Finnish Cancer Registry

The CSF is subject to the processing, classification and data security requirements of the government authorities (Government Decree on Information Security in Central Government 681/2010, the so-called Decree on Information Security).

There is a head of information security responsible for data security at the CSF, a person responsible for technical data security, and a working group that deals with data security issues. Staff are continually trained in handling data and in general information and data security instructions.

Cancer registration

The compilation of cancer cases into a national registry begins with notifications, on the basis of which a case summary for statistical and research purposes is created. Statistics from Statistics Finland on the cause of dealth of individuals and their socio-economic status, profession and education are updated at the Cancer Registry. Data on deaths and domicile information, as well as migration and immigration data of registered persons can be obtained from the Population Information System of the Population Register Center.

The tasks related to the cancer registration process include data collection, recording and encoding, quality assurance and documentation, plus information on lifecycle management and data disclosure.

Health care organisations statutorily provide the Cancer Registry with information on cancer cases. In February 2016, an online cancer notification tool was introduced, but most of the clinical cancer reports were still on paper. Almost all laboratory announcements were received in 2016 as electronic data communications.

Total cancer notifications 2016
Laboratory notifications	Clinical notifications
136 536 (about 700 on paper)	16 934 (online service 989)

With the development of information systems and the digitalisation of health data, the reconfiguration of cancer information and screening declarations has been considered necessary. The THL Code Service Management Team approved new data models in autumn 2016, to be published in spring 2017.

During the year in review, many improvements were made to the quality of the database, such as a storage base in accordance with the new data content specification. Due to the renewed content and staff changes, coding and, in particular, the recording of cancer announcements, clearly slowed down.

A project was launched at the beginning of 2016 to develop a data management plan for the information security and data protection requirements of the Cancer Registry.

Services

 

The Finnish Cancer Registry assists clinical cancer research by providing ready material, guidance in conducting research and analysis, and guidance in selecting the right type of research. The registry provides statistical analysis services to cancer researchers at all stages of research projects. Routine service activities include basic calculations of cancer incidence, mortality, and survival. In 2016, several hospital district-specific cancer forecasts were made.

 

In 2016, a total of 75 data submissions were recorded in the data management of the registry, of which 16 were internal projects (21%). The total number of submissions was 34. The average processing time for data submissions was 22 days. In addition, information was provided on specific requests for information to authorities, experts, researchers and the media. There were 17 requests for statements and 17 notifications for which a statement was given. The average processing time for the statements was 19 days.

 

In 2016, service activities doubled compared to 2015. This was due to the standardisation of price lists and processes and the demand for the invoiced service. A new service price list was published online at the end of 2016 (effective from January 1, 2017), which has opened up the Cancer Registry’s transparent and fee-free information products and services.

Information submissions
N (%)		Invoiced
Project	29 (39%)	28                                                             165
Internal	16 (21%)
Collaborative project	9 (12%)
THL	4 (5%)
Info.submission	17 (23%)	6	58
Total	75 (100%)	34	224

The cancer statistics for 2014 were completed in April 2016. The statistics are available on the Cancer Registry website. Based on the feedback from the field, it was decided that the statistics would only be published in digital format. Interactive cancer statistics were also published at the end of 2016 http://www.cancer.fi/syoparekisteri/en/statistics/

The invitation service that operates at the Mass Screening Registry handles the extraction of women from the population register who are to be invited for screening commissioned by municipalities and sends invitations letters and forms to some municipalities. In 2016, selections from the population register were centralized for the first time at the turn of the year and were made directly from the Population Information System. Also, two extractions were made in the spring of 2016. For paper invitations and forms, municipalities were sent screening material via a security mail in pdf format for printing at the screening points. The new method reduced the number of municipalities subscribing to paper materials to 30.

The Cancer Registry hosted the annual conference of Nordic Cancer Registries in August 2016, in Tuusula. The director of the Registry and the statistician in charge of information submissions participated in the conference of the European Network of Cancer Registries held in Italy in October 2016.

Screenings

Altogether, the Mass Screening Registry collects the invitation and inspection data of about 30 cervical cancer screening laboratories and from breast cancer screenings and the national colorectal cancer screening programme. The information is obtained in electronic format. In 2016, most of the data from screenings were sent to to the registry in time, by the end of July. However, the data deliveries of certain laboratories were delayed, and in the case of some of them, there was a lack of information in the initial phase.

In 2015, about 280 000 women mainly aged 30-60 were invited for cervical cancer screening. About 350 000 women aged 50 – 69 were invited for breast cancer screening. About 20 000 men and women aged 60 – 69 were invited for colorectal cancer screening who had previously received one screening invitation.

In 2016, further efforts were made to reform the data structure of national cancer screenings. The standardised, six-tier and hierarchical structural model was prepared in the autumn and presented in October to the Code Service Management Group of the national Institute for Welfare and Health (THL), which approved it.

The renewal of the SQL databse was transferred to 2017. The new data model will enable the flexible, cutting edge updating of data content, and the registration of non-screening test data outside. The model also seeks to harmonise the screening information in the registry provided by the different actors.

The mass screening statistics for 2013 and 2014 were prepared in April 2016. The statistics are available on the website. In September 2016, an online tool was developed for examining mass screening data in table or graphic format, or for downloading to the desired extent in Excel for further work. The tool will be ready during spring 2017.

In spring 2016, the Ministry of Social Affairs and Health’s working group on screening was given a one-year extension. This was for it to produce information and decisions on how cancer screenings will be implemented in Finland in the future. The working group meetings dealt with screenings for cervical cancer, breast cancer, colorectal cancer and other cancers, the organisation and administration of screenings, and information management requirements. At the end of the year, decisions were made on extending the age groups of cervical and breast cancer screening invitees, and the inclusion of the screening of colorectal cancer in the decree on screening. In addition, it was decided to continue with the development of the screening control structure.

A three-year key project of the Mass Screening Registry started in early 2016, the aim of which is to bring about centralised cancer screening control structures in Finland. These will include a national steering group, program-specific management teams, and a cancer screening centre. The aim is also to focus on the organisation of screening and to ensure screening-type testing and the nationwide registration of related follow-up examinations and treatment measures. In addition, the aim is to create new channels of communication and methods to support screening, guidance and control. During 2016, these goals were promoted by actively participating in the screening workgroup and by coperating with working groups dealing with health care legislative reform (Tireka / the National Archive of Health Information / THL).

The continuation of screening of colorectal cancer in its present form came into question in spring 2016, as no differences were found between Finns in colorectal cancer screening and the control groups. Since screening had proven to be effective in previous randomised trials, and EU recommendations continue to support screening, it was decided in Finland to make a plan to investigate the effectiveness of screening in men and women using new test methods and approaches. The plan received a two-year grant from the Finnish Cancer Society’s Potilaan Polku fund. The aim is to extend the screening of intestinal cancer to a national programme and involve it in legislation in 2017. The research design will be integrated into the startup stage of the new programme.

Statistics production and research

The main focus in 2016 in terms of the study of statistical methodology was to calculate the population share assessment, measure regional variation in the survival of cancer patients and to find a reliable method for a long-term (10-15 years) survival assessment.

A study of figures for long-term survival was published in 2016 in the journal Statistics in Medicine, and gained visibility in national media. The development of a more efficient calculation to measure regional variation in survival figures continues. The registry statistic team developed a method for calculating the inherited variance in population-based family data and applied it to the assessment of the variability in childhood cancers. The computational hierarchical Bayesian method showed that the there are two components in terms of age in the incidence of cervical cancer, due possibly to mutually different cancer risk factors.

Methods of computational statistics

Computational statistical methods have been increasingly used in the statistical and research activities of the Cancer Registry. The R statistical software programme package (popEpi) has been developed to enhance research activity. This calculates the big data from age-adjusted relative rates of cancer and mortality and utilises their statistical modelling. This was introduced for worldwide distribution on the official website of the R programme and the use of the package has been supported by teaching it, such as with international cancer research courses.

Written, graphic and tabular statistical information was produced for the online publication Cancer in Finland using the R software programme, developed at the Cancer Registry. The English version of the online publication was published in August 2016. The Finnish and Swedish sections will be published in 2017. Medical researchers have been supported by developing service activities and consultation on the basis of developed statistical analytical tools. The productisation of this activity was improved and its development continues. In 2016, interactive cancer statistics interactive cancer statistics were released, in which users can view the cancer situation Finland using statistical indicators or graphs (columns, diagrams, and maps).

In 2016, the Cancer Registry participated actively in supporting a number of external cancer studies. Such projects included, among others, survivorship by relatives of people with familial adenomatous polyposis (FAP), prostate cancer survivorship was investigated before and after the introduction of PSA testing in Finland, and the Registry’s research found that neurophibromatosis patients have an increased risk of breast cancer.

Methods were developed in cooperation with third parties. Collaboration with Aalto University was continued on the statistical model of cancer time series, based on modelling of independent non-direct measurable factors. This was applied to the time series of cervical cancer incidence but none of the latent components could be verified by a new method. Research project work continued with the Department of Computer Science of the University of Helsinki, on the application of machine learning to assist cancer coding. Research has continued with the University of Oulu to investigate statistical methods on cancer survival.

The most important of the registry’s own epidemiological research projects was the population-based comprehensive cohort study of cancer risk factors (METCA), which in 2016 provided the first estimates of the population’s share of risk factors in the Finnish population, when competing mortality has been taken into account in statistical analysis. The harmonisation of the material continued throughout the year. Preliminary results were presented at the conference of the Nordic Cancer Registry in August 2016. In addition, a project evaluating the effectiveness of the screening of colorectal tumors was demonstrated, indicating that screening has indirectly contributed to the improvement of the survival of patients not invited for it. For the first time, the frequency of non-organised breast screening in Finland was also studied. It transpired that two-thirds of women had been in mammography before the start of organised screening.

The Cancer Registry research work focuses on identification of causes of cancer at population level, the prevention and early detection of cancer, and factors affecting the survival of cancer patients. The research activities of the Mass Screening Registry are related to the evaluation and renewal of statutory screening and new forms population-level screening methods.

The late effects of cancer treatments in childhood and adolescence were investigated in collaboration with the research group in the University Central Hospital of Turku. Patients bought medicines for cardiovascular disease, hypertension and diabetes more than their healthy siblings. A joint Nordic study on socio-economic late effects started in early 2016 between Danish, Finnish and Swedish research teams.

A Nordic cancer study of twins confirmed that susceptibility to cancer of the skin, prostate, breast, ovaries and uterus is to some extent hereditary. A Nordic study of the whole population found a greater than usual incidence of lymphoma in the relatives of lymphoma patients. The likelihood of patients with neurofibromatosis (NF1) in Finland getting cancer by age 50 was 39%, or 10 times higher than among the rest of the population. In addition, NF1 cancer patients had a worse recovery prognosis rate than the reference population.

A study on nickel-wood workers was published concerning the cancer risk of individual Finnish industries. They found 2-3 times more than the normal incidence of nasal cancer and twice the risk of lung cancer. Cancer risk in some of the other industrial sectors was also investigated, but no abnormalities were observed. Even many decades after exposure to asbestos, there is still an increased risk of mesothelioma.

Although active period of the joint Nordic Occupational Cancer Study (NOCCA), led by Finland, on work related exposure to cancer risks, funded by the Nordic Cancer Union, ended in spring 2015, studies based on this research material and methods were still published at a rate of about one article per month in 2016. The results mentioned the protective effects of work-related physical activity on colorectal cancer, especially in men. However, exposure to solvent was not found to significantly increase the risk of bladder cancer, nor was exposure during pregnancy to heavy metals found to increase the risk of testicular cancer in male children.

A multinational series of follow-up studies on the risk of cancer patients receiving treatment being at later risk of getting another cancer, it was observed that testicular cancer patients receiving radiotherapy were at nearly three times the risk of pancreatic cancer.

There was a connection observed between the use of hormonal IUD in young women and a greater risk of breast cancer, most strongly lobular breast cancer, while on the other hand the risk among the same women of musinosis and endometrial ovarian cancer was almost half of the normal risk. In women with more than five births, there was 30-40% lower morbidity associated with breast, uterine and ovarian cancer, but other cancer morbidity did not differ from women who had had fewer births.

As expected, there was a significantly increased risk of cancer observed due to alcoholic liver cirrhosis or hepatitis due to the use of alcohol. The risk of many other types of cancer was also high, so that the total cancer incidence was almost triple compared to the rest of the population.

Research questions linked to the use of different pharmaceutical were investigated with data from Finnish Randomised Study of Screening for Prostate Cancer (FinRSPC)

A study on the use of nonsteroidal anti-inflammatory drugs (NSAIDs) found that use of prescription NSAIDs was associated with an increased risk of cancer mortality compared to non-users. The finding is likely to be due to the use of NSAIDs alleaviating symptoms in symptomatic prostate cancer. The risk of cancer mortality was not observed when the last few years of use were eliminated from the analysis. A similar result was obtained for the overall cancer mortality rate.

There was no link observed between the use of antiarrhythmic drugs, and in particular digoxin and sotalol, and prostate cancer risk, survival or mortality compared with other antiarrhythmics. However, long-term use of digoxin may have a beneficial effect on the risk of particularly poorly differentiated prostate cancer. Previous studies have been indicative of the protective effect of warfarin concerning prostate cancer. The result from the FinRSPC data was the converse, and the risk of prostate cancer was higher for warfarin users. However, this is limited to short-term and low-dose use. The use of a 5α-reductase inhibitor in the treatment of benign prostatic enlargement was not found to be associated with cancer specific or overall mortality, even in long-term use. The relationship between the statins used to treat hypercholesterolemia in prostate cancer mortality was found to be associated with lowered cancer mortality associated with statins. The use of statins after post-cancer diagnosis had the most significant effect, and the effect on mortality remained for years after use was discontinued. Drug studies also assessed the use of diabetes

drugs in connection with prostate cancer. Metformin users found a decreased risk of prostate cancer, whereas the risk of metastatic disease was elevated in users of sulphonylurea. Sulfonylurea accelerates the production of insulin, so it is possible that high insulin levels are one of the risk factors for prostate cancer.

International joint projects

The EUTOPIA  (Towards Improved Screening for Breast, Cervical and Colorectal Cancer in All of Europe) research project got underway at the Cancer Registry with funding from the EU Horizon 2020 programme. Altogether, there were four sub-reports that defined the benefits and disadvantages of cancer screenings and their indicators; health and social inequalities and indicators; and the target values ​​for the feasibility and quality parameters of the screening for the project. From 2016-2020, the project will assess the benefits, disadvantages and health economics outcomes of cancer screening based on these definitions, together with the project’s other partner countries and stakeholders in the project.

The implementation of the project on the EU countries’ cancer screenings collected data on the most important feasibility and quality indicators for cervical, breast and intestinal cancer screenings. The final report of this project was completed in the latter part of the year for publication in early 2017. The project has been coordinated by the International Agency for Research on Cancer (IARC) under the auspices of the WHO, and has provided expert work on the basis of a cooperation agreement between the CSF and the IARC. In 2016, the development of a joint Nordic screening reporting system (NordScreen) was also launched. Finland is in charge of leading the project.

Work Package 9, on cancer screening, of the pan-EU Cancon Joint Action completed the chapter of the Cancon Guide dealing with developing cancer screening, quality assurance, coordination and evaluation as part of broader guidelines. The study also discussed potential new screening programmes (other than cervical, breast and colorectal cancers).

The CSF organized a closing meeting on the screening package in Helsinki, attended by some 55 participants. The meeting discussed the results achieved and the planned follow-up. The Cancon Guide was released in early 2017.

Breast cancer screening. In 2016, a case-comparison study on the effectiveness of breast cancer screenings in Finland was completed. The study monitored the breast cancer mortality rates of those invited for breast cancer screening from 1991-2012. The impact of the screening was evaluated by participants in the screening and non-participants (ie non-attendants are at greater risk of dying of breast cancer than the rest of the population). According to the study, mammography screening reduced breast cancer mortality by 33% among those screened, as mortality was monitored up to 84 years of age. In the screening age (50-69 years), the decline in breast cancer mortality was 39%. The relative preventive effect of screening for breast cancer mortality has remained unchanged this century.

In summer 2016, the IARC published a Handbook on Breast Cancer Screening, which in a systematic review evaluated the most important benefits and drawbacks of screening. The director of research at the Mass Screening Registry was the vice-chair of the evaluation meeting of the working group concerned with the initiative. The aforementioned results of the effectiveness of breast cancer screenings in Finland were very similar to those of IARC for breast cancer screenings in several other countries (eg Sweden, Denmark, Norway, Netherlands, Italy, Canada, USA) and there is sufficient evidence that the current screening of women of 50-74 years of age reduces mortality.

The disadvantages of breast cancer screening were investigated in 2016, mainly with respect to false positive mammography results found in screenings. In the Finnish screening programme, the average probability of getting a false positive mammography at least once in the age of 50-69 is slightly below 20 percent. The average probability is due to a large variation, for example, due to the diagnosis criteria of the screening centers. Additionally, if a woman had reported significant breast symptoms with screening (lump, secretion or retraction), the probability of false positive test results increased by almost two-fold. On the other hand, with breast symptoms the likelihood of screening cancers increased significantly with the screening visit. During the year, new studies on overdiagnosis of breast cancer screening were also launched in women aged 50-69 years.

In Finland, the breast cancer screening programme is currently carried out between the ages of 50-69, and according to the IARC, there is also sufficient evidence of the effectiveness of screening for 70-74-year-olds. According to regulations on medical radiation and international screening recommendations, breast cancer screening should only be arranged within a well-organised population-based programme. It is likely that mammograms will still be taken from asymptomatic women outside the screening program. In a 2016 survey, it was found that almost two out of three women in the survey had mammograms at least once under the age of 50. Based on this survey, we can estimate that screening-type, non-programmatic opportunistic testing is still relatively common in Finland, and there is little information available on its benefits, adverse effects and cost-effectiveness.

In 2016, a decision was taken to review the material of the survey of breast cancer and quality of life in breast cancer screening. The survey was conducted in such a way that women born in 1963 (n = 5000) and 1964 (n = 5000) were sent a postal survey form about a year before the first breast cancer mammography invitations in 2012 and 2013. The survey was repeated in the first year following the first screening invitation, in other words two years after the previous survey, in 2014 and 2015. The material will be used in the coming years in collaboration with Aalto University’s research team on methodological studies of the concern aroused by breast cancer and related factors, and the relationship between regional and socioeconomic factors in breast cancer screening projects.

Cervical cancer screening

The Cervical Cancer Screening Program evaluated the lifetime cumulative probability of abnormal test results and histologically confirmed cervical cancer precursors in register data over the period1991-2012.

Based on cancer statistics, the cumulative incidence of cervical cancer among women aged 0-84 in Finland today is about 0.5%. Without screening the figure would be about 2.5% (about five times higher). The probability of obtaining a referral for further examination in the screening program at age 30-64 was at least 5.4%, and cervical cancer precursors were found to be slightly above 2%. The result suggests that precursors are also found and handled through screening visits. On the other hand, with the recommended follow-up test (risk group screening, the culmulative probability of screening findings was as high as 34 percent.

In 2016, a prospective research programme for optimal invitation practices was completed. According to a study published under the project, there were no significant differences in the acceptability of sampling methods for home HPV tests (rinsing vs. bristle). This and other previous research results were utilised in preparing the updated version of the Current Care Guidelines concerning the diagnostics of cervical, vaginal, and vulval cellular changes.

In 2016, data collection was also started on smear tests outside the screening programme. Collected material will be utilised in the future to evaluate their impact and cost-effectiveness. In 2016, there was a collaborative project with Aalto University to investigate the impact of socioeconomic status and geographic location on participation in cervical cancer screening.

Colorectal cancer screening

Studies on the screening for bowel cancer cancer are underway to examine the possible correlation of the lifestyle choices of screening participants. The topic is important because, according to randomised screening studies, screening has reduced colorectal cancer mortality, but no corresponding reduction in overall mortality has occurred. It has been suggested that this is due to the selectivity of screening participants and possibly also to the fact that screening participants may experience life-style changes in an unhealthy direction.

In the case of randomised colorectal cancer screening program, lifestyle information was collected by questionnaire in 2010 and 2012 of people born in 1951 (who were screened for the first time in 2011). In 2016 a study was carried out to investigate the randomisation of the screening group and the connection between demographic factors and answering the questionnaire.

A second study combines adult behavioral health information with screening attendance data. According to this study, smokers participated in screening less well than non-smokers (RR 1.3, 95% confidence intervals 1.05-1.67 in men, RR 2.1, 1.61-2.73 in women) but in this context it was not possible to determine possible changes in habits. Although non-screening attendees smoke more than those attending screening, there were no differences in lifestyle that would affect the risk of colorectal cancer.

 Prostate cancer screening research

Prostate cancer is the most common cancer in men and the second most common cause of death from cancer. The randomised screening survey at the monitoring stage covers data from approximately 80,000 men over a period of 20 years. Three serum PSA-based screening trials have been successfully performed. The Finnish project is clearly the largest centre for the European Randomised Study of Prostate Cancer Screening (40% of the total).

In the Finnish Randomised Study of Screening for Prostate Cancer (FINRSPC) data, the post-screening prostate cancer exposure was investigated in relation to the number of screening cycles. The results of the study showed that at least three screening rounds with PSA were required to reduce prostate cancer after screening.

The effect of prostate cancer screening on socioeconomic (SES) differences in prostate cancer incidence and mortality was also studied in the FinRSPC data. The upper socioeconomic position of men had a lower incidence of progressive cancer and a higher incidence of good prognosis cancer. They also participated in screening more actively than men in the lower SES group. The highest education and income group had lower prostate cancer mortality rates in the screening and control group, and screening did not reduce the mortality difference between SES groups.

In 2016 too, the validity of statistics on prostate cancer deaths and the accuracy of statistical death tolls in screening and control groups in Finland were examined. The accuracy of the causes of death was very good, and there were no significant differences between the groups.

The European prostate cancer screening study has shown a 21% reduction in prostate cancer mortality during 13 years of follow-up. The mortality rate achieved through screening varies between different screening centres. According to studies conducted in 2016, differences in the screening protocol did not explain Finland’s lower mortality impact compared to other countries. Nor were there any differences between the screening and control groups in defining cause of death that could explain the differences between the centers in the mortality decline.