Banking on it

Despite growing interest, biobank use still faces a number of hurdles – from IT to governance. Will we ever fully realise the full potential of these hugely important resources? NGP asked Brock Reeve.

The collection of molecular and clinical data in what have come to be known as ‘biobanks’, and the integration of this data, are finding widening recognition as critical in our understanding of the underlying causes of disease, and subsequently in our ability to find effective treatments in the future.

However, despite a growing interest in biobanks as a valuable resource in translational research, their development still faces a number of hurdles – from the IT required to integrate the complex data stored in these biorepositories, through issues relating to governance to the medical and scientific capabilities – that must be broken down if we are to fully realise the full potential of these hugely important resources.

Personalised medicine

“The main interest in biobanks is being driven by their potential in enabling personalised medicine,” says Brock Reeve, COO, Life Science Insights. “The important role that biobanks play is in providing a store of biospecimens and associated data (genomic, clinical and phenotypic data, etc.) – information that is critical throughout the pharmaceutical chain, from drug development through clinical trials and care delivery.

“As a result of the huge potential of biobanks,” Reeve explains, “it is not only drug companies that are showing a keen interest, but also national health systems.” He cites the example of the UK Biobank project, which is conducting population level studies in an effort to understand not only how individual people will respond to certain drugs or treatments, but also the likelihood of individuals contracting various diseases, the impact of the environment on their health, impact of genetic heredity, etc.

Other examples of interest at the government level include the Estonian Genome project, and a number of biobanks in Sweden, which were established decades ago now. “What is new, however,” says Reeve, “is the ability today to carry out genetic analysis on these samples, analysing the genetic components in order to determine why certain people have certain diseases and what impact factors such as environment and heredity have on this.”

Public and private development

Government institutions have also contributed to efforts – such as the National Cancer Institute in the US – while some non-profit organisations, such as patient advocacy groups and disease foundations have started up their own biobanks. “There are also biobanks operating at a private level,” points out Reeve. “For example, pharmaceutical and biotech companies are gathering specimens during the course of their clinical trials, along with the associated data. This helps them to develop their understanding of not only how patients are responding to a drug throughout the course of the trial, but is also often used in research, for early stage drug discovery target identification.”

With so many disparate biobanks being developed concurrently, it would seem there are clear benefits of pooling or sharing the data held. Reeve points out that there are, and will continue to be, a variety of different biobanks to serve different purposes, between which such sharing of information would be less relevant. He agrees, however, on the potential of connecting up networks of biobanks, and points to recent positive signs within the industry towards such a move.

“This is certainly an interesting issue,” says Reeve, “and one that was discussed quite recently at one of several worldwide biobank summits to be organised by IBM.” This networking of biobanks is not a viable proposition today, however. As Reeve explains: “In order to achieve this, there needs to be much more standardisation of data and protocols so that the information we exchange makes sense. For example, there is currently a great deal of variation in how we classify a certain disease, how specimens are collected and stored, and how we maintain records so that the scientific and IT data analysis is consistent. This is quite a major challenge right now.”

On a more positive note, there is, it seems, some movement towards this: for example, discussions at a biobank summit six looked into the type of standards that can be used for annotating the specimens at a clinical level. As Reeve explains, there are many languages currently used, but no set standards on which are the right ones. “Standards are also required on the IT side to determine how we can exchange data and what are the right protocols to follow, adds Reeve. “Standards are going to be key on both the clinical and IT side in order to enable the necessary intercommunication to happen.”

IT: the critical enabler

IT is a critical enabler in making these biorepositories possible, because the amount of data that comes with a large number of specimens is vast – information not only about the sample itself, but potentially genotypic and phenotypic data as well as data on clinical outcomes. However, despite its key role, the IT side of the equation is far from solved, and a number of issues still exist with data storage, analysis and interconnection.

“Even on a more local basis, within the sample collection storage area, there are issues surrounding keeping track of samples in the specialised laboratory systems,” comments Reeve. “Around that, you also have two huge issues – privacy and security. The information must be kept secure and, because of the genetic analysis that is now possible, people are very concerned about the release and availability of their genetic information.”

In response to these concerns, biobanks have ethics policies and privacy guidelines in place to ensure no barriers are crossed concerning the use of samples and sample data. “The challenge faced by the industry is how to de-identify that data, so that when the researcher studies a sample, he or she cannot know who contributed it. The sample and the donor must be de-linked so that diseases can’t be traced back to the individual.” This is a particular concern in the US because of employment or health insurance issues and is, therefore, one of the drivers in the move to international harmonisations – so that the same guidelines and standards are in place on privacy, security and management of data.

A work in progress

Aside from the IT and governance issues, that must still be addressed, it is also worth remembering that however fast it is advancing, the science behind biobanking is a work in progress. “Even though certain tissue samples have been collected for decades, biobanking is at a very early stage; the field of genomics and genetic analysis didn’t exist a few years ago,” says Reeve. “On the scientific front, we’re going to understand a lot more about genetic predisposition to certain disease and how people respond to certain drugs. A major area of scientific research these days is biomarkers, and the biospecimens held in the biobanks are going to be key to understanding and developing those biomarkers. On the scientific front, many people see this as a goldmine.”

One of the main challenges today is coming to a consensus about how samples are gathered and ensuring that they are gained with the proper consent. This is important because the way that samples are stored and collected can have a serious impact on the kind of data you can extract from that sample. It’s vital to ensure that whatever analyses you’re running on the sample are correct. As Reeve adds, the importance of this will become ever more important when we look a few years down the line and consider the kind of analysis that may be possible. “We will find out more about the connection between certain genes, proteins, etc., and as the new scientific and analytic results come out, one of the questions will be how do you apply that to samples gathered over time. The challenge will be how do you gather a sample now for the future, when you don’t know what kind of questions you’ll be asking of that sample or what scientific tools you’ll be using.”

Reeve sees three main areas that need to be addressed in order to move development of biobanks forward: a core organisation and governance issue, in which there are the issues of privacy and security; the scientific aspect; and IT and data standards. “All three of these need to develop in tandem for this to be successful, says Reeve, but is confident that progress is being made on all fronts. “While each will have its own types of hurdles and breakthroughs, the three pieces of the puzzle do have to move along at the same rate.” He adds, however, that one of the big issues on the organisational side is simply getting public acceptance – ensuring that patients trust that any sample given to a third party will be used in the right way and not be used against them.

“Where the three elements move together is through certain programs such as those by the NCI or with Encore UK,” concludes Reeve “or when you have a big national push. There are then also some groups that try to cut across the different areas, trying to connect together some of the national biobanks and national genomics efforts so that they can leverage each other and achieve some of the incredible benefits that are possible in the future.”