Risk stratification, AI and bold new ideas: our latest funded early detection research

We’re using our funding to build a broad and strong early detection research community and we recently invested in 15 new projects and programmes in this field.

Here we highlight some of the proposals supported at our May 2019 Early Detection Committee meeting, which exemplify our strategy to grow the field by supporting international collaborations, leveraging multi-disciplinary collaborations and new technologies, and bolstering early career researchers who are entering the field with new ideas.

Global team will stratify risk of myeloma progression

Kwee Yong, Irene Ghobrial, and Karthik Ramasamy – Early Detection Programme funded for £3.2million for 5 years between University College London, Harvard University and Oxford University

Kwee, Irene and Karthik are seeking to understand characteristics of smouldering myeloma (SMM) indicating low-risk indolent disease or likelihood of progression to multiple myeloma (MM). Bringing together an international team of leading researchers in the field around genomics, imaging, liquid biopsy and immune environment assays, the programme will build an integrative risk model to determine which individuals with SMM need further monitoring to prevent and early detect transition from the precursor condition to MM.

This study combines existing US and UK prospective cohorts (the PROMISE and DEFENSE studies, respectively) and multiple clinical modalities, to generate an evidence base that will enhance and improve the clinical care pathway for these patients. Identification of those who are truly at risk for progression has the potential to improve management and enhance earlier detection, which will benefit prognosis and survival of those who do develop MM.

Using AI to enhance breast cancer screening

Adam Brentnall and Giovanni Montana – Early Detection Project funded for £447,000 over 4 years between Queen Mary University of London and University of Warwick

Adam and Giovanni will develop and test an artificial intelligence (AI) system to aid risk-stratified breast cancer screening to determine if this enhanced approach will be better than the current standard in detecting, minimising false negatives, and determining long-term risk of an interval breast cancer. The team brings together expertise in statistics of breast cancer risk and machine learning, who together will curate a database of mammographic images from nearly 20,000 women with and over 500,000 women without breast cancer across the US, UK, and Russia. While questions remain how the AI algorithms will account for differences in the collection of the images, there is a clear clinical need to improve the efficiency and accuracy of early breast cancer detection in radiographic images.

Testing out a new idea to detect lung cancer

Rachel Evans – Early Detection Primer funded for £98,000 for 1 year at University College London

Rachel is an early career researcher interested in applying her expertise towards early detection. Originally working on developing immunotherapies targeting a protein called ROR1 that is found in immune-driven aggressive tumours with poor prognosis like lung cancer, Rachel thought that this could also be a marker for early disease. This pilot proposal will define ROR1 expression in early stage mouse models of NSCLC and then assess the ability of an engineered protein antibody to bind to ROR1 with high sensitivity and specificity.

Rachel’s work has the potential to build towards developing a ROR1 tracer for monitoring nodules in high risk patients. Before submitting her application, Rachel had participated in our early career focus group and had observed an Early Detection Committee meeting, giving her perspective on how her work might be relevant to the research we are funding in the area and what makes a successful proposal.   

Our May 2019 funded awards in full

Early Detection Programme Awards

Kwee Yong, University College London, Irene Ghobrial, Broad Institute at Harvard University, and Karthik Ramasamy, University of Oxford

Defining risk in smouldering myeloma (SMM) for early detection of multiple myeloma (MM)

Early Detection Project Awards

Matthew Baker and Paul Brennan, University of Strathclyde and University of Edinburgh

Validating serum diagnostics for early diagnosis and stratification of glioma

Kevin Ryan, Cancer Research UK Beatson Institute

Identification of biomarkers for the detection of pre-cancerous lesions associated with pancreatic ductal adenocarcinoma

Adam Brentnall and Giovanni Montana, Queen Mary University of London and University of Warwick         

An artificial intelligence system for real-time risk assessment at mammography screening

Sam Janes, University College London          

ASCENT: analysis of screen-detected lung cancers genomic traits

John Doorbar, University of Cambridge

A novel patch sampling approach for grading and generating cervical disease maps

Early Detection Primer Awards

Karen Pooley, University of Cambridge              

Evaluation of microRNAs as biomarkers in the early detection of breast cancer blood

Rachel Evans, University College London         

Early detection of lung cancer using a molecular tracer

Elizabeth Soilleux, University of Cambridge              

Developing a novel high sensitivity (capture-based) clonality test for the early detection of lymphoid neoplasia, including lymphoma, myeloma and lymphoid leukaemia

Paul Barber, University College London

Exosomal protein dimers as an early detection biomarker for lung cancer

Rosamonde Banks, University of Leeds        

Exploring the use of tumour-associated autoantibodies in the early detection of renal cancer

Helen Coleman and Kathleen Curtius, Queen’s University Belfast and Queen Mary University of London

Understanding the molecular age of Barrett’s oesophagus in a population-representative sample of patients spanning paediatric to older age groups

Daniel Richards and Marta Broto Aviles, Imperial College London             

Developing ultra-sensitive lateral flow immunoassays for the early detection of ovarian cancer

Joseph Jacob, University College London          

Applying an attention-based, time-aware, recurrent neural network model to predict pulmonary nodule evolution in lung cancer screening patients

Sara Zanivan, Cancer Research UK Beatson Institute

Exploring deep plasma redox-proteome profiling for early detection of HCC

Get involved

Do you have an idea for research that could enable earlier detection? Whether you’re early in your career, established in the field, or have never worked on cancer detection before, we’d love to hear from you.