- •Contact us
- •About us
- •Advertise with the FT
- •Terms & Conditions
© The Financial Times Ltd 2012 FT and 'Financial Times' are trademarks of The Financial Times Ltd.
Key words :
Game changing data analysis for the treatment of cancer
21 Jul, 2013 10:30 am
The cancer registration service at Public Health England is creating a highly detailed database of cancer diagnosis and treatment, right from initial symptoms or detection through screening, and through a patient's life and care. Highly detailed analysis and intelligently developed algorithms will entirely change how we treat cancer.
Over the last few years, we have migrated the historical cancer registration systems and vastly improved data collection to create a single database and National Cancer Registration Service. We can now track the complete cancer care pathway, from primary care, to treatments, recovery or palliative care and death. This shows us the entire diagnosis, treatment and disease life-cycle and we can link this to molecular and genomic analysis of the patient and their cancer.
Revolutionising data analysis
Advances in our understanding of cancer biology have shown that it is absolutely crucial that we make sure we are treating each person's cancer in exactly the right way. At the moment we give chemotherapy to patients based on the type of cancer we see down the microscope and how advanced their disease is, but we know that not all patients will respond; what we do know is that in many cases this reflects molecular differences in their own specific cancer. To understand this properly we need to study a very large number of cases and know exactly how each cancer has responded in an individual. Armed with this information clinicians will be able to decide on fine-grain targeted therapies.
"Cancer" embodies many different types of disease, but this personalised data analysis will also be needed for many different illnesses such as cardiovascular disease, diabetes or even dementia.
We have modernised registration over the last five years, in order to enable the data migration to a single system. The database is sizeable, with 350,000 new tumours a year, and 11 million historical cancer registration records.
Each cancer treatment centre already has extensive data on patients and their treatment, and how they respond. The key for us is a liberating this data so that it is available and can be compared nationally. We have been successful because we have kept the process very simple and worked with each provider organisation and local team to make it as easy as possible for them to provide their data on their patients.
The system we have developed is designed to handle data in a wide variety of different formats and we use our 200 expert cancer registration officers to quality assure the data and to reconcile any differences in how the data has been coded so that we can achieve national consistency. We feed back data to the individual trusts so that they can check the accuracy.
The National Cancer Registration Service system, Encore, is based on open source software, using web application framework Ruby on Rails. We have used an Agile project management framework and have managed to deliver effective and quick development without the constraints of a rigid long-term project plan.
Our priority is to ensure that data is secure and personal confidential clinical information on a particular patient is only accessible to their treating clinician. The data available nationally does not name patients, but instead it is tracked according to each cancer and shows how each illness responds.
Behind-the-scenes, we are developing algorithms on how patients best respond to treatment. These will support clinicians in making even better informed decisions on treatment. We already have one site, called predict.nhs.uk, which looks at the best methods of treatment for breast cancer, relying on highly informed predictive analytics.
The next steps will be to give patients access to the system, so that they can update their details and responses to treatment.
Additionally, we want to provide dataset links to genomic medicine: there is an obvious partnership to be had between clinical data and this information. We also want to closely inform clinical research, so that we can really beat cancer using the most intelligent methods.
Meet your experts