Bookmarks
The Oxford Risk Factors And Non Invasive Imaging Study (ORFAN) Arm 4
Safe People
University of Oxford
Academic Institute
No
Safe Projects
DARS-NIC-409610-J6L1F-v0.3
The Oxford Risk Factors And Non Invasive Imaging Study (ORFAN) at large has been operational since 2015. Originally the study was limited to arms 1, 2 and 3. Arm 4 has been developed conceptually from 2018, and came into practical existence from 2019 to both address outstanding scientific questions that cannot be answered with the limited pool of participants in arms 1-3, and to further develop and validate novel biomarkers that have emerged from ORFAN Arm 1-3 and other Ox-HVF studies (see below). ORFAN Arm 4 is funded by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre and The British Heart Foundation (BHF). ORFAN Arm 4 study is a multi-centre observational cohort study involving patients who have had a computed tomography (CT) angiography or CT chest scan. The study is applying to NHS Digital in order to link study participant's patient identifiers with data held by NHS Digital in order to develop and validate cardiovascular disease risk assessment tools that will lead to earlier detection of disease risk and prevent heart attacks and strokes, amongst other cardiovascular diseases. These tools are for usage in the clinical environment and are intended to change patient care for the better. The purpose of the project is to develop new and better biomarkers of cardiovascular disease risk using novel approaches to the analysis of CT scans. The overriding purpose of this work is to reduce the large burden of morbidity and mortality that cardiovascular disease such as heart attack and stroke have in our society. The NHS Digital data requested in this agreement contribute in a fundamental way to the achievement of this purpose by allowing the research team to know what diseases and clinical events had occurred prior to the CT scan of relevance to the study, and what disease and clinical events have occurred for participants following the CT scan. This data is critical to the successful development of image analysis tools as it enables accurate differentiation of participant characteristics, namely imaging characteristics detected from within the CT scans, that can be utilised to predict patient risk. A further major purpose of the ORFAN Arm 4 project is to understand the cardiovascular disease ramifications of COVID-19 infection. The study team know that deleterious impacts on blood vessels and the heart due to acute COVID-19 infection are commonly reported, but they do not know the long-term impacts or how to predict which patients are most at risk of morbidity due to such disease effects. Access to data related to COVID-19 infection and later cardiovascular disease diagnosis and clinical events, along with relevant CT scans of such patients’ hearts and vessels will enable a much more accurate assessment of this risk for individual patients. THE ORFAN STUDY PROTOCOL AND OBJECTIVES The study has two main objectives: 1) To investigate whether biomarkers, including imaging biomarkers, of metabolic risk can predict major adverse cardiovascular events 2) To identify novel biomarkers able to predict cardiovascular disease pathogenesis and extent of pre-existing vascular disease, including in those with COVID-19 infection. The specific primary objective of the ORFAN study is to investigate whether biomarkers of disease risk can predict major adverse cardiovascular events. In order to achieve this objective, the primary outcome measures for the study are: 1. Measurements of plasma/imaging markers of cardiometabolic risk - not directly relevant for this application, although the data requested will be linked to the pseudo-anonymised plasma and imaging data already held by the applicant 2. Atherosclerosis progression by Computerised Tomography - not directly relevant for this application, although the data requested will be linked to the relevant pseudo-anonymised imaging data already held by the applicant 3. Major adverse cardiovascular events over 10 years - relevant for this agreement. The relevant study secondary objective is to identify novel biomarkers able to predict cardiovascular disease pathogenesis and extent of pre-existing vascular disease The secondary outcome measures that are relevant to this application: 1. Measurement of modification to cardiovascular disease pathogenesis and modulation of pre-existing vascular disease that is attributable to infection with coronavirus disease 2019 (COVID-19), and other pre-existing vascular disease. The ORFAN Study was assigned COVID-19 Cardiovascular Disease UK Flagship Project status by the NIHR-BHF in May 2020 and has received ethical approval to pursue scientific inquiry into specific risk that is conveyed by COVID-19 infection in regard to stroke and coronary artery disease. The data from this agreement will form a major part of this inquiry so the study team can understand the heart disease risk that COVID infection confers, and will enable the creation of biomarkers for assessing patients individual risk of heart disease complication following COVID-19 infection. The ORFAN study at large consists of 4 Arms of study. Arms 1, 2 and 3 include a total of 15,500 prospectively recruited participants who are directly consented to be included within the study. The ORFAN Study research arms 1,2 and 3 include directly consented patients for whom the University of Oxford collects and processes data for research purposes. The outcome data of these participants are collected through NHS Digital with whom a data sharing agreement has been signed (DARS-NIC-392669-T1F8B). This agreement herein does not include the collection, processing or storage of any data associated with patients in ORFAN Arms 1, 2 or 3. ORFAN is part of the Oxford cohort for Heart, Vessels and Fat (Ox-HVF), which means the results from this study may be interpreted together with the findings of other Ox-HVF projects. The Ox-HVF cohort is a cluster of clinical studies run from the University of Oxford that together provide results allowing the deployment of a multi-level strategy to understand the mechanisms of cardiovascular disease, specifically heart attack and stroke. This agreement concerns ORFAN Arm 4 only, which is a retrospective study arm which includes up to 100,000 adult participants (75,000 - in the UK; 25,000 – internationally; this application only concerns participants located within England and Wales) who have undergone CT chest, abdomen and pelvis scans for clinical purposes at a participating NHS Trust radiology department. The ORFAN study team are focussed on CT coronary angiograms in this project - scans that only include the heart and surrounding blood vessels and tissues. ORFAN Arm 4 focuses on collecting CT images, patient demographics and clinical information including clinical outcomes and medication usage to aid the development of new CT image analysis algorithms and software tools for the practical application of these algorithms via both traditional and artificial intelligence approaches. ORFAN Arm 4 will provide the required statistical power to allow automation of image analysis processes such as the automated calculation of image analysis techniques developed by the University of Oxford including the perivascular Fat Attenuation Index, the coronary artery Fat Radiomic* Profile and the Atriomic Stroke Algorithm (a novel risk algorithm interrogates the heart atria (the top chambers of the heart) to extract radiomic features, hence ‘atriomic’), as well as the development of new imaging biomarkers - this is the focus of this project. The development of these algorithms and tools has, and will, lead to much improved patient care for those at risk of heart attack and stroke. * In the field of medicine, radiomics is a discipline and collection of methods concerned with the extraction of a large number of statistical features from radiographic medical images using data-characterisation algorithms. These features, termed radiomic features, have the potential to uncover disease characteristics that fail to be appreciated by the naked eye. Following the COVID-19 pandemic, the ORFAN study programme pivoted to include relevant research objectives related to the non-invasive assessment of cardiovascular damage caused by infection with the virus. This work has been designated by the NIHR and the British Heart Foundation as a COVID-19 Cardiovascular Disease UK Flagship Project. The influence of COVID-19 infection on the outcome measures of the study will be explored, and novel tools to assess the impact of COVID-19 infection on blood vessels and the heart will be developed. This agreement is of fundamental importance for the success of ORFAN Arm 4. Access to NHS Digital held data will enable the accurate ascertainment of disease status for all relevant study conditions within all participants. The breadth of relevant diseases and clinical activity for this research is large due to the complexity of accurately adjusting imaging analysis for all conditions. The analysis that will occur will involve the adjustment of patient risk profiles for all clinical events and medication usage prior to the CT scan of interest, and adjustment of all clinical events and medication usage following the CT scan of interest. It is not possible to accurately build risk assessment models without both the prior and post CT scan patient data. As an example, a specific project that will make use of Arm 4 data that has also received significant funding is explained here. This project is the development and clinical translation of the Atriomic Stroke Algorithm, which has received a BHF Translational award (TG/19/2/34831 – ‘Using radiomics and artificial intelligence to predict cardio-embolic stroke’) to validate a novel imaging biomarker for the direct prediction of stroke risk from CT images. This project intends to make use of ORFAN Arm 4 data to provide the most accurate risk assessment of individual stroke risk available. This award has unlocked new technical ability in the ORFAN project as it has funded a very powerful computer capable of processing many thousands of CT scans for deep-learning purposes. Deep-learning is the field of artificial intelligence concerned with the automated interpretation of images, in the same way a computer can identify if a photo portrays a cat or a dog, a computer can also identify if a CT scan portrays a heart or a stomach, to use a rudimentary example. In this project, the computer will learn to identify patients with inflammation around their heart – inflammation that places them at increased risk of stroke. This computer has been purchased and installed at the University of Oxford for this project. The award also funds suitably qualified engineers to work on the ORFAN study utilising this computer for CT scan analysis purposes. The University of Oxford research team who will receive the pseudonymised data from NHS Digital will never receive linkage files that enable the matching of the participants clinical information with their identifiable information. The sole Data Controller is the University of Oxford who also process the data. The team is led by the Professor of Cardiovascular Medicine at the University of Oxford. There is a third-party company, Caristo Diagnostics Ltd, which is a University of Oxford spin-out company from the Antoniades Laboratory. This company is involved in the wider ORFAN study, along with other research studies coordinated from the Professor Antoniades laboratory as part of the Ox-HVF stable of studies. This company will NOT be involved in any processing of NHS Digital data relating to this agreement and will have no data shared with it. Caristo Diagnostics Ltd also do not play any role in determining the means by which any data will be processed under this agreement. There are no funders or commissioners directly involved in the collection or processing of data. The ORFAN study has received funding from a number of sources, however the British Heart Foundation has awarded monies specifically for ORFAN Arm 4. The specific project is the development and clinical translation of the Atriomic Stroke Algorithm via a BHF Translational award (TG/19/2/34831 – ‘Using radiomics and artificial intelligence to predict cardio-embolic stroke’). This project intends to make use of ORFAN Arm 4 data to provide the most accurate risk assessment of individual stroke risk available, as is in keeping with the ORFAN scientific aims. The British Heart Foundation has no control over the methodology of the study nor direct access to NHS Digital data and is therefore not considered a Data Controller. The lawful basis for processing data under GDPR has been reviewed and been assessed as acceptable. The University of Oxford process data under Article 6(1)(e): 'processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller' as they are a Public Authority. Additionally, the University of Oxford process the Special Category Health Data under 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' as the data are required for statistical purposes in the public interest. The public interest relevant to this work is that the research stands to greatly improve the way in which patients – that is, the public of the UK – receive care in relation to heart attack and stroke. These are the top causes of morbidity and mortality in our community and there is huge public interest in improving patient risk assessment and the management of that risk to save lives.
The major outputs from this data processing are twofold: 1) The creation of novel disease risk calculating algorithms algorithms (patient assessment algorithms and the necessary software tools to practically apply the algorithms) for conditions such as heart attack and stroke ready for incorporation into patient care within health services, and 2) the communication of these outputs via scientific and lay-person publications. The research outputs to convey the results will be led by peer reviewed publications in leading international journals, presentations in international and national scientific meetings and subsequent media reporting and public engagement lead by the University of Oxford. Journals being targeted to submit to/publish in: i) The New England Journal of Medicine (NEJM) (impact factor 74.7) ii) The Journal of the American Medical Association (JAMA) (impact factor 45.5) iii) The Lancet (impact factor 60.4) iv) Circulation (impact factor 23.6) v) Journal of the American College of Cardiology (JACC) (impact factor 20.5) vi) British Medical Journal (BMJ) (impact factor 30.2) vii) European Heart Journal (EHJ) (impact factor 22.7) The study team intend major findings of this work to be published in the top-tier general medicine journals (NEJM, JAMA, Lancet, BMJ) as opposed to more cardiac specific findings which will be aimed at the top cardiology journals (Circulation, JACC, EHJ). This enables far greater readership and targeting of impact. All publications in major journals such as those listed here will be open access, meaning any reader from anywhere on Earth can access the full text of the research article without any cost. Congresses and conferences targeted to submit work to: i) Scientific sessions of the American Heart Association ii) Scientific sessions of the European Society of Cardiology iii) Scientific sessions of the American College of Cardiology iv) Scientific sessions of the British Cardiac Society The outputs from this work will be both immediate - with publications in high impact journals within 2-3 years of the commencement of the agreement as well as long-term - when diagnostic biomarkers are implemented in clinical practice. As the ORFAN study is expected to continue collecting outcomes data for at least the next 10 years, the cohort will continue to generate outputs as more events accumulate over time. These long-term impacts have the potential to change clinical practice worldwide and save lives from improved assessment of cardiovascular disease risk. Specific algorithms that the ORFAN Study team wish to create and/or validate through the ORFAN arm 4 study include: 1) An algorithm for the specific risk assessment of ischaemic stroke in those with and without atrial fibrillation (provisionally called the Atriomic Stroke Algorithm) 2) An algorithm for the specific risk assessment of cardiovascular complications in those with COVID-19 infection 3) An algorithm for the likelihood of success of invasive catheter ablation in those with atrial fibrillation 4) Validation of the perivascular Fat Attenuation Index (FAI) algorithm for heart attack risk. The FAI algorithm is an already established algorithm for the accurate assessment of heart attack risk, discovered in the ORFAN Study. FAI involves the assessment of CT scan features, in particular attenuation – the CT term for density, of fat surrounding the coronary vessels. 5) Validation of the Fat Radiomic Profile (FRP) algorithm for heart attack risk. The FRP algorithm is a another already established algorithm for long-term heart attack risk discovered in the ORFAN Study. This algorithm relies upon many radiomic features extracted from CT scans from around the coronary vessels. These features cannot be seen by the naked eye, and so require computer extraction and assessment which is what the FRP computes. 6) Likely other algorithms related to specific at risk population depending on data quality and statistical power All such algorithms are stand-alone patient assessment tools, and any intellectual property created from within the ORFAN Study will be owned by the University of Oxford, and associated intellectual property will be controlled by the University for licensing to health services for inclusion in clinical practice. The ORFAN research team at the University of Oxford has no personal intention to monetise or generate income from the research outputs generated from the ORFAN Arm 4 study. Were discoveries to be made in the ORFAN Arm 4 study, individual researchers within the ORFAN team would be listed inventors on any patents that the University of Oxford may file. The study team expect to be publishing important findings related to the development of novel disease risk assessment for cardiovascular disease within 2-3 years of the DSA commencing. It is possible this may occur earlier if the extraction of data in the first year produces enough scientific power for robust results. They expect novel diagnostic and risk assessment tools to be incorporated into clinical practice within 5 years from the DSA commencing, and further major publications related to the use of the tools developed from within the ORFAN study by the end of the 10 year data retention date. THE LEVEL OF DATA THAT WILL BE CONTAINED IN THE OUTPUTS: The level of data contained in the outputs is aggregate data with small number suppression applied as per the disclosure rules for the various data sets that has been augmented via statistical processes. No individual participants data is ever published individually, with the nature of this scientific work demanding high numbers of participants data for the testing and validation of disease risk calculators. The dissemination activities of the ORFAN Study team are focussed on high-impact peer reviewed journals and oral research presentations at international conferences. Other activities through which the ORFAN study findings are communicated include through NHS Trust clinical grand-round meetings at NHS Trusts who deliver services relevant to those explored in the study, NHS Trust newsletters delivered to staff emails and through high level meetings with NHS executives and clinical leads at NHS Trusts to discuss the suitability for the clinical tools to be tested within the clinical environment at their Trusts. This work also involves audit related work to model the impact of novel tools developed from within ORFAN on current service usage, as has occurred within Oxford University Hospitals NHS Trust already. To target the lay audience, the ORFAN team uses the following approaches and is guided by the Research Services team at the University of Oxford and the Public Engagement team at the Cardiovascular Medicine Division of the University of Oxford: a) Website (www.oxhvf.com); this is updated with all the most up to date information regarding the outputs of the research. This is a public-facing website, and the public can read about the latest outputs of the study. b) Newsletters; when major findings or general outputs are available, the ORFAN team post newsletters both on the website and in hard-copy form to ORFA Arm 1-3 participants. This is not possible for Arm 4, as we do not hold patient identifiable information. Media attention including articles in BBC News, The Guardian, The Financial Times and CBN concerning work published in the prestigious journal ‘Science Translational Medicine’ are the sorts of topics highlighted in newsletters (see example from the Antoniades Lab: https://test188076.files.wordpress.com/2018/01/newsletter-adiporedox-15-09-2017.pdf). c) Press releases; the ORFAN team have an active involvement in outreach activities of the University of Oxford, Oxford University Innovations and the British Heart Foundation communications team, and the major findings from their study lead to press releases, and from there they are distributed to the lay press. d) Workshops and patient and public involvement; the ORFAN team participate in workshops for patients as part of the Biomedical Research Centre in Oxford, and through that they inform the patients about their research and ask for their involvement in the design of protocols, feedback on research procedures and more, through Patient and Public involvement (PPI) panels e) Social media – the ORFAN Study group frequently share results and outputs to their social media presence on Twitter, Facebook and LinkedIn. Exploitation of results/outputs: All intellectual property and knowhow is owned by the University of Oxford, as the sponsor of the ORFAN Study. The University of Oxford, through Oxford University Innovation, maintains the ability to licence any technology that is created within the ORFAN Study for commercial use or for use within health services such as the NHS.
31/03/2021
Safe Data
COVID-19 Hospitalization in England Surveillance System
De-Personalised
Health and Social Care Act 2012 – s261(7); National Health Service Act 2006 - s251 - 'Control of patient information'.
Safe Setting
TRE