Oxford Cancer Biomarkers

Towards A Personalised Treatment of Cancer

Cancer is a complex, heterogeneous disease driven by multiple, often interrelated factors acting at the genetic, epigenetic, transcriptional and translated protein level. The inherent complexity of the factors driving the disease dictates that different cancers, with different driving factors will respond to different treatments and therapies.

Until recently, very few tools have been available to clinicians to allow them to interrogate and characterise different factors driving a given cancer, this has necessitated a largely empirical approach to treatment, i.e. by means of observation or experience rather than theory or pure logic.

The concept of precision oncology is centred on the principle of patient stratification. In practical terms, this means the ability to characterise clinically significant subtypes of cancer and thus tailor treatment to the individual for the best outcomes. In broad terms, it aims to advance the clinical conversation on from ‘how do we best treat cancer?’ towards ‘how do we best treat this individual patient’s cancer?’.

OCB designs, builds and deploys the next generation of precision oncology diagnostic technology which allows clinicians to stratify patent groups that traditionally would have been treated empirically, thus allowing its clients to deliver the benefits of precision oncology to their patients.

Under the Scope – How the Technology Works

As part of the normal treatment for bowel cancer, after surgery, a sample of the patient’s tumour is processed in the pathology laboratory of the hospital and examined under a microscope by a pathologist. In addition to the pathologist’s examination, OCB’s technology takes images of these same tumour samples and subjects them to machine learning based image analysis algorithms. These algorithms identify features in the tumour microenvironment indicating whether the tumour is an aggressive type and therefore more likely to recur. Patients whose tumours have these high-risk features are likely to benefit from adjuvant chemotherapy i.e. they are likely to be one of the 8%.

CSI provided OCB with a high-performance platform designed for AI deep learning in an Infrastructure as a Service (IaaS) model based in the CSI PowerCloud.

OCB’s AI initiatives are supported by an IBM Power System AC922 combining 40 POWER CPUs with four NVIDIA Tesla GPUs. NVLINK2 technology provides direct CPU-GPU communications delivering the high performance required. IBM V7000 delivers 14TB of SAN-attached storage for datasets and AI models.

The IBM Power9 server (AC922) also offers Large Model Support where AI models can be trained on images larger than the memory available on the GPU by using system memory – this allows high-definition images to be used.

Deep learning model training is compute-heavy, but OCB is now able to take advantage of the AC922’s distributed training feature which allows jobs to be broken down and processed in parallel over a cluster of GPUs.

PROJECT GOALS:

A group of cutting-edge oncology researchers wanted to take advantage of the latest artificial intelligence models to develop tools which enable a more tailored or individualised approach to cancer therapy.

---