Defining Personalised Management Strategies in Patients with Abdominal Aortic Aneurysms
Lead Research Organisation:
University of Oxford
Department Name: Surgical Sciences
Abstract
The aorta is the major blood vessel that pumps blood from the heart to the rest of the body. The part of aorta in the belly is called the abdominal aorta, and should be ~20mm in diameter. An abdominal aortic aneurysm (AAA) is an abnormal ballooning of this section to >30mm. AAA mostly affects men in the retirement age (>65 years old), but women are also at risk. Cigarette smoking and genetics are two of the main factors known to cause AAAs. Left untreated, AAA slowly expands and eventually bursts - this results in sudden onset of severe belly pain and internal bleeding. Only 25% of patients will survive a burst AAA - each year about 6,000 people in the UK and 20,000 people worldwide die from burst AAAs.
People with AAAs don't notice symptoms until it bursts. Fortunately, AAAs can be easily detected with an ultrasound scan. In 2009, the NHS established a national AAA screening program (NAAASP) for men over the age of 65. Since the program started, thousands of new AAA patients are diagnosed by the NAAASP each year. In addition, similar numbers of AAAs are detected by chance from scans done for other medical reasons.
There are two types of surgery to treat AAAs: traditional open surgery involves a big cut in the belly to directly repair the AAA and replace it with a synthetic fabric tube; newer "keyhole" surgery involves tiny cuts in the groin to put in stents that cover the AAA from inside.
International guidelines recommend that patients with small AAAs (between 30 to 54mm in diameter) should be monitored by regular ultrasound scans, and for surgery to be considered when an AAA grows larger than 55mm. In combination, these have been shown to decrease the chance of people dying from burst AAAs. To make sure people undergo surgery before the AAA bursts, the NHS spends millions each year to monitor them. This cost will continue to rise as more patients are being monitored. The global market of aneurysm stents is also expanding rapidly (>£3bn by 2024) as more and more stents are being used to treat patients. Alarmingly, many more aortic stent procedures are being done in fee-for-service health care systems (such as Germany and USA) before the AAA reaches the diameter of 55mm.
I recently engaged hundreds of AAA patients at Oxford to see what is most important to them in terms of AAA management. 'Frequency of AAA monitoring' emerged as one of the most important issues during AAA monitoring. I also conducted an online international survey of vascular surgeons, where 'new ways to predict AAA growth' was voted as the top priority for aneurysm research. Such a method did not exist, and would be hugely valuable to the patients and the NHS: It will help clinicians decide how often each patient needs monitoring scans; It will also safeguard patients against unnecessary surgery.
I have developed a method for the prediction of AAA growth with good accuracy. This method involves an additional blood test taken at the same time at the routine AAA monitoring scan. I am also developing a method to predict AAA growth by the analysis of computerised tomography scans.
This proposed project has the primary aim to validate my method of AAA growth prediction. I will assess the accuracy of my prediction method using patient samples collected at different hospitals, internationally. In addition, I will improve the prediction method by doing further analyses on blood samples and CT scans to find additional markers that can be useful. Lastly, I will develop a point of care testing kit (to measure blood) and a medical software (to analyse CT images). These will make our prediction methods accessible to clinical practice.
People with AAAs don't notice symptoms until it bursts. Fortunately, AAAs can be easily detected with an ultrasound scan. In 2009, the NHS established a national AAA screening program (NAAASP) for men over the age of 65. Since the program started, thousands of new AAA patients are diagnosed by the NAAASP each year. In addition, similar numbers of AAAs are detected by chance from scans done for other medical reasons.
There are two types of surgery to treat AAAs: traditional open surgery involves a big cut in the belly to directly repair the AAA and replace it with a synthetic fabric tube; newer "keyhole" surgery involves tiny cuts in the groin to put in stents that cover the AAA from inside.
International guidelines recommend that patients with small AAAs (between 30 to 54mm in diameter) should be monitored by regular ultrasound scans, and for surgery to be considered when an AAA grows larger than 55mm. In combination, these have been shown to decrease the chance of people dying from burst AAAs. To make sure people undergo surgery before the AAA bursts, the NHS spends millions each year to monitor them. This cost will continue to rise as more patients are being monitored. The global market of aneurysm stents is also expanding rapidly (>£3bn by 2024) as more and more stents are being used to treat patients. Alarmingly, many more aortic stent procedures are being done in fee-for-service health care systems (such as Germany and USA) before the AAA reaches the diameter of 55mm.
I recently engaged hundreds of AAA patients at Oxford to see what is most important to them in terms of AAA management. 'Frequency of AAA monitoring' emerged as one of the most important issues during AAA monitoring. I also conducted an online international survey of vascular surgeons, where 'new ways to predict AAA growth' was voted as the top priority for aneurysm research. Such a method did not exist, and would be hugely valuable to the patients and the NHS: It will help clinicians decide how often each patient needs monitoring scans; It will also safeguard patients against unnecessary surgery.
I have developed a method for the prediction of AAA growth with good accuracy. This method involves an additional blood test taken at the same time at the routine AAA monitoring scan. I am also developing a method to predict AAA growth by the analysis of computerised tomography scans.
This proposed project has the primary aim to validate my method of AAA growth prediction. I will assess the accuracy of my prediction method using patient samples collected at different hospitals, internationally. In addition, I will improve the prediction method by doing further analyses on blood samples and CT scans to find additional markers that can be useful. Lastly, I will develop a point of care testing kit (to measure blood) and a medical software (to analyse CT images). These will make our prediction methods accessible to clinical practice.
Organisations
- University of Oxford (Fellow, Lead Research Organisation)
- Jikei University (Collaboration)
- UNIVERSITY OF LEICESTER (Collaboration)
- Karolinska Institute (Collaboration)
- AgPlus Diagnostics Ltd (Collaboration)
- IMPERIAL COLLEGE LONDON (Collaboration)
- Amazon.com (Collaboration)
- UNIVERSITY OF OXFORD (Collaboration)
- UNIVERSITY OF EDINBURGH (Collaboration)
- UNIVERSITY OF GLASGOW (Collaboration)
- Instituto de Investigación Sanitaria La Fe (Collaboration)
- Oxford Computer Consultants (United Kingdom) (Project Partner)
- Protein Simple - a BioTechne Brand (Project Partner)
- NLC Ventures (Project Partner)
- AgPlus Diagnostics (Project Partner)
Publications
Bruijn LE
(2021)
A histopathological classification scheme for abdominal aortic aneurysm disease.
in JVS-vascular science
Chandrashekar A
(2023)
A Deep Learning Approach to Visualize Aortic Aneurysm Morphology Without the Use of Intravenous Contrast Agents.
in Annals of surgery
Chandrashekar A
(2023)
Prediction of Abdominal Aortic Aneurysm Growth Using Geometric Assessment of Computerized Tomography Images Acquired During the Aneurysm Surveillance Period.
in Annals of surgery
Dib L
(2023)
Lipid-associated macrophages transition to an inflammatory state in human atherosclerosis increasing the risk of cerebrovascular complications.
in Nature cardiovascular research
Edsfeldt A
(2022)
Interferon regulatory factor-5-dependent CD11c+ macrophages contribute to the formation of rupture-prone atherosclerotic plaques.
in European heart journal
Kotronias RA
(2022)
Machine learning assisted reflectance spectral characterisation of coronary thrombi correlates with microvascular injury in patients with ST-segment elevation acute coronary syndrome.
in Frontiers in cardiovascular medicine
Lareyre F
(2023)
Applications of artificial intelligence for patients with peripheral artery disease.
in Journal of vascular surgery
Loick P
(2023)
Protective Role for Smooth Muscle Cell Hepcidin in Abdominal Aortic Aneurysm.
in Arteriosclerosis, thrombosis, and vascular biology
Ngetich E
(2022)
The role of dipeptidyl peptidase-IV in abdominal aortic aneurysm pathogenesis: A systematic review.
in Vascular medicine (London, England)
Description | Key milestones for year 1 have been met for the work packages of the fellowship |
Exploitation Route | Fellowship in progress |
Sectors | Healthcare |
Description | My research findings have formed the scientific basis of a new research study protocol and the formation of a new international consortium involving multiple industry, clinical and academic partners. All my international partners enthusiastically support my proposed research and are making contribution in kind in terms of time and effort in seeking their local ethics/regulatory approval to replicate and advance the research. |
First Year Of Impact | 2022 |
Sector | Digital/Communication/Information Technologies (including Software),Environment,Healthcare |
Description | Libin Cardiovascular Institute Aortic Research Prioritization Setting Exercise |
Geographic Reach | North America |
Policy Influence Type | Participation in a guidance/advisory committee |
Description | Vascular Society Special Interest Group Member |
Geographic Reach | National |
Policy Influence Type | Contribution to a national consultation/review |
URL | https://jvsgbi.com/research-priorities-for-aortic-diseases-results-of-the-james-lind-alliance-vascul... |
Description | Determining how bioactive phospholipids regulate development of abdominal aortic aneurysm and coagulation using multiomic approaches |
Amount | £150,000 (GBP) |
Funding ID | RG/F/20/110020 |
Organisation | British Heart Foundation (BHF) |
Sector | Charity/Non Profit |
Country | United Kingdom |
Start | 06/2021 |
End | 06/2025 |
Title | AICT Study protocol |
Description | This is a new research protocol which enables / seeks secondary use of routinely collected CT scan data from clinical sites. |
Type Of Material | Improvements to research infrastructure |
Year Produced | 2022 |
Provided To Others? | No |
Impact | The study protocol has been shared with international research collaborator sites to seek equivalent approval at their institution. This enables the formation of an international research consortium (in preparation) |
Description | Biomedical Image Analysis Lab (Grau Lab) |
Organisation | Imperial College London |
Department | Institute of Biomedical Engineering (IBME) |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Analysis of CT images for the prediction of AAA growth |
Collaborator Contribution | Prof Grau attracted a 4th year engineering student to take up the proposed project as a final year project. We are generating very promising pilot data, and have prepared grant appilications to extend the collaboration. |
Impact | Pilot data to demonstrate feasibility of AAA growth prediction using CT images Subsequently co-invented 4 IP which have been filed as patents and 4x joint publications for which I am the senior author. Two of these patents have been licensed into the spinout company (see relevant section) for which we are academic co-founders. |
Start Year | 2017 |
Description | Industry Partnership (AWS) |
Organisation | Amazon.com |
Department | Amazon Web Services |
Country | United States |
Sector | Private |
PI Contribution | Engagement of AWS in a new niche of research which may increase their revenue (and profile) |
Collaborator Contribution | AWS has provided a letter of support for the AICT consortium grant application (EPSRC Digital Health Hub), which stated a contribution in kind of £100,000 will be made available with a successful funding application. Although this application was not successful, I have formed ongoing communication channels with the AWS team. They also hosted the AICT consortium partners at AWS London headquarters on 7th October 2022 for a workshop with the envision engineering team. |
Impact | No output/outcome as yet - we are having ongoing communications. |
Start Year | 2022 |
Description | Industry Partnership (AgP) |
Organisation | AgPlus Diagnostics Ltd |
Country | United Kingdom |
Sector | Private |
PI Contribution | Providing a new clinical application for their point of care diagnostic platform |
Collaborator Contribution | helped with research grant planning and provided support for my UKRI Future Leaders Fellowship |
Impact | The work package with AgPlus diagnostic is a formal WP for my successful UKRK Future Leaders Fellowship application |
Start Year | 2019 |
Description | International Research Partnership (Jikei) |
Organisation | Jikei University |
Country | Japan |
Sector | Academic/University |
PI Contribution | expanding their research network |
Collaborator Contribution | replicating the clinical recruitment and sample collection process for the OxAAA study Providing additional validation dataset for my research listed as a collaborator for one of the work packages of the UKRI Future Leaders Fellowship |
Impact | data has contributed to DPhil thesis planning for a manuscript |
Start Year | 2018 |
Description | International Research Partnership (KI) |
Organisation | Karolinska Institute |
Country | Sweden |
Sector | Academic/University |
PI Contribution | providing secondary use of their previously collected research data / samples |
Collaborator Contribution | contribution of samples to validate my research output. Included as a collaborator in WP for the UKRI FLF |
Impact | still in progress |
Start Year | 2019 |
Description | International Research Partnership (LaFe) |
Organisation | Instituto de Investigación Sanitaria La Fe |
Country | Spain |
Sector | Charity/Non Profit |
PI Contribution | Engaged the academic site to participate in the AICT consortium |
Collaborator Contribution | The site has contributed significant time locally in seeking regulatory and ethics approval for participating in the AICT consortium. |
Impact | AICT consortium is an inter/multidisciplinary research collaborative to advance AI data science in CT imaging. We have confirmed participation of clinical sites across continents: (England, Scotland, Poland, France, Belgium, Greece, Spain, Brazil) and ongoing discussions with sites in Australia, Sweden, Netherlands and Canada. The underpinning science for this international consortium is described in the AICT Study protocol, which received NHS Human Research Authority approval in 2022 (Ref: 22/HRA/2302). University of Oxford is centrally managing the AICT consortium research agreement (legal and governance framework). This agreement has gone through iterative drafting since early 2022 and is now undergoing final approval by the inaugural member sites (target signing / approval date: Easter 2023). I successfully hosted the Inaugural AICT consortium annual general meeting on 28th Feb 2023 at Vienna. This coincided with the European Congress of Radiology, which is the biggest clinical radiology conference in Europe. Disciplines included: - Clinical: surgery, oncology, radiology - Engineering: computer vision and deep learning - Industry: ML/Ops, clinical site connectivity, data science, software engineering, hospital groups, privacy law - Patient and public stakeholders: PPI, trustworthiness of AI, |
Start Year | 2022 |
Description | National Research Partnership (Ed) |
Organisation | University of Edinburgh |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | providing additional secondary use of their previously collected research data |
Collaborator Contribution | providing an independent dataset to validate my research output |
Impact | Conference presentation x 1 Contribution to DPhil thesis content manuscript in preparation |
Start Year | 2021 |
Description | National Research Partnership (UoG) |
Organisation | University of Glasgow |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Engaged the academic site to participate in the AICT consortium |
Collaborator Contribution | The site investigator is a co-investigator of the AICT study. A statistician at University of Glasgow provided statistics approval of the AICT Study protocol. |
Impact | AICT consortium is an inter/multidisciplinary research collaborative to advance AI data science in CT imaging. We have confirmed participation of clinical sites across continents: (England, Scotland, Poland, France, Belgium, Greece, Spain, Brazil) and ongoing discussions with sites in Australia, Sweden, Netherlands and Canada. The underpinning science for this international consortium is described in the AICT Study protocol, which received NHS Human Research Authority approval in 2022 (Ref: 22/HRA/2302). University of Oxford is centrally managing the AICT consortium research agreement (legal and governance framework). This agreement has gone through iterative drafting since early 2022 and is now undergoing final approval by the inaugural member sites (target signing / approval date: Easter 2023). I successfully hosted the Inaugural AICT consortium annual general meeting on 28th Feb 2023 at Vienna. This coincided with the European Congress of Radiology, which is the biggest clinical radiology conference in Europe. Disciplines included: - Clinical: surgery, oncology, radiology - Engineering: computer vision and deep learning - Industry: ML/Ops, clinical site connectivity, data science, software engineering, hospital groups, privacy law - Patient and public stakeholders: PPI, trustworthiness of AI, |
Start Year | 2021 |
Description | National Research Partnership (UoL) |
Organisation | University of Leicester |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Invited the team to join my research consortium |
Collaborator Contribution | The academic is a senior figure in AAA research and is providing a named mentorship for my UKRI Future Leaders Fellowship Helped me with networking opportunities with other researchers in this field and invited me to join the Vascular Society aortic aneurysm special interest group |
Impact | Included me for international consortium and contributed clinical samples for AAA multiple cohort studies Manuscript x1 in preparation Publication x1 |
Start Year | 2020 |
Description | Oxford University Research Collaboration (HERC) |
Organisation | University of Oxford |
Department | Health Economics Research Centre |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Providing the AAA clinical pathway and my proposed precision management strategy to investigate the |
Collaborator Contribution | Significant time invested to support my grant application and to plan for the delivery of one of my UKRI FLF work packages |
Impact | Successfully supported my UKRI Future Leaders Fellowship (one of the work package is health economics analysis) |
Start Year | 2019 |
Title | ANEURYSM GROWTH RATE ESTIMATION |
Description | A method is provided for estimating a growth rate of an aneurysm. The method comprises receiving a plurality of data points indicative of a motion of an artery in a subject having an aneurysm, the plurality of data points extracted from a plurality of ultrasound images of the artery. The method further comprises fitting a set one or more parameters of a mathematical function to approximate the plurality of data points, the function representative of a model of arterial motion. The method further comprises determining, from at least one parameter of the set of fitted parameters, an estimated growth rate of the aneurysm.Computer-readable media and apparatuses are also described. |
IP Reference | WO2020030890 |
Protection | Patent / Patent application |
Year Protection Granted | 2020 |
Licensed | No |
Impact | Further builds evidence that abdominal aortic aneurysm is a disease with systemic phenotypes and provide evidence to subsequent grant applications |
Title | AORTIC ANEURYSM GROWTH RATE PREDICTION FROM GEOMETRIC ANALYSIS |
Description | A computer-implemented method is provided for predicting a growth rate of an aortic aneurysm. The method comprises analysing one or more geometric measures of a volumetric model of at least a portion of an aorta having an aneurysm. The method further comprises determining, from the analysis, a growth rate prediction for the aortic aneurysm. Computer- readable media and apparatuses are also described. |
IP Reference | WO2022074398 |
Protection | Patent / Patent application |
Year Protection Granted | 2022 |
Licensed | No |
Impact | This IP contributes toward the scientific underpinning of a new research study (AICT study) which has NHS Human research authority approval. Also enabling further international collaboration to further validate the AAA growth prediction algorithm |
Title | FUNCTIONAL IMAGING FEATURES FROM COMPUTED TOMOGRAPHY IMAGES |
Description | Methods, apparatus and computer readable media are provided for identifying functional features from a computed tomography (CT) image.. The CT image may be a contrast-enhanced CT image or a non-contrast CT image. According to some examples, methods, apparatus and computer readable media are also provided for using machine learning to identify functional features from CT images. According to some examples, simulated functional image datasets such as simulated PET images or simulated SUV images are generated from a received CT image. |
IP Reference | WO2021229223 |
Protection | Patent / Patent application |
Year Protection Granted | 2021 |
Licensed | No |
Impact | Underpinning the science for a new research study (AICT study) which has received NHS Human Research Authority Approval in 2022 |
Title | Software as a Medical Device |
Description | This is a software as a medical device being developed by the spinout company (AiSentia) for which I am the co-founder, chief medical officer. |
Type | Diagnostic Tool - Imaging |
Current Stage Of Development | Refinement. Non-clinical |
Year Development Stage Completed | 2023 |
Development Status | Actively seeking support |
Impact | The spinout company (AiSentia) has received innovated UK Smart Grant and SBRI NHS NET ZERO phase 1 award. |
Company Name | AISENTIA LIMITED |
Description | AiSentia is a limited company that manufactures medical software as a device. Our product focus on applying artificial intelligence to computerised tomography imaging to extract higher order details that is otherwise not visualised by human eyes. |
Year Established | 2021 |
Impact | We are at seed funding stage. Secured pre-seed investment from a Dutch venture builder (NLC health) and co-founder investment. Awarded non-dilutive funding from Innovate UK Smart Grant and SBRI NHS NET ZERO award. |
Website | http://www.aisentia.com |