Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Medicine 1100
Analysis of genetic susceptibility
to cervical cancer using candidate
gene and GWAS approaches
IVANA JUKO-PECIREP
ACTA
UNIVERSITATIS
UPSALIENSIS
UPPSALA
2015
ISSN 1651-6206
ISBN 978-91-554-9234-2
urn:nbn:se:uu:diva-248484
Dissertation presented at Uppsala University to be publicly examined in Auditorium Minus,
Gustavianum, Akademigatan 3, Uppsala, Thursday, 28 May 2015 at 09:15 for the degree of
Doctor of Philosophy (Faculty of Medicine). The examination will be conducted in English.
Faculty examiner: Professor Holger Luthman.
Abstract
Juko-Pecirep, I. 2015. Analysis of genetic susceptibility to cervical cancer using candidate
gene and GWAS approaches. Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Medicine 1100. 50 pp. Uppsala: Acta Universitatis Upsaliensis.
ISBN 978-91-554-9234-2.
Cervical cancer is the forth most commonly diagnosed cancer among women worldwide. It is
caused by persistent infection with an oncogenic type of Human Papillomavirus (HPV). The
HPV is a necessary but not sufficient cause of cervical cancer. Environmental factors such as
smoking, high parity and long-term use of oral contraceptives increases the risk of cervical
cancer. Genetic factors also affect the risk of developing the disease. The aim of this thesis is
to search for and evaluate genetic risk factors for cervical cancer using both a candidate gene
approach and a genome-wide association study (GWAS).
Paper I examined the association of genetic variation in three Fanconi Anemia (FA) genes
(FANCA, FANCC and FANCL), involved in DNA repair, with cervical cancer susceptibility
in the Swedish population. No association was observed. Paper II evaluated the association of
genetic variation in the TMC6 and TMC8 genes with susceptibility to cervical cancer in the
Swedish population and an association of two SNPs (rs2290907 and rs16970849) with cervical
cancer was observed. In paper III the first GWAS performed in cervical cancer was reported.
Three independent loci in the major histocompatibility complex (MHC) region at 6p21.3 were
found to affect the susceptibility to cervical cancer. Paper IV examined the sequence variation
in the TMC6 and TMC8 region and its association with cervical cancer. A highly polymorphic
21 bp sequence was identified and found to be repeated 5 to 42 times in both cases and controls.
Lack of this repeat was associated with increased risk of cervical cancer. An intronic SNP
(rs2926778) located in between the TNRC6C and TMC6 genes was also found to be associated
with cervical cancer.
The thesis provides evidence for the importance of genes in the immune system for cervical
cancer susceptibility. The genetic risk factors identified explain only a part of the genetic
susceptibility, implying that other risk factors remains to be identified
Keywords: cervical cancer, association study, human papillomavirus, genetics, complex
disease, TMC6, TMC8, MHC region
Ivana Juko-Pecirep, Department of Immunology, Genetics and Pathology,
Rudbecklaboratoriet, Uppsala University, SE-751 85 Uppsala, Sweden.
© Ivana Juko-Pecirep 2015
ISSN 1651-6206
ISBN 978-91-554-9234-2
urn:nbn:se:uu:diva-248484 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-248484)
Till min kära familj ♥ Mojoj dragoj obitelji
”Promise me you'll always remember: You're braver than you believe, and stronger
than you seem, and smarter than you think”
Winnie The Pooh
List of Papers
This thesis is based on the following papers, which are referred to in the text
by their Roman numerals.
I.
Juko-Pecirep, I., Ivansson L, E., Gyllensten B, U. (2011) Evaluation of Fanconi Anemia genes FANCA, FANCC and FANCL in
cervical cancer susceptibility. Gynecological Oncology, 122
(2011):377–381
II.
Castro, F. A., Ivansson, E. L., Schmitt, M., Juko-Pecirep, I., Kjellberg, L., Hildesheim, A., … Pawlita, M. (2012). Contribution of
TMC6 and 8 (EVER1 and EVER2) variants to cervical cancer
susceptibility. International Journal of Cancer. Journal International Du Cancer, 130(2), 349–355. doi:10.1002/ijc.26016
III.
Chen, D., Juko-Pecirep, I., Hammer, J., Ivansson, I., Enroth, S.,
Gustavsson, I., Feuk, L., Magnusson, K.E.P., McKay, D.J., Wilander, E., Gyllensten, U. (2013) Genome-wide Association Study of
Susceptibility Loci for Cervical Cancer. Journal of the national cancer institute. 2013:105:624-633
IV.
Juko-Pecirep, I., Ameur A., Bunikis I., Höijer I., Feuk L., Gyllensten, U. (2015) Characterization of the TMC6 and TMC8 (EVER1
and EVER2) genes associated with cervical cancer. Manuscript
Reprints were made with permission from the respective publisher.
Related articles
Ivansson, LE., Juko-Pecirep, I., Erlich, HA., Gyllensten, UB. (2011) Pathway-based analysis of genetic susceptibility to cervical cancer in situ: HLADPB1 affects risk in Swedish women. Genes and Immunity. 12 (2011):605614
Contents
Introduction ................................................................................................... 11
Cervical cancer ......................................................................................... 13
Screening and treatment of cervical cancer ......................................... 15
Prevention of cervical cancer ............................................................... 17
Human Papillomavirus ............................................................................. 18
Additional risk factors .............................................................................. 20
Environmental risk factors ................................................................... 20
Genetic risk factors .............................................................................. 20
Genetic analysis of complex diseases ....................................................... 23
Cohort vs. Case-Control study ............................................................. 23
Familial aggregation ............................................................................ 23
Segregation study ................................................................................. 23
Linkage study ....................................................................................... 23
Association study ................................................................................. 24
Technologies in complex disease analysis ............................................... 25
Genotyping methods ............................................................................ 25
Sanger Sequencing ............................................................................... 27
Next Generation Sequencing ............................................................... 27
Present investigation ..................................................................................... 30
Aim ........................................................................................................... 30
Paper I ....................................................................................................... 30
Background .......................................................................................... 30
Results and discussions ........................................................................ 30
Paper II ..................................................................................................... 31
Background .......................................................................................... 31
Results and Discussion ........................................................................ 31
Paper III .................................................................................................... 32
Background .......................................................................................... 32
Results and Discussion ........................................................................ 32
Paper IV .................................................................................................... 34
Background .......................................................................................... 34
Results and Discussion ........................................................................ 35
Conclusion and future perspectives............................................................... 37
Acknowledgements ....................................................................................... 39
References ..................................................................................................... 41
Abbreviations
ASC
ASO
CIS
CIN
CMI
CE
DNA
dNTP
EV
HR
TS
HPV
GWAS
SCC
Pap
LSIL
LR
HSIL
LEEP
RCT
URR
ORF
MHC
HLA
HIV
FANC
PCR
LSO
NGS
MPS
SMRT
Atypical squamous cells
Allele-specific oligos
Cancer In Situ
Cervical intraepithelial neoplasm
Cell-mediated immunity
Capillary electrophoresis
Deoxyribonucleic acid
Deoxynucleotide triphosphates
Epidermodysplasia Verruciformis
High-risk
Tumor suppressor
Human Papillomavirus
Genome-wide association study
Squamous cell carcinoma
Papanicolaou-stain
Low-grade squamous intraepithelial lesions
Low-risk
High-grade squamous intraepithelial lesions
Loop electrosurgical excision procedure
Randomized control trial
Upstream regulatory region
Open reading frame
Major histocompatibility complex
Human leukocyte antigen
Human immunodeficiency virus
Fanconi anemia complementation group
Polymerase chain reaction
Locus-specific oligo
Next generation sequencing
Massively parallel sequencing
Single-molecule real-time
Introduction
Cancer is caused by uncontrolled growth of cells. Abnormal proliferation is
called neoplasia and results in neoplasm, also called tumor. Tumors may be
benign or malign; where the major difference is that the malignant tumors
destroy adjacent normal tissue and sometimes spreads to other parts of the
body (metastases). Cancers are also classified by type of cells that the tumor
starts in and is therefore presumed to be the origin of the tumor. The most
common cancers are those of epithelial origin and are classified as carcinomas. Development of cancer progress is divided in three steps. Dysplasia is
the earliest form of pre-cancerous lesion and can be classified in low-grade
or high-grade. Low-grade dysplasia means that some atypical cell changes
are seen, but not in all cells, while in high-grade dysplasia atypical changes
are seen in many cells as well as abnormal growth. The second step is called
carcinoma in situ (CIS), a Latin word for ”in its place” meaning that the
surrounding tissue has not been invaded. CIS is synonymous with high-grade
dysplasia and the risk of transforming into invasive cancer growth is high.
The final step is invasive carcinoma, commonly called cancer that will, if
left untreated, spread beyond the tissue in which it developed and grow into
surrounding, healthy tissue and may be lethal.
Development of cancer is a multistep process, where normal somatic cells
acquires different abilities in a stepwise fashion and evolves into a cancer
cell. A single mutation is generally not enough to convert a normal cell into
a malignant one, but a number of independent events are necessary in order
for malignant transformation to take place. The cells need to be selfsufficient in growth factors and insensitive to signals that normally inhibit
growth. They also need to avoid apoptosis and overcome the replication
limit that normal cells have. Finally the developing tumor needs the ability to
invade the surrounding tissue and metastasize. For all these changes the
transforming cells need to overcome the natural defense mechanisms against
cancer.
Cancer is a genetic disease, meaning that they result from unnatural function
of one or more genes. It should not be mixed up with the statement that cancer is a hereditary disease. A hereditary disease is passed from the parents to
a child through the inheritance of a defective gene. However, some rare cancers are hereditary, for most of them the genes could be an affecting component for the cancer development and for some of them it just is a by chance
event.
11
The alterations described above require genetic changes in key genes. There
are three categories of genes that are targets for these mutations: oncogenes,
tumor suppressor genes and DNA repair genes. The normal function of oncogenes is to control cell proliferation, apoptosis, or both. The product of
oncogenes can be divided into six broad groups: transcription factors, chromatin remodelers, growth factors, growth factor receptors, signal transducers
and apoptosis regulators. The oncogenes can be activated by chromosomal
rearrangements, point mutations or gene amplification, causing alteration in
oncogene structure or an increase/decrease of its expression. The genetic
changes results in a growth advantage or increased survival of cells carrying
such alteration1. Tumor suppressor (TS) genes protect the cell from events
leading to cancer. In a normally functioning cell some TS gene products
prevent inappropriate cell cycle progression, some steer deviant cells into
apoptosis, while other keep the genome stable and mutation rates low by
ensuring accurate replication, repair and segregation of the cells DNA. Unlike oncogenes, both alleles of a tumor suppressor gene must be inactivated
in order to cause a change in the behavior of the cell. DNA repair proteins
are often classified as TS genes, mainly because mutations in these genes
increase the risk of cancer development.
The work presented in this thesis is focusing on evaluating a common
cancer in women, namely cervical cancer caused by the human papillomavirus (HPV). Using candidate gene and genome wide association studies
(GWAS) approaches we are evaluating the genetic influence on the disease.
12
Cervical cancer
The cervix is found in the lower part of the uterus; the cervical canal connects the uterus and the vagina. (Figure 1) The epithelium of the cervix varies, consisting of two types of epithelial cells, squamous cells that form layers in the epithelium and columnar cells that form the glandular epithelium.
The junction where these two cell types meet is called the transformation
zone2. In this zone the columnar epithelium transforms into squamous epithelium and this is the site where dysplasia, the first step of cervical carcinogenesis, develops. Cervical cancer is divided into two types: Cervical squamous cell carcinoma, which is derived from squamous cells and cervical
adenocarcinoma, arising in the glandular cells of cervix3.
Figure 1. A schematic drawing of the female reproductive organs. The arrow is
indicating the location of the cervix, found between uterus and vagina. Illustration
by Erik Gustavsson
Cervical cancer is the fourth most commonly diagnosed cancer, and the
fourth leading cause of cancer death, among women worldwide. In 2012,
13
528 000 new cancer cases and 266 000 deaths due to cervical cancer was
registered. Around 85% of new cancer cases occurred in the less developed
countries, where also nine out of ten (87%) cervical cancer deaths occurred.
Globally, the highest incidence rates of cervical cancer are reported in Eastern Africa, Melanesia, Southern and Middle Africa and the lowest rates in
Australia/New Zealand and Western Asia4. The average age for developing
cervical cancer is 55 years, however one-fourth of the women diagnosed
with cervical cancer are younger than 40 years5. The cervical squamous cell
carcinoma (SCC) is the most common form, and for a couple of years ago
accounted for 75% of all cervical neoplasms in developing countries6. Adenocarcinomas and adenosquamous cell carcinomas represent 10-25%, where
other rare types comprise the remaining cases7.
The reasons for high incidence rates in developing countries may be due
to different prevalence of risk factors, which will be discussed later in the
thesis. However, the major reason is believed to be the lack of cervical cancer screening and diagnostic methods, which is an important step in detection of precancerous and early stage cervical cancer8.
Figure 2. The incidence of cervical cancer cases (pink) and deaths (red) in Sweden,
has decreased with 50% during the 1960-2012. 14
Screening and treatment of cervical cancer
In Sweden organized cervical cancer screening was implemented in the mid1960s, even before the cause of cervical cancer was known, and has reduced
the numbers of cervical cancer cases with 50%. (Figure 2) According to the
Swedish guidelines all women between ages 23-60 are invited to cytological
screening every 3rd year. Women after 50 between 60 are invited with a 5
years interval. In the screening program cytological differences of cells in
transformation zone of cervix are examined using Papanicolaou-stained
(Pap) smear, that was introduced 19499. In Sweden 680 000 Pap smears are
taken every year10. Of them are more than 30 000 women diagnosed with
cytological abnormalities7. In 2008, 3 450 Swedish women were diagnosed
with cancer in situ of the cervix8. In 2007, 466 women were diagnosed with
cervical cancer, giving an incidence rate of 10.1 cases per 100 0009.
The cells from cervix are collected with a cytobrush, spread out on a microscope slide and stained to enable examination by the microscope. The
Pap-smear is graded depending on the appearance of the cells. The cervical
intraepithelial neoplasm (CIN) system was introduced 1973 and is a scale to
grade cytological differences. Cell abnormalities are graded from one to
three, processing from mild cervical intraepithelial neoplasia (CIN1) to more
moderate or severe degree of neoplasia and micro invasive lesions (CIN2 or
CIN3)11. In the early 1990 the Bethesda system was developed to reflect an
advanced understanding of cervical neoplasia and introduce uniform descriptive diagnostic histological terminology12. The system was modified in 2001
and today the cytological squamous cell differences are classified into four
categories: (1) ASC (atypical squamous cells), (2) LSIL (low-grade squamous intraepithelial lesions) (3) HSIL (high-grade squamous intraepithelial
lesions), and (4) squamous cell carcinoma13.
However, even if Pap smear is used as primary method for detection of
high-risk HPV induced cell changes, the method has its limitations. False
negatives rates as high as 20-30% have been reported and 8% of the samples
received are inadequate14. If the Pap smear shows abnormal findings but the
patients do not have gross cervical lesions, they are often evaluated by colposcopy and colposcopy-directed biopsy. Colposcopy is an optical examination of the mucous membrane in the cervix and helps to establish where the
cell change has occurred and how progressive it is. A biopsy is taken from
the area suspected to be the origin for an examination with microscope. In
the biopsy the presence of HPV DNA or RNA can be shown by in situ hybridization with probes labeled with either radioisotopes or chemically reactive ligands, which are detected by autoradiography, fluorescence or a detection of color reaction. The biopsy is as well examined for dysplasia. If classified as CIN2 or higher (CIN2+) conization is performed, where the mucosal
membrane together with the surrounding tissue in the transformation zone is
removed with a carbon dioxide (CO2) laser, a surgical knife or loop electro15
surgical excision procedure (LEEP). In 95% out of all cases one of the mentioned procedures is necessary to remove and cure cell changes15.
So far the primary diagnostic tools have been cytology and histology, but
new molecular methods detecting HPV DNA in clinical samples are rapidly
replacing cytology as the primary test in screening. The US Food and Drug
Administration (FDA) in 2014 approved HPV DNA to be used for testing of
women age 25 and older. If the woman is positive for HPV 16 or HPV 18 a
colposcopy exam is conducted. Women positive for any of the other 12 high
risk HPVs should have a Pap test to determine the need for a colposcopy16.
Previously only women who’s Pap smear was classified as atypical squamous of unknown significance (ASCUS) were reexamined with an HPV
DNA test17. The importance of HPV testing as support to the screening program has been evaluated in several studies18, 19. The National Board of
Health and Welfare in Sweden are suggesting changes in the Swedish
screening systems where HPV DNA test is recommended as the primary
screening test from 2017. In women age 23-29 a cytology test every third
year should still be considered, because there is no proof that HPV DNA test
would be of a higher effectiveness. This age group has an overall higher
prevalence of HPV infections. That could lead to an unnecessary evaluation
due to that the infection is often self-healing. However, women age 39-49
should be invited every third year for a HPV DNA test because of higher
incidence of cell changes and cervical cancer. In the age group 50-64 the
examination should as well be with an HPV DNA test but every seventh
year. This since the prevalence of cell changes and cervical cancer are not
that common and that the HPV DNA tests have longer duration than the
cytological test20.
Which of the different tools for diagnosis of cervical cancer is to prefer
based on the knowledge so far? Studies comparing the Pap-smear and HPV
DNA testing indicate that HPV DNA testing has better sensitivity and negative predictive values, as well as specificity18, 19, 21, 22. At present the exact
HPV types are not reported clinically, they are mainly used in large epidemiology and research studies. Still, HPV testing can be of high value, due to
that not only the presence of HPV but also the viral load is noted, which has
been proposed as a marker for persistent infection and risk of development
of cervical lesions23.
16
Prevention of cervical cancer
Screening for cell changes and prevention of HPV infection are both important strategies in cervical cancer prevention. The possibility of treating
the disease at an early stage often disrupts the process of neoplasia.
Preventing the HPV infection can prevent cervical cancer. Today this can
be achieved by usage of the HPV prophylactic vaccines. Gardasil® and Cervarix®, both consisting of the major viral coat protein L1, assembled into
macromolecular structures are very similar to the wild-type viron. Still the
vaccines are not identical. Gardasil® that was introduces in 2006 is quadrivalent, protecting the patient against HPV-6, -11, -16 and -18, targeting females 9-26 years of age24. The vaccine has also the potential of preventing
genital warts25. In 2008 a second vaccine was introduced, named Cervarix®
and is a bivalent vaccine protecting against HPV-16 and -18. It is recommended for use for females aged 10-25 years and is believed to protect
against anogenital warts caused by HPV, precancerous lesions and cervical
cancer26. In a large phase III trail conducted for four years with randomized
placebo-controls, both vaccines were shown to be well tolerated and effective in preventing the disease caused by each vaccine HPV type27, 28. In natural infection and vaccination, an immune response and memory of that response protects against infection and disease on re-encounter with the pathogen. It is known that it may take 25-30 years from that a woman is infected
with a carcinogenic HPV type until invasive cervical cancer develops29, 30.
The first pre-invasive cervical lesions are detectible approximately five to
seven years after a carcinogenic HPV infection29. However, cohorts have
been followed for around 10 years for the quadrivalent vaccine and 8.4 years
for the bivalent vaccine31. Where the data suggest that protection is likely to
be long lasting32. However, the issue still remains that the long-term protection is unknown; therefore long-term follow-up studies are necessary.
Today, these two HPV vaccines are approved in more than 100 countries.
The Swedish National Board of Health and Welfare (NBH) decided to include the HPV vaccine in January 2010 in the childhood vaccination targeting girls at age 10-1233. The introduction of the HPV vaccine was approved
only if an extensive follow-up program was implemented. The World Health
Organization (WHO) has also recommendation of follow-up for coverage,
safety and population effectiveness and the aim is to ensure that the screening program continues to be at least as effective as today34.
Scientific evidence and the results of the large phase III randomized control
trial (RCT) indicate that if the HPV vaccines are introduced to girls and
women before the first expose to the HPV virus, it will prevent disease
caused by the HPV types included in the vaccine. Still it is very central to
remember that even if a woman is vaccinated it is important to attend the
screening program. The vaccine protects from the HPV types included, but
17
there are several other high-risk HPV types that can cause cancer. Another
issue is the length of immune memory, up to ten years is known, but what
happens after that? However, if we would like to monitor the population to
see if the prevalence of the HPV types included in the vaccine decrease, it is
necessary to use type specific HPV tests in the screening program.
Human Papillomavirus
In 1976 professor Harald zur Hausen suggested the human papillomavirus to
be the causative agent of cervical cancer35. This was also identified in the
beginning of the 1980s in human genital warts and cervical biopsies36. In
2008 he was awarded with the Nobel Price in Medicine for his discovery.
Papillomaviruses are highly diverse, belonging to the family Papillomaviridae and occur in most mammals and birds. To date over 100 types have been
detected in humans and described based on isolation of the complete genomes. About 40 HPV types can infect the epithelial and mucosa lining of
the anogenital tract and other areas37. The genital and mucosal types belong
to the taxonomy genera alpha-HPV that also contains some cutaneous types,
which cause warts38.
Molecular epidemiologic studies have shown that some types of HPV’s,
around 18 of them39, are the principal cause of invasive cervical cancer and
cervical intraepithelial neoplasia40. Genital HPV types are divided in lowrisk (LR), which are found mainly in genital warts, and high-risk (HR) types,
which are associated with invasive cervical cancer41. In 2005 IARC (WHO
International Agency for research on Cancer) concluded that HPV types 16,
18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, 73 and 82 are carcinogenic in
humans42. Of them HPV 16 and 18 are responsible for around 90% of all
cervical cancer cases40. It is also known that infections with more than one
HR-HPV type can occur and represents a possible reason for development of
the cervical cancer43. Case–control studies, case series, and prevalence surveys have shown that HPV DNA can be detected in 90–100% of all cases.
However, there have been some HPV negative cervical cancers reported.
These finding are believed to be due to an artifact in the current detection
methods, or potentially due to loss of HPV DNA during the progression to
cervical cancer44. In contrast, the prevalence of HPV DNA in women identified as suitable epidemiological controls is between 5–20% in cervical cell
samples45. Based on this infection with at least one oncogenic high-risk HPV
type is regarded as a necessary but not sufficient cause of cervical cancer.
HPV is a common sexually transmitted infection and every second women
will during here lifetime get exposed and infected with HPV46. The prevalence in sexually active young women ranges from 20-46%47 and decreases
with age, but the HPV prevalence is still high in older women and the major
18
risk factor for cervical cancer development48-50. However, the majority (7090%) of women infected will clear the HPV within 12 to 30 months51-53
HPV is a circular double-stranded DNA virus with a size of 8 kb54. (Figure 3) The HPV genome consists of 3 general regions; an upstream regulatory region (URR) containing sequences that control viral transcription and
replication, an early region containing open reading frames (ORFs; e.g. E1,
E2, E4, E5, E6 and E7) encoding proteins that are involved in multiple functions like trans–activation of transcription, transformation, replication, and
viral adaption to different cellular environments, and a third region called
late region that is coding for the L1 and L2 capsid proteins that form the
structure of the viron and make the viral DNA packaging and maturation
possible38,55.
1(3,2
0(,,,
-0
.(,,,
.(,,,
,
!
/(,,,
!
#!
!
$
!
!
!% & )!
Figure 3. A schematic drawing showing the Human Papillomavirus (HPV) genome.
For HPV to successfully infect the host, it requires presence of epidermal
and mucosal epithelial cells that are still able to proliferate56. In the basal cell
layers the virus only express the early genes (E5, E6 and E7), leading to
enhanced proliferation of infected cells and their lateral expansion. The next
step is to infect the suprabasal layer, were the virus express the late genes,
initiating replication of the circular viral genome and formation of the structural proteins. As the virus reaches the upper layers of the epidermis, or mucosa, complete viral particles are assembled and released. In 1989, Munger
et al. showed that only the E6 and E7 genes of HR-HPV types were able to
immortalize human cells in tissue culture57. The E6 and E7 genes are consistently expressed in malignant tissue, and inhibiting their expression blocks
the malignant phenotype of cervical cancer cells58. This is partly achieved by
19
the E6 interaction with p5359 and E7 with RB60, leading to inactivation of
these tumor suppressor pathways.
Additional risk factors
What is a risk factor? A risk factor is anything that changes your chance of
getting a disease, in this case cervical cancer. However, having one or several risk factor does not have to mean that you will develop the diseases. HPV
genital infection is common and HR-HPV is known to be the major risk
factor for cervical cancer development. Approximately 70-90% of all infected women will clear the infection and only a small fraction of HPV infected
women will develop cervical cancer46. Therefore other risk factors beside the
HPV infection must also be involved in determining the outcome of an HPV
infection.
Environmental risk factors
Over the past years, a number of additional environmental risk factors have
been evaluated and associated with the risk of transition from a cervical
HPV infection to cervical malignancy. However, some of these are assumed
to only reflect the risk of acquiring HPV infection such as number of sexual
partners because the HPV is spread through genital contact61, 62. Other risk
factors associated with HPV infection is an early sexual debut (<18 years),
earlier age at first full-time pregnancy (<18 years), high parity63, use of combined hormonal oral contraceptives for longer than 5-years64 as well as previous infections with sexually transmitted diseases. Such as human immunodeficiency virus (HIV), herpes simplex virus 2 and Chlamydia
trachomatis61,65. Use of tobacco, increases the risk of squamous cell cervical
cancer66. The risk rises with quantity of cigarettes smoked per day and number of years smoked61. The explanation could be that smokers are less able to
clear the HPV infection or that smoking causes cancerous progression in
HPV infected cells66.
Genetic risk factors
Another reason why some women develop cervical cancer, while other clear
the HPV infection, could be their genetic constitution. In 1999, Magnusson
et al. reported a significant familial aggregation for cervical cancer. When
comparing incidence of dysplasia and CIS in relatives of women with disease and in controls, they saw that biological relatives have a doubled risk of
developing cervical cancers compared to non-relatives. This provides strong
epidemiological evidence for a genetic link to the development of cervical
cancer67. Later on Magnusson et al. studied adoptive and full mothers, as
20
well as full and adoptive sisters of cases with cervical cancers and estimated
the relative importance of shared genes to 27% and shared familial environments to 2%68. This demonstrates that development of cervical cancer depends to a significant degree on inherited genetic factors. Several other studies have also pointed out the importance of genetic risk factors in cervical
cancer, where first and second-degree relatives and full and half-sibling were
studied. Estimates shows that the familial risk has a heritable component of
71% and an environmental component of 29% in first and second degree
relative, and a 64% heritable component and 36% environmental component
in full and half-siblings69, 70. As mentioned there is large variation in the
susceptibility to developing cervical cancer among women exposed to HRHPV. This could reflect a difference in host susceptibility to either exposure
of HPV, infection following exposure, persistence of the infection or sensitivity to viral oncoproteins71.
Genes of the immune response system have been studied to investigate
the potential association with cervical cancer as well as effect on the susceptibility toward HPV infection and the persistence of the infection. Studies
have been focused on the major histocompatibility complex (MHC) region;
particularly on the human leukocyte antigen (HLA) class II that are a part of
the MHC region72 and several associations have been identified73-76. The
genes found in this region are very polymorphic giving rise to large variety
of expressed MHC molecules. The function of the MHC is to display antigen
peptides on the cell surface to initiate the cell-mediated immunity (CMI).
HLA class I (HLA-A, B and C) are expressed by nucleated cells and present
intracellular antigens to CD8+ T cells where HLA class II (HLA.DR, DQ
and DP) are expressed by dendritic cells, B cells and macrophages and present phagocytosed extracellular antigens to CD4+ T cells. (Figure 4) So in
conclusion a protein can only induce an immune reaction if the MHC molecules present the peptides. If the MHC molecules were absent it would lead
to lack of CMI activation.
#$
!
"!"
Figure 4. Schematic overview of the MHC region on chromosome 6.
21
However, several other candidate genes that are part of different pathways
have been suggested to influence cervical cancer. Genes found in the MHC
regions except of the HLA such as TNF77, LTA78, TAP1 and TAP279. There
are as well other genes in the immunological pathways that have been proposed for instance IL-1080, CCR-281, IFNγ82, KIR83, CD8384, Fas and Fas
ligand85. In 2009 Wang et al. investigated 92 polymorphisms in 49 immune
response and DNA repair genes where they suggested association of the
DNA repair gene Fanconi anemia complementation group A (FANCA) with
cervical cancer susceptibility.86 Following year in 2010 they investigated
several different pathways based on their hypothesized function such as
DNA repair, viral infection and cell entry. Among the different associations
the two genes transmembrane channel-like protein (TMC) 6 and 8 were of
particular interest due to their previous association with Epidermodysplasia
Verruciformis (EV). That is a disease characterized by increased sensitivity
to cutaneous HPV infections and high risk of squamous cell carcinoma of
the skin87. In summary, many studies are found suggesting the immune response involvement in the cervical cancer development as well as nonimmune response genes. Could it be that it is a combination of genes affecting the outcome of an HPV infection rather than single loci? (Figure 5)
Figure 5. A schematic drawing of the risk factors for cervical cancer development.
(From Emma Ivansson)
22
Genetic analysis of complex diseases
Epidemiology studies causes, distribution and control of disease in populations as well as the relationship between causes and effects. Genetic epidemiology focuses on studies of genetic predisposition to disease and the joint
effects of genetic and non-genetic (environmental) effects on disease risk.
This type of research aims to identify links between disease and the factors
that increase the risk of disease. Depending on what kind of scientific question you want to answer, different study approaches are used. The most
common study designs are outlined below.
Cohort vs. Case-Control study
Two basic types of study designs in genetic epidemiology are cohort and
case–control studies. Cohort study examines one or more health effects of
exposure to a single agent. Initially healthy subjects are followed over time
to determine the incidence of health outcomes. Classical case–control study
examines a single disease in relation to exposure to one or more agents. Here
the cases are subjects’ affected by the disease of interest, and the controls are
healthy subjects from the investigated population. The purpose of the control
group is to provide information on the exposure distribution in the population that gave rise to the cases. Researchers obtain and compare exposure
histories of cases as well as controls.
Familial aggregation
Familial aggregation studies evaluate traits, behaviors or disorders within
one family that could be caused by genetic or environmental factors88.
Segregation study
Segregation studies are performed to evaluate whether the pattern of disease
manifestation in families fits to a particular type of inheritance, usually if the
inheritance is dominant or recessive.
Linkage study
In genetic epidemiology, analysis of linkage and association are the two
most common means of testing for genetic loci influencing risk of disease.
Linkage analysis identifies the location of genes that affect the risk of developing a genetic disease. Information on the genetic recombination events in
family-based pedigree structures is the basis of linkage studies. This approach has been shown to be very powerful for simple Mendelian diseases
23
with rare, severe, and high-risk mutations89. The disadvantages of linkage
analyses are that variants with fairly small or moderate effects are not likely
to be detected. Also, if multiple genes are involved in different families,
linkage analyses may not be able to detect them. In complex diseases both
multiple genes and environmental, each with small or moderate effects, as
well as interaction among them, will impact risk of disease and linkage analysis will not be enough to detect these. In this case association studies may
be a better choice.
Association study
Association studies are comparable to case-control studies, the difference is
that the disease associated “exposures” that the researcher is interested in to
identify are different genetic alleles88. Association studies aims to evaluate
the correlation between genetic variants and a disease phenotype in a population. Instead of assessing the recombination events, the linkage disequilibrium (LD) is investigated. LD describes a situation in which some combinations of alleles or genetic markers occur more or less frequently in a population than would be expected from random formation of haplotypes based on
the alleles frequencies. If an association is observed, a particular allele,
genotype or haplotype will be found more often, or less often, than expected
by chance among individuals carrying the trait. Basically the frequencies of
variant alleles among affected individuals are compared to unaffected individuals90.
There are two possible study designs in association studies. The commonly used one is a case-control study were the distribution of genotype or allele
frequencies at a locus in affected subjects is compared with that in unaffected controls, and the statistical analysis is testing if there is significance difference between the two studied groups. If a difference in the frequency of a
genotype or allele is observed between the two groups studied, it indicates
that the locus is involved in disease susceptibility, or could be in LD with a
causal polymorphism. A population-based control study, which is based on
sampling cases and unrelated controls, has the advantage that it is easy to
collect sufficiently large sample sizes and it therefore has a high statistical
power to detect small or moderate effects. However, the drawback with this
study design is that the genotype and haplotype frequencies may vary between ethnic or geographic populations. To avoid this, the cases and controls
most therefore be matched for ethnicity or geographic origin, as well as other
important factors, or else false positive associations can occur because of the
cofounding effects of population stratification.
The second approach is family-based association studies, were parents or
unaffected siblings are used as controls for the case, which is their affected
offspring/sibling. This kind of study is used to avoid the potential cofounding effects of population stratification, mentioned above, giving the ad24
vantage that all individuals in a family pedigree have a common genetic
background. However, avoiding one problem leads to another; family-based
studies are often small because of the difficulty to collect large samples of
well-characterized families and this often leads to lower statistical power91.
Also, family-based studies are prone to the consequences of a common
shared environment. However, comparing the power of the population-based
and the family-based approach in a meta-analysis study, no difference was
observed between them92.
The work presented in this thesis is based on two very common types of
association studies. The first one is the candidate gene study approach93,
which focuses on investigating associations between genetic variation within
pre-specified genes of interest and phenotypes or disease states. Candidate
genes are selected based on prior knowledge of the genes biological functional impact on the trait or disease question, which is the major difficulty
with this approach, or because they have been implicated in disease in previous studies94. The advantage of candidate gene studies is that they are better
suited for detection of genetic effects for common and more complex diseases where the risk associated with any given candidate gene is relatively
small89, 95. The second type of design is the genome-wide association study
(GWAS), where the entire genome is searched for genes that are association
to the disease93. Here 100 000 to million of common genetic variants are
investigated in large set of patients and controls to search for variants associated with a trait. GWAS most commonly focus on finding associations between single nucleotide polymorphisms (SNPs) and disease state96. The
scanning of the genome can be done using various study designs, such as
case-control studies, cohort studies and clinical trials. The drawback of
GWAS is that large samples sizes are necessary for the scanning, due to the
very high number of statistical tests of associations (minimum one per SNP)
that are performed, requiring a very stringent threshold of statistical significance97. Associations discovered by GWAS raise additional questions, particularly because observed effects are typically very small98, and the SNPs
found to be associated represents markers and further investigation are required to identify the true causal variants99.
Technologies in complex disease analysis
Genotyping methods
In a candidate gene study, the SNPs studied for the association with a particular disease can be genotyped using a variety of different typing methods.
The SNPs could be genotyped using a Taqman assay, based on the use of
real-time polymerase chain reaction (PCR) and the 5´nuclease assay. The
25
technique is used to amplify and simultaneously distinguish the two alleles at
the target gene. The assay includes PCR primers that will amplify the targeted site and a set of two probes that are complementary to the targeted sequence but differs at the site where the SNP is found. The probes are allelespecific and are labeled with different reporter dyes. In addition both probes
are labeled with a quencher dye that will block (quench) the fluorescence
from the allele-specific probe. Upon hybridization of the allele-specific oligonucleotide to its complementary sequence, the oligonucleotide is degraded
when the polymerase extends the primer sequence during the PCR, fluorophore is released and the florescence is released, collected and used for allele
calling.
Figure 6. Simplified drawing of the Taqman® genotyping assay. Modified from
Life Technologies100.
The Illumina Goldengate genotyping assay101 allows several loci to be examined in a single reaction through specific extension and amplification steps.
It genotypes the targeted region directly on the genomic DNA and does not
require prior PCR amplification of the genotyping target. The procedure of
the assay works so that the DNA sample is activated for bind to paramagnetic particles. In the hybridization step the assay oligonucleotides, hybridization buffer, and paramagnetic particles are combined with the activated
DNA. For each SNP, three oligonucleotides are designed; two are allelespecific oligos (ASOs) and the third hybridizes 1-20 bp downstream of the
SNP site, and is the locus-specific oligo (LSO). All three oligos contain sequences that are recognized by a set of universal PCR primers. In addition,
the LSO contain a unique sequence complementary to a particular bead type.
26
In this procedure, the assay oligonucleotides hybridize to the genomic
DNA sample that is bound to the paramagnetic particles, upon which a polymerase extends the ASO. The polymerase fills the gap between the ASO
and LSO and drops of when reaching the LSO. The nick between the extended sequence of the ASO and the LSO is sealed by a DNA ligase, forming PCR templates containing the genotype information present at the SNP
site that will in the next step be amplified with universal PCR primers. The
universal primers are labeled with different dyes. After thermal cycling the
dye-labeled DNA products are hybridized to their complement bead type
through their unique address sequence found in the LSO. At the end the Illumina BeadArray Reader is used to analyze the fluorescence signal from the
Array Matrix or BeadChip102. This technology is scalable from genotyping a
single SNP to several thousands.
Sanger Sequencing
To determine the nucleotide order on a particular DNA fragment it is necessary to determine its sequence. A common technique that has been used for
several decades is Dideoxy Sanger sequencing103. The method is based on
hybridization of a sequencing primer to a PCR product. In the next step deoxynucleotide triphosphates (dNTPs) and dideoxynucleotide triphosphates
(ddNTPs) that lack a hydroxyl group at the 3’ end are incorporated. When
the polymerase incorporates a ddNTPs the elongation of the synthesized
strand is terminated, resulting in a range of fragments that differ in length by
one base and are terminated at the point of incorporation of the base ddNTPs
in question. Today, the ddNTPs are fluorescence labeled and the fragments
are size separated using capillary electrophoresis (CE)104, 105. The human
genome was sequenced by fragmentation of DNA, which was cloned into
bacterial and yeast vectors and sequenced using the Sanger technology106.
Figure 7. Chromatogram demonstrating the sequencing results of a part of the
TMC6 gene.
Next Generation Sequencing
The last five years, the Sanger method is being replaced by Next generation
sequencing (NGS) technologies, mainly for large-scale genome analysis.
27
NGS is a collective group of methods characterized by massively parallelization of the sequencing reactions, and therefore also called massively parallel
sequencing (MPS)107. There are different NGS platforms such as Illumina
HiSeq/MiSeq, Life Technologies SOLiD and Ion Torrent/Ion Proton, and
Pacific Biosciences RSII (Table 1). What separates NGS from Sanger sequencing is that NGS methods generate millions to billions of sequence in a
single run. Most of the NGS technologies generate reads between 100-400.
An exception is the Pacific Biosciences RSII with have a read length of 1040 000 bp108. Having an intention to lower the cost of DNA sequencing
compared to the costs of standard dye-terminator methods109. NGS platforms
differs in the achievable read length, accuracy, reads per run, time per run, as
well as the calculated cost per 1 million bases110, 111.
In this thesis two of the NGS sequencing platforms have been used,
namely SOLiD sequencing and Pacific Biosciences RSII. The SOLiD (Sequencing by Oligonucleotide Ligation and Detection) sequencing platform
was introduced in 2008112,113 and uses probes with dual-base encoding and
ligation sequencing. The Pacific Biosciences RSII was introduced in 2009
and is a single-molecule real-time (SMRT) sequencing system114.
In the first step DNA is randomly fragmented and universal nucleic acid
adapters are ligated to the ends. This is generally denoted the sequencing
library. The adapters are used to make the DNA fragments attach to a solid
surface as well as to define the site of the sequencing start. These adapters
look different depending on sequencing method. Linear adapters are used in
SOLiD systems, while so-called bubble adapters are used in the Pacific Biosciences RSII system. In the second step emulsion-PCR is applied in the
SOLiD and Ion systems, where a single enrichment bead and sequencing
library fragment are combined inside an aqueous reaction bubble. A PCR is
then performed inside the bubble resulting in clonal copies of the template
on the surface of the bead. Beads with clonal copies of the DNA are attached
to a glass slide. In the Pacific Biosciences RSII system, the biotinylated
DNA polymerase binds to the bubble-adapted template and this complex is
immobilized on the bottom of a zero mode wave-guide (ZMW) (small well).
The sequencing is performed using DNA polymerase directed synthesis with
fluorescent nucleotides. In the DNA synthesis each nucleotide is labeled by a
fluorophore which gets cleaved of during the incorporation event. Pacific
Biosciences RSII is based on observing the incorporation of nucleotides in
real-time in parallel by recording the light pulses emitted during each incorporation event114. In the SOLiD technology, sequencing is performed by
ligation of fluorophore-labeled oligonucleotides that correspond to a specific
base in the DNA molecule112,113.
28
Table 1. Different Next Generation Sequencing (NGS) platforms
Platform
Company
Sequncing
library
Support
Featured Sequencing
generation
Detection
method
GS FLX
454 Life Science,
Linear adapter
Picotier plate
Emulsion PCR
Synthesis
Pyrosequencing
Linear adapters
Flow cell
Bridge PCR
Synthesis
Roche
Genome
Solexa,
Analyzer
Illumina
SOLiD
Applied
Flurophore labeled
reversible terminator
nucletides
Linear adapters
Flow cell
Emulsion PCR
Ligation
Flurophore labeled
oligonucletide
Biosystems
probes
CGA
Complete
platform
genomics
Circular adapters
DNA
Rolling circle
nanoball
amplification
Ligation
probes
arrays
PacBio RS
Pacific bioscience Bubble adapters
Flurophore labeled
oligonucletide
Zero mode
Single
Synthesis,
Phospholinked
waveguide
molecule
Realtime
flurophore labeled
nucleotide
29
Present investigation
Aim
The aim of this thesis is to study the association of genetic variation with
development of cervical cancer, using both the candidate gene and GWAS
approaches.
Paper I
Evaluation of Fanconi Anemia genes FANCA, FANCC and FANCL in cervical cancer susceptibility
Background
The FANC gene family consists of 13 FA complementation groups (FA-A, B, -C, -D1, -D2, -E, -F, -G, -I, -J, -L, -M and –N). Disruption of the function
in any of them causes Fanconi Anemia (FA)115. FA is rare autosomal recessive DNA repair deficiency disorder, characterized by aplastic anemia and
cellular hypersensitivity to DNA cross-linking agents116. The FA patients are
highly susceptible to different cancer types, including cervical cancer and
other HPV associated tumors117. The increased risk of cancer development
has led to the suggestion that FA genes could play a role in recognition and
repair of DNA damage. Genetic variation in DNA repair genes could affect
identification and repair of DNA damage caused by the action of E6 and E7
HPV oncoproteins118.
In paper I we are evaluating the association of 81 tagSNPs in three of the
thirteen mentioned complementation groups of the FANC pathway: FANCA, FANCC and FANCL. The genes were selected based on earlier reports
of their function or reported association with cervical cancer86.
Results and discussions
TagSNPs were designed from the set of common SNPs genotyped in the
Cacausian (CEU) population samples of the international HapMap project
using the Tagger algorithm in the Haploview software119, 120. Genotypes were
generated for 782 cases (CINIII and ICC) and 775 controls. A nominal asso30
ciation with cervical cancer was observed in FANCA and FANCC gene.
However, none of these associations remained significant after Bonferroni
correction for multiple testing, implying that the study gives no evidence for
an association of the FA genes with cervical cancer in the Swedish population. In a previous study Wang et al. (2009) reported an association of variation at FANCA with cervical cancer in the Costa Rican population86. The
different results between our study and that by Wang could be due to haplotype differences between populations, resulting in different tagged variants.
The Wang et al results may also represent a false positive finding, since no
replication of their findings has been published. TagSNPs are selected as to
cover the genetic variation in the desired genes, but we cannot be certain that
all SNPs are annotated in the HapMap project. Therefore the causal SNP
could be missed and no association with cervical cancer will be found. This
problem can be assessed with sequencing that would capture all genetic variation in the genes of interest.
In conclusion, the pathway still remains biologically relevant and we cannot exclude the possibility that more extensive genetic analyses of the FA
genes would identify a contribution of the FA pathway to cervical cancer.
Paper II
Contribution of TMC6 and 8 (EVER1 and EVER2) variants to cervical cancer susceptibility
Background
The adjacent genes TMC6 and TMC8, also known as EVER1 and EVER2,
have previously been associated with a rare autosomal recessive disease
Epidermodysplasia Verruciformis (EV)121-123. The disease is characterized by
a severe susceptibility to infections by cutaneous HPV of the beta genus and
high risk of squamous cell carcinoma of the skin124. The association between
these two genes and the HPV sensitivity makes them interesting candidate
genes in cervical cancer.
In paper II we are evaluating the association of 22 tagSNPs in the genomic region of TMC6 and TMC8 genes with susceptibility to cervical cancer. The genes were selected based on earlier reports of the functional consequence or reported association with persistent HPV infections and cervical
cancer87.
Results and Discussion
The 22 tagSNPs were selected using data from the HapMap CEU population
in The International HapMap project and the Tagger algorithm in
31
Haploview119, 120. The tagSNPs were genotyped in two Swedish cohorts, in
total 2 989 cases and 2 281 controls117, 125. A pervious study by Wang et al.
investigated the region of interest and reported an association of TMC6 and
TMC8 genes with increased risk of cervical cancer in the Costa Rican population87. In the Swedish population we identified two tagSNPs that were
associated with cervical cancer, located in the intronic region of TNRC6C
(SNP rs2290907), and an adjacent gene upstream of TMC6 and TMC8 gene
(SNP rs16970849). We were not able to replicate the association of the two
associated SNPs from the Costa Rican population in the Swedish cohort.
These results may reflect the fact that we are comparing two ethnically distinct populations. In fact, only 13 tagSNP of the 22 studied here were common to both populations, and most of them vary in allele frequency between
the studies.
Our observation confirms the involvement of TMC6 and TMC8 in the
susceptibility of cervical cancer in the Swedish population as well. However,
further studies of the region are needed to establish the full spectrum of genetic variation, as well as to identify functional variants.
Paper III
Genome-wide association study of susceptibility loci for cervical cancer
Background
GWAS have been successfully used to identify genetic associations across a
wide range of traits and disease entities. In a GWAS the genome is searched
for SNPs that occur more frequently in cases than controls.
In paper III we performed the first GWAS for cervical cancer, in order to
further investigate the genetic basis of cervical cancer in the Swedish population.
Results and Discussion
After quality control the discovery data set included 632 668 SNPs that were
genotyped in 1 034 cases and 3 948 controls117. Testing each SNP individually using the log-additive model, we identified 55 SNPs to be associated
with cervical cancer. All of them were found in the MHC region at 6p21.3,
with the strongest association coming from SNP rs9272143 located between
HLA-DRB1 and HLA-DQA1. This SNP is located within an intergenic region with no obvious functional elements. Conditioning upon SNP
rs9272143 residual association was found at SNP rs2516448, adjacent to the
MICA gene. This gene encodes a membrane–bound protein that acts as ligand for NKG2D activating immunoreceptor. Further statistical conditioning
32
lead to residual association found at an SNP (rs3117027) located in a pseudo
gene of HLA-DPB2 that correlates with DPB1*0301, which has been found
to be associated with increased risk of cervical cancer126. However, further
investigations are needed to determine if the association is driven by
DPB1*0301. We could conclude that the three loci in the MHC region found
to be associated with cervical cancer are independent of each other. One of
the regions most strongly associated with cervical cancer is the MICA gene.
The membrane-bound MICA protein is generally absent in most cells but is
activated upon viral infections and reported to be expressed in epithelial
tumors127. The binding to MICA causes NKG2D to activate cytolitic response of gamma/delta T cells and natural killer cells against transfectants
and tumor cells expressing MICA128. Membrane-bound MICA therefore acts
as a signal during the early immune response against infection or spontaneously arising tumors. The tumor cells on the other hand releases soluble MICA to damage the responsiveness of tumor antigen-specific effector T-cells,
promoting tumor immune evasion as well as compromise the host resistance
to infections127, 129. The transmembrane domain (TMD) of the MICA gene
contains a triplet repeat microsatellite polymorphism consisting of four, five,
six, or nine repetitions of GCT (alleles designed A4, A5, A6, or A9, respectively) or five repetitions of GCT with one additional G nucleotide insertion
(allele designed A5.1)130. The insertion leads to a frameshift mutation that
results in a premature stop codon (TAA), causing a truncation of 10 amino
acids of the TMD as well as the hydrophobic 42-amino acid-long cytoplasmic tail. We found that the T (risk) allele of SNP rs2516448 is in perfect LD
with the A5.1 mutation; as well as the A5.1 allele is associated with less
membrane-bound MICA. The loss of 52 amino acids is believed to affect the
membrane expression of the MICA protein. This may affect the ability of the
cells to alert the immune system of HPV infection or neoplastic change,
leading to impaired immune activation and increased risk of tumor development.
In an independent set of 1 140 case and 1 058 controls we replicated the
association for two of the three SNPs. The effect for rs3117027 was just
above the significance level. We also imputed the genotype alleles for the
classic HLA alleles, and identified an association with cervical cancer with
one HLA class I allele (B*0702) and five HLA class II alleles (DRB1*1301,
DRB1*1501, DQa1*0103, DQB1*0603 and DQB1*0602), consistent with
previously reported HLA association with cervical cancer in European populations72-75, 117, 131. We could also identify DRB1*1301-DQA1*0103DQB1*0603 haplotype to be associated with reduced risk of cervical cancer
and B*0702 and DRB1*1501-DQB1*0602 haplotype with increased risk.
Adjusting for the effect of the classic HLA alleles, the three novel loci remained associated with cervical cancer.
Furthermore, our GWAS associations have been followed up in several
publications. It is known that multiple validations are of importance to con33
firm the GWAS findings and report the genotype-phenotype associations.
Chen et al. performed an additional replication of the three novel loci identified in our first study as well as more extended genotyping of the MICA
exon 5 microsatellite where the frameshift mutation (A5.1) is found. The C
allele of SNP rs9272143 was found to have a protecting effect on cervical
cancer and to be associated with higher expression of HLA-DRB132. This
SNP has been identified as a cis-expression quantitative trait locus (ciseQTL)133. That is loci regulating the levels of the mRNA expression134. The
T allele of SNP rs2516448 was associated with increased susceptibility to
cervical cancer and the same association was shown with MICA-A5.1. The
associations were consistent with our previous finding supporting the role of
HLA-DRB1 and MICA in the pathogenesis of cervical cancer. Despite the
findings in the GWAS no association was seen between the HLA-DPB2
variant SNP rs3117027 and cervical cancer132. However, in a later study
classical HLA-DPB1 alleles based on data from the MHC region were imputed. Allowing a systematic evaluation of the HLA-DP variants in the
Swedish population. The strongest association found was for SNP rs3117027
that seems highly correlated with SNP rs3129294 that may have a putative
regulatory function135.
In conclusion, our GWAS study found risk of HPV-induced CIS to be associated with different polymorphic variations in several regions of the
MHC. The results were also replicated in different cohorts, suggesting the
MHC region as the main host genetic susceptibility factors to cervical cancer. Playing a role in the immune recognition of HPV-infected or –
transformed cervical cells causing increased risk of tumor development.
Still, further studies using for instance genome-wide sequence-based association analysis could be of importance, to fully define the inherited basis of
cervical cancer.
Paper IV
Characterization of the TMC6 and TMC8 genes associated with cervical
cancer.
Background
Previous findings have shown that polymorphisms in TMC6 (EVER1) and
TMC8 (EVER2) are associated with increased risk of cervical cancer, or
susceptibility to HPV or other genital infections87, 136. As previously mentioned these two genes were first been associated with Epidermodysplasia
Verruciformis (EV)121-123. In paper IV we used a variety of different genetic
technologies to characterize the genomic region harboring the TMC6 and
34
TMC8 genes, in order to study their association with cervical cancer in more
details.
Results and Discussion
Using long-range PCR in combination with SOLiD sequencing, the distribution of SNPs across the TMC6 and TMC8 region were first identified in
pools of cases and controls137. Large allele frequency differences were seen
between pools for a number of SNPs, and the 10 SNPs with the largest difference were selected for replication. Replication of the top ten SNPs using
TaqMan genotyping assay resulted in that only one SNP (rs2926778)
showed a significant allele frequency difference between cases and controls.
This SNP is located between TMC6 and the upstream gene, TNRC6C, and
has not been included in the previous tagSNPs analysis by Castro et al
(2013)136. However the polymorphism is located in the intronic region and
the putative functional consequence is unknown. The SNP rs66468151 instead of three different clusters in the Taqman genotyping (homozygote for
A allele, heterozygote for AB alleles and homozygote for B allele) showed 5
different genotype clusters. Using predesigned Taqman CNV assays we then
identified that 3% of the cases have a copy number variation (CNV) in the
examined region. This corresponds to the region with the SNP with the highest allele frequency difference in the pooled sequencing. It is therefore likely
that the failure to validate the top ten SNPs is due to the complexity of the
DNA sequence, such as repeats or other structural variants in this region.
To increase our understanding of the region, we attempted to sequence the
region using Sanger sequencing103. This resulted in complete sequences for
some individuals, but a gap in individuals with the previously noted CNV.
We therefore used a second sequencing approach, based on hybridization
capture of 2-3 kb genomic fragments, and the long-read single-molecule
DNA sequencing of Pacific Biosciences RS II. A de novo assembly of the
sequencing reads was conducted resulting in a complete contig for the region, providing a more detailed understanding of the nature of the variation.
Together with the new hg19 we improved the assembly of TMC6 and TMC8
region and were able to determine the nature of the variation. In the region
where we previously have indications of a CNV, we found a 21 bp sequence
that was be repeated 15 times in the reference sequence and 18 times in the
de novo assembly. Knowing the exact region of the CNV we could identify
39 different alleles in our cervical cancer cases and controls. The alleles
range in size from 0 to 42 repeated units of the 21bp sequence. No significant (p>0.05) allele frequency difference was seen between cases and controls when testing all alleles. However, when comparing the group allele
denoted “0” versus the rest resulted in significantly lower frequency of this
allele in the cases as compared to controls. (p=0.05) This 21 bp sequence is
not found elsewhere in the human genome and does not contain any identi35
fied binding motifs, indicative of its function. At present, we are not able to
suggest a functional role of this repeat. Also, there might exist SNPs and
other types of variations in the region flanking the repeat, within the “0”
allele group that could be of functional importance. Further sequencing of
additional individuals using the capture and Pacific Biosciences method is
needed to address these questions.
In conclusion, our study have resolved the TMC6-TMC8 genomic region
previously associated with cervical cancer, and identified an association for
the SNP rs2926778 located between TMC6 and TNRC6C gene with cervical
cancer. We have also identified structural differences in the DNA sequence
of the TMC6 gene. Lack of repeat units was found to be associated with
increased risk of cervical cancer. However, further studies are needed to
understand the nature of these associations.
*/"/-100))) */"/.*)1+))
*/"/.*,**))
)
.
.
% &
%&
!
*
*
!
Figure 8. Schematic drawing of the study methods used.
36
%&
%&
1$
-#,+
0
Conclusion and future perspectives
Cervical cancer is the fourth most commonly diagnosed cancer, and the
fourth leading cause of cancer death, among women worldwide. In 2012,
528 000 new cancer cases and 266 000 deaths due to cervical cancer was
registered. Around 85% of new cancer cases occurred in the less developed
countries, where also almost nine out of ten (87%) cervical cancer deaths
occurred. This thesis is focusing on evaluating the genetic variation in the
human genome using the GWAS as well as different candidate genes, which
have previously been associated with cervical cancer or other disorders that
are of importance in infection by different HPV's as well as the progression
to cervical cancer.
The first, second and the fourth paper are all based on previous associations found with cervical cancer. The first study could not find any indication of association in the investigated Fanconi anemia (FA) genes with cervical cancer in the Swedish population. The second paper presents association of variation in the region of two genes; TMC6 and TMC8, earlier associated with the HPV-associated disease Epidermodysplasia Verruciformis
(EV) as well as increased risk of HPV associated cervical cancer. In the
fourth paper a new polymorphism was associated with cervical cancer in the
same region as studied in paper three. Upon sequencing of the complex genomic region of TMC6-TMC8, a CNV was identified, and the lack of this
repeat was found to be associated with increased risk of cervical cancer. In
the third paper three novel loci in the major histocompatibility complex
(MHC) region were found to be associated with the disease, highlighting the
importance of variation in this regions for the risk of developing cervical
cancer. In conclusion, the results presented point to the involvement of several genetic components in the development of cervical cancer. The genetic
risk factors that are identified to date have inadequate effect size, and only
explain a portion of the genetic susceptibility, which implies that further
investigations are needed to fully understand the contribution of genetic factors to this disease.
Previous studies have emphasized the importance of genetic predisposition in the disease. Magnusson et al. demonstrated a family clustering of
cervical cancer. He showed that the relative risk in developing cervical cancer was higher among biological mothers compared to the adoptive mothers
to probands. The same was noted for the biological sisters versus adoptive
sisters69. In 2000 Lichtenstein et al. demonstrated that the risk of developing
37
cervical cancer was 0% explained by genetic factors, 20% was explained by
shared environment and 80% was explained by non-shared environmental
factors. In 2006 a Danish group reported family clustering of cervical cancer
in both mono- and dizygotic tweens. They suggested that family factors such
as genes or/and shared environment are important in the development of
cervical cancer138. Taking the information together we are getting different
suggestions of the genetic importance in the development of cervical cancer.
A couple of years ago it was believed that GWAS would help us uncover
and explain some of the genetically diseases. Many of the GWAS of complex disease have indeed pointed out regions with associated genes in
breast139-141, colon142, 143 and prostate cancer144, 145 so there is no doubt that
GWAS have resulted in interesting candidate regions for some of the examined phenotypes. Still, most of the variation that has been identified so far
have small effect and only explain a small part of the heritability. Are we
missing something in our way of studying the human genome, are there still
variants that we have not discovered?
In 2008 the involvement of structural variation, such as copy number variation (CNV), was demonstrated in schizophrenia146. These findings have
lead to an interest in evaluating the potential importance of structural variation in several common human complex diseases. Next generation sequencing (NGS) technologies have given us an opportunity to study the human
genome in even more detail. Perhaps in the future these studies will help us
answer the question of genetic importance in the cervical cancer.
Could it be so that cervical cancer is an immunodeficiency disease? Several studies72, 73 including our GWAS147 have emphasized the involvement of
the immune system genes in the disease. In particular the MHC region, that
plays an important role in presenting foreign antigens to the immune system.
Failure in activating and developing an effective cell-mediated immune response to take care of the HPV infection could result in a persistent infection
as well as increased risk of cervical cancer development148.
It is of importance to remember that cervical cancer is a mixture of an infectious disease and a cancer. It differs from the majority of cancer since
viral infection of the HPV is an essential cause. Knowing this it is of high
importance to prevent the infection. Today two vaccines have been available
for almost a decade, which have proven so far to have a long lasting effect.
However, the issue still remains that the long-term protection is unknown;
therefore long-term follow-up studies are necessary. Another thing is that the
screening should never be allowed to get in the background due to the available vaccination. It is of importance to continue monitoring the prevalence
of different HPV types in infected women, due to that the vaccines only protect against HPV16 and 18 and we do know that there are several other oncogenic HPV types that could affect the cancer development.
38
Acknowledgements
Tack till min handledare Ulf Gyllensten för att jag fått doktorera hos dig.
Tack för all stöd, alla diskussioner och all vägledning. Du är otroligt duktig,
kunnig och påläst, men samtidigt en jordnära och fin människa.
Tack till min vän och kollega, Inger Gustavsson för all ditt stöd under de
här åren vi känt varandra. Alla intressanta diskussioner, råd inom forskningen men också resor och annat som gör livet underbart. Du är fantastisk, duktig, påhittig och kunnig. Jag är glad att jag träffat dig och hoppas att vi aldrig
tappar kontakten. Stor tack till Julia, Ann-Sofi & Pernilla för de trevliga
och roliga fikastunderna, för alla uppmuntrande ord under mitt avhandlingsskrivande. Tack till Malin och Tao för alla givande gruppmöten & Stefan
för att du alltid tar dig tid och svara på mina frågor.
Tack till alla på UGC för hjälpen i de olika projekten. Speciellt Inger J. för
alla svar på mina många funderingar, Anki & Carro för alla fragmentanalys
körningar, Ida för all hård jobb med sekvensering av mina två favoriter
TMC6/TMC8. Adam & Ignas för bioinformatiken.
Tack till Sara, Dijana & Mi för det vi en gång var, för alla fester, diskussioner & vår resa till Milano. Ett extra tack till Mi för alla uppmuntrande ord
under mitt skrivande. Till min Ghazal, min iranska själsfrände. Jag är glad
att jag har mött dig under min doktorandtid för jag vet att jag har hittat en
vän för livet.
Tack till mina gamla som nya vänner, för all stöd under dessa år & alla peppande ord de sista månaderna.
Tack till mina svärföräldrar Vesna & Miro, min mans syster Ivana med
respektive för alla stöttande ord.
♥ Jag vill tacka mina föräldrar Olivera & Ivo för all stöd, för alltid uppmuntrat och hjälpt mig. Jag älskar er så mycket för att ni gett mig ett perfekt
liv, ni är de finaste människorna jag känner och den jag är idag är tack vare
Er. ♥ Mina syskon Darija med familj & Anto, ni har förgyllt mitt liv sedan
den dagen ni föddes. Tack för all ert stöd och alla peppande ord på vägen till
mitt mål. Trojka=Zauvijek. Voli vas seka!
39
“Vi ne birate svoju obitelj. Ona je Božji dar vama, kao što ste i vi njima.”
Desmond Tutu (A vi ste moj najljepsi dar od Boga)
Sist men inte minst vill jag tacka min alldeles egna dream team. ♥ Min man
Anto, min själsfrände, min klippa, min stora kärlek. Tack för all ditt stöd,
din ork vid mina nervösa utbrott, dina uppmuntrande och lugna ord. Jag älskar dig för allt vi klarat oss igenom, för det finns inget vi två inte klarar av
tillsammans. ♥ Noa & Leon, ni är mammas hela värld. Ni är det bästa och
det viktigaste jag uppnått, ni är min stolthet, ni fyller mitt liv med glädje,
skratt och kärlek. Jag älskar Er mer än allt på denna jord.
40
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Cavallo, F., De Giovanni, C., Nanni, P., Forni, G. & Lollini, P.L.
2011: the immune hallmarks of cancer. Cancer Immunol
Immunother 60, 319-26 (2011).
Moscicki, A.B., Schiffman, M., Kjaer, S. & Villa, L.L. Chapter 5:
Updating the natural history of HPV and anogenital cancer. Vaccine
24 Suppl 3, S3/42-51 (2006).
Burghardt, E. & Ostör, A.G. Site and origin of squamous cervical
cancer: a histomorphologic study. Obstet Gynecol 62, 117-27
(1983).
WHO | Human papillomavirus (HPV) and cervical cancer. WHO
(2014).
Livmoderhalscancer. (2015).
Bray, F. et al. Trends in cervical squamous cell carcinoma incidence
in 13 European countries: changing risk and the effects of screening.
Cancer Epidemiol Biomarkers Prev 14, 677-86 (2005).
Vizcaino, A.P. et al. International trends in the incidence of cervical
cancer: I. Adenocarcinoma and adenosquamous cell carcinomas. Int
J Cancer 75, 536-45 (1998).
Jemal, A. et al. Global cancer statistics. CA Cancer J Clin 61, 69-90
(2011).
PAPANICOLAOU, G.N. Cytologic diagnosis of uterine cancer by
examination of vaginal and uterine secretions. Am J Clin Pathol 19,
301-8 (1949).
Dillner, J. Cervical cancer screening in Sweden. Eur J Cancer 36,
2255-9 (2000).
Richart, R.M. Cervical intraepithelial neoplasia. Pathol Annu 8, 30128 (1973).
The Bethesda System for reporting cervical/vaginal cytologic
diagnoses. Report of the 1991 Bethesda Workshop. Am J Surg
Pathol 16, 914-6 (1992).
Solomon, D. et al. The 2001 Bethesda System: terminology for
reporting results of cervical cytology. JAMA 287, 2114-9 (2002).
Burd, E.M. Human papillomavirus and cervical cancer. Clin
Microbiol Rev 16, 1-17 (2003).
Årsrapport_2014.pdf. (2015).
Press Announcements - FDA approves first human papillomavirus
test for primary cervical cancer screening. (2015).
41
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
42
Wright, T.C. et al. 2001 Consensus Guidelines for the Management
of Women with Cervical Cytological Abnormalities. J Low Genit
Tract Dis 6, 127-43 (2002).
Clavel, C. et al. Human papillomavirus testing in primary screening
for the detection of high-grade cervical lesions: a study of 7932
women. Br J Cancer 84, 1616-23 (2001).
Petry, K.U. et al. Inclusion of HPV testing in routine cervical cancer
screening for women above 29 years in Germany: results for 8466
patients. Br J Cancer 88, 1570-7 (2003).
Livmoderhalscancer, screening med cytologi och HPV-test. (2015).
Cuzick, J. et al. HPV testing in primary screening of older women.
Br J Cancer 81, 554-8 (1999).
Schiffman, M. et al. HPV DNA testing in cervical cancer screening:
results from women in a high-risk province of Costa Rica. JAMA
283, 87-93 (2000).
Josefsson, A.M. et al. Viral load of human papilloma virus 16 as a
determinant for development of cervical carcinoma in situ: a nested
case-control study. Lancet 355, 2189-93 (2000).
Quadrivalent vaccine against human papillomavirus to prevent highgrade cervical lesions. N Engl J Med 356, 1915-27 (2007).
Garland, S.M. et al. Quadrivalent vaccine against human
papillomavirus to prevent anogenital diseases. N Engl J Med 356,
1928-43 (2007).
Brown, D.R. et al. The Impact of Quadrivalent Human
Papillomavirus (HPV; Types 6, 11, 16, and 18) L1 Virus-Like
Particle Vaccine on Infection and Disease Due to Oncogenic
Nonvaccine HPV Types in Generally HPV-Naive Women Aged 16–
26 Years. (2009).
Kjaer, S.K. et al. A pooled analysis of continued prophylactic
efficacy of quadrivalent human papillomavirus (Types 6/11/16/18)
vaccine against high-grade cervical and external genital lesions.
Cancer Prev Res (Phila) 2, 868-78 (2009).
Paavonen, J. et al. Efficacy of a prophylactic adjuvanted bivalent L1
virus-like-particle vaccine against infection with human
papillomavirus types 16 and 18 in young women: an interim analysis
of a phase III double-blind, randomised controlled trial. Lancet 369,
2161-70 (2007).
The impact of quadrivalent human papillomavirus (HPV; types 6,
11, 16, and 18) L1 virus-like particle vaccine on infection and
disease due to oncogenic nonvaccine HPV types in sexually active
women aged 16-26 years. J Infect Dis 199, 936-44 (2009).
Evaluating human papillomavirus vaccination programs. Emerg
Infect Dis 10, 1915-23 (2004).
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
Sustained immunogenicity and efficacy of the HPV-16/18 AS04adjuvanted vaccine: up to 8.4 years of follow-up. in Hum Vaccin
Immunother, Vol. 8 390-7 (United States, 2012).
A review of clinical trials of human papillomavirus prophylactic
vaccines. Vaccine 30 Suppl 5, F123-38 (2012).
Vaccinera mot HPV. (2015).
Experts, W.S.A.G.o. Meeting of the immunization Strategic
Advisory Group of Experts, November 2008 – conclusions and
recommendations. (2009).
zur Hausen, H. Condylomata acuminata and human genital cancer.
Cancer Res 36, 794 (1976).
zur Hausen, H. The role of viruses in human tumors. Adv Cancer
Res 33, 77-107 (1980).
Steben, M. & Duarte-Franco, E. Human papillomavirus infection:
epidemiology and pathophysiology. Gynecol Oncol 107, S2-5
(2007).
de Villiers, E.M., Fauquet, C., Broker, T.R., Bernard, H.U. & zur
Hausen, H. Classification of papillomaviruses. Virology 324, 17-27
(2004).
Woodman, C.B., Collins, S.I. & Young, L.S. The natural history of
cervical HPV infection: unresolved issues. in Nat Rev Cancer, Vol.
7 11-22 (England, 2007).
Walboomers, J.M. et al. Human papillomavirus is a necessary cause
of invasive cervical cancer worldwide. J Pathol 189, 12-9 (1999).
Muñoz, N. et al. Epidemiologic classification of human
papillomavirus types associated with cervical cancer. N Engl J Med
348, 518-27 (2003).
Cogliano, V. et al. Carcinogenicity of human papillomaviruses.
Lancet Oncol 6, 204 (2005).
Trottier, H. et al. Human papillomavirus infections with multiple
types and risk of cervical neoplasia. in Cancer Epidemiol
Biomarkers Prev, Vol. 15 1274-80 (United States, 2006).
McGraw, S.L. & Ferrante, J.M. Update on prevention and screening
of cervical cancer. World J Clin Oncol 5, 744-52 (2014).
Bosch, F.X., Lorincz, A., Muñoz, N., Meijer, C.J. & Shah, K.V. The
causal relation between human papillomavirus and cervical cancer. J
Clin Pathol 55, 244-65 (2002).
Baseman, J.G. & Koutsky, L.A. The epidemiology of human
papillomavirus infections. J Clin Virol 32 Suppl 1, S16-24 (2005).
Schiffman, M.H. & Brinton, L.A. The epidemiology of cervical
carcinogenesis. Cancer 76, 1888-901 (1995).
de Villiers, E.M. et al. Human papillomavirus infections in women
with and without abnormal cervical cytology. Lancet 2, 703-6
(1987).
43
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
44
Adam, E. et al. Papillomavirus detection: demographic and
behavioral characteristics influencing the identification of cervical
disease. Am J Obstet Gynecol 182, 257-64 (2000).
Burk, R.D. et al. Declining prevalence of cervicovaginal human
papillomavirus infection with age is independent of other risk
factors. Sex Transm Dis 23, 333-41 (1996).
Ho, G.Y., Bierman, R., Beardsley, L., Chang, C.J. & Burk, R.D.
Natural history of cervicovaginal papillomavirus infection in young
women. N Engl J Med 338, 423-8 (1998).
Moscicki, A.B. et al. The natural history of human papillomavirus
infection as measured by repeated DNA testing in adolescent and
young women. J Pediatr 132, 277-84 (1998).
Rodríguez, A.C. et al. Rapid clearance of human papillomavirus and
implications for clinical focus on persistent infections. J Natl Cancer
Inst 100, 513-7 (2008).
Favre, M. Structural polypeptides of rabbit, bovine, and human
papillomaviruses. J Virol 15, 1239-47 (1975).
Burk, R.D., Chen, Z. & Van Doorslaer, K. Human papillomaviruses:
genetic basis of carcinogenicity. Public Health Genomics 12, 281-90
(2009).
zur Hausen, H. Papillomaviruses and cancer: from basic studies to
clinical application. Nat Rev Cancer 2, 342-50 (2002).
Munger, K., Phelps, W.C., Bubb, V., Howley, P.M. & Schlegel, R.
The E6 and E7 genes of the human papillomavirus type 16 together
are necessary and sufficient for transformation of primary human
keratinocytes. J Virol 63, 4417-21 (1989).
McDougall, J.K. Immortalization and transformation of human cells
by human papillomavirus. Curr Top Microbiol Immunol 186, 101-19
(1994).
Werness, B.A., Levine, A.J. & Howley, P.M. Association of human
papillomavirus types 16 and 18 E6 proteins with p53. Science 248,
76-9 (1990).
Dyson, N., Howley, P.M., Münger, K. & Harlow, E. The human
papilloma virus-16 E7 oncoprotein is able to bind to the
retinoblastoma gene product. Science 243, 934-7 (1989).
Harper, D.M. & Demars, L.R. Primary strategies for HPV infection
and cervical cancer prevention. Clin Obstet Gynecol 57, 256-78
(2014).
Winer, R.L. et al. Condom use and the risk of genital human
papillomavirus infection in young women. in N Engl J Med, Vol.
354 2645-54 (2006 Massachusetts Medical Society., United States,
2006).
Cancer, I.C.o.E.S.o.C. Cervical carcinoma and reproductive factors:
collaborative reanalysis of individual data on 16,563 women with
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
cervical carcinoma and 33,542 women without cervical carcinoma
from 25 epidemiological studies. Int J Cancer 119, 1108-24 (2006).
Smith, J.S. et al. Cervical cancer and use of hormonal
contraceptives: a systematic review. Lancet 361, 1159-67 (2003).
Castellsagué, X., Bosch, F.X. & Muñoz, N. Environmental cofactors in HPV carcinogenesis. Virus Res 89, 191-9 (2002).
Appleby, P. et al. Carcinoma of the cervix and tobacco smoking:
collaborative reanalysis of individual data on 13,541 women with
carcinoma of the cervix and 23,017 women without carcinoma of the
cervix from 23 epidemiological studies. Int J Cancer 118, 1481-95
(2006).
Couto, E. & Hemminki, K. Heritable and environmental components
in cervical tumors. Int J Cancer 119, 2699-701 (2006).
Hemminki, K. & Chen, B. Familial risks for cervical tumors in full
and half siblings: etiologic apportioning. Cancer Epidemiol
Biomarkers Prev 15, 1413-4 (2006).
Magnusson, P.K., Sparén, P. & Gyllensten, U.B. Genetic link to
cervical tumours. Nature 400, 29-30 (1999).
Magnusson, P.K., Lichtenstein, P. & Gyllensten, U.B. Heritability of
cervical tumours. Int J Cancer 88, 698-701 (2000).
Magnusson, P.K. & Gyllensten, U.B. Cervical cancer risk: is there a
genetic component? Mol Med Today 6, 145-8 (2000).
Hildesheim, A. & Wang, S.S. Host and viral genetics and risk of
cervical cancer: a review. in Virus Res, Vol. 89 229-40 (Netherlands,
2002).
Zoodsma, M., Nolte, I.M., Te Meerman, G.J., De Vries, E.G. & Van
der Zee, A.G. HLA genes and other candidate genes involved in
susceptibility for (pre)neoplastic cervical disease. Int J Oncol 26,
769-84 (2005).
Sanjeevi, C.B. et al. Different HLA-DR-DQ haplotypes are
associated with cervical intraepithelial neoplasia among human
papillomavirus type-16 seropositive and seronegative Swedish
women. in Int J Cancer, Vol. 68 409-14 (United States, 1996).
Beskow, A.H., Josefsson, A.M. & Gyllensten, U.B. HLA class II
alleles associated with infection by HPV16 in cervical cancer in situ.
in Int J Cancer, Vol. 93 817-22 (2001 Wiley-Liss, Inc., United
States, 2001).
Ghaderi, M. et al. Risk of invasive cervical cancer associated with
polymorphic HLA DR/DQ haplotypes. Int J Cancer 100, 698-701
(2002).
Deshpande, A. et al. TNF-alpha promoter polymorphisms and
susceptibility to human papillomavirus 16-associated cervical
cancer. in J Infect Dis, Vol. 191 969-76 (United States, 2005).
45
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
46
Niwa, Y. et al. Lymphotoxin-alpha polymorphism and the risk of
cervical cancer in Japanese subjects. in Cancer Lett, Vol. 218 63-8
(Ireland, 2005).
Gostout, B.S. et al. TAP1, TAP2, and HLA-DR2 alleles are
predictors of cervical cancer risk. in Gynecol Oncol, Vol. 88 326-32
(United States, 2003).
Zoodsma, M. et al. Interleukin-10 and Fas polymorphisms and
susceptibility for (pre)neoplastic cervical disease. in Int J Gynecol
Cancer, Vol. 15 Suppl 3 282-90 (United States, 2005).
Coelho, A. et al. Protective role of the polymorphism CCR2-64I in
the progression from squamous intraepithelial lesions to invasive
cervical carcinoma. in Gynecol Oncol, Vol. 96 760-4 (United States,
2005).
Lai, H.C. et al. Genetic polymorphism of the interferon-gamma gene
in cervical carcinogenesis. Int J Cancer 113, 712-8 (2005).
Arnheim, L., Dillner, J. & Sanjeevi, C.B. A population-based cohort
study of KIR genes and genotypes in relation to cervical
intraepithelial neoplasia. in Tissue Antigens, Vol. 65 252-9
(Denmark, 2005).
Zhang, Z. et al. CD83 gene polymorphisms increase susceptibility to
human invasive cervical cancer. in Cancer Res, Vol. 67 11202-8
(United States, 2007).
Sun, T. et al. FASL -844C polymorphism is associated with
increased activation-induced T cell death and risk of cervical cancer.
in J Exp Med, Vol. 202 967-74 (United States, 2005).
Wang, S.S. et al. Common variants in immune and DNA repair
genes and risk for human papillomavirus persistence and
progression to cervical cancer. J Infect Dis 199, 20-30 (2009).
Wang, S.S. et al. Common genetic variants and risk for HPV
persistence and progression to cervical cancer. PLoS One 5, e8667
(2010).
Gauderman, W.J., Witte, J.S. & Thomas, D.C. Family-based
association studies. J Natl Cancer Inst Monogr, 31-7 (1999).
Risch, N. Linkage strategies for genetically complex traits. I.
Multilocus models. Am J Hum Genet 46, 222-8 (1990).
Thomas, D.C. & Thomas, D.C. Statistical methods in genetic
epidemiology. (2004).
Risch, N.J. Searching for genetic determinants in the new
millennium. Nature 405, 847-56 (2000).
Evangelou, E., Trikalinos, T.A., Salanti, G. & Ioannidis, J.P.
Family-based versus unrelated case-control designs for genetic
associations. PLoS Genet 2, e123 (2006).
Ardlie, K.G., Kruglyak, L. & Seielstad, M. Patterns of linkage
disequilibrium in the human genome. in Nat Rev Genet, Vol. 3 299309 (England, 2002).
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
Kwon, J.M. & Goate, A.M. The candidate gene approach. Alcohol
Res Health 24, 164-8 (2000).
Collins, F.S., Guyer, M.S. & Charkravarti, A. Variations on a theme:
cataloging human DNA sequence variation. Science 278, 1580-1
(1997).
Hardy, J. & Singleton, A. Genomewide association studies and
human disease. in N Engl J Med, Vol. 360 1759-68 (United States,
2009).
Hunter, D.J. & Kraft, P. Drinking from the fire hose--statistical
issues in genomewide association studies. in N Engl J Med, Vol. 357
436-9 (United States, 2007).
Witte, J.S. Genome-wide association studies and beyond. Annu Rev
Public Health 31, 9-20 4 p following 20 (2010).
Freedman, M.L. et al. Principles for the post-GWAS functional
characterization of cancer risk loci. in Nat Genet, Vol. 43 513-8
(United States, 2011).
Genotyping. (2015).
Fan, J.B. et al. Highly parallel SNP genotyping. Cold Spring Harb
Symp Quant Biol 68, 69-78 (2003).
Shen, R. et al. High-throughput SNP genotyping on universal bead
arrays. in Mutat Res, Vol. 573 70-82 (Netherlands, 2005).
Sanger, F., Nicklen, S. & Coulson, A.R. DNA sequencing with
chain-terminating inhibitors. Proc Natl Acad Sci U S A 74, 5463-7
(1977).
Pettersson, E., Lundeberg, J. & Ahmadian, A. Generations of
sequencing technologies. in Genomics, Vol. 93 105-11 (United
States, 2009).
Prober, J.M. et al. A system for rapid DNA sequencing with
fluorescent chain-terminating dideoxynucleotides. Science 238, 33641 (1987).
Venter, J.C. et al. The sequence of the human genome. in Science,
Vol. 291 1304-51 (United States, 2001).
Mardis, E.R. New strategies and emerging technologies for
massively parallel sequencing: applications in medical research. in
Genome Med, Vol. 1 40 (2009).
Mardis, E.R. Next-generation sequencing platforms. Annu Rev Anal
Chem (Palo Alto Calif) 6, 287-303 (2013).
Schuster, S.C. Next-generation sequencing transforms today's
biology. in Nat Methods, Vol. 5 16-8 (United States, 2008).
Quail, M.A. et al. A tale of three next generation sequencing
platforms: comparison of Ion Torrent, Pacific Biosciences and
Illumina MiSeq sequencers. in BMC Genomics, Vol. 13 341 (2012).
Liu, L. et al. Comparison of Next-Generation Sequencing Systems. J
Biomed Biotechnol 2012(2012).
47
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
48
Valouev, A. et al. A high-resolution, nucleosome position map of C.
elegans reveals a lack of universal sequence-dictated positioning. in
Genome Res, Vol. 18 1051-63 (United States, 2008).
Smith, A.M. et al. Highly-multiplexed barcode sequencing: an
efficient method for parallel analysis of pooled samples. in Nucleic
Acids Res, Vol. 38 e142 (England, 2010).
Eid, J. et al. Real-time DNA sequencing from single polymerase
molecules. in Science, Vol. 323 133-8 (United States, 2009).
Alpi, A.F. & Patel, K.J. Monoubiquitylation in the Fanconi anemia
DNA damage response pathway. in DNA Repair (Amst), Vol. 8 4305 (Netherlands, 2009).
Mathew, C.G. Fanconi anaemia genes and susceptibility to cancer.
in Oncogene, Vol. 25 5875-84 (England, 2006).
Engelmark, M., Beskow, A., Magnusson, J., Erlich, H. &
Gyllensten, U. Affected sib-pair analysis of the contribution of HLA
class I and class II loci to development of cervical cancer. Hum Mol
Genet 13, 1951-8 (2004).
Spardy, N. et al. The Human Papillomavirus Type 16 E7
Oncoprotein Activates the Fanconi Anemia (FA) Pathway and
Causes Accelerated Chromosomal Instability in FA Cells. in J
Virol, Vol. 81 13265-70 (2007).
de Bakker, P.I. et al. Efficiency and power in genetic association
studies. in Nat Genet, Vol. 37 1217-23 (United States, 2005).
Barrett, J.C., Fry, B., Maller, J. & Daly, M.J. Haploview: analysis
and visualization of LD and haplotype maps. in Bioinformatics, Vol.
21 263-5 (England, 2005).
Ramoz, N. et al. Mutations in two adjacent novel genes are
associated with epidermodysplasia verruciformis. Nat Genet 32,
579-81 (2002).
Ramoz, N., Rueda, L.A., Bouadjar, B., Favre, M. & Orth, G. A
susceptibility locus for epidermodysplasia verruciformis, an
abnormal predisposition to infection with the oncogenic human
papillomavirus type 5, maps to chromosome 17qter in a region
containing a psoriasis locus. J Invest Dermatol 112, 259-63 (1999).
Ramoz, N. et al. Evidence for a nonallelic heterogeneity of
epidermodysplasia verruciformis with two susceptibility loci
mapped to chromosome regions 2p21-p24 and 17q25. in J Invest
Dermatol, Vol. 114 1148-53 (United States, 2000).
Orth, G. Host defenses against human papillomaviruses: lessons
from epidermodysplasia verruciformis. Curr Top Microbiol
Immunol 321, 59-83 (2008).
Castro, F.A. et al. Association of HLA-DRB1, interleukin-6 and
cyclin D1 polymorphisms with cervical cancer in the Swedish
population--a candidate gene approach. Int J Cancer 125, 1851-8
(2009).
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
Ivansson, E.L., Juko-Pecirep, I., Erlich, H.A. & Gyllensten, U.B.
Pathway-based analysis of genetic susceptibility to cervical cancer
in situ: HLA-DPB1 affects risk in Swedish women. in Genes
Immun, Vol. 12 605-14 (England, 2011).
Groh, V., Wu, J., Yee, C. & Spies, T. Tumour-derived soluble MIC
ligands impair expression of NKG2D and T-cell activation. in
Nature, Vol. 419 734-8 (England, 2002).
Bauer, S. et al. Activation of NK cells and T cells by NKG2D, a
receptor for stress-inducible MICA. in Science, Vol. 285 727-9
(United States, 1999).
Salih, H.R., Rammensee, H.G. & Steinle, A. Cutting edge: downregulation of MICA on human tumors by proteolytic shedding. J
Immunol 169, 4098-102 (2002).
Mizuki, N. et al. Triplet repeat polymorphism in the transmembrane
region of the MICA gene: a strong association of six GCT
repetitions with Behcet disease. Proc Natl Acad Sci U S A 94, 1298303 (1997).
Krul, E.J. et al. HLA and susceptibility to cervical neoplasia. in Hum
Immunol, Vol. 60 337-42 (United States, 1999).
Chen, D., Hammer, J., Lindquist, D., Idahl, A. & Gyllensten, U. A
variant upstream of HLA-DRB1 and multiple variants in MICA
influence susceptibility to cervical cancer in a Swedish population.
Cancer Med 3, 190-8 (2014).
Laurila, P.P. et al. Genomic, transcriptomic, and lipidomic profiling
highlights the role of inflammation in individuals with low highdensity lipoprotein cholesterol. in Arterioscler Thromb Vasc Biol,
Vol. 33 847-57 (United States, 2013).
Rockman, M.V. & Kruglyak, L. Genetics of global gene expression.
in Nat Rev Genet, Vol. 7 862-72 (England, 2006).
Chen, D. & Gyllensten, U. Systematic investigation of contribution
of genetic variation in the HLA-DP region to cervical cancer
susceptibility. in Carcinogenesis, Vol. 35 1765-9 (The Author 2014.
Published by Oxford University Press For Permissions, please email:
[email protected]., England, 2014).
Castro, F.A. et al. Contribution of TMC6 and TMC8 (EVER1 and
EVER2) variants to cervical cancer susceptibility. Int J Cancer 130,
349-55 (2012).
Zaboli, G. et al. Sequencing of high-complexity DNA pools for
identification of nucleotide and structural variants in regions
associated with complex traits. Eur J Hum Genet 20, 77-83 (2012).
Thomsen, L.S., Jochumsen, K.M. & Mogensen, O. Carcinoma in
situ cervicis uteri and inheritance--a Danish twin study. in Gynecol
Oncol, Vol. 103 688-91 (United States, 2006).
49
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
50
Stacey, S.N. et al. Common variants on chromosomes 2q35 and
16q12 confer susceptibility to estrogen receptor-positive breast
cancer. in Nat Genet, Vol. 39 865-9 (United States, 2007).
Easton, D.F. et al. Genome-wide association study identifies novel
breast cancer susceptibility loci. in Nature, Vol. 447 1087-93
(England, 2007).
Hunter, D.J. et al. A genome-wide association study identifies
alleles in FGFR2 associated with risk of sporadic postmenopausal
breast cancer. in Nat Genet, Vol. 39 870-4 (United States, 2007).
Broderick, P. et al. A genome-wide association study shows that
common alleles of SMAD7 influence colorectal cancer risk. in Nat
Genet, Vol. 39 1315-7 (United States, 2007).
Tomlinson, I. et al. A genome-wide association scan of tag SNPs
identifies a susceptibility variant for colorectal cancer at 8q24.21. in
Nat Genet, Vol. 39 984-8 (United States, 2007).
Yeager, M. et al. Genome-wide association study of prostate cancer
identifies a second risk locus at 8q24. in Nat Genet, Vol. 39 645-9
(United States, 2007).
Gudmundsson, J. et al. Genome-wide association study identifies a
second prostate cancer susceptibility variant at 8q24. in Nat Genet,
Vol. 39 631-7 (United States, 2007).
Rare chromosomal deletions and duplications increase risk of
schizophrenia. in Nature, Vol. 455 237-41 (England, 2008).
Chen, D. et al. Genome-wide association study of susceptibility loci
for cervical cancer. J Natl Cancer Inst 105, 624-33 (2013).
Stanley, M.A. Immune responses to human papilloma viruses.
Indian J Med Res 130, 266-76 (2009).
Acta Universitatis Upsaliensis
Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Medicine 1100
Editor: The Dean of the Faculty of Medicine
A doctoral dissertation from the Faculty of Medicine, Uppsala
University, is usually a summary of a number of papers. A few
copies of the complete dissertation are kept at major Swedish
research libraries, while the summary alone is distributed
internationally through the series Digital Comprehensive
Summaries of Uppsala Dissertations from the Faculty of
Medicine. (Prior to January, 2005, the series was published
under the title “Comprehensive Summaries of Uppsala
Dissertations from the Faculty of Medicine”.)
Distribution: publications.uu.se
urn:nbn:se:uu:diva-248484
ACTA
UNIVERSITATIS
UPSALIENSIS
UPPSALA
2015
