Will Biobanking Change the World?
Experiencing explosive growth, these highly complex data systems have evolved to store extremely diverse metadata and identify subtle associations

As mentioned in a recent “Hot Topics” discussion, Biobanking is one of those amorphous topics that can mean many things, depending on your background. Not only has its meaning changed over time, it also can be different depending upon the field in which you are working. As such, whenever you are discussing it, you’ll find it prudent to make sure that everyone in the discussion is using the same glossary; otherwise you might well find that biobanking is also a synonym for Tower of Babel! Beyond this, the evolving nature of biobanks has raised a number of ethical questions that it would be wise to consider up front, before you paint yourself into a liability corner from which you can’t get out.

In origin, biobanks, or life banks, arose from the collection of biological samples for reference purposes. While, in its broadest sense, this could incorporate any collection of plant or animal samples, in philosophy, this would probably relate more toward collections aimed at preserving living organisms, such as seed collections and similar operations. Repositories of this type also might be referred to as biorepositories or, depending on the exact type of specimens, tissue banks. With the continuing advancement of the life sciences, these collections tended to become more and more specialized. While biobanks devoted to research on various types of cancer have a high visibility, there also has been an almost explosive growth in biobanks specializing in research on autism, schizophrenia, heart disease, diabetes, and many other diseases.

Additionally, the complexity of the systems used to track these samples and the amount of associated information about the samples has grown significantly. Where, initially, you might be dealing with cabinets of index cards or log books, you are now working with highly complex data systems capable of storing extremely diverse metadata about the samples and identifying subtle associations between them. While considerable research is being performed on how to store these samples at room temperature, due to the perishability of moist tissue samples, most of the samples are currently stored frozen, either in freezers or in liquid nitrogen. The implication of this is that the associated storage facility also is equipped with some type of back-up power system and reserve supplies of liquid nitrogen, along with frequent inspection to ensure the integrity of the whole system.

There has been a growing tendency to use the term biobank specifically to refer to a repository of human samples. Many of these biobanks target medical research and contain information not only on the sample, but on the patient and their medical and genetic history as well. While biobanks are generally associated with currently active research projects, one of the objectives frequently stated in their justification is to provide samples for future projects as well.

As an example, part of a sample taken for current analysis might be stored for future reanalysis when analytical techniques have improved or it’s discovered that there might be a correlating factor that was not analyzed for initially. They also can be used as a ready source of specimens for research projects not yet defined. This nexus of information concerning a patient and their status is becoming a powerful tool in medical research and in producing highly personalized medical therapies. Health Law Review suggests that the best way to understand biobanks is to look at them ‘as research platforms rather than discrete research projects’.1

Software
While some of the software packages used to track biobank samples evolved from other sources, many of the current systems use software based on various LIMS vendors’ system offerings. However, the basic requirements for biobank software are not the same as those for a general LIMS. A LIMS tends to focus on the tracking of samples, their handling, and their generated test results through the lab. Biobank systems, on the other hand, have an enhanced emphasis on the detailed tracking of the physical samples through the lab, including its origin, how it got to the lab, how it was handled in the lab (including how it was stored at every stage in the lab, detailed records of every aliquot taken, and an exhaustive chain of custody for the sample and all of it’s aliquots throughout the lab).

In addition, biobank software also must be able to support the capture of extensive field data information regarding the samples, as well as supporting flexible database queries to allow retrieving samples that meet the studies criteria. It is necessary for both systems to support detailed audit trails and secure electronic signatures.

There are additional data capture requirements not found in the traditional LIMS as well. One of these requirements is for tracking the transportation regulatory requirements for the sample. The hazard classification of the sample may well be different depending on the reason for the analysis and the type of analysis to be performed. It is important for the biobank software to be able to track this information, as the shipping restrictions for hazardous samples, particularly by air, can be quite stringent. Beyond this, the requirements for shipping samples and the associated restrictions also can vary by country, and any software used for large multi-country projects must be able to flag and handle all of these different requirements. Failure to obtain the proper shipping permits and to follow the appropriate local regulations, whether by the biobank operators or the sample collectors can, at the very least, result in significant fines. A good introduction to the transportation of samples can be found on Clinical Trials Online.2 A good listing of some of the best practices, guidelines, and standards relating to biorepositories can be found on the ISBER Web site.3

Regulation and standardization
Unfortunately, there are numerous LIMS instances that meet regulatory and good practices requirements for LIMS only that only meet their own requirements spottily at best. It is not unheard of to find poorly [JRJ1]implemented audit trails with major gaps in the information they capture, and weak electronic signatures, even among some of the more prominent vendors in the field. As such, particularly with the heavy way that biobanks are becoming regulated, it is exceedingly important to ensure that due diligence has been performed in terms of software selection and validation.

Another major point of interest in the biobanking field is how to quantify many of the pieces of metadata acquired. With no set standards in place, the metadata collected with different samples is frequently either inconsistent or difficult to normalize. This type of data frequently includes demographic, dietary, occupational and lifestyle data, in addition to more straight forward items such as date of birth or age. Recommendations have been made to launch efforts to ‘develop a standardized vocabulary for all data types.’4

Ethics
As one might expect, considering the subject matter, one of the biggest concerns resulting from the proliferation of biobanks is the ethical considerations involved. These concerns tend to fall into two broad classes. The first of these is who owns the samples and the second is privacy of the donor.

The former concern is particularly true in instances where medical research is being performed and the possibility that part of a patient’s gene structure might be patented. Some countries have passed laws stating that the samples belong to the person who donated them, particularly if they did so as a patient, other countries laws make them the property of the biobank and the donor waves all rights to them.
The later issue can be viewed in many ways. Some people have raised concerns as to what the effect on their insurance rates might be if they were to be identified as having a particular genetic marker. No matter what the concern, the issue of how to guarantee this privacy has become quite contentious. Proposals range from totally unlinking the samples from any identifying donor information to displaying full detailed patient information.  In between, these extremes many opt for distributing a sample with only a randomized identification number. As the whole point of having a linked sample is to be able to retrieve the identifying information on the donor, there is some appeal to the total disassociation approach, but this also severely limits the usefulness of the database, as there is no way to obtain any additional information from the donor.

Even worse, for samples used for medical research, there is no way to follow up on a patient, say if a therapy for a researched condition was found or to perform a long term study on a selected group of donors. This too has become an ethical concern for some researchers, as the policy for some biobanks, such as the UK Biobank is that, if a condition is identified upon sample intake, the donor will be notified, but if a marker is later identified as a result of continuing research, no such risk notification will take place. Additionally, if a donor were to later reconsider their decision and withdraw their participation in the biobank, the lack of an identifier also would make it impossible for them to withdraw their sample.5 While there are various ways to handle these issues, unless the design opts for totally unlinked data, an ideal system needs to be able to handle this partitioning of the patients meta data securely.

Even without knowing that biobanks made Time magazine’s 10 Ideas Changing the World Right Now list6, it’s easy to understand why they are currently a hot, and complex, topic. However, you probably didn’t need to read all of the above to have figured out how hot they are. After all, they have their own following on Twitter,7 which pretty much says it all!

References
1. Caulfield, T. & Weijer, C. Minimal risk and large-scale biobank and cohort research. – Free Online Library. Health Law Review (2009).www.thefreelibrary.com/Minimal+risk+and+large-scale+biobank+and+cohort+research-a0202134289
2. O’Flaherty, M. Transporting Clinical Trial Samples — Applied Clinical Trials. Clinical Trials Online (2002). appliedclinicaltrialsonline.findpharma.com/appliedclinicaltrials/article/articleDetail.jsp?id=87223
3. International Society for Biological and Environmental Repositories. www.isber.org/index.html
4. Sample and Data Collection and Analysis — Applications of Toxicogenomic Technologies to Predictive Toxicology and Risk Assessment — NCBI Bookshelf. www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=nap12037&part=a20013779ddd00180
5. Knoppers, B.M. & Kharaboyan, L. Deconstructing Biobank Communication of Results. scripted 6, 677-684 (2009).
6. Park, A. Biobanks – 10 Ideas Changing the World Right Now – TIME. (2009). www.time.com/time/specials/packages/article/0,28804,1884779_1884782_1884766,00.html
7. Twitter / People who follow BioBankCentral. twitter.com/BioBankCentral/followers

John Joyce is the LIMS manager for Virginia’s State Division of Consolidated Laboratory Services. He may be contacted at editor@ScientificComputing.com.

Related Resources
“Deconstructing” Biobank Communication of Results	www.law.ed.ac.uk/ahrc/script-ed/vol6-3/knoppers.pdf
A booming banking sector	www.scientific-computing.com/features/feature.php?feature_id=270
Biobank (Wikipedia)	en.wikipedia.org/wiki/Biobank
Biobank Central (Twitter)	twitter.com/biobankcentral
Biobank Definitions (Privileged)	www.privileged.group.shef.ac.uk/projstages/stage1/introduction/biobank-defs
Biobanking and Biomolecular Resources Research Infrastructure (BBMRI)	bbmri.eu
Biobanks – Integration of Human Information to Improve Health	www.vr.se/download/18.235f40c212384f2ca668000177/Biobanks_2008_10.pdf
Biobanks around the World:Current trends and context	www.genomecanada.ca/medias/PDF/en/gps_speakers_presentation/mylene-deschenes.pdf
Biobanks for Medical Research Opinion of the Bioethics Commission at the Federal Chancellery	www.bundeskanzleramt.at/DocView.axd?CobId=25510
Bioethics of Biobank	www.twbiobank.org.tw/conference/conference%20PDF/20070814%20%20PM-ELSI/Ryuichi%20IDA%20%20Bioethics%20of%20Biobank%28Taiwan%29.pdf
Biosample storage (Scientific Computing World)	www.scientific-computing.com/features/feature.php?feature_id=195
Centre for Research Ethics & Bioethics	www.crb.uu.se/research/biobanks/index.html
FasterCures/The Center for Accelerating Medical Solutions	www.fastercures.org
International Society for Biological and Environmental Repositories (ISBER)	www.isber.org
Leading Biobanking Projects & Institutions (Clinfowiki)	www.clinfowiki.org/wiki/index.php/Leading_Biobanking_Projects_%26_Institutions
Legal Pathways (WIKI Legal Platform of BBMRI)	www.legalpathways.eu
OBiBa – Open Source Software for Biobanks (Wikipedia)	en.wikipedia.org/wiki/OBiBa_-_Open_Source_Software_for_Biobanks
Organisation for Economic Co-Operation and Development (OECD) Best Practice Guidelines for Biological Resource Centres	www.oecd.org/dataoecd/7/13/38777417.pdf
Registry of Biological Repositories	www.biorepositories.org
The Future of Biobanks Regulation, ethics, investment and the humanization of drug discovery	www.globalbusinessinsights.com/content/rbdd0026t.pdf
The Genographic Project	https://www3.nationalgeographic.com/genographic/index.html
The Informatics of Biobanking and Genome-Phenome Correlation using Phenotypes Derived from Electronic Medical Records	events.iupui.edu/event/?event_id=1182
The Personal Genome	thepersonalgenome.com
UK Biobank Ethics and Governance Council	www.egcukbiobank.org.uk