Developing and integrating multi-modal techniques to predict oesophageal and gastric cancer behaviour

Mr John Findlay is a Registrar in his final year of training in Upper Gastrointestinal Surgery at the Churchill Hospital (Oxford University Hospitals NHS Foundation Trust) and an Honorary Clinical Lecturer at the Department of Oncology. John’s clinical and academic interest is cancer of the oesophagus and stomach, into which he is currently undertaking postdoctoral research in the Department of Oncology. This focusses on developing new imaging, pathological and molecular metrics and how these can be integrated and used to better predict cancer behaviour and patient outcomes from treatment.

Patients with oesophageal and gastric cancer are usually treated with neoadjuvant therapy followed by surgery. A range of treatment options is available, but all carry substantial risks and implications for patient quality of life. Furthermore, despite optimal treatment cancer recurrence remains common, suggesting the way we assess cancer burden and make decisions about treatment is overly simplistic, and does not represent what is increasingly understood to be very complex underlying biology. The success of these therapies is therefore very variable, however, in the absence of mechanisms to predict this, a ‘one size fits all’ approach needs to be taken. As a result, in hindsight many patients receive ultimately futile and morbid treatment (e.g. chemotherapy and then surgery), whilst others do not receive treatment which would benefit them.

John and his team are looking into ways of understanding why the success of treatment is so variable, and how it could be predicted better. Key to this is the philosophy of integrating data on cancer from different angles and at different time points, rather than viewing modalities, time points and markers in isolation. John is trying to generate and refine new markers of treatment success and outcome with an emphasis on integrating imaging, pathology and molecular techniques in conjunction with machine learning. John hopes this enhanced understanding of the underlying mechanisms will accelerate more personalised patient management, and the development of decision support tools for patients and clinicians to make better and more informed decisions.

In Oxford John is collaborating with the Department of Oncology (Prof Mark Middleton and Prof Fergus Gleeson), the Ludwig Institute for Cancer Research (Oxford Branch) (Prof. Xin Lu), the Big Data Institute (Dr David Wedge), and the Radiology Department of the NHS Foundation Trust Oxford University Hospitals (Dr Kevin Bradley).

Within the UK John’s collaborators are Prof Tim Underwood (University of Southampton), Mr Andrew Davies (Guy’s and St Thomas’), Dr Sheraz Markar (Imperial College London) and other fellow members of the Association of Upper GI Surgeons of Great Britain and Ireland and Royal College of Surgeons Upper GI Research Collaborative.

International collaborators include Dr Don Low  and Dr Fredrik Klevebro (Virginia Mason Medical Centre, Seattle), Ms Jessie Elliot (University College Dublin) and other fellow members of the European Society for Diseases of the Esophagus, Young Investigators Division.

John’s research is funded by the Oxford Biomedical Research Centre and CRUK.

Supporting Local and International Oesophageal Cancer Research Efforts with Clinical Data Infrastructure

Stephen Ash is a Data Manager and clinical statistician at the Ludwig Cancer Centre (Oxford Branch – part of the Nuffield Department of Medicine), and the Churchill Hospital (Oxford University Hospitals NHS Foundation Trust – OUHFT).

Steve acts as a link between data collection within the NHS and data usage for research within the University. He is collecting and storing clinical data from patients with oesophageal or gastric cancer. Together with the team’s Data Assistant Paola Medina Cruz, Steve ensures that researchers working with the dataset are provided with sufficient information about the patient’s cancer and their care, to be able to understand and interpret the analyses which will be carried out on donated blood and tumour samples, whilst maintaining patient confidentiality.

When a patient undergoes investigations for cancer in the NHS, a vast amount of information is generated. These data are collated in a form that can be integrated in the daily administrative work within the NHS, but not necessarily in a format that is suitable for usage within a research environment, especially Oxford’s Cancer Big Data community. Steve develops databases which support a wide range of research in this field, and trains medical staff in how to use them ensuring accuracy and completeness. Steve’s efforts are crucial in informing a portfolio of ongoing research projects studying patient samples that are collected from these same patients. One example is Barbara Braden’s and Adam Bailey’s research on using AI to detect Oesophageal Cancer earlier.

In addition to researchers, the OUHFT surgical team also benefit from the high-quality and comprehensive dataset, using it to analyse their own performance, monitor the occurrence of surgical complications, and look for ways to improve the care they provide to patients.

Steve is also responsible for overseeing Oxford’s contribution to the Esodata project, run by Dr Don Low and the International Society for Diseases of the Esophagus (ISDE), an international collaboration between 30 health institutions around the world. The aim of this project is to be able to better predict post-surgery complications associated with oesophagectomies so that patients at highest risk can be more closely monitored.

Within Oxford, Steve collaborates with (amongst others) the Oxford University Hospitals NHS Foundation Trust (including surgeon and OUHFT Cancer Lead, Mr Nick Maynard), the Ludwig Cancer Centre (Oxford Branch) (including Prof Xin Lu), the Department of Oncology (including Prof Mark Middleton), the Wellcome Trust Centre for Human Genetics (part of the Nuffield Department of Medicine) (including Prof Simon Leedham), the MRC Weatherall Institute for Molecular Medicine (including Simona Fourie), and the Translational Gastroenterology Unit (within the Nuffield Department of Medicine – including Prof Barbara Braden).

Improving Early Detection of Recurring Cancers after Chemoradiotherapy Treatment

Dr. Somnath Mukherjee is part of a team trying to improve outcomes in oesophageal patients treated with chemoradiotherapy. Currently only about a quarter of the patients treated with chemoradiotherapy survive beyond 5 years.

Following treatment cancers can recur both locally (due to a failure of radiotherapy) or in distant parts of the body (due to a failure of chemotherapy). Som is leading in clinical trial (CYTOFLOC) which is investigating whether a minimally invasive technology called cytosponge can reliably detect residual cancer in the  oesophagus following chemoradiotherapy to help earlier detection of locally relapsing cancers which may be targeted with surgery in some patients. In addition, Som is also co-leading in a large multi-centre randomised trial (SCOPE2) which is investigating whether escalating radiotherapy dose or switching chemotherapy based on PET scan response, can improve outcomes in oesophageal cancer.

Within Oxford Som is collaborating with the Department of Oncology (including Prof Maria Hawkins and Prof Mark Middleton). Further collaborators are from the University of Cardiff (Dr Tom Crosby) and the University of Cambridge (Prof Rebecca Fitzgerald).

 

CRUK funded CYTOFLOC and SCOPE2. For the CYTOFLOC study, Europlaz provides the cytosponge device.

Realising the Benefit of Immunotherapy for Oesophageal Cancer Patients

Thomas Carroll is a DPhil student in Prof. Xin Lu’s lab at the Ludwig Institute for Cancer Research (Oxford Branch)  (hosted within the Nuffield Department of Medicine). He did his undergraduate studies in Biochemistry and Classical Studies in Houston (Texas), with a focus on tumour biology, immunology, and bioethics. Thomas’ current project studies the potential benefits of using immunotherapy to treat Oesophageal Adenocarcinoma (OAC) with the aim of addressing the need for improved patient outcomes and more effective therapies in this difficult-to-treat form of cancer. The two main questions in this area of research are: First, is it possible to predict which OAC patients are going to benefit from immunotherapy; and second, how could immunotherapy be augmented to help patients who might not have otherwise benefitted.

Thomas’ work is based around an ongoing OAC immunotherapy clinical trial based at Oxford, and there are a team of researchers developing and delivering this study including trial and data managers, nurses, clinicians and technicians. Thomas is one of several researchers using samples collected during this trial in their research. Specifically, he uses a variety of cutting-edge sequencing technologies to analyse blood and tumour tissue, in the hope of identifying key molecular changes that can be used to identify which future patients will and will not benefit from these new, potentially curative treatments.

By finding predictors of positive responses to immunotherapy, the team hope to implement novel diagnostic tests to make sure patients are getting the best treatment option for their specific cancer and aren’t going through side-effects for a therapy from which they may not benefit. Additionally, being able to identify which patients are unresponsive is the first step in devising new strategies for sensitising them and broadening the benefits of treatment.

Thomas collaborates with a number of groups within Oxford, including the Department of Oncology (Prof. Mark Middleton), the Ludwig Institute for Cancer Research (Oxford Branch) (Prof. Xin Lu), and the Target Discovery Unit (hosted within the Nuffield Department of Medicine, with Prof. Benedikt Kessler.

Using Genetic Information to personalise Treatment

Dr David Wedge is a Group Leader in Cancer Genomics, based at the Big Data Institute in Oxford. David studied Chemistry in Oxford and Computer Science at Manchester Metropolitan University.

David is aiming to better personalise treatment to individual cancer patients by providing additional genetic information to doctors treating patients. Currently, cancer patients are ‘staged’ using a system known as ‘TNM’, which is based on the location of tumours in the original (primary) location, nearby lymph nodes and distant organs (metastases). However, TNM staging does not always allow doctors to distinguish the most aggressive cancers from those that are more benign. By better identifying those tumours that are more likely to spread, treatment could be better personalised to individual patients.

The team around David have sequenced the DNA of patients with metastatic Oesophageal Cancer and compared the results from the primary and numerous secondary tumours. Using this approach David and his team are able to track how the cancers spread from the oesophagus to local lymph nodes and to distant organs. This provides a clearer picture of how the cancers spread and of which mutations may make tumours more aggressive. It is hoped that understanding the mechanisms by which cancer spreads, assays which more effectively stage a cancer can be derived and thus patients could make more informed decisions about treatment options.

David collaborates with researchers from the Addenbrooke’s Hospital in Cambridge (Ayesha Noorani and Rebecca Fitzgerald).

This project was made possible by the PHOENIX autopsy study, which was set up by Ayesha Noorani, and the OCCAMS study, which is led by Rebecca Fitzgerald, with funding from the MRC and Cancer Research UK.

Use of circulating extracellular vesicles for the early detection of Oesophageal Cancer

Elizabeth Bird-Lieberman is a Gastroenterology Consultant in the Translational Gastroenterology Unit (Nuffield Department of Medicine) at the John Radcliffe Hospital (Oxford University NHS Foundation Hospital Trust), and an honorary Senior Clinical Lecturer at the University of Oxford. She trained in medicine at the University of Cambridge and undertook her PhD there.

 

Elizabeth’s research focuses on the development of non-invasive tests to identify those patients at risk of gastrointestinal cancer, and on the development of advanced endoscopic tools to improve the early detection of pre-cancerous change in the gastrointestinal tract.

 

Early- and pre-cancerous changes, which are more easily treated than later stage cancer, are difficult to detect, so the development of better tools to detect those changes is essential to improving patient treatment. The work of Elizabeth and her team enables development of techniques that helps to monitor patients at high risk more intensively, and helps avoid unnecessary procedures for those at lower risk.

Within Oxford Elizabeth collaborates with the Department of Physiology, Anatomy and Genetics (Prof Deborah Goberdhan), trying to develop a non-invasive way of detecting early and pre-cancer of the Oesophagus by looking at blood samples for circulating extracellular vesicles (EVs). EVs are tiny spheres surrounded by a membrane that are released by cells as a way of communicating with each other and we aim to develop a tool that can identify EVs from pre-cancerous cells in the Oesophagus.

 

Elizabeth’s work is funded by Cancer Research UK, NIHR Oxford Biomedical Research Centre, Medical Research Council, Academy of Medical Sciences and National Institute for Health Research.

Machine Learning to help improve Early Detection and Treatment of Barrett’s Oesophagus

Dr Richard Owen is an upper gastrointestinal researcher based in the Ludwig Institute for Cancer Research, Nuffield Department of Medicine, and the Department of Upper GI Surgery, (Oxford University Hospitals NHS Foundation Trust – OUHFT). He studied as an undergraduate at the University of Liverpool, completed his early surgical training in Liverpool and Manchester, and did his DPhil in Oxford. Richard is currently completing his surgical training in Thames Valley and Oxford.

 

Barrett’s Oesophagus (changes in the cell types in the lower oesophagus) is a risk factor for Oesophageal Cancer. Currently, patients with Barrett’s Oesophagus are regularly checked using endoscopic sampling to allow intervention when early signs of cancer develop and the disease is non-life threatening. However, as relatively few Barrett’s Oesophagus patients progress to cancer, the majority needlessly undergoes endoscopic tests. Furthermore, many cancers develop without the proof that Barrett’s was originally present.

 

Using samples donated by affected patients, Richard utilises single cell and small-sample sequencing, and proteomic analysis to explore the mechanisms by which Barrett’s Oesophagus and Oesophageal Cancer develop. The generated data is analysed using big data analytics methods (such as machine learning) to monitor gene expression, protein expression and gene mutations at each stage of cancer development. By combining the clinical outcome with microscopic images of the tissues Richard and his fellow researchers are trying to identify crucial diagnostic features that relate to molecular findings in their sequencing data (MODI-OC). Ultimately the aim of this work is to enable clinicians to identify at an earlier stage which patients will and will not progress to cancer, sparing those without diseases unnecessary screening, and detecting and treating those with disease earlier. Additionally, in patients who progressed to Oesophageal Cancer, similar scientific methods are used to determine if it is possible to identify which patients may benefit from novel cancer treatments, such as immunotherapy, and how this can be used alongside more traditional cancer therapies such as radiation therapy and surgery (LUD2015-005).

 

Within Oxford Richard is collaborating with researchers from the Ludwig Institute for Cancer Research, Oxford Branch (including Professor Xin Lu and Professor Jens Rittscher), the Department of Oncology, University of Oxford (including Professor Mark Middleton), The Wellcome Centre for Human Genetics (including Professor Simon Leedham)and the Target Discovery Institute, (hosted within the  Nuffield Department of Medicine) (including Dr Roman Fischer).
Further collaborators are Dr Claire Palles (Institute of Cancer and Genomic Sciences, University of Birmingham) and Dr Andrew Roth (Department of Computer Science, University of British Columbia).

 

The research is funded by the Ludwig Institute for Cancer Research, CRUK, the Oxford Biomedical Research Centre, the Oxford Health Services Research Committee, and the Oxford University Clinical Academic Graduate School.

Using AI to detect Oesophageal Cancer earlier

Professor Barbara Braden is a Consultant Gastroenterologist and Interventional Endoscopist working at the Translational Gastroenterology Unit (hosted within the Nuffield Department of Medicine) at John Radcliffe Hospital (Oxford University Hospitals NHS Foundation Trust (OUHFT) ), before which she trained at Universities in Cologne and Frankfurt. Barbara is part of a large multi-disciplinary team who are researching better ways of treating and detecting patients with early cancer and pre-cancerous conditions in the oesophagus, using advanced endoscopic imaging methods and endoscopic resection techniques.

Most Oesophageal Cancers are detected by endoscopy when they have reached an advanced stage and endoscopic, surgical, radiotherapy and chemotherapy treatment are less effective and patient prognosis is poor. Detecting early cancer on the other hand, offers a significantly higher chance of cure, as the tumour can be easily removed during an endoscopic examination. Using methods such as radiofrequency ablation to treat pre-cancerous stages can prevent from progressing to cancer.

Unfortunately, during conventional endoscopy the more easily treated pre-cancerous changes and early stage cancers are harder to observe and often missed, especially by less experienced endoscopists. If the endoscopic detection of early neoplastic changes and early Oesophageal Cancers were improved, it would be possible to intervene earlier and dramatically increase the chance of a curative outcome for the patient. Barbara and her colleagues work on a real-time computer algorithm-aided analysis of endoscopic images and videos, which will enable earlier identification of neoplastic changes automatically during endoscopy. The hope is that advanced imaging techniques and computer algorithms will be able to highlight neoplastic changes in the oesophageal lining during the examination, to guide the endoscopist to the best area from which to take biopsies and focus their attention. This new approach could also provide a more standardised high-quality examination that is less dependent on the experience of the endoscopist. With early detection of pre-cancerous changes and early stage cancer, more patients will be offered curative treatment by endoscopic resection, which will reduce the number of people progressing to advanced and potentially incurable cancer.

 

The image above is from an endoscopic oesophageal examination of an individual with Barrett’s Oesophagus, a pre-cancerous condition associated with acid reflux disease.

Barbara is part of a multidisciplinary team of scientists across the UK, trying to bring this technology into the clinic.
In Oxford this team draws from a group of clinicians at the TGU (including Dr. Adam Bailey, Prof. Simon Leedham, and Dr. James East), molecular biologists from the Ludwig Institute for Cancer Research, Oxford Branch (including Prof. Xin Lu), informaticians from the Weatherall Institute for Molecular Medicine (Stephen Taylor) and computer vision expertise from the Big Data Institute (Prof. Jens Rittscher, Dr. Sharib Ali and Dr. Felix Zhou).
Both Prof. Xiaohong Gao (Department of Computer Science at Middlesex University) and Dr. Wei Pang (Department of Computer Science at Aberdeen University) bring additional expertise in computer science to the team.
Barbara’s research is funded by the CRUK Oxford Centre Development Fund and the NIHR Oxford Biomedical Research Centre.

February is Oesophageal Cancer Awareness Month

 

In recognition of February being Oesophageal Cancer Awareness month, the Cancer Research UK Oxford Centre will be posting a series of blog posts on its website highlighting the contribution of Oxford researchers to global efforts aiming to tackle this disease.

In 2014 Cancer Research UK highlighted Oesophageal Cancer as one of their cancers of unmet need and made it one of their funding priorities. Since then, the Centre leadership has made a concerted effort to harness Oxford’s fundamental and translational research expertise to develop novel ways of detecting, diagnosing and treating patients with this disease.

This series of articles will summarises some of the local and national projects our researchers are leading on and contributing to, including:

 

Prof. Mark Middleton – Co-director of the Cancer Research UK Oxford Centre & Head of the Department of Oncology