Personalised medicine? The keys in the bank

Biobanks are likely to be vital in the era of personalised medicine, but the sheer scale of the data involved presents many problems. The key to biobanking success, says Ben Jackson, is coding

HEALTHCARE’S future, we are told, lies in personalised medicine - i.e. the individualised targeting of diagnostics and therapies, rather than the ‘broad brush’ approach that the established model provides. However, our future ability to get the right treatment to the right person at the right time will draw heavily on resources from the present and the past.

Current advancements in genetic and molecular knowledge are helping to drive forward the concept of personalised medicine, but these must be combined with existing information from patient records if the necessary level of clinical understanding is to be achieved. This is where the role of biobanks comes in. By obtaining genotypic and phenotypic data from stored samples of human tissue, body fluids or DNA and combining it with clinical information, biobanks are helping to advance researchers’ understanding of disease at molecular level. And so the move towards personalised medicine can begin.

Mass storage of medical data and material will allow researchers of the future to study in depth how the complex interplay of genes, lifestyle and environment affect our risk of disease. This will enable them to identify even more reliably why some people develop a specific disease and others do not.

The UK Biobank project, aims to shed light on many of the most common life threatening and debilitating diseases such as cancer, heart disease and diabetes, whilst also illuminating many other ‘lower profile’ conditions such as mental illness, Parkinson’s disease, Alzheimer’s disease and motor neuron disease. The project will involve 500,000 individuals currently aged 40 – 69 and from all walks of life, enabling researchers to study many different risk factors together. Backed by Cancer Research UK, the British Heart Foundation, the Juvenile Diabetes Research Foundation and the Parkinson’s Disease Society with the support of many leading scientists, universities and the NHS, the project has already recruited a broad cross-section of participants from Oxford and Manchester. By providing information about their health and lifestyles today and then allowing UK Biobank researchers to follow them through their health records in the coming years, these participants will play a crucial role in the future prevention and treatment of a wide range of conditions. UK Biobank has been set up and is funded by the Wellcome Trust, Medical Research Council, Northwest Regional Development Agency, Department of Health and the Scottish Executive. It is being run as a non-profit charity with initial funding of around £62 million.

From a laboratory point of view, the UK Biobank project is a massive technological undertaking involving specially designed automated systems for the storage, retrieval and tracking of around 10 million biological samples which must be stored in liquid nitrogen environments at between -80ËšC and -220ËšC for several decades. Once the project is running nationwide, samples are expected to be taken from around 1,000 people every day.

The success of such projects will depend upon the integration of molecular and clinical information streams being accomplished in a robust and secure manner. When setting up a biobank, the establishment of effective sample and data management systems is clearly of paramount importance. Longevity of sample storage is just one of the key factors to be considered, along with finding reliable ways of tracking, selecting and retrieving samples which may need to be accessed repeatedly over a period of many years.

Both the UK Biobank and the National Cancer Biobank currently being established by onCore UK have sourced sample management solutions from Thermo Fisher Scientific. Thermo Scientific SmartScan technology has been adopted for the storage of the vast numbers of biological samples being amassed for these two projects, and this now forms an essential part of the biobanks’ overall data architecture.

The SmartScan solution centres around use of special sample tubes which encapsulate a unique 2D coding system for identification purposes. Codes are laser-etched directly on to a specific paper and protected on the  base of each tube behind an optically clear polypropylene window, and can be read either by using an automated high-speed scanner or, alternatively, by eye. The encapsulated codes are resistant to vapour phase liquid nitrogen and the freeze-thaw process, with the polypropylene window providing highly effective protection against abrasion and chemical solvents (including DMSO).  This approach ensures the integrity of sample identification by eliminating the need for labels which can be damaged or peel off.

The 2D coding system was specially developed for the purposes of individual sample identification and tracking, and is designed to hold large amounts of data in a very tiny space. Its 14 x 14 array provides high density storage for up to 3.6 quadrillion unique codes, assuring a virtually endless supply of code combinations for mass storage purposes. Fast and accurate automated reading is guaranteed via the application of Reed-Solomon error correction with data redundancy, and the unique code format ensures there is no overlap with other currently available systems. An alphanumeric code is also included to facilitate the easy identification of individual tubes by eye if the power supply to automated reading facilities should ever fail.

Use of the encapsulated 2D coding system overcomes many of the problems commonly associated with mass sample storage and identification. Other traditional methods of sample labelling such as the use of adhesive labels, marking with a pen or the application of conventional 1D and 2D barcodes are all unsuitable for long-term sample storage because of their susceptibility to damage - especially through exposure to low temperatures or chemicals. By contrast, encapsulated codes cannot fade, peel off or be accidentally erased during processing or storage. This ensures that samples are not rendered worthless through the loss of identification data, whilst also achieving compliance with international quality standards.

Another common pitfall derives from the use of sample plates, whereby specimens are frozen together in a batch and cannot be identified or accessed individually. In such circumstances, an entire plate must be defrosted in order to access a single sample, exposing multiple samples to the risk of potential damage caused by the freeze/thaw process. Storage of samples in individual coded tubes overcomes this problem, thereby enhancing the overall longevity of the entire sample base. In addition, it reduces the potential for errors in sample handling through use of ‘pick and place’ automation systems. Use of encapsulated 2D coding ensures that the ‘picked’ tube can be positively identified so there can be no doubt that the correct sample has been selected. This is easily accomplished using a dedicated reading system to double-check that the 2D code on the sample tube matches that which is held in the sample database.

Experimental history is rapidly becoming a crucial factor in the day to day handling of biobank samples. It is also highly relevant in compound storage applications whereby pharmaceutical companies track the usage of compounds being used at different sites for the purposes of drug discovery. By matching individual sample tube codes with corresponding information held on a central database, users can easily keep track of sample movements, how and when samples have been used - and, in the case of biobank samples - create individual ‘pick lists’ based on patient demographics.

Automated readers for SmartScan products are also available. A dedicated rack reader, SmartScan 96, offers speed and efficiency for tube decoding, utilising dual CCD cameras to decode a 96-place SBS footprint rack in less than six seconds. There is also a single tube reader known as SmartScan Solo which decodes a single tube in less than one second. Both models offer a variety of data export options, enabling sample history databases to be automatically maintained.

The sheer scale of the sample storage facilities maintained by biobanks is such that the need to minimise risks of cross-contamination between samples is of paramount importance. Sample tubes have traditionally been sealed with screw caps, but the risk of cross-contamination between tubes is an ever-present threat. For example, a trace of sample material caught up in a screw cap thread could all too easily splash out while the tube is being opened and contaminate the contents of other open tubes nearby.

To counter this risk, Thermo Scientific Twist-Lock tubes have an internal thread design which ‘locks’ in with the cap in two distinct locations to provide the highest possible sealing integrity. 

An alternative means of sealing sample tubes in biostorage applications is provided by the Thermo Scientific Multisip range of septum plugs which have been developed for high throughput liquid handling applications. They feature a patented split septum design that is pierceable with automated pipetting equipment, providing 100% sealing against the risks of contamination and evaporation even after being pierced 200 times. 

Products such as these underpin the integrity of biobanking projects which are set to become a critical resource for medical research in the 21st century. Speaking recently about the UK Biobank project, Professor Colin Blakemore, Chief Executive of the Medical Research Council has been quoted as saying: “Over the coming years the data from this study will grow into a unique resource for future generations”. His colleague, Dr Mark Walport, Director of the Wellcome Trust, said: “UK Biobank has the capacity to answer questions about health and disease that could benefit future generations”, while Nobel Prize Winner Professor Sydney Brenner FRS has summed up the benefits, saying: “By studying the relationship between genes, lifestyle and environmental factors, the UK Biobank will be the future of medical research”.

The ability to store and identify samples for use in research that is informed by clinical data undoubtedly has the potential to transform our approach to medicine - whether by revealing the genetic processes that cause particular diseases, or identifying molecular markers that may provide early warning signs. Future breakthroughs in research will spring from the secure foundations of effective sample management. The dawning of a new era of personalised medicine may not be far away.

By Ben Jackson. Ben is a Product Manager at Thermo Fisher Scientific.  He has been working in the biotechnology industry for over 7 years and has a PhD in Bacterial Genetics.