping studies. Recently, Rapley and colleagues (on behalf
of the ITCLC) presented evidence for a TGCT susceptibility gene on chromosome Xq27.(105) Their
genome-wide search for linkage in a set of 134 familial TGCT cases yielded a heterogeneity
LOD (hlod) score of 2.01 on chromosome Xq27 using all families (n=99) compatible with X-
linked inheritance and a hlod score of 4.7 on chromosome Xq27, if they included only families
with at least one bilateral TGCT case (n=15) (genome wide significance level p = 0.034). In
addition, 73% (n=14) of the familial TGCT cases with a history of cryptorchism (synonymous
with cryptorchidism), a well known risk factor for TGCT, were linked to locus Xq27. These results
provided evidence for a gene on chromosome Xq27 involved in TGCT susceptibility as well as
in cryptorchism. Two recombination’s, one between markers DXS8043 and DXS8028 on the
centromeric side and one between FRAXA.pcr2 and FMR1Di on the telomeric side, bounded
the identified TGCT1 locus, resulting in an interval of ~4 cM (~2.7 Mb). In this region, three
genes have been reported so far: FMR1, responsible for Fragile X syndrome, a single exon gene
Cxorf1 (expressed in the brain) and a tandemly duplicated gene LOC158813/158812.(262;263) As yet,
germline mutations in any gene in this region have yet to be identified as the cause of increased
risk to develop TGCT.
An alternative approach to linkage studies is searching for TGCT susceptibility genes among
unrelated TGCT cases in founder populations by means of association analyses on a dense
set of markers. These so-called linkage disequilibrium fine-mapping analyses are based on the
hypothesis that patients in founder populations inherited disease mutations from recent and
common ancestors. A previous study showed geographic clustering of TGCT in the northern part
of the Netherlands.(62;95) Another study presented the results of analyses of HLA microsatellite
markers and TGCT in this founder population.(96) The current study includes the previously used
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular
Germ Cell Tumour without cryptorchidism
88 Genetic Predisposition to Testicular Cancer
7
89
study population, expanded with additional TGCT patients from the same founder population.
The aim of this study was to corroborate or refute the previously observed linkage between TGCT
and chromosome Xq27 using association analysis and the Haplotype Sharing Statistic (HSS).
Patients and MethodsPatients and controls
A total of 276 patients were randomly selected from all TGCT patients treated during the period
1977-2001 at the University Medical Center Groningen (UMCG), the Netherlands. The majority
of these patients descended from three provinces (Groningen, Friesland, and Drente) in the
Northern Netherlands, based on information collected about birthplace of the patient’s great-
grandparents. Histological diagnosis was established for all patients by the department of
Pathology of the UMCG.
Through the patients, family members (parents, children and spouse, brothers or sisters) were
asked to participate. For this study on association of the X-chromosome with TGCT, only unaffected
male first-degree family members were used and served as controls (n=169). Mothers of TGCT
patients, if available, were used to check for correct inheritance. Population characteristics are
presented in Table 1. TGCT cases were defined as familial cases when more than one TGCT case
was present in the family. TGCT cases with male-to-male transmission (e.g. father – son) of the
putative TGCT predisposition gene were excluded as these cases were obviously not compatible
with X-linked inheritance and would have obscured the test results. The difference between the
total number of non-seminomas (89%) and pure seminomas (11%) included is in particular due
to different referral patterns for these histological subtypes. Traditionally all patients diagnosed
with a non-seminoma within a defined area of the Comprehensive Cancer Centre in the northern
part of the Netherlands (CCNN) are referred to the UMCG for further management after having
been hemi-orchidectomised at the local hospital. In contrast, the majority of patients diagnosed
with a seminoma are referred to one of the three radiation facilities within the CCNN area
(including UMCG) for radiation treatment. All participants gave their informed consent and the
Medical Ethical Committee of the UMCG approved the study.
Genotyping
High molecular weight genomic DNA was extracted from peripheral lymphocytes from 20
ml EDTA blood using standard protocols.(254) After DNA extraction, a set of 16 polymorphic
microsatellite markers in the Xq27-Xq28 region was genotyped in all patients and their relatives
(see Table 2 for marker details, only markers that meet the quality criteria are shown, see section
Results). Microsatellite markers were selected over a distance of approximately 4.3 Mb in order
to cover the TGCT1 locus. Most markers were obtained from literature and the public databases.
To get an evenly distributed set, the markers starting with XTC0 were newly developed by
searching the downloaded sequences for putative dinucleotide repeats and amplify these loci
with primers selected with the online Primer3 software (http://frodo.wi.mit.edu/cgi-bin/primer3/
primer3_www.cgi). Relative marker positions were obtained from the contigs nt_011681.13 and
nt_019686, and NCBI Map Viewer build 34 version 3 was used to determine the distance between
both contigs. For each polymerase chain reaction (PCR), 0.25 units Taq DNA polymerase (Roche
diagnostics, Mannheim, Germany) was used to amplify the fragments. The reaction volume was
10 µl. Reaction mixtures contained 200 µM of each dNTP, 1.5 mM MgCl2 , 10 mM tris-HCL, 50 mM
KCL and 0.25 µM of each primer (with one primer 5’ labeled with a fluorochrome 6-FAM, HEX
or NED). Cycling was performed on a PTC-225 thermal cycler (MJ research, Waltham, MA, USA).
Amplification consisted of an initial denaturation of 5 min at 95ºC, 35 cycles of 30 sec at 95ºC,
30 sec at 55ºC and 1 min at 72ºC. Post PCR multiplexing was performed by combining 1-10 µl
(based on signal strength) of PCR products. 2.3 µl of the pooled fragments was mixed with 2.5 µl
deionised formamide and 0.2 µl ET-400R size standard (Amersham Pharmacia Biotech, Uppsala,
Sweden) and separated on a MegaBACE 1000 capillary sequencer (Amersham Pharmacia Biotech,
Uppsala, Sweden) according to the manufacturer’s protocol. Results were analyzed using Genetic
Profiler v1.1 (Amersham Pharmacia Biotech, Uppsala, Sweden).
Statistical methods
Only markers that fulfilled the following quality criteria were included in the statistical analysis:
failed genotyping (no results, alleles unknown) in less than 15% of the study population and
erroneously scoring of two alleles in less than 4% of the male sub-population (men of course
carry only one allele for each marker on their single X chromosome). In those few events where
genotyping of an included marker erroneously revealed two alleles in a male, the genotype for
that marker was set to “unknown”. All markers fulfilling the quality criteria were analyzed for
allelic association by a chi-square test using only those alleles with an expected count of at
least three. The data were also analyzed by the Haplotype Sharing Statistic (HSS).(264-267) HSS is a
linkage disequilibrium fine-mapping method that is based on the assumption that patients share
disease mutations inherited from recent common ancestors. It measures the sharing between
Table 1: Population characteristics
Number (%)
Patients 276
Non-seminoma 248 (89%)
Seminoma 28 (11%)
Bilateral TGCT 11 (4%)
Cryptorchism 43 (15%)
Familial TGCT 16 (6%)
Sporadic TGCT without cryptorchism 220 (79.7%)
Controlsa 169 a first degree, male family members
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular
Germ Cell Tumour without cryptorchidism
90 Genetic Predisposition to Testicular Cancer
7
91
a pair of haplotypes at a marker locus as the number of consecutive marker loci carrying the
same alleles starting from the locus under analysis in both telomeric and centromeric direction.
HSS then hypothesises that, in genomic regions containing disease mutations, haplotypes of
patients show more sharing than haplotypes of controls. The association and HSS methods were
also applied to four subgroups of patients, namely to patients with bilateral TGCT, to patients
with familial TGCT, to patients with cryptorchism and to patients with sporadic TGCT without
cryptorchism (Table 1). It should be noted that the number of patients with bilateral TGCT and
the number of patients with familial TGCT were small.
Because multiple markers are analyzed, a multiple testing correction is required. This implicates
that a result is only regarded as significant when the p-value is smaller than 0.05 / 12 = 0.004,
where 12 is the number of markers that passed our quality criteria. A 99.57% confidence interval
corresponds to a 95% CI interval corrected for multiple testing. Power of the study was assessed
by standard statistical theory on normally distributed variables, assuming that the allele counts
follow a binomial distribution, which can be approximated by a normal distribution.
ResultsOf the 16 microsatellite markers that were genotyped, four did not meet our quality criteria (data
not shown). Hence, the analyses were performed on the 12 markers shown in Table2. Allelic
association analysis did not reveal any significant difference, i.e. p<0.004, between general TGCT
patients and controls at any marker (Figure 1). In addition, the HSS did not show a significant
result either (Figure 2). As the sample of Rapley and colleagues(105) consisted of familial TGCT
cases and the linkage evidence became stronger when selecting only cases with bilateral TGCT
or cases with cryptorchism, we also performed analyses on the subsets of patients. The results
are shown in Figures 1 and 2. No significant associations were observed with any of these
subgroups or any marker. We did however observe that for marker DXS1193 the major allele
was less frequent among cases (88.3%) than among controls (96.6%) and that difference was
smaller for cases with cryptorchism (94.7%) and those with familial TGCT (100%). Therefore,
we also analysed the subgroup of sporadic cases without cryptorchism. In this subgroup, both
allelic association and the HSS revealed evidence for a TGCT susceptibility locus (p=0.014 and
p=0.008, respectively), which remain not significant after multiple testing correction (Table
3). However, after the multiple testing correction, a suggestive relative risk was observed for
individuals carrying one of the minor allele at DXS1193 of 3.8 to develop TGCT and a significant
relative risk of 4.7 (99.57% CI: 1.1-19.6) to develop sporadic TGCT without cryptorchism,
compared with carriers of the major allele.
Tabl
e 2:
Cha
ract
eris
tics
of m
arke
rs u
sed
for
this
Xq2
7 as
soci
atio
n sc
reen
(on
ly m
arke
rs tha
t m
eet qu
ality
crite
ria
are
show
n)
Mar
ker
Rela
tive
Prim
ers
Fa
iled
Het
eroz
ygou
s
po
sitio
n (M
b)
Forw
ard
Reve
rsed
PC
R (%
) m
en (%
)
DXS
8043
0.
0000
0a AGTT
CTCA
GAAACA
TTTG
GTT
AGGC
AATT
ATT
GGCA
AAGAGTA
CAGGCA
G
6.5%
3.
50%
DXS
8028
0.
2147
2 a
TGATG
ACA
CTCG
GACT
GC
GAAATA
ATA
ATA
CTTG
CCTT
GCC
T 13
.7%
3.
05%
AFM
a113
zf5
0.50
108 a
AACA
CTGCA
CGATG
AGAA
AGCT
ATC
CTGATT
TTGGTA
CT
4.6%
2.
31%
DXS
8045
1.
4836
0 a
CAGGTA
AATC
TGAGAAATG
TTCT
GC
ACT
GCG
GTG
CTGACT
AGG
7.5%
2.
15%
DXS
1200
1.
7169
3 a
TACA
CACC
AAACA
ACA
GAGCC
T CT
AGGGGCA
CTTG
AAAACA
A
11.5
%
2.30
%
XTC0
08
2.03
909 a
TT
CTGTC
TCACA
AGCC
AGATA
A
CTGATC
CTCT
GACA
GCA
TATA
C 4.
0%
3.21
%
XTC0
221
2.57
815 a
TG
TATC
TGTG
CATG
TACC
TATC
AAGAAGTC
ATC
CACT
GAGTC
TA
2.2%
0.
45%
DXS
998
2.57
935 a
CA
GCA
ATT
TTTC
AAAGGC
AGATC
ATT
CATA
TAACC
TCAAAAGA
0.5%
2.
20%
Frax
ac1
2.95
775 a
GATC
TAATC
AACA
TCTA
TAGACT
TTATT
GATG
AGAGTC
ACT
TGAAGCT
GG
5.5%
0.
70%
DXS
1215
4.
0697
3 b
GGGCA
AAACA
TTAAACC
TCTC
GCC
CTCT
AAGTC
ATT
ACG
CT
4.4%
3.
87%
DXS
1193
2.
9580
6 b
AATT
CTGACT
CTGGGGC
TTATT
TTAAGGTG
AGTA
TGGTG
TGT
12.4
%
1.27
%
DXS
1113
4.
2867
9 b
GGGAGCT
TTAGAGATT
TTGGTA
AAC
ACC
TGTG
GAGGATA
GTA
GTC
TGACT
4.
6%
3.20
%
a lo
cate
d on
con
tig n
t_01
1681
.13
b lo
catio
n on
con
tig n
t_01
9686
, di
stan
ce b
etw
een
cont
igs
dete
rmin
ed u
sing
NCB
I M
ap V
iew
er b
uild
34
vers
ion
3
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular
Germ Cell Tumour without cryptorchidism
92 Genetic Predisposition to Testicular Cancer
7
93
patients is higher than the increase in RR to fathers or sons of TGCT patients, which might
be explained by assuming an X-linked inheritance of a TGCT predisposing trait. In addition,
patients with Klinefelter syndrome (47, XXY constitutional karyotype) have a RR of 67 to develop
mediastinal germ cell tumours. Like TGCT, these tumours are thought to arise from carcinoma in
situ. The presence of an additional copy of the X-chromosome in Klinefelter syndrome suggests
a possible dose-effect of one or more genes on this chromosome, escaping X-inactivation,
on germ cell tumour development. Indeed, cytogenetic studies have revealed that generally
the X-chromosome is overrepresented in TGCT tumour DNA.(76;244;268) Ross and colleagues (269)
determined the sequence of over 99% of the gene-containing region of the X chromosome.
They predicted that nearly 10% of the 1098 genes on the X chromosome are in a class that is
upregulated in testicular and other cancers. Taken together, these observations suggest that
the X chromosome may well harbour TGCT predisposing genes. As yet, these genes and their
mutations remain to be identified.
Table 3: Statistical results for association analyses on marker DXS1193
patients sporadic,
Marker allele controls all Cryptorchism familial no cryptorchism
DXS1193 107 96.6% 88.3% 94.7% 100% 86.1%
105 0.0% 1.3% 0.0% 0.0% 1.7%
109 0.0% 3.9% 2.6% 0.0% 4.4%
110 2.7% 3.5% 2.6% 0.0% 3.9%
112 0.7% 2.6% 0.0% 0.0% 3.3%
114 0.0% 0.4% 0.0% 0.0% 0.6%
p-value 0.059 0.92 0.82 0.014
ORa 3.8 1.6 0.9 4.7
(99.57% CI) (0.9-15.9) (0.1-18.5) (0.0-61.8) (1.1-19.6)a OR (Odds Ratio) for combination of minor alleles Figure 1: Association analyses
Association analysis for alleles. The black squares represent the results at the markers. Lines between
the markers are drawn only for an easier interpretation of the results. The lines distinguish the
different (subgroup) analyses: a thick solid line for all patients, a thick dotted line for cases with
cryptorchism, a thin solid line for the familial case and a thin dotted line for the cases without a
family history of TGCT or cryptorchism. A p-value of < 0.004 is considered significant after multiple
testing correction.
DiscussionIn the current study, we did not find an association between Xq27 and familial TGCT,
cryptorchism or bilateral TGCT. We could therefore not confirm the results found by Rapley
and colleagues.(105) We should however notice that our subgroups of familial or bilateral cases
or cases with cryptorchism were small, which results in sufficient power only when a disease
mutation with a frequency of 5% (or higher) and the putative gene has a large effect (RR>8 for
cryptorchism; RR>20 for bilateral or familial cases). Hence, it cannot be excluded that an X-linked
gene with a smaller effect is involved in familial or bilateral TGCT or cryptorchism.
Interestingly, we did observe an association between the subset of TGCT cases without a family
history of TGCT or cryptorchism and specific alleles for the marker DXS1193 both by allelic
association analysis and by the HSS. The frequency of all minor alleles was increased among
these patients compared to controls: 13.9% versus 3.4%, respectively. The risk to develop
sporadic TGCT without a cryptorchism for an individual who has one of the minor alleles was
estimated to be 4.7 (99.57% CI: 1.1-19.6). This suggests that in our population one or more
low frequent mutations of an Xq27-linked gene contribute to TGCT development but not to
cryptorchism. Alternatively, particular genotypes in this region possibly protect the normal
population from developing TGCT. Further analyses on single nucleotide polymorphism (SNPs)
in candidate genes in this region should be performed to identify the causal gene and to unravel
the nature of its causality.
Several observations had been made that could be interpreted as suggestive of the existence
of a TGCT predisposing gene on the X chromosome. The increase in RR to brothers of TGCT
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular
Germ Cell Tumour without cryptorchidism
94 Genetic Predisposition to Testicular Cancer
7
95
In a study by Rapley and colleagues a significant linkage was reported in 99 X-compatible
pedigrees, in particular in families with cases with bilateral TGCT.(105) Recently Crockford and
colleagues examined an additional 66 pedigrees with two or more cases of TGCT at Xq27. In
contrast to the previous findings, they found no evidence for linkage at this region in this new
set of pedigrees.(262) Moreover three candidate genes from the identified minimal region at
Xq27, FMR1, Cxorf1 and LOC58813/158812 were screened for small deletions, duplications and
missense/non-sense mutations and no pathogenic mutations were observed. Crockford and
colleagues also examined five genomic regions at four other chromosomes and for these regions
no significant results were observed either.(262) Their overall conclusion was that no single
major locus can account for the majority of the familial TGCT cases. They suggest that multiple
susceptibility loci with weak effects contribute to TGCT. Their conclusion is in line with our
results. The region surrounding marker DX1193 could contain one of these susceptibility loci.
Figure 2: HSS
The Haplotype Sharing Statistic. The black squares depict the results at the markers. The different
(subgroup) analyses are represented by different line styles: a thick solid line for all patients, a thin
solid line for the cases with cryptorchism and a thin dashed line for the cases without a family history
of TGCT or cryptorchism. The subgroup of bilateral cases was too small to perform a reliable haplotype
sharing analysis. A p-value of < 0.004 is considered significant after multiple testing correction.
In conclusion, we could not confirm the previously reported association of familial, bilateral
and cryptorchism-associated TGCT with Xq27, but we cannot exclude the presence of an X-
linked gene that slightly or moderately increases risk to develop these particular phenotypes.
Interestingly our data revealed an association between the subset of TGCT cases without a
family history of TGCT or cryptorchism and marker DXS1193. Our findings suggest that in our
population one but possibly more low frequent mutations of an Xq27-linked gene contribute to
TGCT development but not to cryptorchism. It will be interesting to see whether these results can
be confirmed in other populations. Until candidate genes from this region have been identified
and can be checked for mutations, variations and their functional roles, the question of causal
relation or statistical artefact remains unanswered.
Genetic Predisposition to Testicular Cancer
8
97
Chapter 8Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility
MF Lutke Holzik1, HJ Hoekstra1, RH Sijmons2, DTh Sleijfer3,
JEHM Hoekstra-Weebers4,5
Departments of: 1Surgical Oncology, 2Genetics, 3Medical Oncology, 4Wenckebach Institute, University Medical Center Groningen, the Netherlands.5Comprehensive Cancer Center North-Netherlands, Groningen, the Netherlands
Submitted
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility98 Genetic Predisposition to Testicular Cancer
8
99
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility
IntroductionTesticular Germ Cell Tumor (TGCT) is a rare disease. However, it is the most common malignancy
in males between 15 and 40 years of age. The incidence in the Netherlands is approximately
7.5 per 100.000 males and rises slightly each year.(1) Despite many investigations, the etiology
of TGCT remains elusive. In the past couple of years, however, several risk factors have been
identified, with a positive family history being one of the strongest predictors. Brothers of TGCT
patients have a 8-10 fold increased relative risk (RR) to develop TGCT, while the father-son RR is
4-6 times increased.(244) These findings suggest a genetic component in the etiology of TGCT and
it has been postulated that roughly one third of all TGCT cases occur in genetically predisposed
individuals.(89) However, the search for inherited gene mutations predisposing to the TGCT
development has not yet identified high penetrant mutations (i.e. mutations associated with a
high risk of developing TGCT). If such gene mutations would be identified then genetic testing
for hereditary TGCT would be facilitated. This test would either be diagnostic for affected men
suspected of having hereditary TGCT or pre-symptomatic in unaffected men at risk for carrying
the familial mutation. Similar DNA tests are already clinically used for a number of hereditary
tumor syndromes, including hereditary breast-ovarian cancer, several types of hereditary
colorectal cancer and multiple endocrine neoplasia (MEN) syndromes.(244;270)
A review study showed that the percentage of family members who decided to undergo genetic
testing for inherited cancer susceptibility varied between 27% and 80%. This percentage seemed
to depend on several factors including cultural differences (e.g. one study on hereditary breast
cancer showed that British female mutation carriers more often choose prophylactic surgery and
chemoprevention than those from France and Canada), differences in clinical facilities available
(e.g. laboratory facilities) and differences in study methodology (e.g. recruitment source and
response rates).(271;272) It has been reported that participants approached in the clinic are more
often willing to be tested than those recruited through databases and that motivations to
participate differ between these groups.(273)
The willingness to undergo genetic testing seems to depend on whether prevention of disease
is possible or wether the disease resulting from a possible gene mutation can be treated. If
this is not the case, interest in genetic testing is substantially lower, as can be shown by the
lack of interest in genetic testing for Huntington’s Disease and Cystic Fibrosis, compared to the
high patient interest testing for breast cancer and familial colon cancer.(271;274;275) The willingness
seems also to be dependent on the number of family members affected.(276) Factors preventing
people from undergoing DNA testing include an expected unfavorable test result and costs.(277;278)
Additionally, people with a positive family history for cancer frequently overestimate their own
risk of getting cancer themselves, which seem to be a reason to undergo genetic testing.(278)
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility100 Genetic Predisposition to Testicular Cancer
8
101
Information about genetics, predisposition and possible pros and cons should be discussed
before proceeding with DNA testing. Actual knowledge about the different factors that affect the
choices people make for DNA testing is therefore of great importance.(271;279;280)
The majority of studies published focus on the interest of people in mutation analysis for
syndromes for which some of the genes involved have been identified (BRCA1 and BRCA2 for
breast/ovarian cancer, several mismatch repair genes for HNPCC, Lynch syndrome and CDKN2A
for cutaneous melanoma).(278;281-283) The respondents in these studies were mainly women or
older than TGCT patients, with the exception of those in studies that focused on melanoma.
The latter disease mainly affects young people as does TGCT. A difference with familial TGCT is
that families with the most common mutation for hereditary melanoma (in CDKN2A) also face
an increased risk for pancreatic cancer, a tumor with a poor prognosis and no proven early
detection benefit. Men with familial prostate cancer have also been studied. As in TGCT, no high
penetrant germline gene mutations have been identified. However, prostate cancer is common
in elderly men while TGCT is more frequent in young men.(284) A recent publication, the first on
the interest of men in a possible DNA test for hereditary TGCT, showed that the majority of the
participants (66%) expressed interest in this test.(285)
Many studies have examined the interest in genetic testing of patients with suspected or proven
hereditary cancer and their relatives. Less attention has been paid to the interest of men not
confronted with cancer and patients who developed TGCT but whose cancer seems unrelated
to a family history. Comparing these groups may reveal insight into possible differences in
interest in genetic testing. A distinct group of patients are those who become member of a
patient association. The main goal of patient associations is to provide and receive information
about the disease, give members the opportunity to come in contact with fellow patients, and
share experiences and information about TGCT. Members of a patient association may differ in
opinions/attitudes toward genetic testing from patients who do not decide to become members.
TGCT patient association members were reported to be younger, more recently treated, more
highly educated, to have more often experienced a recurrence and consequently received more
treatment, and they reported a worse quality of life compared to non-members.(9)
Anticipating the availability of a possible DNA test for TGCT, we conducted a study into the
interest in and motives for undergoing such a test. We postulated beforehand that more TGCT
patients, and in particular familial TGCT patients, would decide to undergo genetic testing and
that they would like to receive more information about the possibility of having a gene mutation
for cancer than people who had not suffered from cancer. Our thought was that patients with
familial TGCT would suspect they carry a gene mutation more than patients with a negative
family history of TGCT and that men who had not been confronted with cancer would least
suspect they carry a gene mutation. We explored motivations which may play a role in the
decision to undergo genetic testing and differences in motivations between TGCT patients and
men who had not suffered from cancer. Lastly, we expected a positive association between such
motivations for genetic testing and the interest of men to undergo genetic testing.
MethodsRespondents
Four groups of respondents were recruited for this study. Firstly, all familial TGCT patients
(N=44), i.e., patients with two or more confirmed cases of TGCT in the family, were selected from
the database including all TGCT patients/survivors (n=702) treated at the University Medical
Center Groningen between 1977 and 2003. This group is called “familial TGCT”. Secondly, a
similar number of non-familial TGCT patients were selected at random (“sporadic TGCT”) from
the same database. Thirdly, the Dutch TGCT patient association approached their 116 members.
This group is called “patient association”. Finally, the last group consisted of male patients
visiting the Emergency Department of Medisch Spectrum Twente, the Netherlands, following a
minor trauma (e.g. ankle distortion or a small wound). This group is called “controls”. Patients
were approached until the group’s size was comparable to the familial TGCT patient group.
Procedure
A letter with the objectives of the study and a brief explanation about TGCT and its hereditary
aspects, the questionnaire and a prepaid return envelope were sent to the men in groups 1
(familial TGCT) and 2 (sporadic TGCT) by the researchers. The Dutch patient association for TGCT
patients and survivors approached their members by sending the same letter with information,
the questionnaire and return envelope. The control group received the information and the
questionnaire in the Emergency waiting room. All participants gave written informed consent
and the study was approved by the Medical Ethics Committee of both hospitals, the University
Medical Center Groningen (UMCG) and Medisch Spectrum Twente.
Instruments
Data on the following socio-demographic variables were collected: age, educational level, marital
status, employment status, and number of children. Highest educational level completed was
measured on a seven-point scale: (1) primary school, (2) lower vocational, (3) lower secondary,
(4) middle secondary, (5) high secondary, (6) higher vocational degrees and (7) university.
Response categories for current marital status were: (1) married/cohabiting/LAT, (2) single,
(3) divorced, and (4) widower, and for current daily occupation: (1) paid job, (2) student, (3)
unemployed, (4) unable to work, and (5) retired. Also the prevalence/occurrence of other types
of cancer in the family was collected.
One question assessed respondents’ intention to undergo genetic testing for TGCT if this were
possible: The following answers could be given: (1) definitely not, (2) probably not, (3) probably
would, (4) definitely would. Participants were asked to indicate which result they would expect if
genetic testing were possible, using the following answer categories: (1) I have hereditary TGCT
(2) I probably have hereditary TGCT, (3) the chance/risk that I have hereditary TGCT is fifty/fifty,
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility102 Genetic Predisposition to Testicular Cancer
8
103
(4) I probably do not have hereditary TGCT, (5) I do not have hereditary TGCT, (6) I do not have
any expectation about my chance/risk on having hereditary TGCT. Participants could indicate the
degree to which they wanted to have information about “their chance/risk on hereditary TGCT
and its consequences” on a scale ranging from 1 (not wanting to receive any information) to 10
(wanting to know as much as possible).
Fourteen self-constructed questions were used to gain insight into reasons/motivations on why
patients/controls would undergo genetic testing for TGCT/cancer (Table 2) and ten questions
were used to explore reasons why patients/controls would decide against undergoing genetic
testing (Table 3). Response categories ranged from: 1: does not play a role at all in my decision
to undergo genetic testing to 6: plays a very important role in my decision to undergo genetic
testing. Most of these questions were previously used in other studies on women with a genetic
predisposition for breast cancer.(271;278;286;287) The control group men responded to the same
questions but “TGCT” was replaced with the word “cancer”.
Statistical analyses
Statistical analyses were performed using SPSS 14.0. Descriptive statistics were used to
calculate means, frequencies and percentages. Chi square tests and ANOVAs were conducted to
examine differences between groups in socio-demographic variables. For employment status, a
dichotomous variable was created with the categories “not employed for wages” consisting of:
student, being unemployed, unable to work, or retired and “employed for wages” consisting
of paid job. For marital status, a dichotomous variable was created with the categories
“relationship” and “no relationship” consisting of single, divorced and widower.
Chi square tests and ANOVAs followed by post hoc Bonferroni tests were computed to examine
differences between groups in motivations for testing. Pearson’s correlations were calculated
to investigate relationships between variables. Correlation coefficients lower than .30 are
considered weak, those between .30 and .50 are moderately strong, and those higher than .50
are considered strong.(288) Results were considered statistically significant if the probability of
occurrence was <− 0.05.
Results
Of the 44 familial TGCT patients, 23 returned the questionnaire (response 52%); 27 of the 45
sporadic TGCT patients (60%), and 77 of the 116 members of the patient association (66%) did
so. All 42 men in the control group filled in the questionnaire. The study group totaled 169 men.
Mean age was 40.4 years (range 16.4– 69.2). An ANOVA showed a significant effect of group on
age. A Post-hoc Bonferroni test showed that familial TGCT patients were significantly older than
sporadic patients and controls. Mean educational level was relatively high and not significantly
different between the four groups. The vast majority of men had a paid job (86%) and were in
a relationship (79%), and 60% had children. The four groups did not significantly differ in these
variables. TGCT patients mentioned that other types of cancer than TGCT were significantly more
often prevalent in the family (X2 =8.43, p = 0.004) than controls did. (Table 1)
Table 1: Sample characteristics and comparison between groups
Variable Familial Sporadic Patient Controls Test result
TGCT TGCT association (n=42)
(n=23) (n=27) (n=77)
Age: mean (SD) 46.1 (9.9) 36.6 (7.9) 41.1 (8.5) 38.1 (11.9) F= 5.08A, p=.002
Highest education level: 4.5 (1.6) 4.6 (1.4) 5.1 (1.5) 4.7 (1.8) F= 1.1, p=.334
mean (SD)
Marital status X2= 1.45B, p=.69
• Relationship 20 20 61 31
• No relationship
- Single 2 4 10 7
- Divorced 1 3 6 3
- Widower 0 0 0 0
Daily occupation X2 = 3.64C , p=.30
• Employed for wages 18 26 65 36
• Not employed
- student 0 0 2 3
- unemployed 1 1 2 0
- unable to work 1 0 8 3
- retired 3 0 0 0
Children X2=6.95, p=.073
• yes 16 13 51 20
• no 6* 14 26 22
Family members X2=8.43D, p = .01
diagnosed with other
types of cancer?
• yes 16 20 47 25
• no 7 7 29* 17
Genetic testing X2=4.72D, p=.03
intention: n (%)
• Probably/definitely not 3(13) 5(19) 19(25) 16(38)
• Probably/definitely will 20(87) 22(81) 58(75) 26(62)
Perceived risk: n (%) X2=45.38E, p<.001
• (probably) yes 9(39) 2(7) 5(7) 1(2)
• chance/risk is fifty/fifty 7(30) 11(41) 22(29) 2(5)
• (probably) no 2(9) 8(30) 30(40) 26(62)
• no expectations 5(22) 6(22) 19(25) 13(31)
Information wish: 8.6 (1.5) 8.6 (1.7) 8.0 (2.3) 6.9 (2.3) F= 5.12F, p=.002
mean (SD)
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility104 Genetic Predisposition to Testicular Cancer
8
105
A significantly greater percentage of the patients (79%) than of the controls (62%) would decide
to undergo genetic testing (X2= 4.72, p = 0.03). The percentage was highest in the familial
group (87%) but not significantly higher than in the other two patient groups (Table 1). Age was
not, but educational level was significantly but weakly related to interest in genetic testing (r =
-.16, p = .042). The lower the educational level was, the greater the interest in genetic testing.
Men in a relationship, with a paid job, and with children or family members with cancer other
than testicular cancer did not significantly differ from their counterparts in interest in genetic
testing.
Thirty-nine percent of the familial patients believed that they have a hereditary form of TGCT/
cancer, while 7% of the sporadic patients, 7% of the patient association group and 2% of the
controls did so. Only 9% of the familial patients believed that they would not have a hereditary
form of TGCT/cancer, in contrast to 62% of the controls, 40% of the patient association group and
30% of the sporadic patients. A Chi square test, excluding the category: “I have no expectations
about having hereditary TGCT”, revealed that the response pattern was significantly different
between groups (X2=44.16, p<.001) (Table 1). Of the 17 men who suspected that they would
(probably) have a hereditary form of TGCT, 16 (94%) would decide to undergo genetic testing.
Of the 42 men who believed they would have a fifty-fifty chance/risk, 37 (88%) would undergo
testing. 46 of the 66 men (70%) who believed they would (probably) not to have a hereditary
form would undergo testing; and of the 43 men who did not have any expectation about test
results, 27 (63%) would undergo genetic testing (X2=11.56, p=.009).
An ANOVA showed a significant effect of group on the wish for information about hereditary
TGCT/cancer and its consequences. A Bonferroni test showed that controls wanted to receive
significantly less information than patient association members (p = 0.04), sporadic patients (p
= 0.008) and familial patients (p = 0.01). The patients’ mean scores were high, varying from 8.0
to 8.6. The correlation coefficient between the wish for information and intention to undergo
genetic testing was significant (r = 0.54, p < 0.001), meaning that the more information men
Legend Table 1
A: post hoc Bonferroni test: familial TGCT patients significantly older than sporadic patients (p=.004)
and controls (p=.009)
B: X2 relationship versus no relationship
C: X2 paid job versus other
D: all patients versus controls
E: excluding response option “no expectation” from analysis
F: post hoc Bonferroni test: control group significantly lower mean score than familial patients
(p = 0.012 ), sporadic patients (p =0.008) and patient association (p = 0.037)
* information of one respondent is missing
Tabl
e 2:
Mot
ivat
ions
for
gen
etic
tes
ting
for
TGCT
com
parisi
on b
etw
een
grou
ps a
nd r
elat
ions
hip
with
tes
ting
inte
ntio
n
Item
To
tal
fam
ilial
sp
orad
ic
patie
nt
cont
rols
AN
OVA
Co
rrel
atio
nPos
sibl
e ra
nge:
1=
no rol
e–6=
very
gr
oup
patie
nts
patie
nts
asso
-
with
test
ing
impo
rtan
t ro
le ran
king
by
mea
n
ciat
ion
in
tent
ion
scor
e of
tot
al g
roup
m
ean
mea
n m
ean
mea
n m
ean
(SD)
(SD)
(SD)
(SD)
(SD)
F p
r1)
To
supp
ort sc
ient
ific
rese
arch
4.
5 (1
.3)
4.9
(1.1
) 4.
9 (1
.3)
4.5
(1.2
) 3.
9 (1
.6)
4.2
A)
.007
.4
0***
2) T
o ga
in m
ore
certai
nty
abou
t risk
4.
4 (1
.7)
5.0
(1.1
) 4.
6 (2
.0)
4.6
(1.6
) 3.
7 (1
.6)
4.6
B)
.004
.4
9***
of
get
ting
TGCT
/can
cer
3) T
o be
mor
e ce
rtai
n ab
out m
y ch
ildre
n’s
risk
3.
9 (2
.0)
4.1
(2.1
) 4.
0 (2
.0)
4.1
(2.1
) 3.
5 (2
.0)
0.7
.57
.48**
*
4) T
o ge
t re
gula
r m
edic
al c
heck
-ups
3.
9 (1
.7)
4.4
(1.2
) 4.
7 (1
.5)
3.8
(1.5
) 3.
2 (1
.8)
6.2
C)
.001
.4
2***
5) T
o ta
ke p
reve
ntiv
e m
easu
res
3.6
(1.7
) 3.
6 (1
.6)
4.0
(1.9
) 3.
3 (1
.7)
4.0
(1.8
) 1.
6 .2
0 .4
5***
6) B
eing
afrai
d fo
r ca
ncer
rec
urre
nce
3.6
(1.8
) 3.
9 (1
.5)
4.0
(1.6
) 3.
6 (1
.8)
3.1
(1.9
) 2.
0 .1
2 .4
1***
7) A
t m
y pa
rtne
r’s/
child
ren’
s in
sist
ence
3.
4 (1
.8)
3.3
(1.9
) 3.
7 (1
.5)
3.2
(1.8
) 3.
7 (1
.8)
1.0
.39
.32**
*
8) T
o de
crea
se m
y fe
ar o
f ca
ncer
3.
3 (1
.8)
3.3
(1.8
) 3.
9 (1
.4)
3.2
(1.8
) 3.
0 (1
.9)
1.3
.28
.38**
*
9) B
ecau
se the
re is
canc
er in
my
fam
ily
3.1
(1.8
) 4.
3 (1
.3)
3.5
(1.9
) 2.
8 (1
.8)
2.9
(1.7
) 5.
1 D)
.002
.3
1***
10) Ano
ther
fam
ily m
embe
r w
ants
gen
etic
res
earc
h 3.
0 (1
.7)
3.1
(1.8
) 2.
9 (1
.6)
2.8
(1.7
) 3.
5 (1
.9)
1.5
.22
.23**
11)
At a
doct
or’s
ins
iste
nce
2.9
(1.6
) 2.
7 (1
.6)
3.1
(1.5
) 2.
6 (1
.5)
3.4
(1.8
) 2.
3 .0
76
.17*
12) At m
y pa
rent
s’/s
iblin
g(s)
’s ins
iste
nce
2.7
(1.6
) 2.
4 (1
.6)
2.9
(1.5
) 2.
5 (1
.5)
3.0
(1.6
) 1.
4 .2
6 .1
8*
13) To
hel
p de
cide
whe
ther
to
have
chi
ldre
n 2.
7 (1
.9)
2.1
(1.7
) 2.
9 (2
.0)
2.6
(1.9
) 3.
0 (2
.0)
1.3
.27
.21**
14) Gen
eral
fut
ure
plan
ning
(pa
rtne
r, job
) 2.
6 (1
.8)
2.4
(1.7
) 2.
6 (1
.6)
2.6
(1.7
) 2.
8 (1
.9)
0.3
.86
.30**
*
* =
p < 0.
05, **
= p
< 0
.01,
***
= p
< 0
.001
Post
Hoc
Bon
ferron
i te
sts:
A) c
ontrol
gro
ups’
mea
n si
gnifi
cant
ly low
er tha
n th
at o
f sp
orad
ic (p=
0.02
) an
d fa
mili
al p
atie
nts
(p=
0.03
)
B) co
ntro
l gr
oups
’ m
ean
sign
ifica
ntly
low
er tha
n th
at o
f pa
tient
ass
ocia
tion
(p=
0.02
) an
d fa
mili
al p
atie
nts
(p=
0.00
8)
C) c
ontrol
gro
ups’
mea
n si
gnifi
cant
ly low
er tha
n th
at o
f fa
mili
al (p=
0.02
) an
d sp
orad
ic p
atie
nts
(p=
0.00
1)
D) co
ntro
l gr
oups
’ (p
=0.
01) an
d pa
tient
ass
ocia
tion
grou
ps’ m
ean
(p=
0.00
2) s
igni
fican
tly low
er tha
n th
at o
f fa
mili
al p
atie
nts
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility106 Genetic Predisposition to Testicular Cancer
8
107
would like to receive regarding their chance of having hereditary TGCT, the more likely they were
to decide to undergo genetic testing (Table 1).
Based on the mean scores of the total group, the motivations: “to support scientific research”,
“to gain more certainty on the risk of getting TGCT/cancer, “to get regular medical check-ups”
and “to be more certain about my children’s risk” played a relatively important role. The
motivations: “at the insistence of parents/siblings”, “to help decide whether to have children”
and “general future planning” played a relatively small role (Table 2). An ANOVA showed a
significant effect of group on four items. Post-hoc Bonferroni tests showed that: “to gain more
certainty about getting TGCT/cancer” and “because there is cancer in my family” played a
significantly less important role for the control group than for the familial patients (p = 0.008
and p=0.01) and for the patient association group (p = 0.02 en p = 0.002). ”To get regular
medical check-ups” and “to support scientific research” played a significantly less important role
for the controls than for the familial (p=0.02 and p = 0.03 respectively) and sporadic patients
(p=0.001 and p=0.02 respectively).
Correlational analyses of these 14 motivations regarding genetic testing resulted in 10 significant
and moderately strong relationships. The remaining 4 relationships were significant but weak
(Table 2).
Barriers for undergoing genetic testing for TGCT are shown in Table 3. The mean scores appeared
to be relatively low (varying from 1.3 to 2.9). Based on the mean scores of the total group, the
barriers: “possible trouble with insurance company” and “I do not want to undergo surgery”
played a relatively greater role while “it will have implications for my choosing a partner” and
“I do not have faith in such a test” played a smaller role. An ANOVA showed a significant effect
of group on the barrier “I do not want to undergo surgery”. A post-hoc Bonferroni test showed
that this barrier played a significantly less important role for the controls than for the patient
association group (p = 0.005). Correlational analyses between these ten questions and intention
on genetic testing showed two significant but weak correlations. The greater the intention to
undergo genetic testing, the less of a role the barriers “medical check-ups” and “time” played.
Discussion
The first goal of this study was to investigate whether patients treated for TGCT were more
interested in and if they had different motivations for genetic testing for TGCT (if such a test was
available) than people who had not had TGCT. As hypothesized, we found that a significantly
larger percentage of men who had been treated for TGCT would undergo genetic testing for TGCT
(79 %), compared to men (62 %) who had not suffered from cancer. A difference in intention
to undergo genetic testing was also found in a study comparing breast cancer patients with
healthy controls.(289) In that study, the first group was almost six times more likely to express
interest in the genetic test than women without cancer. A second important finding from that
Tabl
e 3:
Mot
ivat
ions
aga
inst
gen
etic
tes
ting
(bar
rier
s) for
TGCT
, co
mpa
risi
on b
etw
een
grou
ps a
nd r
elat
ions
hip
with
tes
ting
inte
ntio
n
Item
To
tal
fam
ilial
sp
orad
ic
patie
nt
cont
rols
AN
OVA
Co
rrel
atio
n
Pos
sibl
e ra
nge:
1=
no rol
e–6=
very
gr
oup
patie
nts
patie
nts
asso
-
with
test
ing
impo
rtan
t ro
le ran
king
by
mea
n
ciat
ion
in
tent
ion
scor
e of
tot
al g
roup
m
ean
mea
n m
ean
mea
n m
ean
(SD)
(SD)
(SD)
(SD)
(SD)
F p
r
1) P
ossi
ble
prob
lem
s w
ith
insu
ranc
e co
mpa
nies
2.
6 (1
.7)
2.6
(1.8
) 2.
5 (1
.6)
2.9
(1.7
) 2.
0 (1
.6)
2.5
.06
-.04
2) I
do n
ot w
ant to
und
ergo
sur
gery
2.
4 (1
.6)
2.7
(1.6
) 2.
3 (1
.6)
2.8
(1.7
) 1.
8 (1
.1)
4.0A
.0
08
-.10
3) I
am a
frai
d fo
r ca
ncer
2.
2 (1
.3)
1.9
(1.0
) 2.
1 (1
.1)
2.3
(1.4
) 2.
2 (1
.3)
0.5
.71
.01
4) I
do n
ot w
ant re
gula
r m
edic
al c
heck
-ups
2.
0 (1
.3)
1.8
(1.0
) 1.
7 (1
.0)
2.2
(1.4
) 1.
7 (1
.1)
2.3
.08
-.22
**
5) It
cou
ld h
ave
impl
icat
ions
for
my
job
1.9
(1.3
) 1.
4 (.7)
2.
0 (1
.5)
2.1
(1.4
) 1.
8 (1
.2)
2.0
.11
-.01
6) It
cou
ld a
ffec
t m
y de
cisi
on to
have
chi
ldre
n 1.
9 (1
.5)
1.6
(1.2
) 1.
9 (1
.6)
1.8
(1.4
) 2.
1 (1
.7)
0.7
.58
.09
7) It
cos
ts m
oney
1.
8 (1
.2)
2.0
(1.3
) 1.
7 (1
.0)
1.9
(1.2
) 1.
6 (1
.2)
0.9
.42
.02
8) It
tak
es tim
e 1.
7 (1
.1)
1.7
(0.8
) 1.
4 (0
.6)
1.9
(1.2
) 1.
5 (0
.9)
2.6
.06
-.16
*
9) I
do n
ot h
ave
faith
in s
uch
a te
st
1.5
(1.0
) 1.
3 (0
.6)
1.4
(0.8
) 1.
6 (0
.9)
1.6
(1.2
) 0.
7 .5
5 -.04
10) It w
ill h
ave
impl
icat
ions
for
my
choi
ce
1.5
(1.1
) 1.
4 (1
.1)
1.3
(0.5
) 1.
6 (1
.1)
1.5
(1.2
) 0.
7 .5
4 -.01
of
a p
artn
er
* =
p < 0.
05, *
* =
p < 0
.01
A: P
ost Hoc
Bon
ferron
i te
st: co
ntro
l gr
oup
sign
ifica
ntly
low
er m
ean
than
pat
ient
ass
ocia
tion
(p =
0.0
05)
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility108 Genetic Predisposition to Testicular Cancer
8
109
study is the revelation that the intention to be tested may not automatically translate into
actually undergoing testing. This observation was also reported by others.(278;290) In contrast to
our expectations, we did not find that significantly more familial TGCT patients would decide to
undergo testing than non-familial TGCT patients. Our finding is not in line with a study focusing
on cutaneous melanoma that reported that interest in genetic testing is greater in patients with
one or more family member with cutaneous melanoma than in patients who do not have family
members with cutaneous melanoma.(282)
It is striking that a large percentage of the male controls (62%) would request genetic testing
for cancer if such a test was available. Anticipated regret could be an important motive for
requesting a genetic test. Anticipated regret means that patients want to prevent feeling regret
in the future because of possibly making a wrong decision at this moment.(291) An even higher
percentage of healthy controls expressed interest in genetic testing in studies on prostate cancer.
One study reported that 92% of (healthy) men who visited a prostate screening clinic expressed
an interest in genetic testing for prostate cancer and 89% indicated that they would undergo
testing if available.(275) This interest of healthy men in an as yet non-existing test was not related
to their having a positive or negative family history of prostate cancer. In contrast, another study
showed that only 27% of healthy men indicated that they would undergo testing for a prostate-
specific antigene (PSA).(292) Approximately 45% of these men knew of the existence of this test
and 15% had already undergone a PSA test. The large contrast in these two studies’ findings
may be due to the fact that a large number of the men in the study of Watson et al.(292) were
well-informed about the test. A review study showed that the more information men have about
the limitations and/or the risk of PSA testing, the less likely they are to decide to be tested.(293)
In the current study, it was found that the lower the educational level, the greater the interest in
genetic testing. It might be that lower educational level is associated with less knowledge about
heredity which may result in a more positive attitude towards testing.
Our expectation that more familial TGCT patients would expect a positive test result than other
TGCT patients and than controls is supported. However, only 39% of the familial TGCT patients
expect a positive test result, which is surprising given that these men have a positive family
history. An explanation could be that although these men were informed in the clinic and the
brief information accompanying the request to participate in the study of the possibility of a
genetic defect, they were also informed that a positive family history is not proof of heredity.
Patients with a negative family history of TGCT are not very likely to have a hereditary form. It
is, therefore, remarkable that 48% of the sporadic TGCT patients assume that they may have a
genetic predisposition for cancer or that their chance is 50%. It is possible that people are not
able to think about risk in “chances”. Many people view their odds of becoming sick as 50-50:
you either get it or you don’t. This is referred to as binary thinking. Our results show that almost
all of the patients who think that they carry a genetic predisposition for TGCT or that their chance
is fifty-fifty, intend to undergo genetic testing.
In line with our expectations, this study showed that men who were not treated for cancer require
less information about their chance/risk on hereditary TGCT and its consequences than men who
were confronted with cancer and its consequent treatment. In contrast to our expectations,
familial patients do not differ on wish for genetic information from sporadic patients and patient
association members. The mean scores on information desire are high, which indicates that
patients are very interested in receiving information on genetic predisposition and available
options. Therefore, it is very important that both patients and controls are given adequate
information about the (possible) hereditary aspects of cancer and the implications of positive
and negative DNA test results, before genetic testing is offered, to prevent misconceptions about
possible genetic testing results. This information may result in less interest for genetic testing,
as was shown in the study from Watson et al.(292) but for TGCT this is not clear yet.
Relatively more important reasons for undergoing genetic testing were: “to support scientific
research”, “to gain more certainty about getting TGCT/cancer”, “to be more certain about my
children’s risk”, “to get regular medical check-ups” and “to take preventive measures”. This
is in line with earlier research in women with breast cancer.(271;278;286) Men in the current study
identified “at the insistence of a doctor”, “at the insistence of parents/siblings”, “to help decide
whether to have children” and “general future planning” as motivations that played a smaller
role which is conform a Dutch study on individuals at increased risk for breast/ovarian cancer
and colon cancer.(294)
Motives to undergo genetic testing differ between patients and controls. For patients, the
motivation: “to support scientific research” plays a significantly more important role than for
controls. It seems that patients have more an altruistic desire to aid genetic and oncology
research than controls. “To get regular medical check-ups” is another motivation that played
a more important role for patients than for controls. Patients diagnosed with TGCT can also
develop a malignancy in the second testicle as women with breast cancer in the other breast.
Medical check-ups may help TGCT patients feel more secure.
This study also examined barriers for undergoing genetic testing. It is remarkable that the mean
scores are low (under median score) suggesting that these motivations play a relatively small
role. Possible problems with insurance companies seemed to be the most important barrier
for the intention to undergo testing. Possible discrimination by insurance companies has been
identified as a concern by others as well.(285;295) The least important barriers were implications for
their choosing a partner and faith in the test. Time and money seem to also play a minor role.
In The Netherlands, the cost of genetic testing is covered by insurance companies, which may
explain why cost was not important in the decision-making process.
All the motivations for genetic testing were significantly related to the intention to undergo
genetic testing (Table 2). The motivations for testing that seemed to play a more important role
Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility110 Genetic Predisposition to Testicular Cancer
8
111
were more highly related to the intention to undergo testing (moderately strong relationships)
than the ones that played a less important role. Only two barriers to undergo genetic testing
are related to the intention to undergo testing (Table 3). “It takes time” and “I do not want
regular medical check-ups” are weakly correlated, meaning that these motivations probably do
not prevent men from genetic testing.
Our study had some limitations which have to be taken into account when interpreting the
results. Firstly, the choice to undergo genetic testing for TGCT was hypothetical. Men do not
have any information about or experience with a genetic test for TGCT. They do not have friends,
family members, or neighbors who have been tested for TGCT with whom they could exchange
information. In addition, it is unknown if/which preventive measures can be taken. This lack
of information/experience may influence the findings, as compared to a genetic test for breast
cancer, which is actually available and fairly common. Secondly, low numbers may have resulted
in a lack of power. The low numbers are the result of the rarity of TGCT. However, 127 TGCT
patients returned the questionnaire. This is the largest number of TGCT patients to have ever
been studied regarding their interest in and motivations for genetic testing, as far as the authors
know. This is a strength of the study. Additionally, the diversity of the studied groups and the
comparison with a healthy control group give this study strength.
When genetic testing for TGCT becomes possible, studies should closely look at the medical,
psychological, ethical and legal value of positive and negative DNA test results as well as the
outcome of monitoring. It will be interesting to examine the actual number of men who would
undergo a genetic test. Additionally, future studies should focus on psychosocial support in
genetic counseling, something the current study did not focus on. Questions that should be
examined include: is it possible to characterize participants/patients who could benefit from
psychosocial support, what kind of support would they require, and who should offer support?
Based on the answers to these questions and with help from this and other studies’ findings,
protocols could be developed regarding patient education and counseling. These protocols could
be used by a multidisciplinary team of doctors, nurses, psychologists and genetic counselors.
ConclusionThis is the second study to investigate the willingness, motivations and expectations of men
treated for TGCT and of men not confronted with cancer regarding genetic testing for TGCT.
Those more inclined to undergo genetic testing are men with lower education; men who have
cancer; men who suspect they have a genetic mutation or those who think the chance is fifty-
fifty; those who want more information about the possibility of carrying a gene mutation; and
men for whom a number of motivations play a greater role. Important motivations in deciding
to undergo genetic testing are: “to support scientific research”, “to gain more certainty on the
risk of getting TGCT/cancer”, “to be more certain about my children’s risk” and “to get medical
check-ups”. Possible insurance problems and not wanting to undergo surgery are relatively
important barriers.
Genetic Predisposition to Testicular Cancer
9
113
Chapter 9
Summary, discussion and future perspectives
Summary, discussion and future perspectives114 Genetic Predisposition to Testicular Cancer
9
115
Summary
Insight into hereditary aspects of Testicular Germ Cell Tumours (TGCT) may lead to the
identification of individuals at increased risk for developing TGCT, increase our understanding
of the mutation pathways that lead to sporadic TGCT, and is expected to contribute to
improvement of TGCT diagnosis (e.g. genetic screening for men at risk) and treatment (e.g. gene
therapy). The current thesis focuses on genetic predisposition for TGCT
In the general introduction on TGCT described in Chapter 1, epidemiological, therapeutical and
clinical aspects of TGCT are discussed. The literature on the genetic aspects of TGCT is reviewed
in Chapter 2. Several observations have suggested that TGCT susceptibility genes exist and
are important in this disease. These include increased TGCT risks associated with a positive
family history, the increased incidence of bilaterality in familial cases and the ethnic and racial
differences that do not change with migration, statistical analysis of observed familial and
nonfamilial cases (e.g. segregation analyses) and possible associations with known hereditary
syndromes and constitutional chromosomal anomalies. In Chapter 3, the literature on these
syndromes and chromosomal anomalies is reviewed. Twenty five of these disorders have
been reported in patients who also developed seminomatous or nonseminomatous testicular
carcinoma. Although these malignancies were too rare to enable the detection of statistically
significant correlations between the chromosomal / hereditary disorder and the testicular cancer,
it was striking that many of the patients had other urogenital abnormalities in addition to their
TGCT. Urogenital abnormalities are caused by disrupted urogenital differentiation and the same
disturbance may have lead to testicular dysgenesis and / or testicular carcinoma. This theory
of shared genetic and environmental risk factors in testicular tissue differentiation has been
postulated by Skakkebaek and colleagues.(12) Tumour cytogenetics has also provided hints that
at least some of the genes involved in the hereditary disorders in question may be causally
related to TGCT as these genes have been mapped to regions that appear to be involved in the
development of sporadic TGCT. Molecular studies on candidate genes will be required to provide
definite answers.
In Chapter 4, a difference is described in the incidence of TGCT between the northern and
eastern part of the Netherlands. Within the Netherlands, which is a relatively small country,
geographic differences in incidence of TGCT exist, with a statistically significant highest incidence
in the northern part of the Netherlands. This geographic clustering of TGCT may be caused by
a stable founder population that likely share a relatively high frequency of genes from common
ancestors, including genes possibly related to TGCT. This founder population is suitable for
searching TGCT susceptibility genes
In search of these candidate genes, a study was undertaken to study the association between
histocompatibility antigens (HLA), in particular class II genes (DQB1, DRB1) and TGCT. This study
Summary, discussion and future perspectives116 Genetic Predisposition to Testicular Cancer
9
117
described in Chapter 5 used genotyping of microsatellite markers in order to confirm or refute
previously published associations. In 151 patients, along with parents or spouses, the HLA-
region (particularly class II) on chromosome 6p21 was genotyped for a set of 15 closely linked
microsatellite markers. In both patients and controls, strong linkage disequilibrium was observed
in the genotyped region indicating that similar haplotypes are likely to be identical by descent.
However, association analysis as well as the transmission disequilibrium test did not show
significant results. An alternative approach to linkage studies is searching for TGCT susceptibility
genes among unrelated TGCT cases in founder populations by means of association analyses on
a dense set of markers. These so-called linkage disequilibrium fine-mapping analyses are based
on the hypothesis that patients in founder populations inherited disease mutations from recent
and common ancestors. This analysis is called haplotype analysis (Haplotype Sharing Statistic,
HSS). Haplotype analysis did not show differences in haplotype sharing between patients and
controls. Therefore, this study did not confirm the previously reported association between HLA
class II genes and TGCT.
Another genomic region thought to harbour TGCT associated gene mutations was the Y
chromosome. Reduced fertility is associated with TGCT and reduced fertility and TGCT might
share genetic risk factors according to the testicular dysgenesis hypothesis (designed by
Skakkebaek et al).(12) Up to 8% of infertility and reduced fertility in the general male population
can be explained by the presence of constitutional deletions of part of the long arm of the Y
chromosome (Yq11), referred to as the azoospermia factor (AZF) region. In Chapter 6, a study
is presented that investigated the frequency of azoospermia factor (AZF) deletions in Dutch
patients with TGCT. In 112 patients with TGCT, screening for constitutional deletions in the AZF
region was performed by multiplex polymerase chain reaction (PCR) analysis in DNA extracted
from peripheral blood lymphocytes. A set of 24 primer pairs of which 20 primer pairs are
homologous to previously identified and mapped sequenced tag sites (STSs) within the AZF
region were used. No deletions in the Yq11 region were detected in any of the 112 patients. The
conclusion of this study is that large Y chromosome microdeletions in the AZF region are not a
major contributor to the development of TGCT and TGCT-associated reduced fertility.
Global efforts have been made to identify TGCT predisposing genes. Although TGCT families have
been reported in the literature, multigenerational pedigrees with several affected cases are rare
and this limits the power of linkage studies. In 1994, the International Testicular Cancer Linkage
Consortium (ITCLC) was formed with the aim to collect TGCT families from all over the world
and to perform genotyping studies. The ITCLC performed genotyping studies and their results
provided evidence for the location of a gene involved in TGCT and cryptorchism susceptibility as
the consortium found a HLOD score for such a gene of 2.01 on chromosome Xq27. In Chapter
7 a genotyping study aimed at this Xq27 region is presented. We used the Haplotype Sharing
Statistic (HSS), which was also used in the HLA screen as described in chapter 5. In 276 patients
and 169 unaffected first-degree male relatives, 12 microsatellite markers covering the candidate
region were genotyped and used to study possible association of TGCT with Xq27 both by
single locus association analysis as well as the HSS. We observed a suggestive association
between the subset of TGCT cases without a family history of TGCT or cryptorchism and marker
DXS1193 both by allelic association (p=0.014) and the HSS (p=0.09), both were not significant
after multiple testing correction. However, when taken all minor alleles at this marker together,
this marker remained significantly different between this subset of TGCT patients (13.9% had
minor allele) and controls (3.4%). An individual without cryptorchism carrying a minor allele
had a 4.7-fold risk to develop sporadic TGCT (99.57% CI 1.1-19.6) compared with carriers of the
major allele. This study could not confirm the previously observed linkage of TGCT to Xq27,
which was particularly observed in families with bilateral cases and cases with cryptorchism, but
we did observe an association between the subset of TGCT cases without a family history of
TGCT or cryptorchism and marker DXS1193. Multiple minor alleles at this locus had an increased
frequency among patients. This suggests that in our population several mutations of an Xq27-
linked gene have a moderate to large effect on TGCT development but not on cryptorchism.
Although there is currently no genetic test available for testing genetic predisposition to TGCT,
this may very well change in the future. In Chapter 8 a study was performed to examine interest
in and motivations for TGCT susceptibility testing if such a test would be available today,
by comparing the opinions of TGCT patients with controls. Four groups of respondents were
recruited. From the database consisting of all TGCT patients/survivors (n=702) treated at the
University Medical Center Groningen between 1977 and 2003, all TGCT patients with an affected
first or second degree relative (familial TGCT) and a randomly selected group of TGCT patients
without a family history (sporadic TGCT) were selected. The third group consisted of TGCT
patients recruited by the Dutch TGCT patients/survivors association (patient association group)
and the fourth of age-matched male patients visiting the hospital’s emergency room for a minor
trauma (controls). A letter with information about the objectives of the study and a questionnaire
was given to all participants. The results: 23/44 familial TGCT patients (response 52%), 27/45
randomly selected sporadic TGCT patients (60%), 77/116 patient association members (66%) and
42 control group men (100%) completed the questionnaire. More patients (79%) than controls
(62%) would undergo genetic testing (X2=4.72, p=0.03). More familial patients (40%) expected
that they would have a genetic mutation than sporadic patients (7.4%), patient association
members (6.6%), and controls (2.4%) (X2=45.4, p<.001). A significant effect of group was found
on 4 of 14 motivations for TGCT genetic testing: “To support scientific research” (F=4.1, p=.007),
“to gain more certainty about risk of getting TGCT/cancer” (F=4.6, p=.004), “to get regular
medical check-ups” (F=6.2, p=.001) and “because there is cancer in my family” (F=5.1, p=0.02).
These motivations played a smaller role for controls than for patients. “Possible insurance
problems” and “I do not want to be operated” are the most important reasons why men would
restrain from genetic testing. We concluded that more patients than controls would choose to
undergo genetic testing if such a test were available. Familial TGCT patients significantly more
often than controls assume that they carry a genetic predisposition. Some motivations play a
Summary, discussion and future perspectives118 Genetic Predisposition to Testicular Cancer
9
119
greater role for patients than for controls in their decision on genetic testing. The greater role a
motivation played the more inclined respondents were to undergo genetic testing
Discussion and future perspectivesThe incidence of TGCT has doubled over the last 40 years. This increase is unlikely to be caused
by genetic predisposition alone and environmental factors most likely play an important role
as well. Although TGCT is an infrequent disease and it is highly curable, it is still important
to unravel the environmental risk factors as well as genetic predisposition for TGCT. Early
identification of men with increased risk for inherited TGCT might lead to early detection and
improved treatment outcome. As mentioned in chapter 1, disseminated TGCT can be classified in
three prognostic groups (good, intermediate and poor prognosis). Survival of TGCT is dependent
on the prognostic group, with 5 years survival rates of 94% (good), 83% (intermediate) and
71% (poor) respectively.(296) In the current literature there are no data on the clinical behaviour
(or prognosis) of TGCT in patients with familial TGCT. Clinical observation does also not point to
a difference in survival. Presently, there is no reason to expect that familial TGCT has a different
outcome than sporadic TGCT. Thus identifying TGCT in genetically predisposed men at an earlier
stage, (i.e. in stage I or in the group with a good prognosis), is expected to improve survival.
Early detection could theoretically be reached by ultrasound of the testes, palpation (routine
testicular self examination) and/or by measuring tumour markers. Presently, however, there is
no proof that men who routinely examine their testes are more likely to detect earlier stage
tumours or improve their prospects for survival.(297) From a therapeutical point of view, treatment
of TGCT may be further optimized using new insights in the oncogenetic pathways involved and
possibly reduce the toxic side-effects of chemotherapy and radiotherapy. Identifying patients
at an earlier stage of dissimenation with favourable prognostic factors, the number of courses
of chemotherapy / radiotherapy can be reduced, resulting in less toxic side effects. Besides
this, the unravelling of the genetic pathway of TGCT may result in the opportunity to develop
gene therapy. There are some trials (in other types of cancer) with promising results. The
use of genetically modified autologous tumour cells appears to be a promising approach for
cancer therapy. Further studies are required to determine whether promising effects on immune
activation will result in an actual clinical benefit for patients.(298)
In chapter 6 we discussed the study of microdeletions in the AZF region of the Y chromosome in
TGCT patients. The AZF deletions are well known deletions of the Y chromosome and account for
8% of the male infertility. This region was studied in TGCT patients because they often present
with abnormal semen characteristics (reduced fertility). We did not find an association between
AZF deletions and TGCT in our study. However, recently a novel Y-chromosome 1.6-MB deletion
was described, named: gr/gr. This deletion was associated with spermatogenic failure as well.
The gr/gr deletion is much smaller than the deleted AZF region studied previously and removes
only part of the AZFc region, including copies of DAZ and a copy of CDY1, as well as other
transcription units (see Chapter 6). Nathason et al recently described an association between
these gr/gr deletions and TGCT. Familial TGCT patients had a threefold increased risk having
these gr/gr deletions.(258) This study focusing on a small part of the AZF region elucidated a small
deletion associated with moderately increased risk for TGCT. Therefore the Y chromosome, in
particular the AZFc region, seems still relevant for further research on genetic predisposition of
TGCT.
Although circumstantial evidence points to the existence of TGCT predisposition, no germ line
gene mutations have yet been identified that confer a high risk to develop TGCT. The question
arises whether such mutations exist at all or whether genetic predisposition for TGCT (as the
gr/gr deletions did) will turn out to consist of a range of mutations, each with only relatively
weak effects. One barrier to resolving this issue is the fact that large families with a strong family
history of TGCT are rare. The international testicular cancer linkage consortium (ITCLC) was
established to collect large number of TGCT families to perform linkage studies. In a previous
study, the ITCLC presented evidence for linkage of TGCT with Xq27 (in a study that was based on
99 pedigrees).(105) However, in their last study, in which 179 pedigrees were studied genome wide,
no regions showed a significant HLOD score of 2 or higher. The genomic regions: 2p23, 3p12,
3q26, 12p13-q21, 18q21-q23 and Xq27 (the Xq27 region was examined in a further 66 pedigrees
compared to their previous study) had a HLOD score between 1 and 2, which is not considered
significant, but neither provides evidence against a TGCT gene in this region.(262) Since the ITCLC
has the largest series of familial TGCT patients (459 pedigrees), and could not identify genomic
regions associated with high TGCT risk, the ITCLC suggests that the susceptibility to TGCT is
determined through multiple loci as opposed to a single locus. Possibly, very strong inherited
TGCT predisposition does not exist. Furthermore, linkage analysis in contrast to association
studies is not a powerful tool for gene detection when frequent genetic variants play a role.(299)
With 459 pedigrees, of which the ITCLC examined only 237, because the rest was considered
not well sampled, only relative risk of 5 or larger for carriers of a genetic variant (frequency
of 15%) can be detected. To detect a relative risk of 2, which is realistic in complex human
diseases, one would need to study at least 2000 pedigrees to have 80% power to detect such
a genetic variant. On the other hand, it cannot be excluded that (part of) the familial clusters
can be attributed to different, rare, strongly predisposing gene mutations present in subsets of
families, that, because of their low frequency in the population, may very well escape detection
in such relatively small linkage studies. From a theoretical point of view, it may be interesting
to perform linkage studies in a subset of familial TGCT patients with urogenital malformations
(like hypospadia, cryptorchism, bilateral disease etc), because in this subgroup of patients
(identical phenotypes), it is very well possible that they share a causative gene which results in
urogenital malformations as well as TGCT (Skakkebaek model). The same approach has been
used for Xq27: In the first study, significant results in a subgroup of familial patients with TGCT:
cryptorchism and/or one case of bilateral disease in the family(12;105)
Summary, discussion and future perspectives120 Genetic Predisposition to Testicular Cancer
9
121
Recently the genetic defect in mouse strain 129.MOLF-Chr19 chromosome substitution strain,
known to develop TGCTs at a high frequency (70-80%), was reported.(300) A germ line mutation
in Dead End Gene (DND1) was found to cause this high tumour risk (and some testicular and
spermatogenic abnormalities).(258) Very little is known about the human homologue of DND1
and it is therefore worthwhile to study its possible association with human TGCT. Therefore the
University Medical Center Groningen (UMCG) is currently screening the gene for mutations in a
series of familial and sporadic TGCT patients and controls.
The collection of more families with TGCT will facilitate genetic studies. Some hospitals
systematically collect DNA from their cancer patients in a prospective way in order to expand
the material for genetic studies. The deCODE initiative (www.decode.com) takes this approach
a large step further by linking genealogical data with disease status and DNA markers in the
Icelandic population. Such approaches are to be encouraged from a pure scientific point of
view; however, privacy and other ethical and legal issues involved in these approaches need to
be addressed very thoroughly. International collaboration will be facilitated if DNA and patient
data collecting will be performed according to (to be developed) international standards. The
HapMap project, a partnership of scientists and funding agencies all over the world to develop
a public resource that will help researchers find genes associated with human disease, is a good
example (www.hapmap.org).
Not only study populations are expanding through collaborative efforts, statistical techniques
and insight in cancer biology and its resulting molecular tools evolve as well. Recently, the scope
of genetic study of TGCT has been extended to include the role of naturally occurring micro
RNAs (miRNAs). Normally miRNAs function as regulators of genetic pathways by manipulating
translational regulation. Some of these miRNAs (miRNA-372 and 373) were shown to allow
tumorigenic growth. As they have also been observed to be expressed in human seminomas and
non-seminomas, but not in normal testicular tissue, it has been suggested that these miRNAs
may represent a new class of oncogenes involved in TGCT development. (18) Time will tell whether
miRNAs play a role in (testicular) cancer predisposition
Should future studies identify clinically important TGCT predisposing mutations, then testing for
these mutations might be welcomed by a substantial subset of TGCT patients and families as
has been the case for many families with common hereditary tumour syndromes. Research on
early detection techniques for TGCT in high risk men will be necessary. Although many studies
have looked into the psychosocial aspects and requirements for genetic testing programmes in
these syndromes, very little is known with respect to testing in the setting of TGCT. Research
on early detection techniques for TGCT in high risk men will be necessary. As in any other new
genetic testing programme, testing would need to be performed in a research setting first, where
medical as well as psychosocial issues would need to be carefully monitored.
Genetic Predisposition to Testicular Cancer
10
123
Chapter 10Nederlandse samenvatting, conclusies en toekomstperspectieven
Nederlandse samenvatting, conclusies en toekomstperspectieven124 Genetic Predisposition to Testicular Cancer
10
125
Nederlandse samenvatting, conclusies en toekomstperspectieven
SamenvattingKanker ontstaat als gevolg van een opeenstapeling van beschadigingen (mutaties) van stukjes
erfelijke aanleg (genen). Deze beschadigingen kunnen zowel erfelijk als niet-erfelijk zijn. Bij
de meeste gevallen van kanker spelen individuele erfelijke genbeschadigingen geen sterke
rol bij het ontstaan van de gezwellen. Als echter erfelijke beschadigingen wel sterk bijdragen
aan het ontstaan van kanker dan spreken we van erfelijk kanker. Ouders met de aanleg voor
erfelijke kanker kunnen die aanleg aan hun kinderen doorgeven. Als deze aanleg bij iemand
aangetoond kan worden nog voordat kanker is ontstaan dan kunnen (afhankelijk van de soort
kanker) soms preventieve maatregelen worden getroffen. Van veel soorten kanker zijn inmiddels
erfelijke vormen beschreven en raken steeds meer details bekend over de aard en de rol van
de betrokken genmutaties. De genen die van belang zijn bij het ontstaan van erfelijke vormen
van kanker blijken vaak ook een rol te spelen bij het ontstaan van niet-erfelijke vormen van
kanker. In dat laatste geval gaat het dan niet om erfelijke maar om niet-erfelijke beschadigingen
van die genen. Dit proefschrift richt zich op het zoeken naar erfelijke factoren die de kans op
zaadbalkanker vergroten. Inzicht in de erfelijke aspecten van zaadbalkanker ( testis carcinoom
(TC)) zou kunnen leiden tot het identificeren van mannen met een erfelijk verhoogd risico op het
krijgen van TC (diagnostiek, evt. genetisch screenen) en tot het vergroten van de kennis over
genmutaties die leiden tot niet-erfelijk (sporadisch) TC.
In een algemene introductie worden in Hoofdstuk 1 epidemiologische, therapeutische en
klinische aspecten van TC beschreven. In Hoofdstuk 2 wordt een overzicht gegeven van de
huidige literatuur over genetische aspecten van TC. Verscheidene observaties suggereren dat
er een genetische predispositie voor TC bestaat en dat de betrokken genen belangrijk zijn bij
het ontstaan c.q. het beloop van deze ziekte. Deze observaties zijn: 1) verhoogd risico op het
krijgen van TC bij familiair voorkomen van TC, 2) verhoogde incidentie van dubbelzijdig TC in
families met TC, 3) de etnische en raciale verschillen in optreden van TC, die niet veranderen
na migratie, 4) resultaten van statistische analyses van familiare en niet-familiaire patiënten
(bijvoorbeeld segregatieanalyse, dit is een statistische techniek die het patroon van het
voorkomen van een aandoening in families vergelijkt met bekende overervingsmodellen) en 5)
mogelijke associatie van TC met bekende erfelijke syndromen en constitutionele chromosomale
afwijkingen. In Hoofdstuk 3 wordt de literatuur over de associatie tussen TC en deze syndromen
en chromosomale afwijkingen samengevat. Vijfentwintig syndromale afwijkingen worden
beschreven waarbij de betreffende patiënten tevens TC hebben ontwikkeld. Aangezien het om
zeer zeldzame combinaties gaat van TC met erfelijke syndromen en constitutionele chromosomale
afwijkingen konden hierbij geen statistische relaties worden aangetoond. Opvallend was dat
veel patiënten naast TC tevens urogenitale afwijkingen hadden (stoornis in de aanleg van
blaas, genitalia etc., bijvoorbeeld cryptorchisme, dit is niet-ingedaalde testikel). Urogenitale
Nederlandse samenvatting, conclusies en toekomstperspectieven126 Genetic Predisposition to Testicular Cancer
10
127
afwijkingen kunnen worden veroorzaakt door een gestoorde urogenitale differentiatie (stoornis
in celuitgroei van bijvoorbeeld organen) en dit zou tevens kunnen leiden tot testiculaire
dysgenesie (niet juiste aanleg) en / of TC. Deze theorie van gedeelde risicofactoren, d.w.z.
genetische en omgevingsrisicofactoren die een rol spelen in de differentiatie van testisweefsel is
beschreven door Skakkebaek et al.(12) De tumorcytogenetische studies (chromosomenonderzoek
van gezwellen) suggereren dat bepaalde genen, betrokken bij de betreffende erfelijke afwijking,
ook gerelateerd zouden kunnen zijn aan TC, aangezien deze genen zich in chromosoomregio’s
bevinden die mogelijk betrokken zijn bij het ontstaan van (sporadisch) TC. Moleculaire studies
naar kandidaat-genen zijn nodig om hier definitieve antwoorden op te geven.
In Hoofdstuk 4 wordt een verschil beschreven in incidentie van TC bij patiënten woonachtig in
Noord- en Oost-Nederland. In een klein land als Nederland bestaan dus geografische verschillen in
incidentie van TC, met een statistisch significante hoger incidentie in het Noorden van Nederland.
Deze geografische clustering van TC zou kunnen worden veroorzaakt door zogenaamde stabiele
founderpopulaties in Noord-Nederland (= “bron populaties” of oorspronkelijke populatie).
Mensen uit een dergelijke founderpopulatie delen waarschijnlijk relatief veel genen van een
klein aantal gemeenschappelijke voorouders, en mogelijk dus ook genen gerelateerd aan TC.
Onderzoek naar TC predisponerende genen in dergelijke founderpopulaties is vanuit theoretisch
oogpunt dan ook aantrekkelijk.
Om kandidaat-genen te onderzoeken werd een studie verricht naar de associatie tussen histo-
compatibiliteitsantigenen (HLA, Humane Leukocyten Antigenen), in het bijzonder klasse II genen
(DQB1, DRB1), en TC. Deze studie, beschreven in Hoofdstuk 5, is een genotyperingsstudie met
microsatellietmerkers (merkers voor korte stukjes DNA) om een eerdere gepubliceerde associatie
te bevestigen dan wel te verwerpen. In het bloed van 151 patiënten samen met de ouders of
eventueel de echtgenote en kinderen vond genotypering plaats van de HLA-regio (klasse II) op
chromosoom 6p21 met behulp van 15 aan elkaar gelinkte microsatellietmerkers. Bij zowel de
patiënten als de controlepersonen was er sprake van sterk linkage disequilibrium (allelen op
verschillende loci erven niet onafhankelijk van elkaar over) in de betreffende regio, hetgeen
betekent dat haplotypes (combinatie van allelen voor meerdere merkers op 1 chromosoom)
gelijk zijn door gemeenschappelijke afkomst (afstamming). Een alternatieve methode voor
koppelingsstudies (linkage-analyse) is het zoeken naar TC predisponerende genen in niet-
verwante TC patiënten in een founderpopulatie met behulp van associatieanalyse op een
set dicht bij elkaar liggende merkers. Deze linkage disequilibrium fine-mapping analyses zijn
gebaseerd op de hypothese dat patiënten in founderpopulaties ziektemutaties hebben geërfd
van gemeenschappelijke voorouders. Deze analyse wordt haplotype-analyse genoemd. Eén zo’n
methode is de Haplotype Sharing Statistic (HSS). In de HLA genotyperingsstudie liet haplotype-
analyse geen verschil zien tussen patiënten en controles, net als klassieke associatieanalyse
en de transmissie disequilibrium test. Deze studie kon daarom de eerder genoemde associatie
tussen HLA klasse II genen en TC niet bevestigen.
Het Y-chromosoom is een andere regio van het genoom waar TC gerelateerde genmutaties zich
zouden kunnen bevinden. Afgenomen fertiliteit is geassocieerd met TC en afgenomen fertiliteit
en TC zouden dus gemeenschappelijke onderliggende genetische risicofactoren kunnen hebben
volgens het eerder genoemde testiculaire dysgenesiemodel (ontworpen door Skakkebaek(12)).
In de algemene populatie kan 8% van de infertiliteit en afgenomen fertiliteit worden verklaard
door de aanwezigheid van constitutionele deleties van een deel van de lange arm van het Y
chromosoom (Yq11), ook wel de azoospermie (AZF) regio genoemd. In Hoofdstuk 6 wordt een
studie beschreven die de prevalentie analyseert van AZF-deleties bij Nederlandse patiënten met
TC. Bij 112 patiënten met TC werd DNA uit bloed geanalyseerd op constitutionele deleties in
de AZF-regio met behulp van multiplex polymerase ketting reacties. Er werd gebruik gemaakt
van 24 primerparen, waarvan 20 primerparen homoloog waren aan eerder geïdentificeerde en
gemapte sequence tag sites (STS) binnen de AZF-regio. Er werden bij deze 112 patiënten geen
deleties gevonden in de Yq11 regio. De conclusie van deze studie is dat (grote) Y-chromosoom
microdeleties in de AZF-regio niet geassocieerd lijken met TC en met TC geassocieerde
afgenomen fertiliteit.
Wereldwijd wordt een zoektocht naar TC predisponerende genen verricht. Ondanks dat er TC
families in de literatuur worden beschreven, zijn families met meerdere aangedane familieleden
schaars en dit belemmert linkage-analyse. In 1994 is het International Testicular Cancer Linkage
Consortium (ITCLC) opgericht met het doel TC families over de gehele wereld te verzamelen en
hierbij genotyperingsstudies te verrichten. Het ITCLC heeft deze studies verricht en de resultaten
toonden aanwijzingen voor een gen gelokaliseerd in de regio Xq27, dat geassocieerd leek te
zijn met TC en cryptorchisme. In Hoofdstuk 7 wordt een genotyperingsstudie gepresenteerd in
onze eigen TC families gericht op deze regio Xq27. Wederom werd HSS gebruikt zoals eerder
beschreven bij de HLA studie (Hoofdstuk 5). Bij 276 patiënten en 169 niet aangedane, mannelijke
eerstelijns verwanten, werd een genotyperingsstudie verricht met 12 microsatellietmerkers
(die de gehele kandidaat-regio op Xq27 dekken) om een mogelijke associatie van TC met de
betreffende Xq27 regio te analyseren(single locus associatieanalyse en HSS). Een suggestieve
associatie werd gevonden tussen een subgroep TC patiënten zonder aangedane familieleden
of zonder cryptorchisme en merker DXS1193, zowel bij allelische associatie (p=0.014) en HSS
(p=0.09), die beide niet significant waren na correctie voor multipele testen. Echter wanneer alle
laagfrequente allelen ter plaatse van deze merker werden samengevoegd bleek deze merker wel
significant verschillend te zijn tussen deze subgroep TC patiënten (13.9% had een laagfrequent
allel) en de controles (3.4%). Een individu zonder cryptorchisme die drager is van een
laagfrequent allel, had een 4.7 keer verhoogd risico om (sporadisch) TC te ontwikkelen (99.57%
betrouwbaarheidsgebied 1.1-19.6) vergeleken met dragers van het frequente allel. Deze studie
kon de eerder gevonden associatie tussen de Xq27 regio en TC patiënten met dubbelzijdig TC
en TC patiënten met cryptorchisme niet bevestigen. We vonden dus echter wel een associatie
tussen een subgroep TC patiënten zonder aangedane familieleden of zonder cryptorchisme en
marker DXS1193. Dit resultaat suggereert dat in onze populatie mutaties van een aan Xq27
Nederlandse samenvatting, conclusies en toekomstperspectieven128 Genetic Predisposition to Testicular Cancer
10
129
gelinkt gen een middelmatig tot groot effect hebben op het ontwikkelen van TC maar niet op
het ontwikkelen van cryptorchisme.
In de toekomst zou een genetische test beschikbaar kunnen komen om genetische predispositie
voor TC te testen. In Hoofdstuk 8 wordt een studie gepresenteerd die de interesse in en motivatie
voor genetisch testen onderzoekt van TC patiënten en controles. Vier groepen respondenten
werden benaderd. Uit alle TC patiënten (n=702), behandeld in het Universitair Medisch Centrum
Groningen tussen 1977 en 2003, werden alle TC patiënten geselecteerd met een aangedaan
eerste of tweede graads familielid (familiair TC; n=44) en een even grote, willekeurige groep TC
patiënten zonder aangedane familieleden (sporadisch). De derde groep bestond uit TC patiënten
van de Nederlandse TC patiëntenvereniging (stichting De Kernzaak) en de vierde groep bestond
uit een aan leeftijd aangepaste controlegroep van mannen die de spoedopvang (EHBO) van
een ziekenhuis bezochten vanwege een klein trauma (verzwikte enkel etc.). De mannen kregen
een informatiebrief over het onderzoek en een vragenlijst. De resultaten toonden aan dat 23/44
familiaire TC patiënten (respons 52%), 27/45 sporadische TC patiënten (60%), 77/116 patiënten
van de patiëntenvereniging (66%) en alle 42 controlemannen de vragenlijst hadden ingevuld.
Meer patiënten (79%) dan controles (62%) zouden overgaan tot genetisch testen (X2=4.72,
p=0.03). Meer familiaire patiënten (40%) dan sporadische TC patiënten (7.4%), patiënten van
de patiëntenvereniging (6.6%) en controles(2.4%) (X2=45.4, p<0.001) veronderstelden dat zij
een genetische predispositie hadden voor TC. Een significant verschil tussen de groepen werd
gevonden bij 4 van de 14 motivaties die een rol kunnen spelen bij mensen in hun beslissing
tot genetisch testen. “Om wetenschappelijk onderzoek verder te helpen (F=4.1, p=.007), “Om
meer zekerheid te verkrijgen over mijn kans op TC/kanker” (F=4.6, p=.004), “Om regelmatig
door een arts te worden gecontroleerd” (F=6.2, p=.001) en “Omdat er kanker voorkomt in
de familie” (F=5.1, p=0.02) spelen een kleinere rol voor de controles dan voor de patiënten.
“Mogelijke problemen met de verzekeringsmaatschappij” en “Ik heb geen zin in operaties” zijn
de voornaamste redenen waarom mannen zouden afzien van genetisch testen. Geconcludeerd
kan worden dat significant meer patiënten dan controle mannen zouden overgaan tot genetisch
testen wanneer dit in de toekomst mogelijk zou zijn. Familiaire TC patiënten veronderstellen
(zoals verwacht) significant vaker dat zij drager zijn van een genetische predispositie. Bepaalde
motivaties spelen een belangrijkere rol voor patiënten dan voor controles. Hoe belangrijker de
motivatie, des te meer geneigd zijn de respondenten om over te gaan tot genetisch testen.
Conclusies en toekomstperspectievenDe incidentie van TC is de afgelopen 40 jaren verdubbeld en neemt nog steeds toe. Deze
toename komt waarschijnlijk niet alleen door een genetische predispositie; omgevings factoren
lijken ook een belangrijke rol te spelen. Ondanks dat TC een zeldzame en goed te genezen ziekte
is, is het toch belangrijk de risico verhogende omgevingsfactoren en de genetische predispositie
voor TC op te helderen. Vroege herkenning van mannen met een verhoogd risico op erfelijke
TC, zou kunnen leiden tot vroege detectie en hierdoor tot een beter behandelbare ziekte en
een betere overleving. Zoals vermeld in Hoofdstuk 1 kan uitgezaaide TC worden onderverdeeld
in drie prognostische groepen (goed, middelmatig en slecht). Overleving is afhankelijk van
de prognostische groep, waarbij de 5 jaars overleving als volgt is: goede prognose: 94%,
middelmatige prognose 83% en de slechte prognose 71%.(296) In de huidge literatuur zijn
geen data bekend over het klinische gedrag cq prognose van specifiek familiair TC. Klinische
observatie toont in ieder geval geen verschil in overleving. Derhalve is er geen aanwijzing om
aan te nemen dat familiair TC zich klinisch anders zou gedragen dan sporadisch TC. Wanneer
bij een man een genetische predispositie voor TC kan worden vastgesteld en hierdoor TC in
een vroeg ziektestadium kan worden opgespoord (stadium 1 of goede prognose groep), dan
zal daarom zoals verwacht ook de overleving van deze patiënt verbeteren. Vroege detectie
van TC zou theoretisch kunnen worden bereikt door palpatie van de testikel (routinematig
zelfonderzoek), echografie van de testikels, en/of door het meten van tumormerkstoffen (zie
Hoofdstuk 1). Naar dit onderwerp is slechts zeer beperkt onderzoek verricht. Tot op heden is er
geen bewijs dat bijvoorbeeld zelfonderzoek van de testikels zal leiden tot het vaststellen van TC
in een vroeg stadium en hiermee tot een betere prognose voor de patiënt.(297)
Theoretisch gezien zou de behandeling van TC verder kunnen worden geoptimaliseerd met behulp
van nieuwe inzichten in de oncogenetische aspecten van TC en mogelijk door het reduceren van
de toxische bijwerkingen van chemotherapie en radiotherapie. Daarnaast zou het ontrafelen van
de genetische aspecten van TC kunnen leiden tot het ontwikkelen van gentherapie. Bij andere
vormen van kanker zijn hiermee al bemoedigende resultaten bereikt. Het gebruik van genetisch
gemodificeerde autologe tumorcellen lijkt een veel belovende behandelingsstrategie te worden.
Verder onderzoek zal moeten bepalen of het veelbelovende effect van immuunactivatie zal
leiden tot een klinisch voordeel voor de patiënt.(298)
In Hoofdstuk 6 is het onderzoek besproken naar microdeleties in de AZF regio op het Y-
chromosoom bij TC patiënten. De AZF deleties zijn bekende deleties van het Y-chromosoom en
zijn verantwoordelijk voor ongeveer 8% van de mannelijke infertiliteit (afgenomen fertiliteit).
Deze regio is bestudeerd in TC patiënten aangezien deze mannen ook vaak minder fertiel zijn
(abnormale semen karakteristieken). Een associatie tussen de AZF-deleties en TC werd niet
gevonden in onze studie. Recent is een nieuwe deletie op het Y-chromosoom beschreven (1.6-
Mb), genoemd: gr/gr. Deze deletie is ook geassocieerd met verminderde fertiliteit. De gr/gr
deletie is veel kleiner dan de microdeleties die bestudeerd werden in de AZF regio en door deze
deletie is slechts een klein gedeelte van de AZFc-regio verwijderd, onder andere kopieën van
DAZ en een kopie van CDY1, evenals andere transcriptie-units (zie Hoofdstuk 6). Nathason et al.
hebben recent een associatie beschreven tussen deze gr/gr deletie en TC. Patiënten met familiair
TC hebben een 3x verhoogd voorkomen van de gr/gr deletie.(258) Deze studie, gericht op een erg
klein gedeelte van de AZF regio toonde een deletie geassocieerd met een gemiddeld verhoogd
risico op TC. Hiermee is het Y-chromosoom, in het bijzonder de AZFc-regio, nog steeds relevant
voor verder onderzoek naar de genetische predispositie van TC.
Nederlandse samenvatting, conclusies en toekomstperspectieven130 Genetic Predisposition to Testicular Cancer
10
131
Hoewel er indirect bewijs bestaat voor een genetische predispositie voor TC, is de kiemlijn-
genmutatie die een hoge kans op het ontstaan van TC veroorzaakt nog niet gevonden. De
vraag doet zich voor of zo’n genmutatie wel bestaat of dat de genetische predispositie voor
TC (zoals de gr/gr deletie) bestaat uit een serie mutaties met elk slechts een zwak effect.
Een van de beperkingen van de mogelijkheden om deze theorie te staven is het feit dat er
slechts zeer weinig families zijn met meerdere familieleden met TC. Het International Testicular
Cancer Linkage Consortium (ITCLC) is opgericht om grote aantallen TC families te verzamelen
en hiermee linkage-studies te verrichten. In een recente studie toonde het ITCLC bewijs voor
linkage met chromosoom Xq27 (gebaseerd op 99 families).(105) Echter in de laatste studie door
het ITCLC, waarbij 179 families met merkers over het gehele genoom werden bestudeerd, werd
geen betekenisvolle regio op het genoom gevonden. De regio’s: 2p23, 3p12, 3q26, 12p13-q21,
18q21-q23 en Xq27 (Xq27 regio werd bij nog eens 66 families onderzocht) hadden een HLOD
score tussen de 1 en 2 (niet significant) en vormen dus geen bewijs voor een genoomregio
geassocieerd met TC. Het is overigens ook geen bewijs dat er in deze regio geen (zeldzaam)
predisponerend gen ligt voor TC.(262) Aangezien het ITCLC over het grootste aantal familiaire
TC patiënten beschikt (459 families) en zij geen genomische regio kunnen aantonen die
geassocieerd is met een verhoogd risico op TC, suggereert de ITCLC dat de predispositie voor
TC wordt bepaald door multipele loci (polygenen) in plaatst van één enkele locus (monogeen).
Het ITLC concludeert met andere woorden dat er waarschijnlijk geen sterke erfelijk monogeen
overervende vorm van TC bestaat. Linkage studies zijn minder krachtig in het detecteren van
een mutant gen dan associatie-studies wanneer frequente genetische varianten een rol spelen
(linkage studie heeft dan minder power en er is dus een veel grotere sample size nodig).(299)
Met 459 families, waarvan de ITCLC er slechts 237 analyseerde (de rest was niet goed in kaart
gebracht) kan slechts een relatief risico van 5 of hoger voor dragers van een genetische variant
(frequentie 15%) worden gedetecteerd. Om een relatief risico van 2 te detecteren (hetgeen
realistisch is in complexe humane ziekten) zijn er tenminste 2000 families nodig om met een
power van 80% een genetische variant te kunnen detecteren. Het kan echter niet worden
uitgesloten dat (gedeelten van) families met TC het gevolg zijn van verschillende, zeldzame,
sterk predisponerende genmutaties, die, vanwege de zeldzaamheid in de populatie, niet worden
gedetecteerd in de relatief kleine linkage-studies. Theoretisch gezien zou het interessant zijn
om linkage-studies te verrichten bij een subgroep van familiaire TC patiënten met urogenitale
afwijkingen (als hypospadie, cryptorchisme, bilateraal TC etc.), aangezien het in deze subgroep
(identieke fenotypen) goed mogelijk is dat zij genen delen die leiden tot urogenitale afwijkingen
en tot TC (Skakkebaek model). Dezelfde benadering is destijds toegepast voor Xq27: In de
eerste studie(105) werden significante resultaten gevonden voor een mogelijk gen in regio Xq27
in een subgroep van familiaire patiënten met TC, deze patiënten hadden allemaal minimaal 1
familielid met dubbelzijdig TC. Daarnaast leek Xq27 tevens geassocieerd met cryptorchisme. (12;105) Het probleem met subgroepanalyses is dat, wanneer je een groot aantal verschillende
subgroepen gaat analyseren, de kans dat er bij toeval één een significant verschil laat zien
aanzienlijk toeneemt. Hiervoor moet statistisch gecorrigeerd worden (multipele-testen-correctie),
wat tot gevolg heeft dat een verschil tussen subgroepen groter moet zijn om significant te zijn
naarmate meer subgroepen worden geanalyseerd. Het is dus belangrijk om vooraf te bepalen
welke subgroepen biologisch gezien relevant zijn om te testen. Welke subgroepen je wilt
analyseren, zal moeten afhangen van de functie van het gen dat je onderzoekt. In het geval
van een analyse van het gehele chromosoom is dat niet mogelijk en zul je dus voor een groot
aantal testen moeten corrigeren wat het onderscheidende vermogen van de analyse niet ten
goede komt.
Recent is het genetische defect in een muizenstam 129.MOLF-Chr19 chromosoom substitutiestam,
die bekend staat om de hoge frequentie (70-80%) van TC ontwikkeling, gerapporteerd.(300) Een
kiemlijn mutatie in het Dead End Gene (DND1) bleek verantwoordelijk voor dit hoge risico op
TC (en ook op semenafwijkingen).(258) Er is erg weinig bekend over het humane homoloog van
DND1 en dit maakt het de moeite waard om een mogelijke associatie tussen DND1 en humaan TC
te bestuderen. In het Universitair Medisch Centrum Groningen (UMCG) wordt momenteel in een
serie familiaire en sporadische TC patiënten en controles het DND1 gen gescreend op mutaties.
Het verzamelen van meer TC families zal genetisch onderzoek naar TC vergemakkelijken. Er
zijn ziekenhuizen die systematisch DNA verzamelen (prospectief) om zo materiaal in handen
te krijgen voor genetische studies. Het deCODE initiatief (www.decode.com) gaat nog een stap
verder en verbindt in de populatie op IJsland de genealogische data met ziekte-status en DNA
merkers. Wetenschappelijk gezien moeten deze projecten zeer sterk worden aangemoedigd,
echter privacy, ethische en wettelijke aspecten moeten hierbij niet uit het oog worden verloren.
Internationale samenwerking zal worden vergemakkelijkt wanneer de verzameling van DNA en
patiëntengegevens volgens (nog te ontwikkelen) internationale standaarden wordt verricht. Het
HapMap project, (een samenwerkingsverband van wetenschappers over de gehele wereld dat
een openbare verzameling gegevens beschikbaar maakt, die behulpzaam zijn bij het vinden van
genen geassocieerd met menselijke ziekten), is een goed voorbeeld (www.hapmap.org).
Niet alleen studiepopulaties zullen zich door samenwerking uitbreiden maar ook de statistische
technieken, de inzichten in kankerbiologie en de hieruit voortvloeiende moleculaire technieken
zullen zich verder ontwikkelen. Recent is het genetisch onderzoek naar TC uitgebreid in de
richting van de rol van het, in het genoom natuurlijk voorkomende, micro RNA (miRNA). Normaal
gesproken functioneert miRNA als een regulator van genetische wegen door de translatieregulatie
te manipuleren. Enkele van deze miRNAs (miRNA-372 and 373) blijken oncogeen te zijn. Deze
miRNAs zijn ook al aangetoond in humane seminomen en non-seminomen, maar niet in het
normaal testisweefsel. Dit wekt de suggestie dat deze miRNAs een nieuw soort oncogenen zijn
die betrokken lijken te zijn bij TC ontwikkeling.(18) De tijd zal leren of miRNAs een rol spelen in
de predispositie voor TC.
Nederlandse samenvatting, conclusies en toekomstperspectieven132 Genetic Predisposition to Testicular Cancer
10
133
Wanneer in de toekomst klinisch relevante TC predisponerende mutaties worden geïdentificeerd,
dan zal waarschijnlijk een groot aantal TC patiënten (en families) zich hierop laten testen,
zoals thans ook het geval is voor families met bekende erfelijke tumorsyndromen (bijvoorbeeld
erfelijke borstkanker en erfelijke darmkanker). Alhoewel veel studies de psychosociale aspecten
van genetische testen (en naar de benodigdheden / voorzorgsmaatregelen hiervoor) bij de al
bekende tumorsyndromen hebben onderzocht, is hierover erg weinig bekend bij TC. Onderzoek
naar vroege opsporingstechnieken bij mannen met een hoog risico op TC blijft nodig. Zoals bij elk
ander nieuw genetisch testprogramma, moet het genetisch testen eerst in een onderzoekssetting
plaatsvinden zodat medische en psychologische kwesties goed bestudeerd kunnen worden.
Genetic Predisposition to Testicular Cancer 135
Reference List
Reference List136 Genetic Predisposition to Testicular Cancer 137
Reference List
(1) Cancer in the Netherlands,trends,prognoses and implications for call of health. 2004
www.kwfkankerbestrijding.nl/content/documents/SCK_rapport_Kanker_in_Nederland.pdf
(2) Albers P, Albrecht W, Algaba F, Bokemeyer C, Cohn-Cedermark G, Horwich A et al.
Guidelines on testicular cancer. Eur Urol 2005; 48(6):885-894.
(3) Schmoll HJ, Souchon R, Krege S, Albers P, Beyer J, Kollmannsberger C et al. European
consensus on diagnosis and treatment of germ cell cancer: a report of the European
Germ Cell Cancer Consensus Group (EGCCCG). Ann Oncol 2004; 15(9):1377-1399.
(4) International Germ Cell Consensus Classification: a prognostic factor-based staging
system for metastatic germ cell cancers. International Germ Cell Cancer Collaborative
Group. J Clin Oncol 1997; 15(2):594-603.
(5) Abouassaly R, Klein EA, Raghavan D. Complications of surgery and chemotherapy for
testicular cancer. Urol Oncol 2005; 23(6):447-455.
(6) Meinardi MT, Gietema JA, van der Graaf WT, van Veldhuisen DJ, Runne MA, Sluiter WJ et
al. Cardiovascular morbidity in long-term survivors of metastatic testicular cancer. J Clin
Oncol 2000; 18(8):1725-1732.
(7) Nuver J, Smit AJ, Wolffenbuttel BH, Sluiter WJ, Hoekstra HJ, Sleijfer DT et al.
The metabolic syndrome and disturbances in hormone levels in long-term survivors of
disseminated testicular cancer. J Clin Oncol 2005; 23(16):3718-3725.
(8) Nuver J, Smit AJ, Sleijfer DT, van Gessel AI, van Roon AM, van der Meer J et al. Left
ventricular and cardiac autonomic function in survivors of testicular cancer. Eur J Clin
Invest 2005; 35(2):99-103.
(9) Fleer J. Quality of life of testicular cancer survivors.
Thesis: http://irs.ub.rug.nl/ppn/291569412. University of Groningen, The Netherlands 2006.
(10) Hoei-Hansen CE, Rajpert-De Meyts E, Daugaard G, Skakkebaek NE. Carcinoma in situ
testis, the progenitor of testicular germ cell tumours: a clinical review. Ann Oncol 2005;
16(6):863-868.
(11) Oosterhuis JW, Looijenga LH. Testicular germ-cell tumours in a broader perspective. Nat
Rev Cancer 2005; 5(3):210-222.
(12) Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an
increasingly common developmental disorder with environmental aspects. Hum Reprod
2001; 16(5):972-978.
(13) von Eyben FE. Chromosomes, genes, and development of testicular germ cell tumors.
Cancer Genet Cytogenet 2004; 151(2):93-138.
(14) Reuter VE. Origins and molecular biology of testicular germ cell tumors. Mod Pathol
2005; 18 Suppl 2:S51-S60.
(15) Dearnaley D, Huddart R, Horwich A. Regular review: Managing testicular cancer. BMJ
2001; 322(7302):1583-1588.
Reference List138 Genetic Predisposition to Testicular Cancer 139
(16) Korkola JE, Houldsworth J, Chadalavada RS, Olshen AB, Dobrzynski D, Reuter VE et al.
Down-regulation of stem cell genes, including those in a 200-kb gene cluster at
12p13.31, is associated with in vivo differentiation of human male germ cell tumors.
Cancer Res 2006; 66(2):820-827.
(17) McIntyre A, Summersgill B, Grygalewicz B, Gillis AJ, Stoop J, van Gurp RJ et al.
Amplification and overexpression of the KIT gene is associated with progression in the
seminoma subtype of testicular germ cell tumors of adolescents and adults. Cancer Res
2005; 65(18):8085-8089.
(18) Voorhoeve PM, le Sage C, Schrier M, Gillis AJ, Stoop H, Nagel R et al. A genetic screen
implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Cell
2006; 124(6):1169-1181.
(19) van Basten JP, Schraffordt Koops H, Sleijfer DT, Pras E, van Driel MF, Hoekstra HJ.
Current concepts about testicular cancer. Eur J Surg Oncol 1997; 23(4):354-360.
(20) Hain SF, O’Doherty MJ, Timothy AR, Leslie MD, Partridge SE, Huddart RA.
Fluorodeoxyglucose PET in the initial staging of germ cell tumours. Eur J Nucl Med
2000; 27(5):590-594.
(21) Gels ME, Hoekstra HJ, Sleijfer DT, Marrink J, De Bruijn HW, Molenaar WM et al. Detection
of recurrence in patients with clinical stage I nonseminomatous testicular germ cell
tumors and consequences for further follow-up: a single-center 10-year experience. J Clin
Oncol 1995; 13(5):1188-1194.
(22) Oliver RT, Mason MD, Mead GM, von der Maase H, Rustin GJ, Joffe JK et al. Radiotherapy
versus single-dose carboplatin in adjuvant treatment of stage I seminoma:
a randomised trial. Lancet 2005; 366(9482):293-300.
(23) de Wit R, Stoter G. Treatment of metastatic testicular carcinoma according to prognosis;
new development. Ned Tijdschr Geneeskd 2001; 145(25):1194-1199.
(24) de Wit R, Roberts JT, Wilkinson PM, de Mulder PH, Mead GM, Fossa SD et al.
Equivalence of three or four cycles of bleomycin, etoposide, and cisplatin chemotherapy
and of a 3- or 5-day schedule in good-prognosis germ cell cancer: a randomized study
of the European Organization for Research and Treatment of Cancer Genitourinary Tract
Cancer Cooperative Group and the Medical Research Council. J Clin Oncol 2001; 19(6):
1629-1640.
(25) De Santis M, Becherer A, Bokemeyer C, Stoiber F, Oechsle K, Sellner F et al. 2-18
fluoro-deoxy-D-glucose positron emission tomography is a reliable predictor for viable
tumor in postchemotherapy seminoma: an update of the prospective multicentric
SEMPET trial. J Clin Oncol 2004; 22(6):1034-1039.
(26) Lutke Holzik MF, Hoekstra HJ, Mulder NH, Suurmeijer AJ, Sleijfer DT, Gietema JA.
Non-germ cell malignancy in residual or recurrent mass after chemotherapy for
nonseminomatous testicular germ cell tumor. Ann Surg Oncol 2003; 10(2):131-135.
(27) Fizazi K, Tjulandin S, Salvioni R, Germa-Lluch JR, Bouzy J, Ragan D et al. Viable
malignant cells after primary chemotherapy for disseminated nonseminomatous
germ cell tumors: prognostic factors and role of postsurgery chemotherapy-results from
an international study group. J Clin Oncol 2001; 19(10):2647-2657.
(28) Bosl GJ, Motzer RJ. Testicular germ-cell cancer. N Engl J Med 1997; 337(4):242-253.
(29) Giwercman A, Muller J, Skakkebaek NE. Prevalence of carcinoma in situ and other
histopathological abnormalities in testes from 399 men who died suddenly and
unexpectedly. J Urol 1991; 145(1):77-80.
(30) Jacobsen GK, Henriksen OB, von der Maase H. Carcinoma in situ of testicular tissue
adjacent to malignant germ-cell tumors: a study of 105 cases. Cancer 1981; 47(11):
2660-2662.
(31) Boyle P. Testicular cancer: the challenge for cancer control. Lancet Oncol 2004; 5(1):56-61.
(32) Skakkebaek NE. Possible carcinoma-in-situ of the testis. Lancet 1972; 2(7776):516-517.
(33) Jones RH, Vasey PA. Part I: testicular cancer--management of early disease. Lancet
Oncol 2003; 4(12):730-737.
(34) Aetiology of testicular cancer: association with congenital abnormalities, age at puberty,
infertility, and exercise. United Kingdom Testicular Cancer Study Group. BMJ 1994;
308(6941):1393-1399.
(35) Forman D, Oliver RT, Brett AR, Marsh SG, Moses JH, Bodmer JG et al. Familial testicular
cancer: a report of the UK family register, estimation of risk and an HLA class 1 sib-pair
analysis. Br J Cancer 1992; 65(2):255-262.
(36) Osterlind A, Berthelsen JG, Abildgaard N, Hansen SO, Hjalgrim H, Johansen B et al. Risk
of bilateral testicular germ cell cancer in Denmark: 1960-1984. J Natl Cancer Inst 1991;
83(19):1391-1395.
(37) Wanderas EH, Fossa SD, Tretli S. Risk of a second germ cell cancer after treatment of a
primary germ cell cancer in 2201 Norwegian male patients. Eur J Cancer 1997; 33(2):
244-252.
(38) Brown LM, Pottern LM, Hoover RN. Testicular cancer in young men: the search for
causes of the epidemic increase in the United States. J Epidemiol Community Health
1987; 41(4):349-354.
(39) Swerdlow AJ, Huttly SR, Smith PG. Testicular cancer and antecedent diseases. Br J
Cancer 1987; 55(1):97-103.
(40) Swerdlow AJ, Huttly SR, Smith PG. Prenatal and familial associations of testicular
cancer. Br J Cancer 1987; 55(5):571-577.
(41) Swerdlow AJ, Higgins CD, Pike MC. Risk of testicular cancer in cohort of boys with
cryptorchidism. BMJ 1997; 314(7093):1507-1511.
(42) Hasle H, Mellemgaard A, Nielsen J, Hansen J. Cancer incidence in men with Klinefelter
syndrome. Br J Cancer 1995; 71(2):416-420.
(43) Verp MS, Simpson JL. Abnormal sexual differentiation and neoplasia. Cancer Genet
Cytogenet 1987; 25(2):191-218.
(44) Fossa SD, Aass N, Heilo A, Daugaard G, Skakkebaek E, Stenwig AE et al. Testicular
carcinoma in situ in patients with extragonadal germ-cell tumours: the clinical role of
pretreatment biopsy. Ann Oncol 2003; 14(9):1412-1418.
Reference List140 Genetic Predisposition to Testicular Cancer 141
(45) Harland SJ, Cook PA, Fossa SD, Horwich A, Mead GM, Parkinson MC et al. Intratubular
germ cell neoplasia of the contralateral testis in testicular cancer: defining a high risk
group. J Urol 1998; 160(4):1353-1357.
(46) Jacobsen R, Bostofte E, Engholm G, Hansen J, Olsen JH, Skakkebaek NE et al. Risk
of testicular cancer in men with abnormal semen characteristics: cohort study. BMJ
2000; 321(7264):789-792.
(47) Moller H, Skakkebaek NE. Risk of testicular cancer in subfertile men: case-control study.
BMJ 1999; 318(7183):559-562.
(48) Tollerud DJ, Blattner WA, Fraser MC, Brown LM, Pottern L, Shapiro E et al. Familial
testicular cancer and urogenital developmental anomalies. Cancer 1985; 55(8):
1849-1854.
(49) Dieckmann KP, Pichlmeier U. The prevalence of familial testicular cancer: an analysis of
two patient populations and a review of the literature. Cancer 1997; 80(10):1954-1960.
(50) Dong C, Lonnstedt I, Hemminki K. Familial testicular cancer and second primary cancers
in testicular cancer patients by histological type. Eur J Cancer 2001; 37(15):1878-1885.
(51) Heimdal K, Olsson H, Tretli S, Flodgren P, Borresen AL, Fossa SD. Familial testicular
cancer in Norway and southern Sweden. Br J Cancer 1996; 73(7):964-969.
(52) Kruse C. Bilateral and familial occurrence of germ cell tumors of the testis. Urologe A
1987; 26(2):61-62.
(53) Ondrus D, Kuba D, Chrenova S, Matoska J. Familial testicular cancer and developmental
anomalies. Neoplasma 1997; 44(1):59-61.
(54) Patel SR, Kvols LK, Richardson RL. Familial testicular cancer: report of six cases and
review of the literature. Mayo Clin Proc 1990; 65(6):804-808.
(55) Polednak AP. Familial testicular cancer in a population-based cancer registry. Urol Int
1996; 56(4):238-240.
(56) Sonneveld DJ, Sleijfer DT, Schraffordt Koops H, Sijmons RH, van der Graaf WT, Sluiter
WJ et al. Familial testicular cancer in a single-centre population. Eur J Cancer 1999;
35(9):1368-1373.
(57) Westergaard T, Olsen JH, Frisch M, Kroman N, Nielsen JW, Melbye M. Cancer risk in
fathers and brothers of testicular cancer patients in Denmark. A population-based study.
Int J Cancer 1996; 66(5):627-631.
(58) Goss PE, Bulbul MA. Familial testicular cancer in five members of a cancer-prone
kindred. Cancer 1990; 66(9):2044-2046.
(59) Forman D, Gallagher R, Moller H, Swerdlow TJ. Aetiology and epidemiology of testicular
cancer: report of consensus group. Prog Clin Biol Res 1990; 357:245-253.
(60) Khoury MJ, Beaty TH, Liang KY. Can familial aggregation of disease be explained by
familial aggregation of environmental risk factors? Am J Epidemiol 1988; 127(3):674-683.
(61) Heimdal K, Fossa SD. Genetic factors in malignant germ-cell tumors. World J Urol 1994;
12(4):178-181.
(62) Sonneveld DJ, Schaapveld M, Sleijfer DT, te Meerman GJ, van der Graaf WT, Sijmons
RH et al. Geographic clustering of testicular cancer incidence in the northern part of The
Netherlands. Br J Cancer 1999; 81(7):1262-1267.
(63) Senturia YD. The epidemiology of testicular cancer. Br J Urol 1987; 60(4):285-291.
(64) Parkin DM, Iscovich J. Risk of cancer in migrants and their descendants in Israel: II.
Carcinomas and germ-cell tumours. Int J Cancer 1997; 70(6):654-660.
(65) Geddes M, Buiatti E. Review of studies of cancer in Italian migrants. IARC Sci Publ 1993;
123:24-30.
(66) Osterlind A, Berthelsen JG, Abildgaard N, Hansen SO, Jensen H, Johansen B et al.
Incidence of bilateral testicular germ cell cancer in Denmark, 1960-84: preliminary
findings. Int J Androl 1987; 10(1):203-208.
(67) Sonneveld DJ, Schraffordt Koops H, Sleijfer DT, Hoekstra HJ. Bilateral testicular germ
cell tumours in patients with initial stage I disease: prevalence and prognosis--a single
centre’s 30 years’ experience. Eur J Cancer 1998; 34(9):1363-1367.
(68) Tian Q, Frierson HF, Jr., Krystal GW, Moskaluk CA. Activating c-kit gene mutations in
human germ cell tumors. Am J Pathol 1999; 154(6):1643-1647.
(69) Przygodzki RM, Hubbs AE, Zhao FQ, O’Leary TJ. Primary mediastinal seminomas:
evidence of single and multiple KIT mutations. Lab Invest 2002; 82(10):1369-1375.
(70) Looijenga LH, de Leeuw H, van Oorschot M, van Gurp RJ, Stoop H, Gillis AJ et al. Stem
cell factor receptor (c-KIT) codon 816 mutations predict development of bilateral
testicular germ-cell tumors. Cancer Res 2003; 63(22):7674-7678.
(71) Rapley EA, Hockley S, Warren W, Johnson L, Huddart R, Crockford G et al. Somatic
mutations of KIT in familial testicular germ cell tumours. Br J Cancer 2004; 90(12):
2397-2401.
(72) Kroman N, Frisch M, Olsen JH, Westergaard T, Melbye M. Oestrogen-related cancer risk
in mothers of testicular-cancer patients. Int J Cancer 1996; 66(4):438-440.
(73) Heimdal K, Olsson H, Tretli S, Flodgren P, Borresen AL, Fossa SD. Risk of cancer in
relatives of testicular cancer patients. Br J Cancer 1996; 73(7):970-973.
(74) Nichols CR, Heerema NA, Palmer C, Loehrer PJ, Sr., Williams SD, Einhorn LH. Klinefelter’s
syndrome associated with mediastinal germ cell neoplasms. J Clin Oncol 1987; 5(8):
1290-1294.
(75) Houldsworth J. Genetics and biology of male germ cell tumors. Chest Surg Clin N Am
2002; 12(4):629-643.
(76) Lutke Holzik MF, Sijmons RH, Sleijfer DT, Sonneveld DJ, Hoekstra-Weebers JE, Echten-
Arends J et al. Syndromic aspects of testicular carcinoma. Cancer 2003; 97(4):984-992.
(77) Levin HS. Tumors of the testis in intersex syndromes. Urol Clin North Am 2000; 27(3):
543-51.
(78) Hasle H, Clemmensen IH, Mikkelsen M. Risks of leukaemia and solid tumours in
individuals with Down’s syndrome. Lancet 2000; 355(9199):165-169.
Reference List142 Genetic Predisposition to Testicular Cancer 143
(79) Satge D, Sasco AJ, Cure H, Leduc B, Sommelet D, Vekemans MJ. An excess of testicular
germ cell tumors in Down’s syndrome: three case reports and a review of the literature.
Cancer 1997; 80(5):929-935.
(80) Gallagher RP, Huchcroft S, Phillips N, Hill GB, Coldman AJ, Coppin C et al. Physical
activity, medical history, and risk of testicular cancer (Alberta and British Columbia,
Canada). Cancer Causes Control 1995; 6(5):398-406.
(81) Pettersson A, Richiardi L, Nordenskjold A, Kaijser M, Akre O. Age at surgery for
undescended testis and risk of testicular cancer. N Engl J Med 2007; 356(18):1835-1841.
(82) Braun MM, Caporaso NE, Page WF, Hoover RN. Prevalence of a history of testicular
cancer in a cohort of elderly twins. Acta Genet Med Gemellol (Roma ) 1995; 44(3-4):
189-192.
(83) Swerdlow AJ, De Stavola BL, Swanwick MA, Maconochie NE. Risks of breast and
testicular cancers in young adult twins in England and Wales: evidence on prenatal and
genetic aetiology. Lancet 1997; 350(9093):1723-1728.
(84) Adami HO, Bergstrom R, Mohner M, Zatonski W, Storm H, Ekbom A et al. Testicular
cancer in nine northern European countries. Int J Cancer 1994; 59(1):33-38.
(85) Bergstrom R, Adami HO, Mohner M, Zatonski W, Storm H, Ekbom A et al. Increase in
testicular cancer incidence in six European countries: a birth cohort phenomenon. J Natl
Cancer Inst 1996; 88(11):727-733.
(86) Giwercman A, Carlsen E, Keiding N, Skakkebaek NE. Evidence for increasing incidence of
abnormalities of the human testis: a review. Environ Health Perspect 1993; 101 Suppl 2:
65-71.
(87) Sharpe RM. The ‘oestrogen hypothesis’- where do we stand now? Int J Androl 2003; 26:
2-15.
(88) Dieckmann KP, Endsin G, Pichlmeier U. How valid is the prenatal estrogen excess
hypothesis of testicular germ cell cancer? A case control study on hormone-related
factors. Eur Urol 2001; 40:677-683.
(89) Nicholson PW, Harland SJ. Inheritance and testicular cancer. Br J Cancer 1995; 71(2):
421-426.
(90) Heimdal K, Olsson H, Tretli S, Fossa SD, Borresen AL, Bishop DT. A segregation analysis
of testicular cancer based on Norwegian and Swedish families. Br J Cancer 1997; 75(7):
1084-1087.
(91) Knudson AG, Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad
Sci U S A 1971; 68(4):820-823.
(92) Moller H. Trends in sex-ratio, testicular cancer and male reproductive hazards: are they
connected? APMIS 1998; 106(1):232-238.
(93) Petersen PM, Skakkebaek NE, Giwercman A. Gonadal function in men with testicular
cancer: biological and clinical aspects. APMIS 1998; 106(1):24-34.
(94) Sonneveld DJ, Hoekstra HJ, van der Graaf WT, Sluiter WJ, Mulder NH, Willemse PH et al.
Improved long term survival of patients with metastatic nonseminomatous testicular
germ cell carcinoma in relation to prognostic classification systems during the cisplatin
era. Cancer 2001; 91(7):1304-1315.
(95) Lutke Holzik MF, Sonneveld DJ, Hoekstra HJ, te Meerman GJ, Sleijfer DT, Schaapveld M.
Do the eastern and northern parts of The Netherlands differ in testicular cancer? Urology
2001; 58(4):636-637.
(96) Sonneveld DJ, Lutke Holzik MF, Nolte IM, Sleijfer DT, van der Graaf WT, Bruinenberg M
et al. Testicular carcinoma and HLA Class II genes. Cancer 2002; 95(9):1857-1863.
(97) Bateman AC, Howell WM. Human leukocyte antigens and cancer: is it in our genes? J
Pathol 1999; 188(3):231-236.
(98) de Jong MM, Nolte IM, de Vries EG, Schaapveld M, Kleibeuker JH, Oosterom E et al.
The HLA class III subregion is responsible for an increased breast cancer risk. Hum Mol
Genet 2003; 12(18):2311-2319.
(99) Birkeland SA, Storm HH, Lamm LU, Barlow L, Blohme I, Forsberg B et al. Cancer risk
after renal transplantation in the Nordic countries, 1964- 1986. Int J Cancer 1995; 60(2):
183-189.
(100) Spano JP, Atlan D, Breau JL, Farge D. AIDS and non-AIDS-related malignancies: a
new vexing challenge in HIV-positive patients. Part II. Cervical and anal squamous
epithelial lesions, lung cancer, testicular germ cell cancers, and skin cancers. Eur J Intern
Med 2002; 13(4):227-232.
(101) Corzo D, Salazar M, Granja CB, Yunis EJ. Advances in HLA genetics. Exp Clin
Immunogenet 1995; 12(3):156-170.
(102) Özdemir E, Kakehi Y, Mishina M, Ogawa O, Okada Y, Özdemir D et al. High-resolution
HLA-DRB1 and DQB1 genotyping in Japanese patients with testicular germ cell
carcinoma. Br J Cancer 1997; 76(10):1348-1352.
(103) Rapley EA, Crockford GP, Easton D, Stratton MR, Bishop DT. Localisation of
susceptibility genes for familial testicular germ cell tumour. APMIS 2003; 111:128-135.
(104) Rapley EA, Crockford GP, Easton D, Stratton MR, Bishop DT. The Genetics of Testicular
Germ Cell Tumours in proceedings of: Germ Cell Tumours V. Harden, P, Joffe,
JK and Jones, W G Eds 2002;1-22.
(105) Rapley EA, Crockford GP, Teare D, Biggs P, Seal S, Barfoot R et al. Localization to Xq27
of a susceptibility gene for testicular germ-cell tumours. Nat Genet 2000; 24(2):197-200.
(106) DeWolf WC, Lange PH, Einarson ME, Yunis EJ. HLA and testicular cancer. Nature 1979;
277(5693):216-217.
(107) Carr BI, Bach FA. Possible association between HLA-AW24 and metastatic testicular
germ-cell tumours. Lancet 1979; 1(8130):1346-1347.
(108) Majsky A, Abrahamova J, Korinkova P, Bek V. HLA system and testicular germinative
tumours. Oncology 1979; 36(5):228-231.
(109) Pollack MS, Vugrin D, Hennessy W, Herr HW, Dupont B, Whitmore WF, Jr. HLA antigens
in patients with germ cell cancers of the testis. Cancer Res 1982; 42(6):2470-2473.
Reference List144 Genetic Predisposition to Testicular Cancer 145
(110) Oliver RT, Stephenson CA, Parkinson MC, Forman D, Atkinson A, Bodmer J et al. Germ
cell tumours of the testicle as a model of MHC influence on human malignancy. Lancet
1986; 1(8496):1506.
(111) Aiginger P, Kuzmits R, Kratzig C, Schwarz HP, Zielinski CC, Kuhbock J et al. HLA antigens
and germ-cell tumours. Lancet 1987; 1(8527):276-277.
(112) Dieckmann KP, von Keyserlingk HJ. HLA typing in familial testicular cancer. Eur Urol
1989; 16(5):361-365.
(113) Ryberg D, Heimdal K, Fossa SD, Borresen AL, Haugen A. Rare Ha-ras1 alleles and
predisposition to testicular cancer. Int J Cancer 1993; 53(6):938-940.
(114) Dieckmann KP, Klan R, Bunte S. HLA antigens, Lewis antigens, and blood groups in
patients with testicular germ-cell tumors. Oncology 1993; 50(4):252-258.
(115) Heimdal KR, Lothe RA, Fossa SD, Borresen AL. Association studies of a polymorphism in
the Wilms’ tumor 1 locus in Norwegian patients with testicular cancer. Int J Cancer 1994;
58(4):523-526.
(116) Olie RA, Looijenga LH, Boerrigter L, Top B, Rodenhuis S, Langeveld A et al. N- and
KRAS mutations in primary testicular germ cell tumors: incidence and possible biological
implications. Genes Chromosomes Cancer 1995; 12(2):110-116.
(117) Heimdal K, Andersen TI, Skrede M, Fossa SD, Berg K, Borresen AL. Association studies
of estrogen receptor polymorphisms in a Norwegian testicular cancer population. Cancer
Epidemiol Biomarkers Prev 1995; 4(2):123-126.
(118) Harries LW, Stubbins MJ, Forman D, Howard GC, Wolf CR. Identification of genetic
polymorphisms at the glutathione S- transferase Pi locus and association with
susceptibility to bladder, testicular and prostate cancer. Carcinogenesis 1997; 18(4):
641-644.
(119) Sakashita S, Koyanagi T, Tsuji I, Arikado K, Matsuno T. Congenital anomalies in children
with testicular germ cell tumor. J Urol 1980; 124(6):889-891.
(120) Dexeus FH, Logothetis CJ, Chong C, Sella A, Ogden S. Genetic abnormalities in men with
germ cell tumors. J Urol 1988; 140(1):80-84.
(121) Sijmons R H., Burger G.T N. Familial cancer database: a clinical aide-mémoire. Familial
Cancer 2001; 1(1):51-55.
(122) Rescorla FJ, Breitfeld PP. Pediatric germ cell tumors. Curr Probl Cancer 1999; 23(6):
257-303.
(123) Savage MO, Lowe DG. Gonadal neoplasia and abnormal sexual differentiation.
Clin Endocrinol (Oxf) 1990; 32(4):519-533.
(124) Weiss GR, Garnick MB. Testicular cancer in a Russell-Silver dwarf. J Urol 1981; 126(6):
836-837.
(125) Boisen KA, Main KM, Rajpert-De Meyts E, Skakkebaek NE. Are male reproductive
disorders a common entity? The testicular dysgenesis syndrome. Ann N Y Acad Sci 2001;
948:90-99.
(126) Cortes D, Visfeldt J, Moller H, Thorup J. Testicular neoplasia in cryptorchid boys at
primary surgery: case series. BMJ 1999; 319(7214):888-889.
(127) Depue RH, Pike MC, Henderson BE. Cryptorchidism and testicular cancer. J Natl Cancer
Inst 1986; 77(3):830-833.
(128) Pike MC, Chilvers C, Peckham MJ. Effect of age at orchidopexy on risk of testicular
cancer. Lancet 1986; 1(8492):1246-1248.
(129) Pottern LM, Brown LM, Hoover RN, Javadpour N, O’Connell KJ, Stutzman RE et al.
Testicular cancer risk among young men: role of cryptorchidism and inguinal hernia. J
Natl Cancer Inst 1985; 74(2):377-381.
(130) Fearon ER. Human cancer syndromes: clues to the origin and nature of cancer. Science
1997; 278(5340):1043-1050.
(131) Cendron M, Wein AJ, Schwartz SS, Murtagh F, Livolsi VA, Tomaszewski JE. Germ cell
tumor of testis in a patient with von Hippel-Lindau disease. Urology 1991; 37(1):69-71.
(132) Hartley AL, Birch JM, Kelsey AM, Marsden HB, Harris M, Teare MD. Are germ cell tumors
part of the Li-Fraumeni cancer family syndrome? Cancer Genet Cytogenet 1989; 42(2):
221-226.
(133) de Jong B, van Echten J, Looijenga LH, Geurts vK, Oosterhuis JW. Cytogenetics of the
progression of adult testicular germ cell tumors. Cancer Genet Cytogenet 1997; 95(1):
88-95.
(134) Oosterhuis JW, Looijenga LH, van Echten J, de Jong B. Chromosomal constitution and
developmental potential of human germ cell tumors and teratomas. Cancer Genet
Cytogenet 1997; 95(1):96-102.
(135) Mostert M, Rosenberg C, Stoop H, Schuyer M, Timmer A, Oosterhuis W et al.
Comparative genomic and in situ hybridization of germ cell tumors of the infantile
testis. Lab Invest 2000; 80(7):1055-1064.
(136) Summersgill B, Osin P, Lu YJ, Huddart R, Shipley J. Chromosomal imbalances associated
with carcinoma in situ and associated testicular germ cell tumours of adolescents and
adults. Br J Cancer 2001; 85(2):213-220.
(137) Qiao D, Zeeman AM, Deng W, Looijenga LH, Lin H. Molecular characterization of hiwi,
a human member of the piwi gene family whose overexpression is correlated to
seminomas. Oncogene 2002; 21(25):3988-3999.
(138) Kratzik C, Aiginger P, Kuzmits R, Spona J, Kirnbauer M, Seiser A et al. HLA-antigen
distribution in seminoma, HCG-positive seminoma and non- seminomatous tumours of
the testis. Urol Res 1989; 17(6):377-380.
(139) Kume H, Tachikawa T, Teramoto S, Isurugi K, Kitamura T. Bilateral testicular tumour in
neurofibromatosis type 1. Lancet 2001; 357(9253):395-396.
(140) Foster K, Osborne RJ, Huddart RA, Affara NA, Ferguson-Smith MA, Maher ER. Molecular
genetic analysis of the von Hippel-Lindau disease (VHL) tumour suppressor gene in
gonadal tumours. Eur J Cancer 1995; 31A(13-14):2392-2395.
Reference List146 Genetic Predisposition to Testicular Cancer 147
(141) Ye DW, Zheng J, Qian SX, Ma Y, Zheng X, Li D et al. p53 gene mutations in Chinese
human testicular seminoma. J Urol 1993; 150(3):884-886.
(142) Bartkova J, Falck J, Rajpert-De Meyts E, Skakkebaek NE, Lukas J, Bartek J. Chk2 tumour
suppressor protein in human spermatogenesis and testicular germ-cell tumours.
Oncogene 2001; 20(41):5897-5902.
(143) Heidenreich A, Gaddipati JP, Moul JW, Srivastava S. Molecular analysis of P16(Ink4)/
CDKN2 and P15(INK4B)/MTS2 genes in primary human testicular germ cell tumors. J Urol
1998; 159(5):1725-1730.
(144) Drescher B, Lauke H, Hartmann M, Davidoff MS, Zumkeller W. Immunohistochemical
pattern of insulin-like growth factor (IGF) I, IGF II and IGF binding proteins 1 to 6 in
carcinoma in situ of the testis. Mol Pathol 1997; 50(6):298-303.
(145) Drescher B, Zumkeller W, Lauke H, Hartmann M, Davidoff MS. Insulin-like growth
factor-binding protein (IGFBP)-5 in human testicular tubules. Adv Exp Med Biol 1997;
424:153-154.
(146) Ghebranious N, Donehower LA. Mouse models in tumor suppression. Oncogene 1998;
17(25):3385-3400.
(147) Skotheim RI, Monni O, Mousses S, Fossa SD, Kallioniemi OP, Lothe RA et al.
New insights into testicular germ cell tumorigenesis from gene expression profiling.
Cancer Res 2002; 62(8):2359-2364.
(148) Jiang LI, Nadeau JH. 129/Sv mice--a model system for studying germ cell biology and
testicular cancer. Mamm Genome 2001; 12(2):89-94.
(149) Kash K, Dabney M, Holland JC, Osborne MP, Miller D. Familial cancer and genetics:
Psychosocial and ethical aspects. In: L.Baider, CL Cooper, A.Kaplan De-Nour (Eds.).
Cancer and the Family. Chichester, Engeland: John Wiley., 1997: 389-403.
(150) Melzer D, Zimmern R. Genetics and medicalisation. BMJ 2002; 324(7342):863-864.
(151) Press NA, Yasui Y, Reynolds S, Durfy SJ, Burke W. Women’s interest in genetic testing
for breast cancer susceptibility may be based on unrealistic expectations. Am J Med
Genet 2001; 99(2):99-110.
(152) Sakai N, Yamada T, Asao T, Baba M, Yoshida M, Murayama T. Bilateral testicular tumors
in androgen insensitivity syndrome. Int J Urol 2000; 7(10):390-392.
(153) Collins GM, Kim DU, Logrono R, Rickert RR, Zablow A, Breen JL. Pure seminoma arising
in androgen insensitivity syndrome (testicular feminization syndrome): a case report and
review of the literature. Mod Pathol 1993; 6(1):89-93.
(154) Kocak M, Yalvac S, Pata O, Turan H, Haberal A. A seminoma case which occurred in a
patient with familial testicular feminization syndrome. Acta Obstet Gynecol Scand 2000;
79(10):890-891.
(155) Muller J, Skakkebaek NE. Testicular carcinoma in situ in children with the androgen
insensitivity (testicular feminisation) syndrome. Br Med J (Clin Res Ed) 1984; 288(6428):
1419-1420.
(156) Cassio A, Cacciari E, D’Errico A, Balsamo A, Grigioni FW, Pascucci MG et al. Incidence of
intratubular germ cell neoplasia in androgen insensitivity syndrome. Acta Endocrinol
(Copenh) 1990; 123(4):416-422.
(157) Rutgers JL, Scully RE. The androgen insensitivity syndrome (testicular feminization):
a clinicopathologic study of 43 cases. Int J Gynecol Pathol 1991; 10(2):126-144.
(158) Kaisary AV, Williams G. A case of bilateral intra-abdominal testes and seminomas in a
previously unrecognised hermaphrodite. Br J Urol 1982; 54(3):323.
(159) Rutgers JL, Scully RE. Pathology of the testis in intersex syndromes. Semin Diagn Pathol
1987; 4(4):275-291.
(160) Hellwinkel OJ, Bassler J, Hiort O. Transcription of androgen receptor and
5alpha-reductase II in genital fibroblasts from patients with androgen insensitivity
syndrome. J Steroid Biochem Mol Biol 2000; 75(4-5):213-218.
(161) Adesokan A, Adegboyega PA, Cowan DF, Kocurek J, Neal DE, Jr. Testicular “tumor” of
the adrenogenital syndrome: a case report of an unusual association with myelolipoma
and seminoma in cryptorchidism. Cancer 1997; 80(11):2120-2127.
(162) Rutgers JL, Young RH, Scully RE. The testicular “tumor” of the adrenogenital syndrome.
A report of six cases and review of the literature on testicular masses in patients with
adrenocortical disorders. Am J Surg Pathol 1988; 12(7):503-513.
(163) Phillips HA, Howard GC. Testicular seminoma in a patient with ataxia-telangiectasia.
Clin Oncol (R Coll Radiol ) 1999; 11(1):63-64.
(164) Parra G, Seery W, Buchbinder M, Cole AT. Congenital total hemihypertrophy and
carcinoma of undescended testicle: a case report. J Urol 1977; 118(2):343-344.
(165) Dieckmann KP, Rube C, Henke RP. Association of Down’s syndrome and testicular
cancer. J Urol 1997; 157(5):1701-1704.
(166) Yang Q, Rasmussen SA, Friedman JM. Mortality associated with Down’s syndrome in the
USA from 1983 to 1997: a population-based study. Lancet 2002; 359(9311):1019-1025.
(167) Jacobsen GK, Henriques UV. A fetal testis with intratubular germ cell neoplasia (ITGCN).
Mod Pathol 1992; 5(5):547-549.
(168) Satge D, Jacobsen GK, Cessot F, Raffi F, Vekemans M. A fetus with Down syndrome and
intratubular germ cell neoplasia. Pediatr Pathol Lab Med 1996; 16(1):107-112.
(169) Martin MM, Wu SM, Martin AL, Rennert OM, Chan WY. Testicular seminoma in a patient
with a constitutively activating mutation of the luteinizing hormone/chorionic
gonadotropin receptor. Eur J Endocrinol 1998; 139(1):101-106.
(170) Avril MF, Chompret A, Verne-Fourment L, Terrier-Lacombe MJ, Spatz A, Fizazi K et al.
Association between germ cell tumours, large numbers of naevi, atypical naevi and
melanoma. Melanoma Res 2001; 11(2):117-122.
(171) Hemminki A. The molecular basis and clinical aspects of Peutz-Jeghers syndrome.
Cell Mol Life Sci 1999; 55(5):735-750.
(172) Raghavan D, Zalcberg JR, Grygiel JJ, Teriana N, Cox KM, McCarthy W et al. Multiple
atypical nevi: a cutaneous marker of germ cell tumors. J Clin Oncol 1994; 12(11):2284-2287.
Reference List148 Genetic Predisposition to Testicular Cancer 149
(173) Sigg C, Pelloni F. Dysplastic nevi and germ cell tumors of the testis--a possible
further tumor in the spectrum of associated malignancies in dysplastic nevus syndrome.
Dermatologica 1988; 176(3):109-110.
(174) Flechon A, Droz JP. Hereditary persistence of alpha-fetoprotein in testis disease. J Urol
2001; 165(6 Pt 1):2004.
(175) Schefer H, Mattmann S, Joss RA. Hereditary persistence of alpha-fetoprotein. Case report
and review of the literature. Ann Oncol 1998; 9(6):667-672.
(176) Staples J. Alphafetoprotein, cancer, and benign conditions. Lancet 1986; 2(8518):1277.
(177) Cochran PK, Chauvenet AR, Hart PS, de Graaf SS, Cushing B, Kroovand L et al.
Hereditary persistence of alpha-fetoprotein in a child with testicular germ cell tumor.
Med Pediatr Oncol 1999; 32(6):436-437.
(178) Albers DD, Males JL. Seminoma in hypogonadotropic hypogonadism associated with
anosmia (Kallmann’s syndrome). J Urol 1981; 126(1):57-58.
(179) Bebb GG, Grannis FW, Jr., Paz IB, Slovak ML, Chilcote R. Mediastinal germ cell tumor in
a child with precocious puberty and Klinefelter syndrome. Ann Thorac Surg 1998; 66(2):
547-548.
(180) Carroll PR, Morse MJ, Koduru PP, Chaganti RS. Testicular germ cell tumor in patient with
Klinefelter syndrome. Urology 1988; 31(1):72-74.
(181) Isurugi K, Imao S, Hirose K, Aoki H. Seminoma in Klinefelter’s syndrome with 47, XXY,
15s+ karyotype. Cancer 1977; 39(5):2041-2047.
(182) Kurabayashi A, Furihata M, Matsumoto M, Sonobe H, Ohtsuki Y, Aki M et al. Primary
intrapelvic seminoma in Klinefelter’s syndrome. Pathol Int 2001; 51(8):624-628.
(183) Muller J, Skakkebaek NE, Ratcliffe SG. Quantified testicular histology in boys with sex
chromosome abnormalities. Int J Androl 1995; 18(2):57-62.
(184) Schimke RN, Madigan CM, Silver BJ, Fabian CJ, Stephens RL. Choriocarcinoma,
thyrotoxicosis, and the Klinefelter syndrome. Cancer Genet Cytogenet 1983; 9(1):1-7.
(185) Sogge MR, McDonald SD, Cofold PB. The malignant potential of the dysgenetic germ
cell in Klinefelter’s syndrome. Am J Med 1979; 66(3):515-518.
(186) Epenetos AA, Collis CH. Seminoma in Marfan’s syndrome. Postgrad Med J 1979;
55(648):762-764.
(187) Beheshti M, Hardy BE, Mancer K, McLorie G, Churchill BM. Neoplastic potential in
patients with disorders of sexual differentiation. Urology 1987; 29(4):404-407.
(188) Donahoe PK, Crawford JD, Hendren WH. Mixed gonadal dysgenesis, pathogensis, and
management. J Pediatr Surg 1979; 14(3):287-300.
(189) Gourlay WA, Johnson HW, Pantzar JT, McGillivray B, Crawford R, Nielsen WR. Gonadal
tumors in disorders of sexual differentiation. Urology 1994; 43(4):537-540.
(190) Kulkarni JN, Kamat MR, Borges AM. Bilateral synchronous tumors in testes in
unrecognized mixed gonadal dysgenesis: a case report and review of literature. J Urol
1990; 143(2):362-364.
(191) Takazawa R, Kawakami S, Kageyama Y, Kihara K, Oshima H. Mixed gonadal dysgenesis
detected after finding a large seminoma mimicking a uterine inguinal hernia. BJU Int
2001; 87(9):903.
(192) Groot-Loonen JJ, Voute PA, de Kraker J. Testicular tumor concomitant with von
Recklinghausen’s disease. Med Pediatr Oncol 1988; 16(2):116-117.
(193) Aggarwal A, Krishnan J, Kwart A, Perry D. Noonan’s syndrome and seminoma of
undescended testicle. South Med J 2001; 94(4):432-434.
(194) Tartaglia M, Mehler EL, Goldberg R, Zampino G, Brunner HG, Kremer H et al. Mutations
in PTPN11, encoding the protein tyrosine phosphatase SHP-2, cause Noonan syndrome.
Nat Genet 2001; 29(4):465-468.
(195) Berkmen F. Persistent mullerian duct syndrome with or without transverse testicular
ectopia and testis tumours. Br J Urol 1997; 79(1):122-126.
(196) Duenas A, Saldivar C, Castillero C, Flores G, Martinez P, Jimenez M. A case of bilateral
seminoma in the setting of persistent mullerian duct syndrome. Rev Invest Clin 2001;
53(2):193-196.
(197) Kulkarni JN, Bhansali MS, Tongaonkar HB, Kamat MR, Borges AM. Carcinoma in the third
testis in a case of polyorchidism and persistent mullerian structure syndrome. Eur Urol
1992; 22(2):174-176.
(198) Melman A, Leiter E, Perez JM, Driscoll D, Palmer C. The influence of neonatal orchiopexy
upon the testis in persistent Mullerian duct syndrome. J Urol 1981; 125(6):856-858.
(199) Wu HC, Chen JH, Lu HF, Shen WC. Persistent mullerian duct syndrome wtih seminoma:
CT findings. AJR Am J Roentgenol 2000; 174(1):102-104.
(200) Eastham JA, McEvoy K, Sullivan R, Chandrasoma P. A case of simultaneous bilateral
nonseminomatous testicular tumors in persistent mullerian duct syndrome. J Urol 1992;
148(2 Pt 1):407-408.
(201) Robinson AC, Jones WG. Prader Willi syndrome and testicular tumour. Clin Oncol (R Coll
Radiol ) 1990; 2(2):117.
(202) Uehling D. Cryptorchidism in the Prader-Willi syndrome. J Urol 1980; 124(1):103-104.
(203) Hornstein L, Bove KE, Towbin RB. Linear nevi, hemihypertrophy, connective tissue
hamartomas, and unusual neoplasms in children. J Pediatr 1987; 110(3):404-408.
(204) Parra RO, Cummings JM, Palmer DC. Testicular seminoma in a long-term survivor of the
prune belly syndrome. Eur Urol 1991; 19(1):79-80.
(205) Massad CA, Cohen MB, Kogan BA, Beckstead JH. Morphology and histochemistry of
infant testes in the prune belly syndrome. J Urol 1991; 146(6):1598-1600.
(206) Miller RW, Rubinstein JH. Tumors in Rubinstein-Taybi syndrome. Am J Med Genet 1995;
56(1):112-115.
(207) Cohen PR, Kurzrock R. Miscellaneous genodermatoses: Beckwith-Wiedemann syndrome,
Birt-Hogg- Dube syndrome, familial atypical multiple mole melanoma syndrome,
hereditary tylosis, incontinentia pigmenti, and supernumerary nipples. Dermatol Clin
1995; 13(1):211-229.
Reference List150 Genetic Predisposition to Testicular Cancer 151
(208) Goedert JJ, McKeen EA, Javadpour N, Ozols RF, Pottern LM, Fraumeni JF, Jr. Polythelia
and testicular cancer. Ann Intern Med 1984; 101(5):646-647.
(209) Mehes K. Association of supernumerary nipples with other anomalies. J Pediatr 1979;
95(2):274-275.
(210) Lykkesfeldt G, Bennett P, Lykkesfeldt AE, Micic S, Rorth M, Skakkebaek NE et al. Testis
cancer. Ichthyosis constitutes a significant risk factor. Cancer 1991; 67(3):730-734.
(211) Lykkesfeldt G, Hoyer H, Lykkesfeldt AE, Skakkebaek NE. Steroid sulphatase deficiency
associated with testis cancer. Lancet 1983; 2(8365-66):1456.
(212) Chaganti RS, Houldsworth J. Genetics and biology of adult human male germ cell
tumors. Cancer Res 2000; 60(6):1475-1482.
(213) Hutson JM, Hasthorpe S, Heyns CF. Anatomical and functional aspects of testicular
descent and cryptorchidism. Endocr Rev 1997; 18(2):259-280.
(214) Prener A, Engholm G, Jensen OM. Genital anomalies and risk for testicular cancer in
Danish men. Epidemiology 1996; 7(1):14-19.
(215) Strader CH, Weiss NS, Daling JR, Karagas MR, McKnight B. Cryptorchism, orchiopexy,
and the risk of testicular cancer. Am J Epidemiol 1988; 127(5):1013-1018.
(216) Swerdlow AJ, De Stavola BL, Swanwick MA, Mangtani P, Maconochie NE. Risk factors for
testicular cancer: a case-control study in twins. Br J Cancer 1999; 80(7):1098-1102.
(217) Ulbright TM. Testis risk and prognostic factors. The pathologist’s perspective. Urol Clin
North Am 1999; 26(3):611-626.
(218) Swerdlow AJ, Huttly SR, Smith PG. Testis cancer: post-natal hormonal factors, sexual
behaviour and fertility. Int J Cancer 1989; 43(4):549-553.
(219) Moss AR, Osmond D, Bacchetti P, Torti FM, Gurgin V. Hormonal risk factors in testicular
cancer. A case-control study. Am J Epidemiol 1986; 124(1):39-52.
(220) Wilkinson TJ, Colls BM. Testicular cancer and age at puberty. BMJ 1994; 309(6959):955.
(221) Spermon JR, Witjes JA, Nap M, Kiemeney LA. Cancer incidence in relatives of patients
with testicular cancer in the eastern part of The Netherlands. Urology 2001; 57(4):747-752.
(222) Harland SJ. Conundrum of the hereditary component of testicular cancer. Lancet 2000;
356(9240):1455-1456.
(223) Akre O, Ekbom A, Hsieh CC, Trichopoulos D, Adami HO. Testicular nonseminoma and
seminoma in relation to perinatal characteristics. J Natl Cancer Inst 1996; 88(13):883-889.
(224) Mills PK, Newell GR, Johnson DE. Familial patterns of testicular cancer. Urology 1984;
24(1):1-7.
(225) Newell GR, Mills PK, Johnson DE. Epidemiologic comparison of cancer of the testis and
Hodgkin’s disease among young males. Cancer 1984; 54(6):1117-1123.
(226) Candidate regions for testicular cancer susceptibility genes. The International Testicular
Cancer Linkage Consortium. APMIS 1998; 106(1):64-70.
(227) Leahy MG, Tonks S, Moses JH, Brett AR, Huddart R, Forman D et al. Candidate regions
for a testicular cancer susceptibility gene. Hum Mol Genet 1995; 4(9):1551-1555.
(228) Murty VV, Bosl GJ, Houldsworth J, Meyers M, Mukherjee AB, Reuter V et al. Allelic loss
and somatic differentiation in human male germ cell tumors. Oncogene 1994; 9(8):
2245-2251.
(229) Skotheim RI, Kraggerud SM, Fossa SD, Stenwig AE, Gedde-Dahl JT, Danielsen HE et al.
Familial/bilateral and sporadic testicular germ cell tumors show frequent genetic
changes at loci with suggestive linkage evidence. Neoplasia 2001; 3(3):196-203.
(230) Bodmer WF. The HLA system: structure and function. J Clin Pathol 1987; 40(9):948-958.
(231) Complete sequence and gene map of a human major histocompatibility complex. The
MHC sequencing consortium. Nature 1999; 401(6756):921-923.
(232) Boon M, Nolte IM, Bruinenberg M, Spijker GT, Terpstra P, Raelson J et al. Mapping of a
susceptibility gene for multiple sclerosis to the 51 kb interval between G511525 and
D6S1666 using a new method of haplotype sharing analysis. Neurogenetics 2001; 3(4):
221-230.
(233) Foissac A, Crouau-Roy B, Faure S, Thomsen M, Cambon-Thomsen A. Microsatellites in
the HLA region: on overview. Tissue Antigens 1997; 49(3 Pt 1):197-214.
(234) Beckmann L, Fischer C, Deck KG, Nolte IM, te MG, Chang-Claude J. Exploring haplotype
sharing methods in general and isolated populations to detect gene(s) of a complex
genetic trait. Genet Epidemiol 2001; 21 Suppl 1:S554-S559.
(235) Dorum A, Moller P, Kamsteeg EJ, Scheffer H, Burton M, Heimdal KR et al. A BRCA1
founder mutation, identified with haplotype analysis, allowing genotype/phenotype
determination and predictive testing. Eur J Cancer 1997; 33(14):2390-2392.
(236) Levinson DF, Kirby A, Slepner S, Nolte I, Spijker GT, te MG. Simulation studies of
detection of a complex disease in a partially isolated population. Am J Med Genet 2001;
105(1):65-70.
(237) Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the
insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet
1993; 52(3):506-516.
(238) Bernardi D, Salvioni R, Vaccher E, Repetto L, Piersantelli N, Marini B et al. Testicular
germ cell tumors and human immunodeficiency virus infection: a report of 26 cases.
Italian Cooperative Group on AIDS and Tumors. J Clin Oncol 1995; 13(11):2705-2711.
(239) Calista D. Five cases of melanoma in HIV positive patients. Eur J Dermatol 2001; 11(5):
446-449.
(240) Hentrich MU, Brack NG, Schmid P, Schuster T, Clemm C, Hartenstein RC. Testicular germ
cell tumors in patients with human immunodeficiency virus infection. Cancer 1996;
77(10):2109-2116.
(241) Timmerman JM, Northfelt DW, Small EJ. Malignant germ cell tumors in men infected with
the human immunodeficiency virus: natural history and results of therapy. J Clin Oncol
1995; 13(6):1391-1397.
(242) Carrington M. Recombination within the human MHC. Immunol Rev 1999; 167:245-256.
Reference List152 Genetic Predisposition to Testicular Cancer 153
(243) Roychoudhury AK, Nei M. Human polymorphic genes world distribution. New York:
Oxford university press 1988;234-239.
(244) Lutke Holzik MF, Rapley EA, Hoekstra HJ, Sleijfer DT, Nolte IM, Sijmons RH. Genetic
predisposition to testicular germ-cell tumours. Lancet Oncol 2004; 5(6):363-371.
(245) Jorgensen N, Carlsen E, Nermoen I, Punab M, Suominen J, Andersen AG et al.
East-West gradient in semen quality in the Nordic-Baltic area: a study of men from the
general population in Denmark, Norway, Estonia and Finland. Hum Reprod 2002; 17(8):
2199-2208.
(246) Nijman JM, Schraffordt Koops H, Kremer J, Willemse PH, Sleijfer DT, Oldhoff J. Fertility
and hormonal function in patients with a nonseminomatous tumor of the testis.
Arch Androl 1985; 14(2-3):239-246.
(247) Kent-First M, Muallem A, Shultz J, Pryor J, Roberts K, Nolten W et al. Defining regions
of the Y-chromosome responsible for male infertility and identification of a fourth AZF
region (AZFd) by Y-chromosome microdeletion detection. Mol Reprod Dev 1999; 53(1):
27-41.
(248) Layman LC. Human gene mutations causing infertility. J Med Genet 2002; 39(3):153-161.
(249) Vogt PH, Edelmann A, Kirsch S, Henegariu O, Hirschmann P, Kiesewetter F et al. Human
Y chromosome azoospermia factors (AZF) mapped to different subregions in Yq11.
Hum Mol Genet 1996; 5(7):933-943.
(250) Foresta C, Moro E, Garolla A, Onisto M, Ferlin A. Y chromosome microdeletions in
cryptorchidism and idiopathic infertility. J Clin Endocrinol Metab 1999; 84(10):3660-3665.
(251) Herrinton LJ, Zhao W, Husson G. Management of cryptorchism and risk of testicular
cancer. Am J Epidemiol 2003; 157(7):602-605.
(252) Echten van J, Jong de B. Cytogenetics of testicular germ cell tumors of adults. The
Cancer Journal 1998; 11(5):242-246.
(253) Bianchi NO, Richard SM, Peltomaki P, Bianchi MS. Mosaic AZF deletions and
susceptibility to testicular tumors. Mutat Res 2002; 503(1-2):51-62.
(254) Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from
human nucleated cells. Nucleic Acids Res 1988; 16(3):1215-1216.
(255) Aknin-Seifer IE, Touraine RL, Lejeune H, Laurent JL, Lauras B, Levy R. A simple, low cost
and non-invasive method for screening Y-chromosome microdeletions in infertile men.
Hum Reprod 2003; 18(2):257-261.
(256) Y chromosome deletion detection system version 2.0. technical manual no 248 2003;
(Promega Corporation USA).
(257) Frydelund-Larsen L, Vogt PH, Leffers H, Schadwinkel A, Daugaard G, Skakkebaek NE et
al. No AZF deletion in 160 patients with testicular germ cell neoplasia. Mol Hum Reprod
2003; 9(9):517-521.
(258) Nathanson KL, Kanetsky PA, Hawes R, Vaughn DJ, Letrero R, Tucker K et al. The Y
eletion gr/gr and susceptibility to testicular germ cell tumor. Am J Hum Genet 2005;
77(6):1034-1043.
(259) Repping S, Skaletsky H, Brown L, van Daalen SK, Korver CM, Pyntikova T et al.
Polymorphism for a 1.6-Mb deletion of the human Y chromosome persists through
balance between recurrent mutation and haploid selection. Nat Genet 2003; 35(3):
247-251.
(260) Richiardi L, Bellocco R, Adami HO, Torrang A, Barlow L, Hakulinen T et al. Testicular
cancer incidence in eight northern European countries: secular and recent trends. Cancer
Epidemiol Biomarkers Prev 2004; 13(12):2157-2166.
(261) Hemminki K, Li X. Familial risk in testicular cancer as a clue to a heritable and
environmental aetiology. Br J Cancer 2004; 90(9):1765-1770.
(262) Crockford GP, Linger R, Hockley S, Dudakia D, Johnson L, Huddart R et al. Genome-wide
linkage screen for testicular germ cell tumour susceptibility loci. Hum Mol Genet 2006;
15(3):443-451.
(263) Redolfi E, Montagna C, Mumm S, Affer M, Susani L, Reinbold R et al. Identification of
CXorf1, a novel intronless gene in Xq27.3, expressed in human hippocampus. DNA Cell
Biol 1998; 17(12):1009-1016.
(264) Nolte IM, Spijker G, Boon M, et al. Haplotype Sharing Statistic: fine-mapping of disease
gene loci by comparing patients and controls for the length of haplotype sharing. In:
Nolte I.M. Statistics and population genetics of haplotype sharing as a tool for
fine-mapping of disease gene loci. Groningen, PhD thesis, 2003, 67-85.
(265) Te Meerman GJ, Van der Meulen MA, Sandkuijl LA. Perspectives of identity by descent
(IBD) mapping in founder populations. Clin Exp Allergy 1995; 25 Suppl 2:97-102.
(266) Van der Meulen MA, Te Meerman GJ. Association and haplotype sharing due to Identity
by Descent, with an application to genetic mapping. In: Pawlowitzki IH, Edwards JH,
Thompson E (eds), editors. Genetic mapping of disease genes. London: Academic Press,
1997: 115-135.
(267) Van der Meulen MA, te Meerman GJ. Haplotype sharing analysis in affected individuals
from nuclear families with at least one affected offspring. Genet Epidemiol 1997; 14(6):
915-920.
(268) Kawakami T, Okamoto K, Sugihara H, Hattori T, Reeve AE, Ogawa O et al. The roles
of supernumerical X chromosomes and XIST expression in testicular germ cell tumors.
J Urol 2003; 169(4):1546-1552.
(269) Ross MT, Grafham DV, Coffey AJ, Scherer S, McLay K, Muzny D et al. The DNA sequence
of the human X chromosome. Nature 2005; 434(7031):325-337.
(270) Honrado E, Benitez J, Palacios J. The molecular pathology of hereditary breast cancer:
genetic testing and therapeutic implications. Mod Pathol 2005; 18(10):1305-1320.
(271) Bleiker EM, Hahn DE, Aaronson NK. Psychosocial issues in cancer genetics-current status
and future directions. Acta Oncol 2003; 42(4):276-286.
(272) Julian-Reynier CM, Bouchard LJ, Evans DG, Eisinger FA, Foulkes WD, Kerr B et al.
Women’s attitudes toward preventive strategies for hereditary breast or ovarian
carcinoma differ from one country to another: differences among English, French, and
Canadian women. Cancer 2001; 92(4):959-968.
Reference List154 Genetic Predisposition to Testicular Cancer 155
(273) Geller G, Doksum T, Bernhardt BA, Metz SA. Participation in breast cancer susceptibility
testing protocols: influence of recruitment source, altruism, and family involvement on
women’s decisions. Cancer Epidemiol Biomarkers Prev 1999; 8(4 Pt 2):377-383.
(274) Craufurd D, Dodge A, Kerzin-Storrar L, Harris R. Uptake of presymptomatic predictive
testing for Huntington’s disease. Lancet 1989; 2(8663):603-605.
(275) Miesfeldt S, Jones SM, Cohn W, Lippert M, Haden K, Turner BL et al. Men’s attitudes
regarding genetic testing for hereditary prostate cancer risk. Urology 2000; 55(1):46-50.
(276) Bottorff JL, Ratner PA, Balneaves LG, Richardson CG, McCullum M, Hack T et al.
Women’s interest in genetic testing for breast cancer risk: the influence of
sociodemographics and knowledge. Cancer Epidemiol Biomarkers Prev 2002; 11(1):89-95.
(277) de Wert G. Ethics of predictive DNA-testing for hereditary breast and ovarian cancer.
Patient Educ Couns 1998; 35(1):43-52.
(278) Pasacreta JV. Psychosocial issues associated with genetic testing for breast and ovarian
cancer risk: an integrative review. Cancer Invest 2003; 21(4):588-623.
(279) Bluman LG, Rimer BK, Regan SK, Lancaster J, Clark S, Borstelmann N et al. Attitudes,
knowledge, risk perceptions and decision-making among women with breast and/or
ovarian cancer considering testing for BRCA1 and BRCA2 and their spouses.
Psychooncology 2003; 12(5):410-427.
(280) Shiloh S, Ilan S. To test or not to test? Moderators of the relationship between risk
perceptions and interest in predictive genetic testing. J Behav Med 2005; 28(5):467-479.
(281) Guillem JG, Wood WC, Moley JF, Berchuck A, Karlan BY, Mutch DG et al. ASCO/SSO
review of current role of risk-reducing surgery in common hereditary cancer syndromes.
J Clin Oncol 2006; 24(28):4642-4660.
(282) Kasparian NA, Meiser B, Butow PN, Soames Job RF, Mann GJ. Anticipated uptake
of genetic testing for familial melanoma in an Australian sample: An exploratory study.
Psychooncology 2007; 16(1):69-78.
(283) Riedijk SR, de Snoo FA, van Dijk S, Bergman W, van Haeringen A, Silberg S et al.
Hereditary melanoma and predictive genetic testing: why not? Psychooncology 2005;
14(9):738-745.
(284) Cormier L, Valeri A, Azzouzi R, Fournier G, Cussenot O, Berthon P et al. Worry and
attitude of men in at-risk families for prostate cancer about genetic susceptibility and
genetic testing. Prostate 2002; 51(4):276-285.
(285) Peters JA, Vadaparampil ST, Kramer J, Moser RP, Court LJ, Loud J et al. Familial testicular
cancer: interest in genetic testing among high-risk family members. Genet Med 2006;
8(12):760-770.
(286) Bruno M, Tommasi S, Stea B, Quaranta M, Schittulli F, Mastropasqua A et al. Awareness
of breast cancer genetics and interest in predictive genetic testing: a survey of a
southern Italian population. Ann Oncol 2004; 15 Suppl 1:I48-I54.
(287) Jager M, Oosterwijk B, Middel B, Hoekstra-Weebers JEHM. Motivatie voor genetische
counseling en angst voor kanker bij een verhoogd familair risico op erfelijke borst- en
eierstokkanker.(motivations regarding genetic counseling and fear for cancer in women
having an increased risk for familial breast and ovarian cancer). Gedrag en Gezondheid
2004; 32:18-27.
(288) Cohen J. Statistical power analysis for the behavioral sciences. Erlbaum, Hillsdale,
Nj 1988.
(289) Cappelli M, Surh L, Humphreys L, Verma S, Logan D, Hunter A et al. Psychological and
social determinants of women’s decisions to undergo genetic counseling and testing for
breast cancer. Clin Genet 1999; 55(6):419-430.
(290) Lerman C, Narod S, Schulman K, Hughes C, Gomez-Caminero A, Bonney G et al. BRCA1
testing in families with hereditary breast-ovarian cancer. A prospective study of patient
decision making and outcomes. JAMA 1996; 275(24):1885-1892.
(291) Tijmstra TJ. The imperative character of medical technology and the significance of
‘anticipated decision regret’. Ned Tijdschr Geneeskd 1987; 131(26):1128-1131.
(292) Watson E, Hewitson P, Brett J, Bukach C, Evans R, Edwards A et al. Informed decision
making and prostate specific antigen (PSA) testing for prostate cancer: a randomised
controlled trial exploring the impact of a brief patient decision aid on men’s knowledge,
attitudes and intention to be tested. Patient Educ Couns 2006; 63(3):367-379.
(293) Evans R, Edwards A, Brett J, Bradburn M, Watson E, Austoker J et al. Reduction in
uptake of PSA tests following decision aids: systematic review of current aids and their
evaluations. Patient Educ Couns 2005; 58(1):13-26.
(294) Bleiker EM, Aaronson NK, Menko FH, Hahn DE, van Asperen CJ, Rutgers EJ et al. Genetic
counseling for hereditary cancer: a pilot study on experiences of patients and family
members. Patient Educ Couns 1997; 32(1-2):107-116.
(295) Harel A, Abuelo D, Kazura A. Adolescents and genetic testing: what do they think about
it? J Adolesc Health 2003; 33(6):489-494.
(296) van Dijk MR, Steyerberg EW, Habbema JD. Survival of non-seminomatous germ cell
cancer patients according to the IGCC classification: An update based on meta-analysis.
Eur J Cancer 2006; 42(7):820-826.
(297) Austoker J. Screening for ovarian, prostatic, and testicular cancers. BMJ 1994;
309(6950):315-320.
(298) Moiseyenko VM, Danilov AO, Baldueva IA, Danilova AB, Tyukavina NV, Larin SS et al.
Phase I/II trial of gene therapy with autologous tumor cells modified with tag7/
PGRP-S gene in patients with disseminated solid tumors: miscellaneous tumors.
Ann Oncol 2005; 16(1):162-168.
(299) Risch N, Merikangas K. The future of genetic studies of complex human diseases.
Science 1996; 273(5281):1516-1517.
(300) Youngren KK, Nadeau JH, Matin A. Testicular cancer susceptibility in the 129.MOLF-Chr19
mouse strain: additive effects, gene interactions and epigenetic modifications.
Hum Mol Genet 2003; 12(4):389-398.
Genetic Predisposition to Testicular Cancer 157
Addendum
Addendum158 Genetic Predisposition to Testicular Cancer 159
Genetische termen
DNA
Deoxyribonucleic acid ofwel DNA is een dubbele helix van nucleotiden (ook wel basen
genoemd), die alle genetische informatie van een cel bevat. Er bestaan vier nucleotiden die
worden aangeduid met A, C, G en T.
Chromosoom
Een chromosoom is een compact DNA-molecuul in associatie met eiwitten. Elk mens heeft 46
chromosomen, om precies te zijn, 23 chromosomenparen. Van elk paar is er één geërfd van de
vader en één van de moeder.
Genoom
Het genoom is de verzameling van chromosomen van een organisme.
Merker
Een merker is een locatie op het genoom (d.i. locus) die varieert tussen mensen. Alle mensen
hebben voor ~99,9% hetzelfde DNA. Verschillen tussen mensen worden bepaald door ~0,1%
van het DNA. Aangezien het genoom bestaat uit zo’n 3,5×109 baseparen (bp), betekent dit nog
steeds ongeveer 3,5×106 verschillen. Wij werken in ons laboratorium hoofdzakelijk met twee
verschillende soorten merkers: microsatellietmerkers en single nucleotide polymorfismes (SNPs).
Een microsatelliet is een polymorf locus met een DNA-volgorde bestaande uit een variabel aantal
van oligonucleotide (di-, tri-, tetra- of pentanucleotide) eenheden. Een SNP is een locus van één
nucleoide dat kan variëren. (voor een betekenis van een polymorfisme, zie mutatie)
Allel
Een allel is een specifieke vorm van een merker. Bijvoorbeeld, voor een microsatelliet-merker
duidt men het allel aan met het aantal herhaalde oligonucleotide-eenheden.
Haplotype
Een haplotype is een combinatie van allelen voor meerdere merkers op één chromosoom.
Genotype
Het genotype is de genetische samenstelling van allelen op één of meerdere loci voor beide
chromosomen uit een chromosomenpaar.
Addendum - Verklarende woordenlijst
Addendum160 Genetic Predisposition to Testicular Cancer 161
Fenotype
Het fenotype is de uiterlijke verschijning van een ziekte of een specifiek kenmerk, d.w.z. de
fysieke expressie van genen / allelen). Dit in tegenstelling tot het genotype.
Fase van haplotypes
Fase voor haplotypes afleiden wil zeggen dat bepaald wordt welke allelen bij elkaar op hetzelfde
chromosoom liggen. Omdat iedereen twee allelen per merker heeft en het laboratorium niet kan
zeggen welke allelen voor naburige merkers bij elkaar horen op een chromosoom, zijn er voor
twee merkers twee mogelijke combinaties. Door te kijken welke allelen in de ouders aanwezig
zijn, kan worden bepaald welke allelen bij elkaar op hetzelfde chromosoom liggen.
Heterozygositeit of informativiteit
Personen die op een merkerlocus twee dezelfde allelen hebben, worden homozygoot genoemd.
Als een persoon twee verschillende allelen heeft, dan is hij heterozygoot. De heterozygositeit
van een merker geeft aan welk percentage van de mensen heterozygoot is.
Mutatie
Een mutatie is een verandering in het DNA die bijvoorbeeld een ziekte of afwijking tot gevolg
heeft. Als zo’n verandering geen aanwijsbare invloed heeft op het fenotype, dan wordt hij
‘neutraal’ genoemd en heet hij een polymorfisme of (genetische) variant.
Recombinatie
Een recombinatie beschrijft het verschijnsel dat, tijdens het ontstaan van geslachtscellen (d.i. een
meiose), de twee chromosomen van een ouder stukken met elkaar uitwisselen zodat allelen, die
bij de ouder nog op twee veschillende chromosomen aanwezig waren, bij het kind op hetzelfde
chromosoom terechtkomen. Per meiose komt een recombinatie redelijk vaak voor. De lengte van
een gebied, dat 1% kans heeft op een recombinatie, wordt centiMorgan (cM) genoemd.
Linkage disequilibrium
Linkage disequilibrium wil zeggen dat allelen op verschillende loci niet onafhankelijk van elkaar
overerven. Als gevolg daarvan kunnen ze statistisch met elkaar gecorreleerd zijn. Dit ontstaat
doordat er geen of weinig recombinaties optreden tussen loci die dicht bij elkaar op het
chromosoom liggen.
Penetrantie
De penetrantie van een mutatie is de kans dat iemand die deze mutatie draagt daadwerkelijk
ziek wordt.
Fenokopie
Met een fenokopie wordt iemand bedoeld die niet de benodigde mutatie(s) draagt, maar die
toch de ziekte of een kenmerk heeft.
Statistische termen
Type I fout
De type I fout geeft bij het testen van een nul-hypothese tegen een alternatieve hypothese de
kans weer dat de nul-hypothese onterecht wordt verworpen.
Onderscheidingsvermogen of power
Het onderscheidingsvermogen ofwel power geeft de kans weer dat terecht de nul-hypothese
wordt verworpen.
Associatieanalyse
Associatieanalyse vergelijkt voor één of meerdere merkers de frequenties van de verschillende
allelen, genotypes of haplotypes tussen onafhankelijke patiënten en controlepersonen. Als een
merker het ziektegen is of in linkage disequilibrium met het ziektegen is, wordt verwacht dat het
ook geassocieerde allel, genotype of haplotype in een verhoogde frequentie aanwezig is in een
steekproef van patiënten vergeleken met een steekproef van controlepersonen. Het kan echter
ook zijn dat de aanwezigheid van een bepaald allel beschermt tegen het krijgen van een ziekte.
Zo’n beschermend of protectief allel zal dan een lagere frequentie hebben in een steekproef van
patiënten in vergelijking tot een steekproef van controlepersonen.
Transmission/disequilibrium test
De transmissie/disequilibrium test bekijkt trio’s bestaande uit een patiënt met beide ouders en
analyseert of ouders een bepaald allel van een merker vaker doorgeven aan hun aangedane kind
dan andere allelen. Voor een merker die niet gelinkt (d.w.z. gekoppeld) is met de te onderzoeken
ziekte, wordt verwacht dat beide allelen, aanwezig in de ouder van de patiënt, even veel kans
(d.i. 50%) hebben om doorgegeven te worden. Beide allelen moet wel verschillend zijn, d.w.z.
de ouder moet heterozygoot zijn.
Hardy-Weinberg equilibrium test
Het Hardy-Weinberg evenwicht geeft weer hoeveel homo- en heterozygoten verwacht worden
in de populatie. Bijvoorbeeld, als een allel een frequentie heeft van 10% in de populatie,
dan verwacht je 10%×10% =1% homozygote personen voor dit allel en 2×10%×90%=18%
heterozygote personen in de populatie.
In ons laboratorium wordt voor elke merker getest of het Hardy-Weinberg evenwicht aanwezig
is. Een afwijking van het evenwicht duidt meestal op een systematische fout in het scoren van
de genotypes. Zoals wij de test gebruiken, is het dus een test om de kwaliteit van de geleverde
genotype-data te checken.
Overgenomen uit: Statistics and population genetics of haplotype sharing as a tool for fine-
mapping of disease gene loci. Proefschrift Dr. I.M. Nolte, Groningen 2003.
Genetic Predisposition to Testicular Cancer 163
Dankwoord
Dankwoord164 Genetic Predisposition to Testicular Cancer 165
Dankwoord
Dit onderzoek was niet mogelijk geweest zonder de bereidheid van alle patiënten en controle
personen om DNA-materiaal af te staan. Ik wil mijn oprechte waardering hiervoor uitspreken.
Het onderzoek dat in dit proefschrift wordt beschreven werd uitgevoerd bij de afdelingen
Chirurgische Oncologie van het Universitair Medisch Centrum Groningen, in samenwerking met
de afdeling Genetica, Medische Oncologie, Medische Biologie, Pathologie, Wenckebach Instituut
en het Integraal Kankercentrum Noord-Nederland(IKN). Velen hebben mij in staat gesteld mijn
promotie-onderzoek tot een eind te brengen. Iedereen die hieraan heeft bijgedragen wil ik
hartelijk bedanken voor zijn of haar inzet. Natuurlijk zal ik een aantal mensen in het bijzonder
noemen.
Graag wil ik de promotores Prof. dr. H.J. Hoekstra en Prof. dr. D.Th. Sleijfer en de copromotores
Dr. R.H. Sijmons en Dr. J.E.H.M. Hoekstra-Weebers bedanken voor het in mij gestelde vertrouwen,
de steun en begeleiding.
Prof. dr. H.J. Hoekstra, Beste Harald, drijvende kracht achter het zaadbalkanker onderzoek.
Jij bood mij de kans om promotie-onderzoek te verrichten. Je enthousiasme, begeleiding en
laagdrempelige mogelijkheden voor overleg zijn van grote waarde. De inzet waarmee jij op
persoonlijke wijze (vele) onderzoekers begeleid is erg motiverend en lovenswaardig. Bovendien
beschik jij over een vermogen om grote vaart in de wetenschap te houden waardoor de
onderzoeksmolen blijft draaien en mijn enthousiasme alleen maar groter werd. Harald ik heb de
afgelopen jaren veel van je geleerd en ik ben je hiervoor zeer dankbaar.
Prof. dr. D.Th. Sleijfer, Beste Dirk, veel heb ik geleerd van je gestructureerde en kritische
werkwijze. Manuscripten kwamen vaak zeer snel gecorrigeerd retour met zeer verhelderend
commentaar. Bovendien tipte je mij geregeld als er weer een “interessant” stuk was gepubliceerd
over zaadbalkanker. Je kritische blik is van belangrijke waarde geweest voor de totstandkoming
van mijn proefschrift. Ik wil je hiervoor bedanken, maar ook voor de bemoedigende woorden
die af en toe nodig waren.
Dr. R.H. Sijmons, Beste Rolf, jij hebt me kennis laten maken met de “ins en outs” van de
Genetica. Het enthousiasme waarmee jij je kennis overbrengt in heldere begrijpelijke taal is
bewonderenswaardig. Voor overleg was je altijd bereikbaar en veel tijd hebben we samen
doorgebracht (in jullie noodgebouw waarin je veel te lang was gehuisvest) bij de totstandkoming
van de manuscripten en het kritisch analyseren van gepubliceerde artikelen. Wanneer ik iets
“definitiefs” inleverde ging jij er nog even “met het stof kammetje door”, wat resulteerde in
fraaie zinnen met mooie nuanceringen. Je verhelderende benaderingswijze en sturende inbreng
heeft grote invloed gehad op de totstandkoming van dit proefschrift. Ik ben je voor dit alles
zeer dankbaar.
Dankwoord166 Genetic Predisposition to Testicular Cancer 167
Dr. J.E.H.M. Hoekstra-Weebers, Beste Josette, Het was voor mij natuurlijk ideaal om een
“echtpaar” als onderzoeksbegeleiders te hebben. Met één telefoontje naar jullie huis, had ik
jullie beiden aan de telefoon en kon ik “dingen” regelen of vragen, 24 uur per dag, 7 dagen per
week. Jij hebt me veel geleerd over “kwaliteit van leven” en “psychosociaal” onderzoek en het
ontwerpen en verwerken van vragenlijsten. In het beoordelen van mijn laatste manuscript was
je kritisch en het heeft me veel moeite gekost om aan je eisen te voldoen maar je bleef geduldig
en nam ruim de tijd voor heldere uitleg. Psychosociaal onderzoek was volledig nieuw voor mij
en ik ben blij dat ik hier ervaring mee heb opgedaan. Ik heb hier veel van geleerd en ik wil je
hiervoor hartelijk bedanken.
De leden van de beoordelingscommissie, Prof. dr. S. Horenblas, Prof. dr. J.W. Oosterhuis en
Prof. dr. P.H.B. Willemse wil ik bedanken voor de beoordeling van het manuscript.
Daanaast ben ik Dr. I.M. Nolte veel dank verschuldigd voor de hulp bij de vele statistiek. Beste
Ilja, genetische statistiek is een vak appart en ik ben je zeer dankbaar voor je vele hulp hierbij.
Je hebt veel tijd genomen om mij uitleg te geven en je hebt de manuscripten kritisch beoordeeld
en de statistiek geweldig verzorgd.
Het laboratorium onderzoek heb ik kunnen uitvoeren met hulp van medewerkers op het lab van
de afdeling Medische Biologie van het Universitair Medisch Centrum Groningen. In het bijzonder
wil ik hier bedanken Dr. ir. G.J. te Meerman, Marcel Bruinenberg en Gerrit van der Steege.
“Mijn onderzoeksgroep / kamergenoten / collega-onderzoekers”: Imi Veldman, Joke Fleer,
Eric Sonneveld, Rudi Komdeur, Marrit Tuinman en David Cobben. Ik ben blij dat ik jullie heb
leren kennen. We hebben veel tijd doorgebracht in onze kamer / op de gang / tijdens de koffie.
Ik heb veel van jullie geleerd (en jullie hadden meer discipline dan ik en dat benijd ik). Gelukkig
was er ook tijd voor ontspanning en “social talk” veel dank hiervoor.
Mijn paranimfen: Stephan Lutke Holzik, mijn broertje. Vaak vroeg je wanneer ik nou eens klaar
zou zijn met mijn onderzoek. Welnu, dat stadium is bereikt. Ik ben blij en trots dat je de taak
van paranimf op je wilt nemen. Deepu Daryanani, collega en vriend, jij weet ook wat het is om
onderzoek en opleiding te combineren. Als collega’s hebben we veel contact en we zijn goede
vrienden geworden, veel dank dat je mijn paranimf wilt zijn.
Mijn ouders, pap en mam, jullie stonden garant voor een onbezorgde jeugd en jullie hebben mij
alle kansen gegeven om mijzelf te ontwikkelen tot wie ik nu ben. Jullie onvoorwaardelijke steun
en interesse zijn onmisbaar, daarom draag ik dit boekje ook aan jullie op.
Lieve Marjolijn, ik vind het echt geweldig dat jij mij de ruimte hebt gegeven om in Groningen
te promoveren terwijl jij daar de nodige kilometers voor hebt moeten reizen. Jij hebt mij altijd
weten aan te moedigen en je stimuleert me om van elke situatie iets te leren en iets positiefs
mee te nemen. Bedankt voor je steun en liefde en de allergrootste vreugde die je ons hebt
gegeven met Louise. Lieve Louise, je bent zo vrolijk en nog zo klein, jij laat zien waar het in het
leven om gaat. Marjolijn en Louise dit is ook jullie boekje!
Genetic Predisposition to Testicular Cancer 169
List of Publications
List of Publications170 Genetic Predisposition to Testicular Cancer 171
List of publications
Lutke Holzik MF, Hoekstra HJ, Sijmons RH, Sleijfer DTh, Fleer J, Hoekstra-Weebers JEHM.
Interest in and motivations regarding genetic testing for testicular germ cell tumour
susceptibility, Submitted 2007
Lutke Holzik MF, Hoekstra HJ, Sijmons RH, Sonneveld DJ, van der Steege G, Sleijfer DTh, Nolte IM.
Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with
sporadic testicular germ cell tumour without cryptorchidism.
Eur J Cancer. 2006 Aug; 42(12):1869-74.
Nuver J, Lutke Holzik MF, van Zweeden M, Hoekstra HJ, Meijer C, Suurmeijer AJ, Groen HJ,
Hofstra RM, Sluiter WJ, Groen H, Sleijfer DTh, Gietema JA.
Genetic variation in the bleomycin hydrolase gene and bleomycin-induced pulmonary toxicity
in germ cell cancer patients.
Pharmacogenet Genomics. 2005 Jun; 15(6):399-405.
Lutke Holzik MF, Storm K, Sijmons RH, D’hollander M, Arts EG, Verstraaten ML, Sleijfer DTh,
Hoekstra HJ.
Absence of constitutional Y chromosome AZF deletions in patients with testicular germ cell
tumors.
Urology. 2005 Jan; 65(1):196-201
Lutke Holzik MF, Rapley EA, Hoekstra HJ, Sleijfer DTh, Nolte IM, Sijmons RH.
Genetic predisposition to testicular germ-cell tumours.
Lancet Oncol. 2004 Jun; 5(6):363-71.
Lutke Holzik MF
A Non-Germ Cell Malignancy in a Recurrent Retroperitoneal Tumor Mass After Combined
Treatment for a Nonseminomatous Testicular GermCell Tumor
The American Journal of Urology Review. 2003 Nov/Dec; 1(6):2-5
Lutke Holzik MF, Hoekstra HJ, Mulder NH, Suurmeijer AJ, Sleijfer DTh, Gietema JA.
Non-germ cell malignancy in residual or recurrent mass after chemotherapy for
nonseminomatous testicular germ cell tumor.
Ann Surg Oncol. 2003 Mar; 10(2):131-5.
List of Publications172 Genetic Predisposition to Testicular Cancer 173
Lutke Holzik MF, Sijmons RH, Sleijfer DTh, Sonneveld DJ, Hoekstra-Weebers JEHM,
van Echten-Arends J, Hoekstra HJ.
Syndromic aspects of testicular carcinoma.
Cancer. 2003 Feb 15; 97(4):984-92.
Fleer J, Hoekstra HJ, Sleijfer DTh, Lutke Holzik MF, Hoekstra-Weebers JEHM.
Postmortem diagnosis of testicular cancer.
Lancet. 2002 Nov 9; 360(9344):1511-2 (letter)
Sonneveld DJ, Lutke Holzik MF,Nolte IM, Sleijfer DTh, van der Graaf WT,
Bruinenberg M, Sijmons RH, Hoekstra HJ, Te Meerman GJ.
Testicular carcinoma and HLA Class II genes.
Cancer. 2002 Nov 1; 95(9):1857-63.
Lutke Holzik MF, Sonneveld DJ, Hoekstra HJ, te Meerman GJ, Sleijfer DTh,
Schaapveld M.
Do the eastern and northern parts of The Netherlands differ in testicular cancer? (letter)
Urology. 2001 Oct; 58(4):636-7
Lutke Holzik MF, Mastboom, WJB
Galsteenileus. Nederlands tijdschrift voor geneeskunde (studenten editie). 2000; (3):48-50
Genetic Predisposition to Testicular Cancer 175
Curriculum Vitae
Curriculum Vitae176 Genetic Predisposition to Testicular Cancer 177
Curriculum Vitae
Martijn Frederik Lutke Holzik werd op 25 februari 1974 te Enschede geboren. Hij bezocht het
Jacobus College in Enschede waar hij in 1991 zijn HAVO diploma en in 1993 zijn VWO diploma
behaalde. Aansluitend studeerde hij geneeskunde aan de Rijksuniversiteit Groningen. Na het
doctoraal examen in 1998 vonden de co-schappen plaats in het Medisch Spectrum Twente
te Enschede. In februari 2000 behaalde hij (cum laude) zijn artsen titel. Aansluitend werd hij
arts-assistent chirurgie in het Sophia ziekenhuis locatie Weezenlanden te Zwolle. Medio 2000
werd hij voor de opleiding chirurgie aangenomen. Voorafgaand aan zijn opleiding chirurgie
startte hij vanaf eind 2000 als arts-onderzoeker bij de afdeling chirurgische oncologie in het
Academisch Ziekenhuis Groningen (thans Universitair Medisch Centrum Groningen). Gedurende
ruim anderhalf jaar werd de basis voor dit proefschrift gelegd. Eind 2002 begon hij de opleiding
tot chirurg in het Universitair Medisch Centrum Groningen (Prof. Dr. H.J. ten Duis) en in 2004
werd dit vervolgd in het Medisch Spectrum Twente te Enschede (Dr. W.J.B. Mastboom) alwaar hij
zijn opleiding ook zal afronden. Martijn is getrouwd met Marjolijn en hun dochter Louise werd
14 januari 2007 geboren.