Abstract
I
Despite the clear need for high-quality biospecimens, acquiring them can be difficult. Generally, the collection, processing, and storage procedures are not standardized, and the degree and type of data annotation vary. This becomes further complicated because samples are collected for distinct purposes that may not be appropriate for other uses, collection practices vary, and tissues are collected after surgery or autopsy, where sample degradation can be an issue. Because many samples are collected for diagnostic purposes, collection of normal control samples is not routine, making comparisons difficult. There are also costs to procure these samples, and the activities surrounding collection are often not reimbursed. Consent issues, including for the future use of biospecimens, can be insurmountable as well.
Toward High-Quality, Well-Annotated Biospecimens
One important step toward obtaining high-quality biospecimens is following best practices that provide state-of-the-science guidance and harmonize procedures for collection, processing, storage, and distribution of samples and data. The International Society for Biological and Environmental Repositories has developed Best Practices for Repositories, focusing on the collection, storage, retrieval, and distribution of biological materials for research (Pitt et al., 2008). The National Cancer Institute (NCI) has developed NCI Best Practices (most recently updated in 2011) that examine the scientific evidence for collection, annotation, processing, and storage of biospecimens (Office of Biorepositories and Biospecimen Research 2011). This resource also includes best practices for ethical, legal, and policy issues.
Another import step is to collect data on how biospecimens are collected, processed, and stored. Biospecimen Reporting for Improved Study Quality (BRISQ) (Moore et al., 2011) and Standard PREanalytical Code (SPREC) (Betsou et al., 2010) are classification schemas that can be used to help capture preanalytical variation.
Finally, collecting appropriate clinical data, using data standards, and messaging when possible improves sample annotation. Systematized Nomenclature of Medicine—Clinical Terms (SNOMED-CT) is one commonly used clinical terminology.
Permission to Share
Collecting high-quality, well-annotated biospecimens is only the beginning. It is often difficult to share data and samples across projects, and certainly across institutions, for many reasons. Permission to share must also be enabled. Consent from participants is most often solicited (or not, in the case of residual clinical samples in surgeries) for a single study. If aware of the limitations, these participants are sometimes frustrated that their data and samples are not shared widely. Individuals who contribute samples expect that these samples will be used widely and to the greatest extent of their value.
By and large, researchers continue to work in silos where little specimen and data sharing exists. Each institution typically has its own program composed of individual studies with small collections that may or may not be adequately powered. Researchers are frustrated they are unable to obtain the quantity or quality of samples needed for their research. Providing permission to share allows high-quality biospecimens and data to be used by others and has the potential to advance research.
As future research protocols are developed, participants should be given the option to share their data and samples widely, as well as be re-contacted to provide additional information or samples. Fortunately, policies for data and sample access that promote sharing are being created. An excellent example of a novel informed consent process is the Portable Legal Consent (PLC), the brainchild of several attendees of the 2011 Sage Bionetworks Commons Congress. It became clear that if data are to be shared, particularly in bold commons projects such as the Sage Synapse database, then consent must be attached to the data. The legal inspiration for PLC is the informed consent process developed by the Personal Genome Project, and some of the literal text is drawn from their consent form. PLC also builds on the considerable achievement of Creative Commons to create “human readable” interfaces for complex legal documents. Genetic Alliance plans to launch a campaign to enable individuals to move clinical data into Synapse.
This effort, or any like it, will require a culture shift at the institution level, with policies and incentives that reward sharing (Kaye et al., 2012). Returning results and incidental findings to participants must also be considered (Wolf et al., 2012). By promoting a culture of sharing, we have the potential to make high-quality, well-annotated biospecimens widely available. Continuing to operate in silos will not make precision medicine and better health a reality.