United Kingdom Childhood Cancer Study (UKCCS)

The University of York requires access to NHS England data for the purpose of the following research project: The United Kingdom Childhood Cancer Study (UKCCS).

UKCCS is one of the largest and most comprehensive childhood cancer case-control studies ever undertaken.

With the initial aim of investigating causes of cancer diagnosed before 15 years of age, the UKCCS was set-up in the early 1990s. This ground-breaking study, one of the largest and most comprehensive ever undertaken, examined the potential causes of the development of the disease and the role of a wide range of biological, physical, and chemical agents acting on parents before the child’s conception, and the child during pregnancy and in early life. In brief, the UKCCS was specifically designed to make maximum use of routinely compiled administrative data; linking all childhood cancer cases and their potential controls to birth notifications using pre-1996 NHS numbers, and area-based census measures using residential postcode. In addition, rather than relying solely on information reported by parents at interview, the study accessed medical records to answer key questions relating to prescription drugs, illness histories and co-morbidities. In the 2000s, UKCCS extended its remit to investigate health outcomes and patterns further along the patient pathway, demonstrating the utility of UKCCS’s population-based national data to examine survival variations by cancer type; it is this area which offers promise for future research.

Survival of children diagnosed with cancer has improved dramatically in recent decades, exceeding 80% in the UK and other economically developed countries. Indeed, there are currently about half a million people who have recovered from having cancer as a child in Europe alone; and this will increase as therapy improves and populations age. However, whilst the majority of cancers that occur in children are more responsive to chemotherapy than those occurring in adults, treatments are often more aggressive, and it is well known that this can lead to adverse health problems later in life (e.g. second cancers, cardiac conditions, and bone problems). The UKCCS is a population-based epidemiological study that has, and continues, to inform on potentially avoidable causes of childhood cancer. With linkage to secondary healthcare data, the UKCCS aims to inform the public health agenda of the future by providing information about the health and healthcare needs of adults diagnosed and treated for cancer as children.

Diagnostic-specific comparisons between cases and controls will be used to examine long-term health conditions including secondary cancers, skeletal-related problems, cardiac conditions, respiratory illnesses and fertility issues, as well as investigating hospital activity, survival and whether those who had cancer as children ever return to background levels. Comparisons will be made with UKCCS controls, who being born in the same year and of the same sex as those diagnosed with cancer as children, are an appropriate population comparator.

The following is a summary of the aims of the research project provided by the applicant on behalf of the University of York:

Diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine:

(i) Secondary care activity patterns (hospital inpatient and outpatient)

(ii) Reasons for inpatient admissions including surgical interventions

(iii) Outcome variations (mortality, cancer and secondary care) using area-based measures of deprivation.

(iv) The potential impact of selection-bias at the time of the original study in relation to i-iii above

The following NHS England Data will be accessed:

• Linked Hospital Episode Statistics

o Admitted Patient Care (HES APC)

o Accident & Emergency (HES AE)

o Outpatients (HES OP)

These will be used to examine health and healthcare activity patterns: comparing patterns of those who had cancer as a child to those who did not.

• Emergency Care Data Set (ECDS) – necessary because these data will be used with HES AE to examine whether people who had cancer as a child have different emergency care activity patterns to those who did not have cancer as a child.

• Civil Registrations Mortality – necessary because the mortality of childhood cancer survivors will be examined relative to the general population controls as early mortality is one poor outcome childhood cancer survivors may experience. For the analysis of health and healthcare patterns, it is also important to know if, and when, study participants have died as this will inform their longitudinal follow-up.

• Cancer Registration – necessary to examine what subsequent cancers childhood cancer survivors may develop and whether these are different to those which occur among the general population controls. Cancer registration data may also inform the patterns of health and healthcare activity UKCCS plan to examine among the childhood cancer survivors and their population controls.

• Demographics – necessary because as the study is following study participants over time, it would be useful to know if the study members are no longer registered with a general practice in NHS England; this information being recorded in the reason for removal and reason for removal date fields. This will inform the longitudinal follow-up as participants that have moved out of England are not likely to have secondary healthcare activity within HES.

The level of data will be at the individual person-level and will have the unique study identifier attached to each record. No identifiable data are being requested at present.

The Data will be minimised as follows:

• Limited to a study cohort identified by the University of York. The original UKCCS had a case-control design: each child diagnosed with cancer aged under 15 years old (n=4430) during the study period (1991-1996) was matched on sex, date of birth (month and year), and area of residence (Family Health Service Authority in England & Wales, and Health Board in Scotland) to at least two children of the same age who did not have cancer. To ascertain children diagnosed with cancer, proactive notification systems were established across England, Scotland, and Wales; the majority of cases being notified at the point of diagnosis directly to the study by paediatric oncologists or haematologists. For each case, control children (matched on sex, year and month of birth) were randomly selected from the same Family Health Service Authority (FHSA – replaced by Health Authorities in 1995) in England and Wales or Health Board (HB) in Scotland. In total, 4430 cases and 9758 controls were registered. Around a third of the 4430 children diagnosed with cancer have since died, as have some of the control children. The total cohort size is now 1264.

The University of York is the research sponsor and the controller as the organisation responsible for ensuring that the data will only be processed for the purpose described above.

The lawful basis for processing personal data under the UK GDPR is:

Article 6(1)(e) - processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller.

The lawful basis for processing special category data under the UK GDPR is:

Article 9(2)(j) - processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject.

This processing is in the public interest because it adheres to the UK Policy Framework for Health and Social Care Research, which protects and promotes the interests of patients, service users and the public, and aims to produce generalisable and publicly available information to inform future decisions over patients’ treatments or care. UKCCS compare the health of childhood cancer survivors to that of their peers as they age, making it uniquely placed to investigate the long-term healthcare needs of UK participants who have recovered from having cancer during childhood.

Cancer Research UK (CRUK) funds the UKCCS; the funder has no involvement in the conduct of the research, and no data controller or processing responsibilities. CRUK are the only funder and manage all aspects of the project award. Funding was first awarded in July 2024, and is specifically for the study described.

The funder(s) have no ability to suppress or otherwise limit the publication of findings.

The University of York is the only data processor and the only data controller.

Data will only be accessed by substantive employees of the University of York.

The acceptability of using patient identifiable information to link to routinely compiled administrative health data without explicit consent and using the resulting data for research purposes, has been tested in several forums with both patient and parent representation from the National Alliance of Childhood Cancer Parent Organisations (now the Childhood Cancer Parents Alliance http://ccpa.org.uk/) prior to linkage to death notifications and tracking; and more recently, with regard to linkage to secondary care data, with patient groups in UKCCS's sister study the Yorkshire and Humberside Haematology Network (https://yhhn.org), and clinicians who treat children, adolescents and young adults. Collection of data for linkage was fully supported.

As part of this study, UKCCS are initiating an active patient partnership, in line with all studies conducted by UKCCS research group (see https://sites.google.com/york.ac.uk/bloodcanceropenday/home). As part of the CRUK funding for UKCCS, PPI are recruiting a small team of adults who had cancer as a child to contribute, develop and comment on aspects of the study ranging from research questions to materials for dissemination of outputs. A committee will oversee the patient partnership, whose members will be drawn from among clinicians, patients and their relatives, specialist nurses, and researchers with relevant experience. Preliminary discussions with paediatric oncologists and nurse consultants as well as a number of people diagnosed with cancer in their childhood (who are now young adults) have already taken place. Headed by an independent chair, this advisory group will advise on UKCCS public/patient facing profile, including the production of literature and development of the website.

The expected outputs of the processing will be:

• Submissions to peer reviewed journals with at least three publications anticipated over the next three years.

• A report of findings will be disseminated via patient groups, UKCCS websites, charity partners and social media.

• Presentations at UK Childhood Cancer Conferences, National Cancer Research Institute Conferences, and the Pan-European Network for Care of Survivors after Childhood and Adolescent Cancers Conferences.

The outputs will not contain person level NHS England Data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived.

The outputs will be communicated to relevant recipients through the following dissemination channels:

• Journals

• Co-hosted events e.g. conferences including UK Childhood Cancer Conference, National Cancer Research Institute Conference, and Pan-European Network for Care of Survivors after Childhood and Adolescent Cancers forum

• Reports aimed at patients, including via social media, newsletters and at meetings of patient-public interest groups.

Target dates for outputs on the topics is within 3 years of receiving the data. By way of example, the analysis on therapy-related associations will commence three months after receipt of the linked data and will continue for 26 months when findings will be ready for submission to scientific journals.

A major aim of UKCCS is to improve care and outcomes for those who have recovered from cancer following a diagnosis as a child. UKCCS, an internationally recognized study with a longstanding record of informing on the possible causes of childhood cancer, is ideally placed to provide national data about the long-term health and healthcare needs of these participants. Indeed, the study has already provided population-based data on childhood cancer outcomes, its variation by cancer type, and socio-demographic characteristics. Importantly, the UKCCS is truly population-based and results are generalizable. Hence, findings have the potential to help inform policy and commissioning of services.

Thus far, the UKCCS has used linked HES, cancer and death data to examine morbidity and mortality among survivors of childhood acute lymphoblastic leukaemia through their teenage and young adult years (BMJ Open doi: 10.1101/2021.07.27.21261102). Reporting on the types of conditions that occur following a diagnosis of the most common childhood cancer, this is an example of the type of information that the study can provide to non-cancer specialists, general practitioners, and patients. Not only will this piece of research inform on outcomes for patients diagnosed in the early 1990s- the UKCCS’s period of recruitment- but the study can also provide baseline data against which the potential impact of treatment de-escalation and other protocol modifications of later years can be compared.

UKCCS has also published the design of the follow-up of childhood cancer survivors and the comparison cohort (International Journal of Epidemiology doi: 10.1136/bmjopen-2023-073712). This research shows that not only does the study have virtually complete coverage of childhood cancer survivors who were diagnosed in Britain in the early 1990s, but also of a population-representative sample with which the health of survivors can be compared. With additional data available for both cohorts such as socioeconomic status, this research also shows that study will be able to answer more specific analyses than is possible for survivor cohorts which rely on population rates for comparison; an example where the UKCCS can provide useful information is whether any inequalities in health among survivors is different to that expected among the population controls.

This Data Sharing Agreement permits the retention of the data provided under previous iterations of this Agreement for an interim period. This is a pragmatic approach to provide an active Agreement whilst enabling The Institute of Cancer Research to complete the necessary actions to enable a subsequent application to extend the Agreement meeting all applicable data sharing standards as published in NHS England’s website (see: https://digital.nhs.uk/services/data-access-request-service-dars/dars-guidance).

1) Aim and Purpose of the Application

Building on one of the largest and most comprehensive childhood cancer case-control studies ever undertaken- the United Kingdom Childhood Cancer Study, UKCCS, this project aims to link to national datasets to investigate a range of questions relating to the long-term health and healthcare needs of participants who had cancer when they were children. Diagnostic-specific comparisons between cases and controls will be used to examine a wide range of factors, including known and suggested therapy-related associations, as well as second cancers and late mortality. In addition, secondary care activity patterns will be examined to investigate whether those who had cancer as children ever return to background levels.

a. Justification of processing data

Article 6(1) of the GDPR (lawfulness of processing), the reference in point (e) to processing of personal data that is necessary for the performance of a task carried out in the public interest or in the exercise of the controller’s official authority includes processing of personal data that is necessary for… (c) the exercise of a function conferred on a person by an enactment or rule of law.

Article 9(2)(j)

(processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject)

Survival of children diagnosed with cancer has improved dramatically in recent decades, exceeding 80% in the UK and other economically developed countries. Indeed, by 2020 it is estimated that there will be half a million people who have recovered from having cancer as children in Europe alone; and this will increase as therapy improves and populations age. However, whilst the majority of cancers that occur in children are more responsive to chemotherapy than those occurring in adults, treatments are often more aggressive, and it is well known that this can lead to adverse health problems (e.g. second cancers, and cardiac conditions, and bone problems) several years later. The UKCCS is a population-based epidemiological study that has, and continues, to inform on potentially avoidable causes of childhood cancer. Now seeking to expand its remit, the UKCCS, with linkage to secondary healthcare data, aims to inform the public health agenda of the future by providing information about the health and healthcare needs of adults diagnosed and treated for cancer as children.

b. Explanation of public task

The majority of studies examining the long-term health outcomes of those who have recovered from having cancer as children have relied on subjects’ participation and self-reported health outcomes, rather than electronic health records, which can offer comprehensive follow-up data. Furthermore, the interpretation of study findings in this area of research can be limited since information on background levels of health events in the population is often missing from UK studies, which have tended to use hospital admission data to generate comparison populations, meaning that individuals with no admissions are excluded. Hence, with a view to providing a unique and valuable resource for childhood cancer studies, cases and controls from the UKCCS are followed as two cohorts by linking individuals to routinely compiled NHS data. This enables the UKCCS to put the health of childhood cancers into the context of their peers as they age, making it uniquely placed to investigate the long-term healthcare needs of UK participants who have recovered from having cancer during childhood.

During the UKCCS data collection phase (1991-1996), consent was sought from the study subjects’ parents, in line with the rules and regulations at the time. This consent allowed participants and family members to fill in survey's and questionnaires - from which the data was used in the study analysis. The study subjects were children (aged under 15 years) during this period, and no contact with families have been maintained. The children themselves, who are now adults (aged 23 to 42 in 2019), have never been approached and asked to consent to the use of individual identifying information for linkage purposes.

Section 251 support was historically granted under the reference number ECC 5-02(FT5)/2009 to allow NHS England to disseminate mortality data on the cohort it had flagged.

Around a third of the 4430 children diagnosed with cancer have since died, as have some of the control children; and the only way to attempt to trace and obtain consent from living cases and controls (many of whom may be unaware of their involvement) would be to use the identifiable data UKCCS already hold, which is neither justifiable or practicable. In its absence, the acceptability of using patient identifiable information to link to routinely compiled administrative health data without explicit consent and using the resulting data for research purposes has been tested in several forums with patient and parent representation. Consensus was to support linkage following discussions in 2004/5 with representatives from the National Alliance of Childhood Cancer Parent Organisations (now the Childhood Cancer Parents Alliance http://ccpa.org.uk/) prior to linkage to death notifications and tracking; and more recently, with regard to linkage to secondary care data, with patient groups in UKCCS' sister study the Yorkshire and Humberside Haematology Network (https://yhhn.org), and clinicians who treat children, adolescents and young adults. In 2018, UKCCS was granted Section 251 support (24th July 2018, 18/CAG/0066) after gaining renewed ethical approval (27th April 2018, REC 18/YH/0135) to continue linkage to national death notifications, cancer registrations and movements in/out of the NHS, and to add linkage to secondary healthcare datasets. This approval superceded the historical approval ECC 5-02(FT5)/2009.

2) How will the data requested achieve the aim identified?

Building on existing infrastructure, the principal aim of the study is to examine the long-term health and healthcare needs of those who have recovered after having cancer as a child. The Long-term health conditions UKCCS intend to investigate include secondary cancers, skeletal-related problems, cardiac conditions, respiratory illnesses and fertility issues, as well as investigating hospital activity and survival. Comparisons will be made with UKCCS controls, who being born in the same year and of the same sex as those diagnosed with cancer as children, are an appropriate population comparator. Importantly, the data requested offers contemporaneously recorded information, available for all study subjects and by its nature free from differential recall bias; this design often produces very different results from those emanating from studies reliant on participation and self-report. In UKCCS analyses, data from the different administrative sources will be used together, for instance censoring the occurrence of certain events or hospital activity at a defined timepoint, whether at secondary cancer, movement abroad, or death for example, is clearly relevant in longitudinal studies. As for the events of interest, UKCCS know from published literature and UKCCS' own feasibility work conducted in a dataset with less follow-up post-cancer diagnosis than UKCCS (https://yhhn.org/research/comparison), that these adverse health events can be identified in Hospital Episode Statistics (HES). Some detectable differences between cases and controls are anticipated within the twenty years of data that can currently be accessed; as study subjects are presently aged 23-42 years, other health events will become more common as they age, so continuing tracking of patients offers opportunity for further research as the data matures.

3) Background of study

With the initial aim of investigating causes of cancer diagnosed before 15 years of age, the United Kingdom Childhood Cancer Study (UKCCS) was set-up in the early 1990s. This ground-breaking study, one of the largest (4430 cases and 9758 controls) and most comprehensive ever undertaken, examined the potential causes of the development of the disease and the role of a wide range of biological, physical, and chemical agents acting on parents before the child’s conception, and the child during pregnancy and in early life. Predicated on NHS infrastructures, the UKCCS remains unique in several important respects; most notably in relation to its inbuilt methodological design features and medical note abstractions, both of which have enabled the study to address certain research questions more rigorously than other similar investigations. In brief, the UKCCS was specifically designed to make maximum use of routinely compiled administrative data; linking all childhood cancer cases and their potential controls to birth notifications using pre-1996 NHS numbers, and area-based census measures using residential postcode. In addition, rather than relying solely on information reported by parents at interview, the study accessed medical records to answer key questions relating to prescription drugs, illness histories and co-morbidities. In the 2000s, UKCCS extended its remit to investigate health outcomes and patterns further along the patient pathway, demonstrating the utility of UKCCS’s population-based national data to examine survival variations by cancer type and sociodemographic characteristics; it is this area which offers promise for future research.

4) Relationship between the proposed project under this application and the wider project and collaboration

The proposed project, which requires linked secondary healthcare and mortality data provided by NHS England, aims to investigate the long-term health of participants who have recovered after having cancer as children. Researchers working on this study will be based at the University of York, and data will not be shared with any third parties. The proposed project is predicated on the UK Childhood Cancer Study (UKCCS) that was carried out in the 1990s. The two elements have separate aims and data management structures. Furthermore, the data for the historical study did not include data from NHS England (all information being collected by consent).

5) Description of the study subjects

The original UKCCS had a case-control design: each child diagnosed with cancer aged under 15 years old (n=4430) during the study period (1991-1996) was matched on sex, date of birth (month and year), and area of residence (Family Health Service Authority in England & Wales, and Health Board in Scotland) to at least two children of the same age who did not have cancer. To ascertain children diagnosed with cancer, proactive notification systems were established across England, Scotland and Wales; the majority of cases being notified at the point of diagnosis directly to the study by paediatric oncologists or haematologists. For each case, control children (matched on sex, year and month of birth) were randomly selected from the same Family Health Service Authority (FHSA – replaced by Health Authorities in 1995) in England and Wales or Health Board (HB) in Scotland. In total, 4430 cases and 9758 controls were registered. These registrants comprise the individuals included in the proposed study; the NHS numbers (pre-1996 style) along with sex and date of birth will be used by NHS England to link to routine administrative databases.

6) Purpose of the request

a. Detailed information about the purpose of the project under this application

The majority of cancers that occur in children are more responsive to chemotherapy than those that occur in adults; but treatments are aggressive and, for some individuals, adverse health problems may occur many years later. This project will link a nationwide epidemiological case-control study of childhood cancer, the UKCCS, to national datasets to investigate a range of questions relating to the longer-term health and healthcare needs of participants who had cancer as children. Diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine:

(i) Known and suggested therapy-related associations, including cardiovascular and bone problems, as well as reproductive/fertility events (women only).

(ii) Second cancers and late mortality.

(iii) Broad secondary care activity patterns (hospital inpatient & outpatient), to investigate whether those participants who had cancer as children return to general population levels.

(iv) Outcome variations (mortality, cancer & secondary care) with area-based measures of deprivation.

(v) The potential impact of selection-bias at the time of the original study in relation to 1-4 above.

The planned timetable for these analysis is over three years following receipt of the data, to explore the research questions in sufficient depth.

The protocol prepared for this study and provided as evidence for this agreement overviews the full study and was prepared in accordance with guidelines for Research Ethics Committee (ethical approval) and the Confidentiality Advisory Group (s251 support). The historical UKCCS study, which is touched on in the protocol for contextual purposes and background only, is entirely separate from this new phase of the study (purpose of this agreement) - which looks at long term health.

b. Data required

In a previous version of this agreement UKCCS established tracking through NHS England's MRIS Reports, and would like to continue receiving death notifications, cancer registrations and movements in/out of the NHS following the last download of data despatched around the end of 2016. UKCCS would also like to extend the data linkage of UKCCS cases and controls to include secondary healthcare datasets within HES, namely HES Admitted Patient Care (HES-APC) from 1997/98; Outpatients (HES-OP) from 2003/04; and Accident and Emergency (HES-A&E) from 2007/08 until the latest financial year available.

c. Why the data is required for this purpose?

UKCCS would like to link to secondary healthcare datasets (HES-APC, HES-OP and HES-A&E) as well as death and cancer registrations in order to study the health of children previously diagnosed with cancer (during their childhood) relative to that of a similarly aged population. A range of hypotheses, requiring individual level data, will be examined (see examples in 6a above). The study has a longitudinal (follow-up) design, and so UKCCS are requesting all available years of data. UKCCS is a nationwide study including all childhood cancers diagnosed over a restricted time-period, together with a representative sample of the general population. The total number of individuals on whom UKCCS require information is relatively small (around 12,684 subjects). This proposal offers a unique data source; and there are no alternative means of obtaining the data necessary to answer the research questions

d. Details of efforts to minimise data required

The linkage will involve the UKCCS cohorts (12684 participants) in total; only the variables necessary to perform analyses are requested.

7) Organisations involved

a. Who is the Data Controller (sole, joint) and who is the Data Processor?

The University of York is the sole Data Controller and also the Data Processor.

b. Are there any organisations which are involved in the wider project (but not processing the data) and if so, the nature of their role?

As explained in 4 above, (this arm of the study subject to this agreement) which receives data from NHS England, is completely separate from the historical study; no organisations external to the University of York are involved.

There are no non-NHS sites involved in the collection of data.

c. Are there any funders/commissioners involved and if so, the nature of their role?

Bloodwise funds the UKCCS and confirms they have no involvement in the conduct of the research, including no data controller or processing responsibilities.

Over the first 3 years following receipt of the data, diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine a range of questions, as described above. The University of York will produce several outputs on these topics, publishing the findings as papers and reports, as well as conference presentations. All peer-reviewed articles will be published in well-respected journals such as the British Journal of Cancer, International Journal of Epidemiology, Cancer Epidemiology, European Journal of Cancer, Cancer, and PLoS One. In addition, findings will be disseminated at conferences including UK Childhood Cancer Conference, National Cancer Research Institute Conference, and Pan-European Network for Care of Survivors after Childhood and Adolescent Cancers Conference.

Across all outputs, data are only ever made available in aggregated form, with small numbers suppressed

As part of this study, UKCCS will be initiating an active patient partnership, in line with all studies conducted by UKCCS research group (see https://yhhn.org/partnership as an example). A committee will oversee the patient partnership, whose members will be drawn from among clinicians, patients and their relatives, specialist nurses, and researchers with relevant experience. Preliminary discussions with paediatric oncologists and nurse consultants as well as a number of people diagnosed with cancer in their childhood (who are now young adults) have already taken place; the committee will be formalised following approval of this application. Headed by an independent chair, this advisory group will advise on UKCCS public/patient facing profile, including the production of literature and development of the website.

Core funding comes from a Bloodwise programme grant (Apr 2016-Mar 2021; ref 15037; the epidemiology of haematological malignancies: determinants, prognostics, treatment and survivorship). Funders receive annual interim reports, and final reports will be provided at the end of the funding periods. It is anticipated that the findings of the study could be used to help develop guidelines and improve clinical practice (e.g. monitoring) and those who have recovered from cancers' wellbeing; with outputs on survivorship dependent on this agreement it is worth noting that previous UKCCS research are referred to in NICE guidelines, two examples being signs and symptoms of brain tumours (https://www.nice.org.uk/guidance/ng12) and whether intramuscular administration of vitamin K at birth is linked to childhood cancer (https://www.nice.org.uk/guidance/cg37). Impacts will be maximised via national/international dissemination to NHS practitioners/commissioners, politicians/policy makers, and third-sector organisations.

To ensure accessibility to peer-reviewed outputs, all reports will be published under creative commons attribution 4.0 licence (CC BY); support for this is routinely included in all grant applications and the study’s website (www.ukccs.org) will provide links to these open access publications, conference proceedings, and copies of reports summarizing the findings.

The UKCCS website (www.UKCCS.org) will be updated to reflect all activities, and will contain information about the current proposal (e.g. fair processing, the data linkages and the purpose of these) and ongoing research; and social media forums (e.g. a study-specific twitter feed) will be established. Outputs will be promoted through the study’s Twitter account (to be established) and via the study’s funders (@bloodwise_uk (31,900 followers)). Lay summaries of findings will be provided on the UKCCS website (www.ukccs.org). UKCCS also engages with national charities who publish study findings on websites and in their magazines; an example of such dissemination can be found on the following link: https://bloodwise.org.uk/blog/children-less-affluent-backgroundsmore-likely-die-following-treatment-leukaemia-uk.

The UKCCS is an observational epidemiological study and as such, there are no plans- nor are any foreseen- to develop algorithms, testing and development of tools and new technologies.

Target dates for outputs on the topics is within 3 years of receiving the data. By way of example, the analysis on therapy-related associations will commence three months after receipt of the linked data and will continue for 26 months when findings will be ready for submission to scientific journals.

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

A major aim of UKCCS is to improve care and outcomes for those who have recovered from cancer following a diagnosis as a child. UKCCS, an internationally recognized study with a longstanding record of informing on the possible causes of childhood cancer, is ideally placed to provide national data about the long-term health and healthcare needs of these participants. Indeed, the study has already provided population-based data on childhood cancer outcomes, its variation by cancer type, and socio-demographic characteristics. Importantly, the UKCCS is truly population-based and results are generalizable. Hence, findings have the potential to help inform policy and commissioning of services.

This Data Sharing Agreement permits the retention of the data provided under previous iterations of this Agreement for an interim period. This is a pragmatic approach to provide an active Agreement whilst enabling The Institute of Cancer Research to complete the necessary actions to enable a subsequent application to extend the Agreement meeting all applicable data sharing standards as published in NHS Digital’s website (see: https://digital.nhs.uk/services/data-access-request-service-dars/dars-guidance).

1) Aim and Purpose of the Application

Building on one of the largest and most comprehensive childhood cancer case-control studies ever undertaken- the United Kingdom Childhood Cancer Study, UKCCS, this project aims to link to national datasets to investigate a range of questions relating to the long-term health and healthcare needs of participants who had cancer when they were children. Diagnostic-specific comparisons between cases and controls will be used to examine a wide range of factors, including known and suggested therapy-related associations, as well as second cancers and late mortality. In addition, secondary care activity patterns will be examined to investigate whether those who had cancer as children ever return to background levels.

a. Justification of processing data

Article 6(1) of the GDPR (lawfulness of processing), the reference in point (e) to processing of personal data that is necessary for the performance of a task carried out in the public interest or in the exercise of the controller’s official authority includes processing of personal data that is necessary for… (c) the exercise of a function conferred on a person by an enactment or rule of law.

Article 9(2)(j)

(processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject)

Survival of children diagnosed with cancer has improved dramatically in recent decades, exceeding 80% in the UK and other economically developed countries. Indeed, by 2020 it is estimated that there will be half a million people who have recovered from having cancer as children in Europe alone; and this will increase as therapy improves and populations age. However, whilst the majority of cancers that occur in children are more responsive to chemotherapy than those occurring in adults, treatments are often more aggressive, and it is well known that this can lead to adverse health problems (e.g. second cancers, and cardiac conditions, and bone problems) several years later. The UKCCS is a population-based epidemiological study that has, and continues, to inform on potentially avoidable causes of childhood cancer. Now seeking to expand its remit, the UKCCS, with linkage to secondary healthcare data, aims to inform the public health agenda of the future by providing information about the health and healthcare needs of adults diagnosed and treated for cancer as children.

b. Explanation of public task

The majority of studies examining the long-term health outcomes of those who have recovered from having cancer as children have relied on subjects’ participation and self-reported health outcomes, rather than electronic health records, which can offer comprehensive follow-up data. Furthermore, the interpretation of study findings in this area of research can be limited since information on background levels of health events in the population is often missing from UK studies, which have tended to use hospital admission data to generate comparison populations, meaning that individuals with no admissions are excluded. Hence, with a view to providing a unique and valuable resource for childhood cancer studies, cases and controls from the UKCCS are followed as two cohorts by linking individuals to routinely compiled NHS data. This enables the UKCCS to put the health of childhood cancers into the context of their peers as they age, making it uniquely placed to investigate the long-term healthcare needs of UK participants who have recovered from having cancer during childhood.

During the UKCCS data collection phase (1991-1996), consent was sought from the study subjects’ parents, in line with the rules and regulations at the time. This consent allowed participants and family members to fill in survey's and questionnaires - from which the data was used in the study analysis. The study subjects were children (aged under 15 years) during this period, and no contact with families have been maintained. The children themselves, who are now adults (aged 23 to 42 in 2019), have never been approached and asked to consent to the use of individual identifying information for linkage purposes.

Section 251 support was historically granted under the reference number ECC 5-02(FT5)/2009 to allow NHS Digital to disseminate mortality data on the cohort it had flagged.

Around a third of the 4430 children diagnosed with cancer have since died, as have some of the control children; and the only way to attempt to trace and obtain consent from living cases and controls (many of whom may be unaware of their involvement) would be to use the identifiable data UKCCS already hold, which is neither justifiable or practicable. In its absence, the acceptability of using patient identifiable information to link to routinely compiled administrative health data without explicit consent and using the resulting data for research purposes has been tested in several forums with patient and parent representation. Consensus was to support linkage following discussions in 2004/5 with representatives from the National Alliance of Childhood Cancer Parent Organisations (now the Childhood Cancer Parents Alliance http://ccpa.org.uk/) prior to linkage to death notifications and tracking; and more recently, with regard to linkage to secondary care data, with patient groups in UKCCS' sister study the Yorkshire and Humberside Haematology Network (https://yhhn.org), and clinicians who treat children, adolescents and young adults. In 2018, UKCCS was granted Section 251 support (24th July 2018, 18/CAG/0066) after gaining renewed ethical approval (27th April 2018, REC 18/YH/0135) to continue linkage to national death notifications, cancer registrations and movements in/out of the NHS, and to add linkage to secondary healthcare datasets. This approval superceded the historical approval ECC 5-02(FT5)/2009.

2) How will the data requested achieve the aim identified?

Building on existing infrastructure, the principal aim of the study is to examine the long-term health and healthcare needs of those who have recovered after having cancer as a child. The Long-term health conditions UKCCS intend to investigate include secondary cancers, skeletal-related problems, cardiac conditions, respiratory illnesses and fertility issues, as well as investigating hospital activity and survival. Comparisons will be made with UKCCS controls, who being born in the same year and of the same sex as those diagnosed with cancer as children, are an appropriate population comparator. Importantly, the data requested offers contemporaneously recorded information, available for all study subjects and by its nature free from differential recall bias; this design often produces very different results from those emanating from studies reliant on participation and self-report. In UKCCS analyses, data from the different administrative sources will be used together, for instance censoring the occurrence of certain events or hospital activity at a defined timepoint, whether at secondary cancer, movement abroad, or death for example, is clearly relevant in longitudinal studies. As for the events of interest, UKCCS know from published literature and UKCCS' own feasibility work conducted in a dataset with less follow-up post-cancer diagnosis than UKCCS (https://yhhn.org/research/comparison), that these adverse health events can be identified in Hospital Episode Statistics (HES). Some detectable differences between cases and controls are anticipated within the twenty years of data that can currently be accessed; as study subjects are presently aged 23-42 years, other health events will become more common as they age, so continuing tracking of patients offers opportunity for further research as the data matures.

3) Background of study

With the initial aim of investigating causes of cancer diagnosed before 15 years of age, the United Kingdom Childhood Cancer Study (UKCCS) was set-up in the early 1990s. This ground-breaking study, one of the largest (4430 cases and 9758 controls) and most comprehensive ever undertaken, examined the potential causes of the development of the disease and the role of a wide range of biological, physical, and chemical agents acting on parents before the child’s conception, and the child during pregnancy and in early life. Predicated on NHS infrastructures, the UKCCS remains unique in several important respects; most notably in relation to its inbuilt methodological design features and medical note abstractions, both of which have enabled the study to address certain research questions more rigorously than other similar investigations. In brief, the UKCCS was specifically designed to make maximum use of routinely compiled administrative data; linking all childhood cancer cases and their potential controls to birth notifications using pre-1996 NHS numbers, and area-based census measures using residential postcode. In addition, rather than relying solely on information reported by parents at interview, the study accessed medical records to answer key questions relating to prescription drugs, illness histories and co-morbidities. In the 2000s, UKCCS extended its remit to investigate health outcomes and patterns further along the patient pathway, demonstrating the utility of UKCCS’s population-based national data to examine survival variations by cancer type and sociodemographic characteristics; it is this area which offers promise for future research.

4) Relationship between the proposed project under this application and the wider project and collaboration

The proposed project, which requires linked secondary healthcare and mortality data provided by NHS Digital, aims to investigate the long-term health of participants who have recovered after having cancer as children. Researchers working on this study will be based at the University of York, and data will not be shared with any third parties. The proposed project is predicated on the UK Childhood Cancer Study (UKCCS) that was carried out in the 1990s. The two elements have separate aims and data management structures. Furthermore, the data for the historical study did not include data from NHS Digital (all information being collected by consent).

5) Description of the study subjects

The original UKCCS had a case-control design: each child diagnosed with cancer aged under 15 years old (n=4430) during the study period (1991-1996) was matched on sex, date of birth (month and year), and area of residence (Family Health Service Authority in England & Wales, and Health Board in Scotland) to at least two children of the same age who did not have cancer. To ascertain children diagnosed with cancer, proactive notification systems were established across England, Scotland and Wales; the majority of cases being notified at the point of diagnosis directly to the study by paediatric oncologists or haematologists. For each case, control children (matched on sex, year and month of birth) were randomly selected from the same Family Health Service Authority (FHSA – replaced by Health Authorities in 1995) in England and Wales or Health Board (HB) in Scotland. In total, 4430 cases and 9758 controls were registered. These registrants comprise the individuals included in the proposed study; the NHS numbers (pre-1996 style) along with sex and date of birth will be used by NHS Digital to link to routine administrative databases.

6) Purpose of the request

a. Detailed information about the purpose of the project under this application

The majority of cancers that occur in children are more responsive to chemotherapy than those that occur in adults; but treatments are aggressive and, for some individuals, adverse health problems may occur many years later. This project will link a nationwide epidemiological case-control study of childhood cancer, the UKCCS, to national datasets to investigate a range of questions relating to the longer-term health and healthcare needs of participants who had cancer as children. Diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine:

(i) Known and suggested therapy-related associations, including cardiovascular and bone problems, as well as reproductive/fertility events (women only).

(ii) Second cancers and late mortality.

(iii) Broad secondary care activity patterns (hospital inpatient & outpatient), to investigate whether those participants who had cancer as children return to general population levels.

(iv) Outcome variations (mortality, cancer & secondary care) with area-based measures of deprivation.

(v) The potential impact of selection-bias at the time of the original study in relation to 1-4 above.

The planned timetable for these analysis is over three years following receipt of the data, to explore the research questions in sufficient depth.

The protocol prepared for this study and provided as evidence for this agreement overviews the full study and was prepared in accordance with guidelines for Research Ethics Committee (ethical approval) and the Confidentiality Advisory Group (s251 support). The historical UKCCS study, which is touched on in the protocol for contextual purposes and background only, is entirely separate from this new phase of the study (purpose of this agreement) - which looks at long term health.

b. Data required

In a previous version of this agreement UKCCS established tracking through NHS Digital's MRIS Reports, and would like to continue receiving death notifications, cancer registrations and movements in/out of the NHS following the last download of data despatched around the end of 2016. UKCCS would also like to extend the data linkage of UKCCS cases and controls to include secondary healthcare datasets within HES, namely HES Admitted Patient Care (HES-APC) from 1997/98; Outpatients (HES-OP) from 2003/04; and Accident and Emergency (HES-A&E) from 2007/08 until the latest financial year available.

c. Why the data is required for this purpose?

UKCCS would like to link to secondary healthcare datasets (HES-APC, HES-OP and HES-A&E) as well as death and cancer registrations in order to study the health of children previously diagnosed with cancer (during their childhood) relative to that of a similarly aged population. A range of hypotheses, requiring individual level data, will be examined (see examples in 6a above). The study has a longitudinal (follow-up) design, and so UKCCS are requesting all available years of data. UKCCS is a nationwide study including all childhood cancers diagnosed over a restricted time-period, together with a representative sample of the general population. The total number of individuals on whom UKCCS require information is relatively small (around 12,684 subjects). This proposal offers a unique data source; and there are no alternative means of obtaining the data necessary to answer the research questions

d. Details of efforts to minimise data required

The linkage will involve the UKCCS cohorts (12684 participants) in total; only the variables necessary to perform analyses are requested.

7) Organisations involved

a. Who is the Data Controller (sole, joint) and who is the Data Processor?

The University of York is the sole Data Controller and also the Data Processor.

b. Are there any organisations which are involved in the wider project (but not processing the data) and if so, the nature of their role?

As explained in 4 above, (this arm of the study subject to this agreement) which receives data from NHS Digital, is completely separate from the historical study; no organisations external to the University of York are involved.

There are no non-NHS sites involved in the collection of data.

c. Are there any funders/commissioners involved and if so, the nature of their role?

Bloodwise funds the UKCCS and confirms they have no involvement in the conduct of the research, including no data controller or processing responsibilities.

Over the first 3 years following receipt of the data, diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine a range of questions, as described above. The University of York will produce several outputs on these topics, publishing the findings as papers and reports, as well as conference presentations. All peer-reviewed articles will be published in well-respected journals such as the British Journal of Cancer, International Journal of Epidemiology, Cancer Epidemiology, European Journal of Cancer, Cancer, and PLoS One. In addition, findings will be disseminated at conferences including UK Childhood Cancer Conference, National Cancer Research Institute Conference, and Pan-European Network for Care of Survivors after Childhood and Adolescent Cancers Conference.

Across all outputs, data are only ever made available in aggregated form, with small numbers suppressed

As part of this study, UKCCS will be initiating an active patient partnership, in line with all studies conducted by UKCCS research group (see https://yhhn.org/partnership as an example). A committee will oversee the patient partnership, whose members will be drawn from among clinicians, patients and their relatives, specialist nurses, and researchers with relevant experience. Preliminary discussions with paediatric oncologists and nurse consultants as well as a number of people diagnosed with cancer in their childhood (who are now young adults) have already taken place; the committee will be formalised following approval of this application. Headed by an independent chair, this advisory group will advise on UKCCS public/patient facing profile, including the production of literature and development of the website.

Core funding comes from a Bloodwise programme grant (Apr 2016-Mar 2021; ref 15037; the epidemiology of haematological malignancies: determinants, prognostics, treatment and survivorship). Funders receive annual interim reports, and final reports will be provided at the end of the funding periods. It is anticipated that the findings of the study could be used to help develop guidelines and improve clinical practice (e.g. monitoring) and those who have recovered from cancers' wellbeing; with outputs on survivorship dependent on this agreement it is worth noting that previous UKCCS research are referred to in NICE guidelines, two examples being signs and symptoms of brain tumours (https://www.nice.org.uk/guidance/ng12) and whether intramuscular administration of vitamin K at birth is linked to childhood cancer (https://www.nice.org.uk/guidance/cg37). Impacts will be maximised via national/international dissemination to NHS practitioners/commissioners, politicians/policy makers, and third-sector organisations.

To ensure accessibility to peer-reviewed outputs, all reports will be published under creative commons attribution 4.0 licence (CC BY); support for this is routinely included in all grant applications and the study’s website (www.ukccs.org) will provide links to these open access publications, conference proceedings, and copies of reports summarizing the findings.

The UKCCS website (www.UKCCS.org) will be updated to reflect all activities, and will contain information about the current proposal (e.g. fair processing, the data linkages and the purpose of these) and ongoing research; and social media forums (e.g. a study-specific twitter feed) will be established. Outputs will be promoted through the study’s Twitter account (to be established) and via the study’s funders (@bloodwise_uk (31,900 followers)). Lay summaries of findings will be provided on the UKCCS website (www.ukccs.org). UKCCS also engages with national charities who publish study findings on websites and in their magazines; an example of such dissemination can be found on the following link: https://bloodwise.org.uk/blog/children-less-affluent-backgroundsmore-likely-die-following-treatment-leukaemia-uk.

The UKCCS is an observational epidemiological study and as such, there are no plans- nor are any foreseen- to develop algorithms, testing and development of tools and new technologies.

Target dates for outputs on the topics is within 3 years of receiving the data. By way of example, the analysis on therapy-related associations will commence three months after receipt of the linked data and will continue for 26 months when findings will be ready for submission to scientific journals.

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

A major aim of UKCCS is to improve care and outcomes for those who have recovered from cancer following a diagnosis as a child. UKCCS, an internationally recognized study with a longstanding record of informing on the possible causes of childhood cancer, is ideally placed to provide national data about the long-term health and healthcare needs of these participants. Indeed, the study has already provided population-based data on childhood cancer outcomes, its variation by cancer type, and socio-demographic characteristics. Importantly, the UKCCS is truly population-based and results are generalizable. Hence, findings have the potential to help inform policy and commissioning of services.

1) Aim and Purpose of the Application

Building on one of the largest and most comprehensive childhood cancer case-control studies ever undertaken- the United Kingdom Childhood Cancer Study, UKCCS, this project aims to link to national datasets to investigate a range of questions relating to the long-term health and healthcare needs of participants who had cancer when they were children. Diagnostic-specific comparisons between cases and controls will be used to examine a wide range of factors, including known and suggested therapy-related associations, as well as second cancers and late mortality. In addition, secondary care activity patterns will be examined to investigate whether those who had cancer as children ever return to background levels.

a. Justification of processing data

Article 6(1) of the GDPR (lawfulness of processing), the reference in point (e) to processing of personal data that is necessary for the performance of a task carried out in the public interest or in the exercise of the controller’s official authority includes processing of personal data that is necessary for… (c) the exercise of a function conferred on a person by an enactment or rule of law.

Article 9(2)(j)

(processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject)

Survival of children diagnosed with cancer has improved dramatically in recent decades, exceeding 80% in the UK and other economically developed countries. Indeed, by 2020 it is estimated that there will be half a million people who have recovered from having cancer as children in Europe alone; and this will increase as therapy improves and populations age. However, whilst the majority of cancers that occur in children are more responsive to chemotherapy than those occurring in adults, treatments are often more aggressive, and it is well known that this can lead to adverse health problems (e.g. second cancers, and cardiac conditions, and bone problems) several years later. The UKCCS is a population-based epidemiological study that has, and continues, to inform on potentially avoidable causes of childhood cancer. Now seeking to expand its remit, the UKCCS, with linkage to secondary healthcare data, aims to inform the public health agenda of the future by providing information about the health and healthcare needs of adults diagnosed and treated for cancer as children.

b. Explanation of public task

The majority of studies examining the long-term health outcomes of those who have recovered from having cancer as children have relied on subjects’ participation and self-reported health outcomes, rather than electronic health records, which can offer comprehensive follow-up data. Furthermore, the interpretation of study findings in this area of research can be limited since information on background levels of health events in the population is often missing from UK studies, which have tended to use hospital admission data to generate comparison populations, meaning that individuals with no admissions are excluded. Hence, with a view to providing a unique and valuable resource for childhood cancer studies, cases and controls from the UKCCS are followed as two cohorts by linking individuals to routinely compiled NHS data. This enables the UKCCS to put the health of childhood cancers into the context of their peers as they age, making it uniquely placed to investigate the long-term healthcare needs of UK participants who have recovered from having cancer during childhood.

During the UKCCS data collection phase (1991-1996), consent was sought from the study subjects’ parents, in line with the rules and regulations at the time. This consent allowed participants and family members to fill in survey's and questionnaires - from which the data was used in the study analysis. The study subjects were children (aged under 15 years) during this period, and no contact with families have been maintained. The children themselves, who are now adults (aged 23 to 42 in 2019), have never been approached and asked to consent to the use of individual identifying information for linkage purposes.

Section 251 support was historically granted under the reference number ECC 5-02(FT5)/2009 to allow NHS Digital to disseminate mortality data on the cohort it had flagged.

Around a third of the 4430 children diagnosed with cancer have since died, as have some of the control children; and the only way to attempt to trace and obtain consent from living cases and controls (many of whom may be unaware of their involvement) would be to use the identifiable data UKCCS already hold, which is neither justifiable or practicable. In its absence, the acceptability of using patient identifiable information to link to routinely compiled administrative health data without explicit consent and using the resulting data for research purposes has been tested in several forums with patient and parent representation. Consensus was to support linkage following discussions in 2004/5 with representatives from the National Alliance of Childhood Cancer Parent Organisations (now the Childhood Cancer Parents Alliance http://ccpa.org.uk/) prior to linkage to death notifications and tracking; and more recently, with regard to linkage to secondary care data, with patient groups in UKCCS' sister study the Yorkshire and Humberside Haematology Network (https://yhhn.org), and clinicians who treat children, adolescents and young adults. In 2018, UKCCS was granted Section 251 support (24th July 2018, 18/CAG/0066) after gaining renewed ethical approval (27th April 2018, REC 18/YH/0135) to continue linkage to national death notifications, cancer registrations and movements in/out of the NHS, and to add linkage to secondary healthcare datasets. This approval superceded the historical approval ECC 5-02(FT5)/2009.

2) How will the data requested achieve the aim identified?

Building on existing infrastructure, the principal aim of the study is to examine the long-term health and healthcare needs of those who have recovered after having cancer as a child. The Long-term health conditions UKCCS intend to investigate include secondary cancers, skeletal-related problems, cardiac conditions, respiratory illnesses and fertility issues, as well as investigating hospital activity and survival. Comparisons will be made with UKCCS controls, who being born in the same year and of the same sex as those diagnosed with cancer as children, are an appropriate population comparator. Importantly, the data requested offers contemporaneously recorded information, available for all study subjects and by its nature free from differential recall bias; this design often produces very different results from those emanating from studies reliant on participation and self-report. In UKCCS analyses, data from the different administrative sources will be used together, for instance censoring the occurrence of certain events or hospital activity at a defined timepoint, whether at secondary cancer, movement abroad, or death for example, is clearly relevant in longitudinal studies. As for the events of interest, UKCCS know from published literature and UKCCS' own feasibility work conducted in a dataset with less follow-up post-cancer diagnosis than UKCCS (https://yhhn.org/research/comparison), that these adverse health events can be identified in Hospital Episode Statistics (HES). Some detectable differences between cases and controls are anticipated within the twenty years of data that can currently be accessed; as study subjects are presently aged 23-42 years, other health events will become more common as they age, so continuing tracking of patients offers opportunity for further research as the data matures.

3) Background of study

With the initial aim of investigating causes of cancer diagnosed before 15 years of age, the United Kingdom Childhood Cancer Study (UKCCS) was set-up in the early 1990s. This ground-breaking study, one of the largest (4430 cases and 9758 controls) and most comprehensive ever undertaken, examined the potential causes of the development of the disease and the role of a wide range of biological, physical, and chemical agents acting on parents before the child’s conception, and the child during pregnancy and in early life. Predicated on NHS infrastructures, the UKCCS remains unique in several important respects; most notably in relation to its inbuilt methodological design features and medical note abstractions, both of which have enabled the study to address certain research questions more rigorously than other similar investigations. In brief, the UKCCS was specifically designed to make maximum use of routinely compiled administrative data; linking all childhood cancer cases and their potential controls to birth notifications using pre-1996 NHS numbers, and area-based census measures using residential postcode. In addition, rather than relying solely on information reported by parents at interview, the study accessed medical records to answer key questions relating to prescription drugs, illness histories and co-morbidities. In the 2000s, UKCCS extended its remit to investigate health outcomes and patterns further along the patient pathway, demonstrating the utility of UKCCS’s population-based national data to examine survival variations by cancer type and sociodemographic characteristics; it is this area which offers promise for future research.

4) Relationship between the proposed project under this application and the wider project and collaboration

The proposed project, which requires linked secondary healthcare and mortality data provided by NHS Digital, aims to investigate the long-term health of participants who have recovered after having cancer as children. Researchers working on this study will be based at the University of York, and data will not be shared with any third parties. The proposed project is predicated on the UK Childhood Cancer Study (UKCCS) that was carried out in the 1990s. The two elements have separate aims and data management structures. Furthermore, the data for the historical study did not include data from NHS Digital (all information being collected by consent).

5) Description of the study subjects

The original UKCCS had a case-control design: each child diagnosed with cancer aged under 15 years old (n=4430) during the study period (1991-1996) was matched on sex, date of birth (month and year), and area of residence (Family Health Service Authority in England & Wales, and Health Board in Scotland) to at least two children of the same age who did not have cancer. To ascertain children diagnosed with cancer, proactive notification systems were established across England, Scotland and Wales; the majority of cases being notified at the point of diagnosis directly to the study by paediatric oncologists or haematologists. For each case, control children (matched on sex, year and month of birth) were randomly selected from the same Family Health Service Authority (FHSA – replaced by Health Authorities in 1995) in England and Wales or Health Board (HB) in Scotland. In total, 4430 cases and 9758 controls were registered. These registrants comprise the individuals included in the proposed study; the NHS numbers (pre-1996 style) along with sex and date of birth will be used by NHS Digital to link to routine administrative databases.

6) Purpose of the request

a. Detailed information about the purpose of the project under this application

The majority of cancers that occur in children are more responsive to chemotherapy than those that occur in adults; but treatments are aggressive and, for some individuals, adverse health problems may occur many years later. This project will link a nationwide epidemiological case-control study of childhood cancer, the UKCCS, to national datasets to investigate a range of questions relating to the longer-term health and healthcare needs of participants who had cancer as children. Diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine:

(i) Known and suggested therapy-related associations, including cardiovascular and bone problems, as well as reproductive/fertility events (women only).

(ii) Second cancers and late mortality.

(iii) Broad secondary care activity patterns (hospital inpatient & outpatient), to investigate whether those participants who had cancer as children return to general population levels.

(iv) Outcome variations (mortality, cancer & secondary care) with area-based measures of deprivation.

(v) The potential impact of selection-bias at the time of the original study in relation to 1-4 above.

The planned timetable for these analysis is over three years following receipt of the data, to explore the research questions in sufficient depth.

The protocol prepared for this study and provided as evidence for this agreement overviews the full study and was prepared in accordance with guidelines for Research Ethics Committee (ethical approval) and the Confidentiality Advisory Group (s251 support). The historical UKCCS study, which is touched on in the protocol for contextual purposes and background only, is entirely separate from this new phase of the study (purpose of this agreement) - which looks at long term health.

b. Data required

In a previous version of this agreement UKCCS established tracking through NHS Digital's MRIS Reports, and would like to continue receiving death notifications, cancer registrations and movements in/out of the NHS following the last download of data despatched around the end of 2016. UKCCS would also like to extend the data linkage of UKCCS cases and controls to include secondary healthcare datasets within HES, namely HES Admitted Patient Care (HES-APC) from 1997/98; Outpatients (HES-OP) from 2003/04; and Accident and Emergency (HES-A&E) from 2007/08 until the latest financial year available.

c. Why the data is required for this purpose?

UKCCS would like to link to secondary healthcare datasets (HES-APC, HES-OP and HES-A&E) as well as death and cancer registrations in order to study the health of children previously diagnosed with cancer (during their childhood) relative to that of a similarly aged population. A range of hypotheses, requiring individual level data, will be examined (see examples in 6a above). The study has a longitudinal (follow-up) design, and so UKCCS are requesting all available years of data. UKCCS is a nationwide study including all childhood cancers diagnosed over a restricted time-period, together with a representative sample of the general population. The total number of individuals on whom UKCCS require information is relatively small (around 12,684 subjects). This proposal offers a unique data source; and there are no alternative means of obtaining the data necessary to answer the research questions

d. Details of efforts to minimise data required

The linkage will involve the UKCCS cohorts (12684 participants) in total; only the variables necessary to perform analyses are requested.

7) Organisations involved

a. Who is the Data Controller (sole, joint) and who is the Data Processor?

The University of York is the sole Data Controller and also the Data Processor.

b. Are there any organisations which are involved in the wider project (but not processing the data) and if so, the nature of their role?

As explained in 4 above, (this arm of the study subject to this agreement) which receives data from NHS Digital, is completely separate from the historical study; no organisations external to the University of York are involved.

There are no non-NHS sites involved in the collection of data.

c. Are there any funders/commissioners involved and if so, the nature of their role?

Bloodwise funds the UKCCS and confirms they have no involvement in the conduct of the research, including no data controller or processing responsibilities.

Over the first 3 years following receipt of the data, diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine a range of questions, as described above. The University of York will produce several outputs on these topics, publishing the findings as papers and reports, as well as conference presentations. All peer-reviewed articles will be published in well-respected journals such as the British Journal of Cancer, International Journal of Epidemiology, Cancer Epidemiology, European Journal of Cancer, Cancer, and PLoS One. In addition, findings will be disseminated at conferences including UK Childhood Cancer Conference, National Cancer Research Institute Conference, and Pan-European Network for Care of Survivors after Childhood and Adolescent Cancers Conference.

Across all outputs, data are only ever made available in aggregated form, with small numbers suppressed

As part of this study, UKCCS will be initiating an active patient partnership, in line with all studies conducted by UKCCS research group (see https://yhhn.org/partnership as an example). A committee will oversee the patient partnership, whose members will be drawn from among clinicians, patients and their relatives, specialist nurses, and researchers with relevant experience. Preliminary discussions with paediatric oncologists and nurse consultants as well as a number of people diagnosed with cancer in their childhood (who are now young adults) have already taken place; the committee will be formalised following approval of this application. Headed by an independent chair, this advisory group will advise on UKCCS public/patient facing profile, including the production of literature and development of the website.

Core funding comes from a Bloodwise programme grant (Apr 2016-Mar 2021; ref 15037; the epidemiology of haematological malignancies: determinants, prognostics, treatment and survivorship). Funders receive annual interim reports, and final reports will be provided at the end of the funding periods. It is anticipated that the findings of the study could be used to help develop guidelines and improve clinical practice (e.g. monitoring) and those who have recovered from cancers' wellbeing; with outputs on survivorship dependent on this agreement it is worth noting that previous UKCCS research are referred to in NICE guidelines, two examples being signs and symptoms of brain tumours (https://www.nice.org.uk/guidance/ng12) and whether intramuscular administration of vitamin K at birth is linked to childhood cancer (https://www.nice.org.uk/guidance/cg37). Impacts will be maximised via national/international dissemination to NHS practitioners/commissioners, politicians/policy makers, and third-sector organisations.

To ensure accessibility to peer-reviewed outputs, all reports will be published under creative commons attribution 4.0 licence (CC BY); support for this is routinely included in all grant applications and the study’s website (www.ukccs.org) will provide links to these open access publications, conference proceedings, and copies of reports summarizing the findings.

The UKCCS website (www.UKCCS.org) will be updated to reflect all activities, and will contain information about the current proposal (e.g. fair processing, the data linkages and the purpose of these) and ongoing research; and social media forums (e.g. a study-specific twitter feed) will be established. Outputs will be promoted through the study’s Twitter account (to be established) and via the study’s funders (@bloodwise_uk (31,900 followers)). Lay summaries of findings will be provided on the UKCCS website (www.ukccs.org). UKCCS also engages with national charities who publish study findings on websites and in their magazines; an example of such dissemination can be found on the following link: https://bloodwise.org.uk/blog/children-less-affluent-backgroundsmore-likely-die-following-treatment-leukaemia-uk.

The UKCCS is an observational epidemiological study and as such, there are no plans- nor are any foreseen- to develop algorithms, testing and development of tools and new technologies.

Target dates for outputs on the topics is within 3 years of receiving the data. By way of example, the analysis on therapy-related associations will commence three months after receipt of the linked data and will continue for 26 months when findings will be ready for submission to scientific journals.

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

A major aim of UKCCS is to improve care and outcomes for those who have recovered from cancer following a diagnosis as a child. UKCCS, an internationally recognized study with a longstanding record of informing on the possible causes of childhood cancer, is ideally placed to provide national data about the long-term health and healthcare needs of these participants. Indeed, the study has already provided population-based data on childhood cancer outcomes, its variation by cancer type, and socio-demographic characteristics. Importantly, the UKCCS is truly population-based and results are generalizable. Hence, findings have the potential to help inform policy and commissioning of services.

The study is not in support of a PhD/post graduate research study.

1) Aim and Purpose of the Application

Building on one of the largest and most comprehensive childhood cancer case-control studies ever undertaken- the United Kingdom Childhood Cancer Study, UKCCS, this project aims to link to national datasets to investigate a range of questions relating to the long-term health and healthcare needs of participants who had cancer when they were children. Diagnostic-specific comparisons between cases and controls will be used to examine a wide range of factors, including known and suggested therapy-related associations, as well as second cancers and late mortality. In addition, secondary care activity patterns will be examined to investigate whether those who had cancer as children ever return to background levels.

a. Justification of processing data

Article 6(1) of the GDPR (lawfulness of processing), the reference in point (e) to processing of personal data that is necessary for the performance of a task carried out in the public interest or in the exercise of the controller’s official authority includes processing of personal data that is necessary for… (c) the exercise of a function conferred on a person by an enactment or rule of law.

Article 9(2)(j)

(processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject)

Survival of children diagnosed with cancer has improved dramatically in recent decades, exceeding 80% in the UK and other economically developed countries. Indeed, by 2020 it is estimated that there will be half a million people who have recovered from having cancer as children in Europe alone; and this will increase as therapy improves and populations age. However, whilst the majority of cancers that occur in children are more responsive to chemotherapy than those occurring in adults, treatments are often more aggressive, and it is well known that this can lead to adverse health problems (e.g. second cancers, and cardiac conditions, and bone problems) several years later. The UKCCS is a population-based epidemiological study that has, and continues, to inform on potentially avoidable causes of childhood cancer. Now seeking to expand its remit, the UKCCS, with linkage to secondary healthcare data, aims to inform the public health agenda of the future by providing information about the health and healthcare needs of adults diagnosed and treated for cancer as children.

b. Explanation of public task

The majority of studies examining the long-term health outcomes of those who have recovered from having cancer as children have relied on subjects’ participation and self-reported health outcomes, rather than electronic health records, which can offer comprehensive follow-up data. Furthermore, the interpretation of study findings in this area of research can be limited since information on background levels of health events in the population is often missing from UK studies, which have tended to use hospital admission data to generate comparison populations, meaning that individuals with no admissions are excluded. Hence, with a view to providing a unique and valuable resource for childhood cancer studies, cases and controls from the UKCCS are followed as two cohorts by linking individuals to routinely compiled NHS data. This enables the UKCCS to put the health of childhood cancers into the context of their peers as they age, making it uniquely placed to investigate the long-term healthcare needs of UK participants who have recovered from having cancer during childhood.

During the UKCCS data collection phase (1991-1996), consent was sought from the study subjects’ parents, in line with the rules and regulations at the time. This consent allowed participants and family members to fill in survey's and questionnaires - from which the data was used in the study analysis. The study subjects were children (aged under 15 years) during this period, and no contact with families have been maintained. The children themselves, who are now adults (aged 23 to 42 in 2019), have never been approached and asked to consent to the use of individual identifying information for linkage purposes.

Section 251 support was historically granted under the reference number ECC 5-02(FT5)/2009 to allow NHS Digital to disseminate mortality data on the cohort it had flagged.

Around a third of the 4430 children diagnosed with cancer have since died, as have some of the control children; and the only way to attempt to trace and obtain consent from living cases and controls (many of whom may be unaware of their involvement) would be to use the identifiable data UKCCS already hold, which is neither justifiable or practicable. In its absence, the acceptability of using patient identifiable information to link to routinely compiled administrative health data without explicit consent and using the resulting data for research purposes has been tested in several forums with patient and parent representation. Consensus was to support linkage following discussions in 2004/5 with representatives from the National Alliance of Childhood Cancer Parent Organisations (now the Childhood Cancer Parents Alliance http://ccpa.org.uk/) prior to linkage to death notifications and tracking; and more recently, with regard to linkage to secondary care data, with patient groups in UKCCS' sister study the Yorkshire and Humberside Haematology Network (https://yhhn.org), and clinicians who treat children, adolescents and young adults. In 2018, UKCCS was granted Section 251 support (24th July 2018, 18/CAG/0066) after gaining renewed ethical approval (27th April 2018, REC 18/YH/0135) to continue linkage to national death notifications, cancer registrations and movements in/out of the NHS, and to add linkage to secondary healthcare datasets. This approval superceded the historical approval ECC 5-02(FT5)/2009.

2) How will the data requested achieve the aim identified?

Building on existing infrastructure, the principal aim of the study is to examine the long-term health and healthcare needs of those who have recovered after having cancer as a child. The Long-term health conditions UKCCS intend to investigate include secondary cancers, skeletal-related problems, cardiac conditions, respiratory illnesses and fertility issues, as well as investigating hospital activity and survival. Comparisons will be made with UKCCS controls, who being born in the same year and of the same sex as those diagnosed with cancer as children, are an appropriate population comparator. Importantly, the data requested offers contemporaneously recorded information, available for all study subjects and by its nature free from differential recall bias; this design often produces very different results from those emanating from studies reliant on participation and self-report. In UKCCS analyses, data from the different administrative sources will be used together, for instance censoring the occurrence of certain events or hospital activity at a defined timepoint, whether at secondary cancer, movement abroad, or death for example, is clearly relevant in longitudinal studies. As for the events of interest, UKCCS know from published literature and UKCCS' own feasibility work conducted in a dataset with less follow-up post-cancer diagnosis than UKCCS (https://yhhn.org/research/comparison), that these adverse health events can be identified in Hospital Episode Statistics (HES). Some detectable differences between cases and controls are anticipated within the twenty years of data that can currently be accessed; as study subjects are presently aged 23-42 years, other health events will become more common as they age, so continuing tracking of patients offers opportunity for further research as the data matures.

3) Background of study

With the initial aim of investigating causes of cancer diagnosed before 15 years of age, the United Kingdom Childhood Cancer Study (UKCCS) was set-up in the early 1990s. This ground-breaking study, one of the largest (4430 cases and 9758 controls) and most comprehensive ever undertaken, examined the potential causes of the development of the disease and the role of a wide range of biological, physical, and chemical agents acting on parents before the child’s conception, and the child during pregnancy and in early life. Predicated on NHS infrastructures, the UKCCS remains unique in several important respects; most notably in relation to its inbuilt methodological design features and medical note abstractions, both of which have enabled the study to address certain research questions more rigorously than other similar investigations. In brief, the UKCCS was specifically designed to make maximum use of routinely compiled administrative data; linking all childhood cancer cases and their potential controls to birth notifications using pre-1996 NHS numbers, and area-based census measures using residential postcode. In addition, rather than relying solely on information reported by parents at interview, the study accessed medical records to answer key questions relating to prescription drugs, illness histories and co-morbidities. In the 2000s, UKCCS extended its remit to investigate health outcomes and patterns further along the patient pathway, demonstrating the utility of UKCCS’s population-based national data to examine survival variations by cancer type and sociodemographic characteristics; it is this area which offers promise for future research.

4) Relationship between the proposed project under this application and the wider project and collaboration

The proposed project, which requires linked secondary healthcare and mortality data provided by NHS Digital, aims to investigate the long-term health of participants who have recovered after having cancer as children. Researchers working on this study will be based at the University of York, and data will not be shared with any third parties. The proposed project is predicated on the UK Childhood Cancer Study (UKCCS) that was carried out in the 1990s. The two elements have separate aims and data management structures. Furthermore, the data for the historical study did not include data from NHS Digital (all information being collected by consent).

5) Description of the study subjects

The original UKCCS had a case-control design: each child diagnosed with cancer aged under 15 years old (n=4430) during the study period (1991-1996) was matched on sex, date of birth (month and year), and area of residence (Family Health Service Authority in England & Wales, and Health Board in Scotland) to at least two children of the same age who did not have cancer. To ascertain children diagnosed with cancer, proactive notification systems were established across England, Scotland and Wales; the majority of cases being notified at the point of diagnosis directly to the study by paediatric oncologists or haematologists. For each case, control children (matched on sex, year and month of birth) were randomly selected from the same Family Health Service Authority (FHSA – replaced by Health Authorities in 1995) in England and Wales or Health Board (HB) in Scotland. In total, 4430 cases and 9758 controls were registered. These registrants comprise the individuals included in the proposed study; the NHS numbers (pre-1996 style) along with sex and date of birth will be used by NHS Digital to link to routine administrative databases.

6) Purpose of the request

a. Detailed information about the purpose of the project under this application

The majority of cancers that occur in children are more responsive to chemotherapy than those that occur in adults; but treatments are aggressive and, for some individuals, adverse health problems may occur many years later. This project will link a nationwide epidemiological case-control study of childhood cancer, the UKCCS, to national datasets to investigate a range of questions relating to the longer-term health and healthcare needs of participants who had cancer as children. Diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine:

(i) Known and suggested therapy-related associations, including cardiovascular and bone problems, as well as reproductive/fertility events (women only).

(ii) Second cancers and late mortality.

(iii) Broad secondary care activity patterns (hospital inpatient & outpatient), to investigate whether those participants who had cancer as children return to general population levels.

(iv) Outcome variations (mortality, cancer & secondary care) with area-based measures of deprivation.

(v) The potential impact of selection-bias at the time of the original study in relation to 1-4 above.

The planned timetable for these analysis is over three years following receipt of the data, to explore the research questions in sufficient depth.

The protocol prepared for this study and provided as evidence for this agreement overviews the full study and was prepared in accordance with guidelines for Research Ethics Committee (ethical approval) and the Confidentiality Advisory Group (s251 support). The historical UKCCS study, which is touched on in the protocol for contextual purposes and background only, is entirely separate from this new phase of the study (purpose of this agreement) - which looks at long term health.

b. Data required

In a previous version of this agreement UKCCS established tracking through NHS Digital's MRIS Reports, and would like to continue receiving death notifications, cancer registrations and movements in/out of the NHS following the last download of data despatched around the end of 2016. UKCCS would also like to extend the data linkage of UKCCS cases and controls to include secondary healthcare datasets within HES, namely HES Admitted Patient Care (HES-APC) from 1997/98; Outpatients (HES-OP) from 2003/04; and Accident and Emergency (HES-A&E) from 2007/08 until the latest financial year available.

c. Why the data is required for this purpose?

UKCCS would like to link to secondary healthcare datasets (HES-APC, HES-OP and HES-A&E) as well as death and cancer registrations in order to study the health of children previously diagnosed with cancer (during their childhood) relative to that of a similarly aged population. A range of hypotheses, requiring individual level data, will be examined (see examples in 6a above). The study has a longitudinal (follow-up) design, and so UKCCS are requesting all available years of data. UKCCS is a nationwide study including all childhood cancers diagnosed over a restricted time-period, together with a representative sample of the general population. The total number of individuals on whom UKCCS require information is relatively small (around 12,684 subjects). This proposal offers a unique data source; and there are no alternative means of obtaining the data necessary to answer the research questions

d. Details of efforts to minimise data required

The linkage will involve the UKCCS cohorts (12684 participants) in total; only the variables necessary to perform analyses are requested.

7) Organisations involved

a. Who is the Data Controller (sole, joint) and who is the Data Processor?

The University of York is the sole Data Controller and also the Data Processor.

b. Are there any organisations which are involved in the wider project (but not processing the data) and if so, the nature of their role?

As explained in 4 above, (this arm of the study subject to this agreement) which receives data from NHS Digital, is completely separate from the historical study; no organisations external to the University of York are involved.

There are no non-NHS sites involved in the collection of data.

c. Are there any funders/commissioners involved and if so, the nature of their role?

Bloodwise funds the UKCCS and confirms they have no involvement in the conduct of the research, including no data controller or processing responsibilities.

Over the first 3 years following receipt of the data, diagnostic-specific comparisons between those who had cancer as a child and those who did not will be used to examine a range of questions, as described above. The University of York will produce several outputs on these topics, publishing the findings as papers and reports, as well as conference presentations. All peer-reviewed articles will be published in well-respected journals such as the British Journal of Cancer, International Journal of Epidemiology, Cancer Epidemiology, European Journal of Cancer, Cancer, and PLoS One. In addition, findings will be disseminated at conferences including UK Childhood Cancer Conference, National Cancer Research Institute Conference, and Pan-European Network for Care of Survivors after Childhood and Adolescent Cancers Conference.

Across all outputs, data are only ever made available in aggregated form, with small numbers suppressed

As part of this study, UKCCS will be initiating an active patient partnership, in line with all studies conducted by UKCCS research group (see https://yhhn.org/partnership as an example). A committee will oversee the patient partnership, whose members will be drawn from among clinicians, patients and their relatives, specialist nurses, and researchers with relevant experience. Preliminary discussions with paediatric oncologists and nurse consultants as well as a number of people diagnosed with cancer in their childhood (who are now young adults) have already taken place; the committee will be formalised following approval of this application. Headed by an independent chair, this advisory group will advise on UKCCS public/patient facing profile, including the production of literature and development of the website.

Core funding comes from a Bloodwise programme grant (Apr 2016-Mar 2021; ref 15037; the epidemiology of haematological malignancies: determinants, prognostics, treatment and survivorship). Funders receive annual interim reports, and final reports will be provided at the end of the funding periods. It is anticipated that the findings of the study could be used to help develop guidelines and improve clinical practice (e.g. monitoring) and those who have recovered from cancers' wellbeing; with outputs on survivorship dependent on this agreement it is worth noting that previous UKCCS research are referred to in NICE guidelines, two examples being signs and symptoms of brain tumours (https://www.nice.org.uk/guidance/ng12) and whether intramuscular administration of vitamin K at birth is linked to childhood cancer (https://www.nice.org.uk/guidance/cg37). Impacts will be maximised via national/international dissemination to NHS practitioners/commissioners, politicians/policy makers, and third-sector organisations.

To ensure accessibility to peer-reviewed outputs, all reports will be published under creative commons attribution 4.0 licence (CC BY); support for this is routinely included in all grant applications and the study’s website (www.ukccs.org) will provide links to these open access publications, conference proceedings, and copies of reports summarizing the findings.

The UKCCS website (www.UKCCS.org) will be updated to reflect all activities, and will contain information about the current proposal (e.g. fair processing, the data linkages and the purpose of these) and ongoing research; and social media forums (e.g. a study-specific twitter feed) will be established. Outputs will be promoted through the study’s Twitter account (to be established) and via the study’s funders (@bloodwise_uk (31,900 followers)). Lay summaries of findings will be provided on the UKCCS website (www.ukccs.org). UKCCS also engages with national charities who publish study findings on websites and in their magazines; an example of such dissemination can be found on the following link: https://bloodwise.org.uk/blog/children-less-affluent-backgroundsmore-likely-die-following-treatment-leukaemia-uk.

The UKCCS is an observational epidemiological study and as such, there are no plans- nor are any foreseen- to develop algorithms, testing and development of tools and new technologies.

Target dates for outputs on the topics is within 3 years of receiving the data. By way of example, the analysis on therapy-related associations will commence three months after receipt of the linked data and will continue for 26 months when findings will be ready for submission to scientific journals.

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

A major aim of UKCCS is to improve care and outcomes for those who have recovered from cancer following a diagnosis as a child. UKCCS, an internationally recognized study with a longstanding record of informing on the possible causes of childhood cancer, is ideally placed to provide national data about the long-term health and healthcare needs of these participants. Indeed, the study has already provided population-based data on childhood cancer outcomes, its variation by cancer type, and socio-demographic characteristics. Importantly, the UKCCS is truly population-based and results are generalizable. Hence, findings have the potential to help inform policy and commissioning of services.

The study is not in support of a PhD/post graduate research study.