Nordic Summer School of Cancer Epidemiology, phase III
Virrat Winter Symposium 2016
Virrat, Finland, 29-31 January 2016
Coordinated by Eero Pukkala
Finnish Cancer Registry (www.cancerregistry.fi) &
School of Health Sciences, University of Tampere (www.uta.fi/hes/en)
Program
I
Nordic Summer School of Cancer Epidemiology
Phase III: Virrat Winter Symposium 2016
Virrat, Finland, 29-31 January 2016
PROGRAM
FRIDAY, 29 January
16:00
Bus from Tampere Railway station at
16:00 => Airport at about 16:25
18:15
Arrival to Virrat, Room occupation
18:45
Welcome meal (restaurant)
19:15
OPENING SESSION (priest house)
Eero Pukkala: Opening of the Virrat Winter Symposium;
practical information
Solveig Hofvind: Results of chairperson election; tasks of a chairman of a
scientific symposium (max 5 minute presentation)
19:30-21:00
SESSION I – Descriptive epidemiology
Chairperson: Petra Makkonen (co-chair: Gerda Engholm)
Student presentations: 15 minutes plus 5 minutes discussion time
Mithila Shrestha: INCIDENCE
TRENDS OF GLIOMAS IN
FINLAND, 1970-2013
Christiana Opokuaah Appiah:
CERVICAL CANCER IN
NORWAY AND FINLAND: A
COMPARISON OF INCIDENCE
RATES AND TRENDS
Gerda Engholm: What else you
can do with NORDCAN?
21:00
Poster hanging
II
SATURDAY, 30 January
8.00
Breakfast (restaurant)
8.30-10:10
SESSION II – Aetiology of cancer
Chairperson: Gunvor Gipling Wåde (co-chair: Bendix Carstensen)
Adalbjorg Kristbjornsdottir: CANCER INCIDENCE AND USE OF
GEOTHERMAL HOT WATER FOR HEATING AND BATHING
Sofie Have Hoffmann: RISK OF TESTICULAR CANCER WITH
CRYPTORCHIDISM - A DANISH COHORT STUDY
Eva María Guðmundsdóttir: CANCER INCIDENCE AMONG ICELANDIC
PILOTS
Shadi Azam: THE ASSOCIATION OF PHYSICAL ACTIVITY WITH
MAMMOGRAPHIC DENSITY
Lene Lerche: PHYSICAL FITNESS IN THE DANISH COHORT “DIET,
CANCER AND HEALTH – NEXT GENERATIONS” – A VALIDATION
STUDY
Jadwiga Buchwald: NICOTINE METABOLISM RATE AND TOBACCO
CONSUMPTION AS RISK FACTORS OF LUNG CANCER
10:30
Coffee break
10:45-11:30
SESSION III – Poster highlights
Guided walk and discussions in the
poster room
Chairperson: Þórdís Jónsdóttir
(co-chair: Hans Storm)
Deependra Singh: FALSE-POSITIVE RESULTS IN RELATION TO BREAST
SYMPTOMS IN FINNISH MAMMOGRAPHY SCREENING PROGRAM
Lovísa Kristín Sigurjónsdóttir: AN ESTIMATION OF COMPLETENESS IN
NORDIC CANCER REGISTRIES
Sushmita Katuwal: CASE CONTROL STUDIES ON RISK FACTORS OF
BREAST CANCER
Viivi Seppänen: INDUCED ABORTIONS IN FINNISH CHILDHOOD
CANCER SURVIVORS
III
11:45
Lunch (restaurant)
12.50-14:50
SESSION IV – Screening
Chairperson: Anne Mohr Drewes (co-chair: Solveig Hofvind)
Maiju Pankakoski: CUMULATIVE PROBABILITY OF ABNORMALITIES IN
ORGANIZED CERVICAL CANCER SCREENING
Petra Makkonen: EFFECT OF ORGANIZED SCREENING AND
OPPORTUNISTIC TESTING IN CERVICAL CANCER IN FINLAND AMONG
YOUNG WOMEN
Nataliia Moshina: MAMMOGRAPHIC DENSITY AND PERFORMANCE
MEASURES IN THE NORWEGIAN BREAST CANCER SCREENING
PROGRAM
Gunvor Gipling Waade: COMPRESSION FORCE USED IN THE
NORWEGIAN BREAST CANCER SCREENING PROGRAM
Linda Werner Hartman: RISK-REDUCING SURGERY AND CANCER
INCIDENCE IN A BRCA1/2-POSITIVE COHORT IN SWEDEN
Þórdís Jónsdóttir: ATTITUDES OF ICELANDIC
WOMEN TOWARDS GENETIC COUNSELING
AND GENETIC TESTING REGARDING
HEREDITARY CANCER
14:50
Coffee break
15:30
Multicultural deeply scientific discussions &
outdoor games
Main organiser: Ivalu Katajavaara Sørensen
18:00
Dinner (Mikontalo, built 1892)
20:00
SAUNA SESSION – (river shore sauna)
Sauna, ice-hole swimming, sauna disco, fire
place sausages
IV
SUNDAY, 31 January
8.50
Breakfast (restaurant)
9:20-10:50
SESSION V – Events after cancer diagnosis
Chairperson: Linda Hartman (co-chair: Esa Läärä)
Annemarie Brusen Jensen: VENOUS THROMBOEMBOLISM IN SMALL
CELL LUNG CANCER
Anne M. Drewes & Maria E. Møller: RISK OF GETTING A SECOND
PRIMARY BRAIN TUMOUR IN A PRIMARY BREAST CANCER
POPULATION (20 min)
Melanie Poganitsch: PROBABILITY OF FATHERHOOD AND MARRIAGE
STATUS AFTER CHILDHOOD CANCER
Mette Vestergaard Jensen: ENDOCRINE LATE EFFECTS IN SURVIVORS
OF ADOLESCENT AND YOUNG ADULT CANCER: A DANISH
POPULATION-BASED COHORT STUDY
11:00
Lunch
11:30
CLOSURE SESSION
Esa Läärä: Guidelines for article writing
Eero Pukkala: How to submit a paper and get it published: what happens the
journal editorial process
Hans Storm: Future of cancer epidemiology
Certificates and closing remarks
Packing, leaving the rooms
13:00
Bus leaves towards Tampere.
Stops at Railway Station at
14:45, then makes city
sightseeing and finally reaches
Tampere Airport at about 16.
V
ABSTRACTS
(in alphabetical order according to the first name)
VI
CANCER INCIDENCE AND USE OF GEOTHERMAL HOT WATER FOR HEATING
AND BATHING
Adalbjorg Kristbjornsdottir, MPH, Centre of Public Health Sciences, University of
Iceland, Stapi, Reykjavik, Iceland.
Vilhjalmur Rafnsson, MD, Department of Preventive Medicine, University of Iceland,
Stapi Hringbraut, Reykjavik, Iceland.
Background: Residents of geothermal areas have increased incidence and excess
mortality of non-Hodgkin’s lymphoma, breast, prostate, and kidney cancers. These
populations are exposed to chronic low level ground gas emissions and various
pollutants from the geothermal hot water. The aim was to assess whether utilization of
geothermal hot water is associated with risk of cancer according to length of
residence.
Methods: Those 5-64 years of age were obtained from the census 1981 and followed
through 1981-2013. Personal identifier was used in record linkage with nation-wide
emigration, death and cancer registries. The exposed population was defined as
inhabitants of communities with geothermal hot-water supply systems since 1972.
Reference populations were defined according to different degree of
volcanic/geothermal activity. Hazard ratio (HR), 95% confidence intervals (CI), without
and with stratification on cumulative years of residence, was estimated in Cox-model.
Results: Adjusted HRs for all cancer was 1.17 (95%CI 1.09-1.26) without stratification
on residence and 1.21 (95%CI 1.12-1.30) with stratification. Analyses of
corresponding HRs for breast cancer was 1.42 (1.18-1.72) and 1.49 (1.23-1.80); for
prostate cancer 1.43 (1.19-1.72); and 1.48 (1.23-1.78), and for lymphoid and
haematopoietic tissue 1.49 (1.17-1.91), and 1.55 (1.21-1.98). These patterns of
positive dose-response relationship were observed between all incidences of cancers
and length of residence.
Conclusion: The increased cancer incidence is consistent with previous findings in
geothermal area. Positive dose-response relationship between incidence of cancers,
length of residence, and gradient of geothermal/volcanic activity need urgent
consideration. More information on water and air pollution is needed in further studies.
VII
Title: RISK OF GETTING A SECOND PRIMARY BRAIN TUMOUR IN A PRIMARY
BREAST CANCER POPULATION
Authors: Anne M. Drewes1, Maria E. Møller2, Gerda Engholm3, Rasmus HertzumLarsen3 & Hans Storm3
1. University of Aarhus, Aarhus, Denmark, 2. University of Copenhagen, Denmark, 3. The Danish Cancer Society, Denmark.
Affiliation: The Danish Cancer Society, Copenhagen, Denmark.
Introduction: In the USA, a trend indicated that women under the age of 40, who
have had a primary breast cancer, have an increased risk of a new primary tumour
within the brain. On the contrary, a protecting effect was observed in women over the
age of 50 years with a previous history of breast cancer. This is likely due to a
different hormonal status with a lower oestrogen level in postmenopausal women.
Purpose: In a primary breast cancer population in Denmark, the aim of this study
were to compare premenopausal women (<50 years) to postmenopausal women’s
risk of getting primary brain cancer.
Methods: Our study cohort included all women who were diagnosed with a primary
breast cancer between 1978-2003. A total of 34,936 women were identified. In this
cohort 187 developed a brain tumour. Standardized incidence ratios (SIRs) were
used to calculate the risk of getting a second primary brain tumour in the breast
cancer population.
Results: No increased risk for secondary brain tumours was found for women
younger than 50 years of age. However, women at the age 50-59 had a significant
increased risk of developing a brain tumour within 5-10 years after their breast
cancer diagnosis in year 1993-2003, with a SIR of 1,61 (95 % CI 1.07-2,33). By
contrast, no correlation was found for women in this age group, diagnosed between
1978 and 1992. Furthermore no overall trend towards an increased risk was found
between 1978-2003 in this primary breast cancer population.
Conclusion: In a previous breast cancer population, the results did not indicate an
overall increased risk of developing a brain cancer for premenopausal as compared
to postmenopausal women. However, a subgroup at the age 50-59 years had an
increased risk. The study was partly initiated due to the increasing amount of
publications investigating the effect of exogenous hormones to both breast and brain
cancer. More and detailed studies including exposure to exogenous hormones in
relation to the menopause are needed to explain if brain cancer in breast cancer
patients is linked to hormone changes.
VIII
VENOUS THROMBOEMBOLISM IN SMALL CELL LUNG CANCER
Annemarie Brusen Jensen1, 2, Anne Flou Kristensen2
1
School of Medicine and Health, University of Aalborg, Denmark
2
Department of Clinical Biochemistry, Aalborg University Hospital, Denmark
Background and aims: Patients suffering from Small Cell Lung Cancer (SCLC) have
an increased risk of developing venous thromboembolism (VTE). The occurrence of
VTE is a major clinical challenge as it increases patient morbidity and mortality. The
aim of this study was to create a profile of lung cancer in general in the Nordic
countries and to evaluate the hypercoagulative state of SCLC in order to investigate
the risk of VTE in SCLC patients.
Methods: A lung cancer profile from 2008-2013 was assessed using NORDCAN.
Plasma samples from 239 SCLC patients, provided from the RASTEN study (Lund
University Hospital, Sweden), were collected at three timepoints; baseline, during
treatment and follow-up, and compared to 60 healthy age-related controls collected
at Aalborg University Hospital, Denmark. The patients received standard treatment
+/- the anticoagulant enoxaparin and was further divided into limited and extensive
disease stage. The coagulative status was assessed using a procoagulant
phospholipid activity assay, which determines the clotting time and a thrombin
generation assay, which determine the Endogenous Thrombin Potential (ETP) and
the maximum thrombin generation (peak height).
Results and conclusions: In 2009-2013 the average annual numbers of new cases of
lung cancer were 7523 for males and 6229 for females for all Nordic countries
combined. The age-standardised incidence rates (W) for males and females were
29.2 and 22.5 per 100,000 per year, respectively. A total of 14100 males and 15577
females are living with the lung cancer diagnose at the end of year 2013 and a
average of 11637 people in the Nordic countries died annually (2008-2012) from
lung cancer, including 6565 males and 5072 females. The five-year agestandardised relative survival is 12.8% for males and 18% for females (2009-2013).
Patients with SCLC had a significant increased thrombin generation shown by higher
levels of ETP (1272.95±292.01 nM•min) and peak height (223.19±72.95 nM) when
compared to healthy controls (ETP 944.62±268.72 nM•min, U= 1705.5, p< 0.001;
peak height 81.66±37.25 nM, U= 475, p < 0.001). In addition, these patients had a
higher procoagulant phospholipid activity, shown by a shorter clotting time
(34.86±11.12 sec.) compared to controls (59.65±8.18 sec.), U= 596.5, p<0.001. No
statistical significant difference between limited and extensive disease stage was
found. The findings indicate that SCLC patients generate a faster and greater
amount of thrombin leading to the assumption that SCLC have a higher risk of VTE.
The abstract should not be cited or reproduced without written agreement with the authors.
IX
CERVICAL CANCER IN NORWAY AND FINLAND: A COMPARISON OF
INCIDENCE RATES AND TRENDS
Christiana Opokuaah Appiah 1 and Bo Terning Hansen2
1
Department of Laboratory Medicine, Childrens and Womens Health, NTNU
2
Department of Research, Cancer Registry of Norway, Oslo, Norway
Background: Cervical cancer is the fourth most commonly diagnosed cancer in
women worldwide, with about 527,600 new cases recorded in 2012. It is prevented by
Pap smear, which was introduced in the 1960s. At that time, Norway and Finland had
similar incidence rates for cervical cancer, at about 15/100000 women. Organized
screening programs against cervical cancer were established in Finland and Norway
in the early 1960s and 1995, respectively. The purpose of this study is to compare the
incidence rates and trends between Norway and Finland for approximately 60 years
(1953-2013).
Methods: Cervical cancer age-adjusted and age-specific incidence rates for both
countries were obtained for the period 1953 – 2013 from the NORDCAN database
(Version 5.3, 04.2013). Incidence trends were studied by joinpoint regression.
Results: Norway experienced a significant annual growth in age standardized cervical
cancer incidence rates of 1.3% for the period 1953-1975, then a decline of 3.8% and
3.5% for 1975-1988 and 1991-2000, respectively. The age standardized incident rates
for Finland decreased steadily at 8.5% per year for 1967–1976 and 5.2% for 1976–
1991. For the remaining calendar year intervals investigated, the incidence trends did
not change significantly in any of the countries. In all years investigated, the incidence
rate was higher in Norway than in Finland. Age specific incidence rates differed by
age groups, with the rates for older age groups being lower.
Conclusions: The present study shows that cervical cancer screening has reduced
the incidence in cancer rates in Finland and Norway. However, the reduction started
earlier and was greater in Finland than in Norway. The differences between the
countries may be associated with differences in the screening programs. Also,
national differences in factors related to the background risk of cervical cancer may
have influenced the differences in the incidence rates and trends.
X
CHARACTERIZATION OF SECONDARY ACUTE MYELOID LEUKEMIA PATIENTS
USING DATA FROM THE SWEDISH ACUTE LEUKEMIA REGISTRY
Edda Blümel1, Christer Nilsson2, Sören Lehmann1,2
1
2
Department of Medical Sciences, Uppsala University, Sweden
Department of Medicine, Karolinska Institute, Sweden
Background: Acute myeloid leukemia (AML) is a cancer of the myeloid lineage of
blood cells and it affects mostly the elderly. The first line treatment is chemotherapy
and the responsiveness is largely influenced by the cytogenetic risk profile of the
patient. AML patients are categorized into (a) favorable, (b) intermediate and (c)
adverse risk groups. Secondary AML (sAML) is a heterogeneous and poorly defined
subgroup of AML. It includes patients with antecedent hematological diseases (AHD)
and patients that have previously been treated with cytotoxic therapies for another
disease - therapy-related AML patients (t-AML) (1). The aim of this study was to
investigating whether sAML constitutes an independent prognostic factor in AML.
Materials: Data from the Swedish Acute Leukemia Registry was utilized, including
patients diagnosed with AML between 1997 and 2013.
Methods: Primary AML patients were compared to sAML patients with regard to
general patient characteristics and survival.
Results: The studied cohort consisted of 5,881 AML patients, of whom 72% were
primary AML (n=4,233), 19% AHD (n=1,098) and 9% t-AML (n=550). AHD showed
male predominance (p-value=0.0003, 95% CI, 0.68-0.89), while t-AML displayed
female predominance (p-value=0.0018, 95% CI, 1.1-1.6). The median age at
diagnosis was 71 years (range: 17-100 yrs.). AHD patients were significantly older
compared to primary AML patients (70 yrs.–73 yrs., p-value<0.0001). A favorable
cytogenetic risk profile was uncommon in AHD patients (2%) and adverse
cytogenetics was the most frequent cytogenetic profile for t-AML patients (47%).
Primary AML patients with intermediate and high cytogenetic risk profile were
significantly more likely to achieve complete remission than AHD patients with the
same risk profiles (p-value=0.03; p-value=0.014). Survival analyses of intensively
treated patients showed superior survival for primary AML in all age groups (pvalue<0.0001). Cox-regression analyses showed that age had the strongest adverse
impact (HR 2.19, 95% CI, 2.00-2.39), while a favorable cytogenetic risk had positive
impact (HR 0.48, 95% CI, 0.40-0.58). AHD and t-AML showed both to be independent
negative prognostic factors when comparing to primary AML patients (AHD vs.
primary AML: HR 1.70, 95% CI, 1.51-1.93; t-AML vs. primary AML: HR 1.76, 95% CI,
1.51-2.04).
Conclusion: In this study we have carried out the up to date largest population based
analysis of sAML and primary AML patients, showing that sAML constitutes a distinct
subtype of AML, displaying a poor outcome.
1. Estey E, Döhner H. Acute myeloid leukaemia. Lancet. 2006;368(9550):1894-907.
XI
CANCER INCIDENCE AMONG ICELANDIC PILOTS.
Eva María Guðmundsdóttir1, Vilhjálmur Rafnsson2.
1
Center of Public Health Sciences, University of Iceland.
Department of Preventive Medicine, University of Iceland.
2
Aircrews have higher cancer incidence, predominantly skin cancer and
breast cancer than the general population. However, the causes are not yet known.
The aim of the study is to investigate cancer risk among pilots flying on international
routes compared with other pilots. This is a cohort study of 562 male pilots, who
were licensed commercial pilots. Information on employment time, for each pilot,
was obtained from the Icelandair. Information on cancer was obtained by record
linkage of personal identifier with the Icelandic Cancer Registry. The inclusion period
of the cohort is from 1944 to the end of year 2003. The follow-up period for cancer is
from 1st of January 1955 to 31st of October 2014. The cohort was divided into two
groups: Pilots flying on international routes and other pilots. Cox regression analysis
was used to estimate the hazard ratio (HR) between the two groups. The statistics
were performed in R 3.2.2. The group of pilots flying internationally was younger
than the group of other pilots, and age in years was adjusted for in the Cox
regression analysis, introducing age as a continuous variable. The HR for all cancers
was 1.84, 95% confidence interval 1.16 – 2.92. It is known from previous studies that
pilots flying international routes (if operating jets) are exposed to higher dose of
cumulative cosmic radiation than other pilots, because of higher altitude and longer
duration flown.
Next steps are to calculate the accumulated cosmic radiation, based on the
type of aircraft, and air hours per individual pilot, and per calendar year. Then the
cancer risk will be evaluated, depending on the amount of estimated cosmic
radiation exposure (internal analysis of the Icelandair pilots). Moreover, standardized
incidence ratio will be calculated in order to compare the incidence of cancer among
the pilots with that of the general population.
XII
COMPRESSION FORCE USED IN THE NORWEGIAN BREAST CANCER
SCREENING PROGRAM
Gunvor Gipling Waade1, Sofie Sæbuødegård2, Peter Hogg3,4, Nataliia Moshina2,
Solveig Hofvind1,2
1
Oslo and Akershus University College of Applied Sciences, Oslo, Norway
Cancer Registry of Norway, Oslo, Norway
3
School of Health Sciences, University of Salford, Salford, UK
4
Karolinska Institute, Stockholm, Sweden
2
Background: Compression force is applied during mammography to reduce breast
thickness, which reduces radiation dose and improves image quality. There are no
evidence-based recommendations regarding the optimum compression force to
apply. In this study we investigate compression force used within and between
different screening centers in the Norwegian Breast Cancer Screening Program
(NBCSP), as a first step towards establishing evidence-based recommendations for
compression force in the program.
Material and Methods: We investigated the applied compression force used on
19 487 randomly selected women screened at fifteen breast centers in the NBCSP
during January-March 2014. The study included 38 974 craniocaudal (CC) and
38 974 mediolateral-oblique (MLO) views, 77 948 images in total, performed by 218
radiographers. Variation in compression force between and within breast centers,
vendors and age and body mass index (BMI) of the screened women were assessed
and deceptively described.
Results: The range in average force between the breast centers was 63N for CC and
56N for MLO, while it was 90N and 89N, for CC and MLO respectively, between the
radiographers. The compression force increased with increasing BMI, but did not
differ statistically significantly by the women’s age groups or vendor.
Conclusion: We identified a wide variation in compression force between and within
the NBCSP breast centers, larger between than within the centers. Futures studies
are needed to investigate the impact of these findings on radiation dose, patient
experience and image quality.
XIII
NICOTINE METABOLISM RATE AND TOBACCO CONSUMPTION AS RISK
FACTORS OF LUNG CANCER
Jadwiga Buchwald
University of Helsinki, Finland
Background and Aim: Smoking is the major risk factor for lung cancer and
cardiovascular disease, and is thus of high public health relevance world-wide.
Roughly every fifth adult globally smokes and the costs of smoking to the society are
heavy. The majority of smokers desire to quit but find it remarkably difficult as nicotine
is a highly addictive neuro-stimulant.
Nicotine metabolism rate is strongly correlated to total nicotine clearance rate which
influences smoking behaviour. The ratio of 3-hydroxycotinine/cotinine (i.e. nicotine
metabolite ratio, NMR) is an established biomarker of nicotine metabolism rate. Higher
NMR is associated with increased smoking and decreased rates of quitting. Thus,
those with faster nicotine metabolism should be at higher risk for lung cancer. In a
genome-wide association study on NMR the strongest genetic association was found
for CYP2A6, which is the primary metabolic enzyme for nicotine as well as being
involved in the metabolism of carcinogens. Therefore, high NMR might possibly be
linked to faster carcinogen metabolism, which would lead to reduced exposure and
potentially reduced cancer risk.
The aim of this paper is to study the effect of NMR on lung cancer risk and to gain a
holistic picture of the associations between NMR, smoking and lung cancer.
Understanding of these relationships along with the genetics of NMR is key for
designing personalized smoking cessation interventions.
Material and Methods: Genetic risk score (GRS) of NMR was used as a proxy for
NMR and genetic variation in CYP2A6. Cox proportional hazard models and
cumulative hazard curves were used to assess NMR and cumulative exposure to
cigarette smoke (cigarette packs smoked during the lifetime) as risk factors for lung
cancer. Mendelian randomization (MR) analysis was employed to study the causal
relationships between NMR, smoking and smoking induced diseases.
In the first phase of the study, the NAG-FIN sample of 1171 subjects with 15 nonprevalent lung cancer cases during the follow up time period of around 11 years was
utilized to plan the study, and as a basis for power calculations. The NAG-FIN sample
consists of ever-smoking adult twins born in 1938–1957. After this, a larger sample
from the National FINRISK Study 1992–2012 with genome-wide genotype data
available for N>20000 will be used in collaboration with the National Institute for
Health and Welfare.
Results and Conclusions: Initial analyses of the NAG-FIN dataset did not reach
statistical significance. Based on power calculations the larger dataset would have
enough power to detect the effect sizes indicated in the analyses of the NAG-FIN
sample.
XIV
PHYSICAL FITNESS IN THE DANISH COHORT “DIET, CANCER AND HEALTH –
NEXT GENERATIONS” – A VALIDATION STUDY
Lene Lerche1, Anja Olsen1
1
Unit of Diet, Genes and Environment, Danish Cancer Society Research Center, Denmark
Background
Physical fitness comprises several components, of which, cardiorespiratory fitness
(VO2max (ml O2 /kg/min)) is of particularly interest due to its strong inverse association
with risk of chronic diseases, including cardiovascular disease, diabetes and cancer.
However, standard test of cardiorespiratory fitness are cost prohibitive and impractical
in large epidemiological studies. As a result, less complicated methods including
submaximal exercise testing and non-exercise questionnaire based methods are most
often used in epidemiological research.
Obejctives
The overall obejctive was to evaluate the validity of a submaximal fitness test The
Danish Step Test (TDST) and a simple self-reported question (SSRQ) as methods for
estimating physical fitness in the Danish prospective cohort study “Diet, Cancer and
Health – Next Generations”. This was done by comparing these two methods,
respectively, with a VO2maxtest, which is considered the gold standard.
Methods
125 participants aged 19-67 years were recruited from the cohort. Participants
completed a VO2maxtest, TDST and answered the SSRQ rating their physical fitness
level. Pearson product-moment-correlation-coefficients were calculated to assess the
relationship between the VO2maxtest and TDST. The VO2maxtest, TDST and the SSRQ
were grouped into five categories. The degree of misclassification across categories
between TDST and the SSRQ, respectively, in relation to the VO 2maxtest were
investigated.
Results
Moderate correlations between the VO2max test and TDST were found (men: r=0.555,
n=60, p<0.05, women: r=0.658, n=65, p<0.05).
When comparing the categories of physical fitness from TDST with the VO 2maxtest, on
average only 6% of the women were classified outside the same (±1) category.
However, for men there was a higher degree of misclassification with 38% outside the
same (±1) category, where TDST especially seemed to underestimate physical
fitness. When comparing the categories from the SSRQ with the results from the
VO2maxtest, only 9% of the women were misclassified. Among men, only 13% fell
outside the same (±1) category.
Conclusions
The SSRQ was found to be a superior method to TDST when estimating physical
fitness in a less cost prohibitive and more practical way than the VO2maxmethod.
XV
RISK-REDUCING SURGERY AND CANCER INCIDENCE IN A BRCA1/2POSITIVE COHORT IN SWEDEN
Linda Werner Hartman1,2, Martin Nilsson2,3, Karin Henriksson4, Niklas Loman2,3
1
Regional Cancer Center South, Lund, Sweden
Division of Oncology and Pathology, Department of Clinical Sciences Lund, Lund University,
Sweden
3
Department of Oncology, Skåne University Hospital, Lund/Malmö, Sweden.
4
Department of Clinical genetics, Skåne University Hospital, Lund/Malmö, Sweden
2
Background: Since the breast cancer genes BRCA1 and BRCA2 were
identified in the mid-1990s, there is controversy about the breast cancer risk
associated to mutations in these genes. The purpose of this study is to assess the
risk of breast and ovarian cancer, as well as frequency and effect of prophylactic
surgery, among presymptomatic BRCA1 and BRCA2 mutation carriers identified
through genetic counselling in Sweden.
Material: A population based study inviting all Swedish women with a
mutation in BRCA1 or BRCA2, found at presymptomatic testing at a clinical
oncology centre from the first tests in 1994 until 20061231. The women were
identified through the international CIMBA-study. Follow-up-data for cancer was
obtained by matching with the National cancer registry in 2009, whereas data on
prophylactic surgery was collected from medical records in 2006 to 2008.
Methods: The frequency of prophylactic mastectomy and prophylactic
oophorectomy was assessed. Further, the age-specific incidence of breast
cancer, was assessed using Kaplan-Meier estimates, censoring the follow-up at
prophylactic mastectomy.
Results: The study consists of 298 women that were at age 17-85 years
(median 37 years) when identified as BRCA1 (242 women) and BRCA2 (56
women) mutation carriers. At the end of the follow up 144 (48%) had performed
prophylactic bilateral mastectomy and 160 (54%) had removed their ovaries. In
total, 20 women developed breast cancer (including breast cancer in-situ) during
the follow up (total 1397 person-years, median 4.2 years), and 7 developed
ovarian cancer.
Conclusions: In this series of Swedish BRCA1 and BRCA2 mutation
carriers from the first decade of genetic testing, a majority of the women had
performed prophylactic surgery at the end of follow-up. Unfortunately the follow-up
time is too short to get good estimates of age-specific incidences of breast and
ovarian cancer, and to assess the gain from prophylactic surgery. We have
therefor initiated data collection for long term follow-up.
XVI
CUMULATIVE PROBABILITY OF ABNORMALITIES IN ORGANIZED CERVICAL
CANCER SCREENING
Maiju Pankakoski1, Sirpa Heinävaara1, Tytti Sarkeala1, Ahti Anttila1
1
Mass Screening Registry, Finnish Cancer Registry, Helsinki, Finland
Background
Borderline cervical changes detected by screening rarely lead to a severe disease.
In Finland approximately 5-6 % of smears are classified as borderline annually. We
investigated the individual-level screening outcome for women screened at aged 30
to 60.
Material
Screening histories of organized screening during 1991-2012 were collected for all
women from the Mass Screening Registry of the Finnish Cancer Registry. Data
consisted of approximately 5.0 million routine screening invitations and 3.6 million
visits among 1.8 million women. In addition there were 310 000 invitations and 230
000 visits for follow-up screening.
Methods
Screening results in 5-year screening rounds were presented by age groups. If there
were follow-up screens within the screening round, the most severe result of the
round was reported. Cumulative probabilities of borderline cervical changes and
more severe abnormalities for women aged 30 to 60 were also analyzed.
Preliminary results and conclusions will be presented.
XVII
PROBABILITY OF FATHERHOOD AND MARRIAGE STATUS AFTER CHILDHOOD
CANCER
Melanie Poganitsch
Children´s Hospital, University of Helsinki, Helsinki University Central Hospital, FI-00029 Helsinki, Finland
BACKGROUND:
A number of comprehensive reviews of the late adverse effects after childhood cancer
treatment have been published. Testicular damage is a well-known effect of childhood cancer
therapy. However, only few reports provide information about effects of childhood cancer
therapy on fertility and marriage status. In this study we aimed to study how diagnosis of
childhood cancer, type of cancer therapy and the measured testicular function during pubertal
development influence on marriage status, likelihood to live alone, divorce or parent a child at
adulthood.
METHODS:
The cohort of 263 adult male survivors who were treated for childhood (<18y) cancer during
1964-2000 in the Children’s Hospital in Helsinki and were surviving more than 5 years after
diagnosis were identified through hospital records. The clinical characteristics of cancer
diagnosis, treatment and the follow-up data of pubertal testicular function, including
occurrence of signs of spontaneous puberty, testicular growth, testicular serum markers and
the use of testosterone substitution were collected from the hospital records. The clinical
information was linked to the Population Register data on offspring, marriage status and
information if person was living alone. Results were compared with those of controls. Five
age, gender and area of residence matched controls were identified for each survivor from the
Finnish Population Register Center and formed the control group.
Results and conclusion are not yet available.
XVIII
AN ESTIMATION OF COMPLETENESS IN NORDIC CANCER REGISTRIES
Lovísa Kristín Sigurjónsdóttir1, Gerda Engholm1
1
The Danish Cancer Society, Copenhagen, Denmark
Introduction: To make sure that the incidence rates and survival proportions for
cancers featured in a cancer registry database are as close to their true value as
possible it is really important that there is a high degree of completeness in the
registry. The term completeness is defined as the extent to which all incident
cancers in a population are included in a cancer registry. One of the methods that
can be used to estimate completeness is the mortality/incidence ratio (M/I), which
can give an indication of completeness relative to other registries or over time. When
the quality of the mortality data is good the M/I ratio is related to the case fatality and
will equal 1 – the survival probability. The M/I ratio should in general be less than
one, and a ratio greater than expected suggests incomplete registration.
The aim of this project was to estimate the completeness of cancer registration in
three Nordic registries by using the M/I ratio method.
Material and Methods: The data used in this study was found in the NORDCAN
database. In this database the numbers for incidence, mortality and survival of 36
different cancers for males and 37 for females were found from the cancer registries
in Denmark, Norway and Sweden. The M/I ratio for each specific cancer was
calculated for all three registries and plotted vs. the 1 – survival probability. The M/I
ratios were also compared between registries and tested for significant differences.
Results: The testing for significant differences has not been finished yet but looking
at the raw data it shows that overall the M/I ratio for most cancers in the registries is
more or less equal to the 1 – survival probability. In Sweden the M/I ratio for
gallbladder and pancreas cancer was >1 for both males and females. The M/I ratio
for liver cancer was >1 for females in all three countries. In general Sweden had the
highest number of M/I ratios greater than expected for both males and females.
There also seems to be a higher number of ratios that are greater than expected for
female cancers in all three countries.
Conclusions: When interpreting the results one must keep in mind that they only give
an indication of the completeness, not the actual completeness. Also the evaluation
of the M/I ratio is dependent on stable incidence and survival rates and good quality
of the mortality data. That being said, so far one can conclude that since the M/I ratio
for most cancers is more or less equal to the 1 – survival probability this indicates
good completeness in the registries overall. The anomalies found for a few of the
cancers can indicate incomplete registration but can also be explained by other
factors, such as a rapid de- or increase in incidence relative to mortality. To
determine the exact causes this requires further examination.
XIX
ENDOCRINE LATE EFFECTS IN SURVIVORS OF ADOLESCENT AND YOUNG
ADULT CANCER: A DANISH POPULATION-BASED COHORT STUDY
Mette Vestergaard Jensen1,2, Kathrine Rugbjerg1, Jeanette Falck Winther1,
Christoffer Johansen1,2
1
2
Danish Cancer Society Research Center, Survivorship Unit, Copenhagen, Denmark
University of Copenhagen, Copenhagen, Denmark
Background: Survivors of adolescent and young adult cancer (AYA cancer) are at
risk for a broad range of therapy-related late effects. However, late effects in this
specific group of cancer survivors are understudied. In a nation-wide populationbased cohort study, we aimed to investigate the lifetime risk for endocrine late
effects in survivors of AYA cancer.
Methods: In the Danish Cancer Registry we identified 43,062 1-year survivors of
cancer diagnosed at ages 15-39 years. From the Danish Civil Registration system
we randomly chose 259,350 cancer free comparison subjects matched on year of
birth and sex. By linkage to the National Patient Register we identified all hospital
contacts for endocrine diseases (ICD-8-codes: 240-258, ICD-10-codes E01-E35 and
E89). We compared observed and expected numbers of hospital contacts and
calculated standardized hospitalization rate ratios (RR’s) and absolute excess risks
(AER’s)
Results: Of the AYA-cancer survivors, 3605 (8.4%) had at least one hospital
contact for an endocrine disease, while 2406 would have been expected (RR, 1.50,
95% confidence interval (CI), 1.45-.155). RR’s were highest for diagnoses of
testicular hypofunction (RR, 64.6, 95% CI, 41.6-100.4), ovarian hypofunction (RR,
11.8, 95% CI, 7.1-19.9) and pituitary hypofunction (RR, 9.7, 95 % CI, 7.4-12.8). Yet,
the leading causes for hospital contacts were diseases of the thyroid gland, diabetes
mellitus and testicular dysfunction, which constituted 39.1%, 18.2% and 13.3% of
total AER respectively. The RR for any endocrine disease were highest among
survivors of leukemia (RR, 3.7, 95% CI, 2.9-4.7), while survivors of Hodgkin
Lymphoma had the highest disease-specific AER for hypothyroidism (AER, 328, 95
% CI, 264-393).
Conclusion: Survivors of AYA cancer are at increased risk for endocrine diseases
later in life. With these findings, we contribute with knowledge to improve the basis
for patient counseling and follow-up and allow better planning of future preventive
intervention and surveillance strategies.
XX
INCIDENCE TRENDS OF GLIOMAS IN FINLAND, 1970-2013
Mithila Shrestha, Anssi Auvinen1, 2
1
School of Health Sciences, University of Tampere, Tampere, Finland
2
STUK – Radiation and Nuclear Safety Authority, Helsinki, Finland
Objective: The objective of this study is to investigate the incidence trends of gliomas
during a period of 1970 through 2013 using the records from Finnish Cancer Registry.
Background: In the Nordic countries, mobile phone use was started in the mid-1980s
and increased sharply in the mid-1990s. Thus, the time trends of gliomas during 1970
to 2013 in Finland may provide useful information regarding the possible risks
associated with radiofrequency electromagnetic fields (RF-EMF) emitted from mobile
phones.
Methods: The current analysis used data from Finnish cancer registry for the glioma
cases and the size of the population was obtained from the Statistics Finland web
pages. The approximate Poisson method was used to calculate the relative risks for
gliomas and to determine a 95% confidence interval. Time-trend analysis was
conducted using join-point regression analysis.
Results: The relative risk of glioma was 1.26 for men compared with women.
Significantly rising trends were noticed for gliomas during 1970-1990 (APC 2.5%; 95%
CI, 1.8-3.3) among men and 1970-1989 (APC 2.3%; 95% CI, 1.5-3.2) among women.
However, during 1990-2013, the incidence has remained relatively stable for both
men and women. An earlier increasing trend is very likely due to increased histological
confirmation and therefore more accurate diagnosis for histological subtypes such as
astrocytoma. The interaction between year of diagnosis and age at diagnosis was
noted.
Conclusion: In Finland, there has been no uniform increase in incidence rates of
gliomas during study period. Higher incidence of gliomas among male is consistent
with the other previous studies. But clearly, there was a lack of increase in glioma
incidence after 1990, during the period with increasing prevalence of mobile phone
users.
XXI
MAMMOGRAPHIC DENSITY AND PERFORMANCE MEASURES IN THE
NORWEGIAN BREAST CANCER SCREENING PROGRAM
Nataliia Moshina1, Sofie Sebuødegård1, and Solveig Hofvind1
1
Cancer Registry of Norway, Oslo, Norway
Background: High mammographic density (MD) is known to mask breast tumors and
decrease sensitivity of breast cancer screening programs. We aimed to investigate
how performance measures, including positive predictive values (PPVs) and
histopathologic tumor characteristics, are associated with MD among women who
underwent recall examinations in the Norwegian Breast Cancer Screening Program
(NBCSP).
Material: We used information about 28,826 recall examinations from 26,951
subsequently screened women aged 50-69 years, 1996-2010, to analyze PPVs.
Furthermore, we used data about 7,366 invasive breast cancers diagnosed among
prevalently and subsequently screened women in the same period.
Methods: The radiologists subjectively classified MD on the mammograms at the
recall examination into three categories: fatty (<30% fibroglandular tissue); medium
dense (30-70%) and dense (>70%). PPV-1 was defined as the probability of breast
cancer among women recalled due to abnormal findings on the screening
mammograms. PPV-2 was defined as the probability of breast cancer among women
underwent invasive diagnostic procedures at the recall examination. We examined
PPVs by MD. We used logistic regression to estimate odds ratio (OR) of screen
detected breast cancer associated with MD among the recalled women. Chi-square
test was used to compare the distribution of tumor characteristics by MD. OR of tumor
characteristics by MD was estimated by logistic regression analysis. We used
screening mode (screen-film mammography and digital mammography) and age as
adjusting variables in regression analyses.
Results: PPV-1 and PPV-2 decreased by increasing MD (p for trend <0.05 for both).
Compared with women with fatty breasts, the adjusted OR was 0.90 (95% CI: 0.840.96) for those with medium dense breasts and 0.85 (95% CI: 0.76-0.95) for those
with dense breasts. Mean and median tumor size of invasive breast cancers was
13.8 and 12 mm, respectively, for women with fatty breasts, and 16.2 and 14 mm for
those with dense breasts. Lymph node positive tumors were identified among 20.6%
of women with fatty breasts compared with 27.2% of those with dense breasts
(P<0.001). Women with dense breasts had an increased risk of large (OR, 1.44;
95% CI, 1.18–1.73) and lymph node positive tumors (OR, 1.26; 95% CI, 1.05–1.51)
compared with women with fatty and medium dense breasts.
Conclusions: Fewer women needed to be recalled or undergo an invasive procedure
to detect one breast cancer among those with fatty versus dense breasts in the
NBCSP, 1996-2010. High mammographic density was positively associated with
larger tumor size and lymph node positive tumors.
XXII
ASSOCIATION BETWEEN POOR ORAL HEALTH AND STOMACH CANCER RISK:
A COHORT STUDY IN SWEDEN
Nelson Gichora1, Zhiwei Liu1, Weimin Ye1
1
Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, 17177, Stockholm,
Sweden
Background
Poor oral health might be involved in the pathogenesis of gastrointestinal cancers,
particularly esophageal squamous cell carcinoma and gastric cancer. However,
previous studies have reported inconsistent results.
We aimed to explore the association between poor oral health and the risk of stomach
cancer.
Material
A cohort of 20212 participants with a document of index oral examination was
assembled in 1973-74 for a population-based prevalence study of oral mucosal lesion
in Uppsala County in central Sweden. Cancer cases, migration, death on the
participants were obtained by linking to the national registers in Sweden through
personal identification numbers.
Methods
Person-time will be calculated from the date of entry, occurrence of any cancer,
death, migration out of Sweden, or 31st of December 2010, whichever occurred first.
Cox proportional hazards regression models will be used to estimate hazard ratios
(HRs) with corresponding 95% confidence intervals (CIs) as measures of relative risk,
adjusted for age, gender, place of residence, cigarette smoking, and alcohol
consumption.
Results
Project Status (18th Jan 2016): The project is currently ongoing. Linkage to emigration
registers was finally accomplished and data cleanup performed. Analysis is currently
ongoing.
Conclusions
XXIII
EFFECT OF ORGANIZED SCREENING AND OPPORTUNISTIC TESTING IN CERVICAL
CANCER IN FINLAND AMONG YOUNG WOMEN
Petra Makkonen1, Sirpa Heinävaara1,3, Anni Virtanen1, Tytti Sarkeala1, Ahti Anttila1
1
The Finnish Cancer Registry, Finland
2
Faculty of Medicine, University of Helsinki, Finland
3
The Department of Public health, University of Helsinki, Finland
Background: The effectiveness of the organized screening programme for cervical cancer has
been shown in several studies. However, screening under 25 years old women has been
shown to have only little or no impact on the risk of cervical cancer and clear effects have
been observed among women aged over 40 years. In Finland there currently exists in
addition to the organized screening programme an extensive opportunistic screening policy
that concentrates especially on younger women. According to current understanding, a wellorganized screening programme is more effective, costs less and results in less harm than
opportunistic testing but the significance of the opportunistic testing in preventing cervical
cancer calls for further research. The aim of this study is to provide more information about
the effect of opportunistic testing in addition to organized screening in cervical cancer among
young women.
Material: The original data from a previous study included 1,546 cervical cancer cases
registered in the Finnish Cancer Registry and diagnosed during the years 2000-2009, and
9,276 age-matched controls drawn from the population register. This data was linked with
screening histories in mass screening register and further linked with opportunistic testing
data collected from the Social Insurance Institution of Finland (Kela), Finnish Student Health
Service (YTHS), local public health services in Turku region and HUSLAB which provides
laboratory services in the Uusimaa region. Finally, we restricted the data to women screened
under the age of 40 years. Eventually our study included 462 cases and 2772 controls,
altogether 3234 women.
Methods: OR’s and 95% confidence intervals for the association of cervical cancer diagnosis
and participation in organized screening and opportunistic screening 5 or 10 years before the
diagnosis were estimated using unconditional logistic regression. Results will be corrected for
self-selection bias.
Results: Preliminary results and crude OR’s will be presented.
Conclusion: If opportunistic testing can’t be shown to bring any considerable additional
benefit on reducing the risk of cervical cancer and screening in general is less effective in
younger age groups, the current screening policy in Finland should be seriously questioned
and revised.
XXIV
FALSE-POSITIVE RESULTS IN RELATION TO BREAST SYMPTOMS IN FINNISH
MAMMOGRAPHY SCREENING PROGRAM
Deependra Singh1,2, Janne Pitkäniemi1, Nea Malila1,2, Ahti Anttila1
1
Finnish Cancer Registry, Helsinki
2
School of Health Sciences, University of Tampere, Tampere, Finland
Background: Mammography has been found effective as the primary screening test for breast
cancer. We estimated the cumulative probability of false-positive screening test results in
women with and without breast symptoms reported at screen.
Methods: A retrospective cohort study was done using individual screening data of 413611
women aged 50-69 years with 2627256 invitations for mammography screening, in 19922012 in Finland. Symptoms (lump, etc.) were reported at 56805 visits and 48873 visits were
tested false-positive. Generalized linear models were used to estimate the cumulative ‘ever’
and ‘first’ false-positive including false-positive referral to surgery and cancer detection
probability after 10 screening visits. The estimates were compared among women with and
without symptoms separately along with combined symptoms, at each screening visit.
Results: The estimated cumulative probabilities were 16% and 14% for ever and first- falsepositive results respectively. Similarly, the cumulative probabilities of false-positive referral
and cancer detection were 1.5% and 5% respectively. In women with a history of lump, the
cumulative ever false-positive and cancer detection probabilities were 41% and 15%
respectively, compared to 14% and 4% with no lump. Likewise, women who had a history of
‘lump and retraction’ and ‘lump and secretion’ had the cumulative false-positive probabilities
of 55% (95% CI, 45-68) and 51% (95% CI, 43-61) respectively.
Conclusion: In conclusion, information on breast symptoms affect the balance of absolute
benefits and harms. Even though with high cumulative false-positives probability, detection
of invasive cancers suggests that symptom information is useful for further assessment.
XXV
THE ASSOCIATION OF PHYSICAL ACTIVITY WITH MAMMOGRAPHIC DENSITY
Shadi Azam1, Arja R. Aro1, My von Euler-Chelpin2, Ilse Vejborg3, Anne Tjønneland4,
Elsebeth Lynge2, Zorana J. Andersen2
1
Unit for Health Promotion, Department of Public Health, University of Southern Denmark, Esbjerg,
Denmark
2
Center for Epidemiology and Screening, Department of Public Health, University of Copenhagen,
Copenhagen, Denmark
3
Diagnostic Imaging Centre, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
4
Danish Cancer Research Center, Danish Cancer Society, Copenhagen, Denmark
Background: Physical activity is recognized as a modifiable lifestyle risk factor in
prevention of breast cancer. Mammographic density (MD) is increasingly used as a
biomarker of breast cancer risk, as it is one of the strongest risk factors. Studies on
the association between physical activity and MD have reported inconsistent results.
The purpose of this study is to evaluate the association of leisure, transport related
and occupational physical activity with MD.
Material and Methods: For 5,703 (1,202 pre- and 4,501 postmenopausal) women
who participated in the Danish Diet, Cancer and Health cohort (1993-1997) and
attended mammographic screening in Copenhagen (1993-2001), we used MD
assessed at the first screening after cohort entry. MD was defined as either
mixed/dense or fatty. Participation and duration (hours/week) in leisure-time physical
activities: sports, walking, cycling (leisure and to-and from work), and gardening; as
well as occupational physical activity (sedentary, standing, manual, heavy manual),
and potential confounders were assessed by questionnaire at the cohort baseline.
The association between physical activities and MD was analysed by logistic
regression with adjustment for confounders.
Results: 56.3% women had mixed/dense MD, 47.5% participated in sports, 70.1%
cycled, 52.2% did gardening and 92.7% walked. We found a significant positive
association between participation in sports (1.17; 1.05-1.30) and cycling (1.19; 1.061.33) and odds of having mixed/dense MD, which attenuated (1.08; 0.96-1.22 and
1.10; 0.97-1.25) in a fully adjusted model. We found no association between walking
and MD (0.98; 0.79-1.22) and weak association between gardening and MD (1.06;
0.94-1.18) in fully adjusted model. We found no dose-response relationship between
time spent on activities and MD, and no association of occupational physical activity
with MD.
Conclusions: Physical activity does not affect MD. Our results suggest that the
protective effect of physical activity on breast cancer risk is not mediated by MD.
XXVI
RISK OF TESTICULAR CANCER WITH CRYPTORCHIDISM
A Danish cohort study
Sofie Have Hoffmann, Gerda Engholm1, Rasmus Hertzum-Larsen1 and Lene
Østerballe2
1
2
Danish Cancer Register, Copenhagen, Denmark
Rigshospitalet, Copenhagen, Denmark
Background
Previous studies have found that cryptorchidism, which is usually treated surgically, is
associated with an increased risk of testicular cancer, but whether age at surgery has
an effect on the risk is still unclear.
The aim of this study is to investigate the risk of testicular cancer in a cohort of Danish
men who have had surgery for cryptorchidism, compared to the general male
population in Denmark, with a special interest to the association between age at
surgery and risk of testicular cancer.
Material
Through the Pathology Department 1274 Danish boys, born before 1996, were
identified as operated for cryptorchidism at Rigshospitalet in the period 1997 to 2008.
Cohort subjects have been followed-up for occurrence of testicular cancer, death or
migration, which ever occurred first, in the Danish Cancer Register, up until end of
follow-up the 31st of December 2013. Testicular incidence rates in the general Danish
population was accessed through the Nordic Cancer Statistics Database NORDCAN.
Methods
At first, the observed number of testicular cancers among cohort members was
compared to the expected number based on the person years at risk in the study
population and the testicular cancer incidence rates for the total male population in
Denmark. Secondly the cohort was subdivided by age at surgery for cryptorchidism
and the risks estimated separately for boys operated before puberty (before age 13)
and after.
Results
Compared to the Danish general male population, the risk of testicular cancer among
men who underwent surgery for cryptorchidism was 2.85 (95% CI, 1.66 to 4.56).
If operated after the age of 13, men had an increased but insignificant risk of 2.37
(95% CI, 0.87 to 5.17), whereas the risk was 3.19 (95% CI, 1.59 to 5.71) if operated
before the age of 13.
Conclusions
Our findings confirm that cryptorchidism is associated with an increased risk of
testicular cancer, but results stratified on age at surgery (13+/-) were inconclusive.
XXVII
CASE CONTROL STUDIES ON RISK FACTORS OF BREAST CANCER
Sushmita Katuwal, University of Tampere
Background
The etiology of breast cancer is not completely understood. The aim of this population based
case control study is to get more insight on interactions of various risk factors stratified
according to characteristics of breast cancer such as histology, stage and age at diagnosis.
Reproductive factors (age at first delivery, parity, birth intervals, age at last delivery) play an
important role in the breast development and transformations. Early age at first delivery and
parity have shown to be protective for breast cancer. Pregnancy has dual effect on breast
cancer risk with transient increase in the risk for short time followed by long term protection.
Short interval between pregnancies increases exposure to steroids and incomplete
differentiation of breast cells. Hormonal therapy, most importantly oestrogen-progestagen
combinations, increases breast cancer risk. Obesity and disease conditions such as benign
breast disease, endometriosis, hypertensive diseases and lipid metabolism are associated with
increased breast cancer risk. Physical activity shows protective effect on breast cancer risk.
Mammographic screening leads to the increased diagnoses of breast cancer.
Materials and methods
All 23,000 cases of breast cancer diagnosed between 2009 and 2013 in Finland are obtained
from the Finnish Cancer Registry and five matched controls are selected for each case from
the National Population Register. The risk information is obtained with register linkages
based on the personal identity codes assigned to all residents in Finland. Multivariate
conditional logistic regression method is used to see the association of different factors with
breast cancer.
Before starting the main study – a separate case control study nested in an existing cohort of
Finnish grand multiparous (GM) women (N women), i.e., women with at least five children
will be conducted. An earlier study suggested higher risk of breast cancer if the interval
between 1st and 2nd birth was shorter than one year (RR=5.3; 95% CI 2.0–14), and the risk
was especially high during the three first year after the last birth (RR=2.4; 95% CI 1.3–4.3),
but the power of that study was limited. Utilization of data on all over 4,000 breast cancer
cases in this cohort, diagnosed between 1974 and 2014, gives strong results on the role of
reproductive events and birth intervals in etiology of breast cancer in this specific population
of GM women.
XXVIII
ATTITUDES TOWARDS GENETIC COUNSELING AND TESTING: A TESTIMONY
OF ICELANDIC WOMEN
Þórdís Jónsdóttir1,, Unnur A. Valdimarsdóttir2, Laufey Tryggvadóttir3, Heiðdís B.
Valdimarsdóttir4
1
2
3
4
School of Public Health Sciences, University of Iceland
The Centre of Public Health Sciences, University of Iceland
Icelandic Cancer Registry, Reykjavik, Iceland
University of Reykjavik, Iceland / Mount Sinai School of Medicine, New York
Background: In Iceland around 0,8% of the population carries a BRCA2 mutation.
The company DeCode genetics has information about the mutation carriers and a
long debate has been going on if the carriers should be contacted in any way. Little
is known about the interest of the Icelandic population regarding genetic counseling
and genetic testing. In a research from 1998, 74% of women were interested in
genetic testing. The aim of the study is to research the interest of Icelandic women
regarding genetic information and see if it has changed over the last eighteen years.
Material: Women attending screening at the Cancer Detection Clinic in Iceland from
the 12th of October until the 20th of November 2015 got an invitation to participate in
the research. Eligible participants were 1626 women, with 61 e-mails not working.
1565 participants got sent a short online survey and 1123 finished the research. The
response rate was 72%.
Methods: The survey included questions about background, family history of cancer
as well as knowledge about genetic counseling and genetic testing. Descriptive
statistics were used to evaluate the women’s attitudes.
Results: The mean age of participants was 49 years, with the youngest being 21 and
the oldest 76. Most of them got a college education or more (56,1%) and 80,3%
were married or with a partner, 77,1% work full time. The interest in getting a genetic
counseling was 78,8% (most certainly/likely) and 82,7% for genetic testing (most
certainly/likely). The women showed a very positive attitude regarding using existing
data about mutation carriers with 96,7% saying they would most certainly or likely
want to know the results.
Conclusion: Icelandic women show great interest in genetic counseling and genetic
testing. These results show that the interest might have increased over the last
eighteen years. These results could help with legislation and show the need for a
change in the health care system.
XXIX
INDUCED ABORTIONS IN FINNISH CHILDHOOD CANCER SURVIVORS
Viivi Seppänen1,2, Laura-Maria Madanat-Harjuoja1, Tiina Hakanen1, Janne
Pitkäniemi1, Nea Malila1,3
1
The Finnish Cancer Registry, Helsinki, Finland
University of Helsinki, Faculty of Medicine, Helsinki, Finland
3
School of Health Sciences, University of Tampere, Tampere, Finland
2
BACKGROUND: Improvements in cancer therapy have contributed to high survival of
childhood cancer patients. Due to adverse effects of cancer treatments, childhood
cancer survivors worry about health of their offspring and risks of having children.
Results of previous studies exploring induced abortions among childhood cancer
survivors have been inconsistent. Our population-based cohort study aimed to
investigate whether female childhood cancer survivors are more likely to terminate
their pregnancies by choice compared to the population controls.
MATERIAL: Using registry data of four registers, we identified induced abortions and
deliveries (livebirths and stillbirths) in 1,575 female childhood cancer survivors
(diagnosed with cancer between 1971 and 2012 at 0-14 years of age in Finland) and
in 10,503 female population controls. The Finnish Cancer Registry and the Central
Population Register were used to identify the study cohorts. Induced abortions and
deliveries were identified using the Finnish Medical Birth Register and the Register of
Induced Abortions.
METHODS: Incidence rates of induced abortions and pregnancies were compared
between childhood cancer survivors and the female population controls over the
follow-up period between 1987 and 2013. Associations between induced abortions
and cancer exposure of the parent were evaluated using Poisson regression
modelling, yielding rate ratios (RRs) with 95% confidence intervals. The generalized
estimating equation (GEE) models were used to account for dependency between
multiple pregnancy outcomes of the same mother. The variables examined in
multivariate analyses were age of mother at time of pregnancy, calendar time of
pregnancy, and parity. We evaluated if there was a change in the risk of induced
abortions over time.
RESULTS AND CONCLUSIONS: We identified 228 induced abortions and 1,173
pregnancies in 553 female childhood cancer survivors, and 1,747 induced abortions
and 10,772 pregnancies among 4,764 population controls. The study is still in process
and the results of the analyses are not yet available due to time-consuming data
management (multiple outcomes of the same mother over time) to include personyears into the model.
XXX