Availability and Quality of Human
Biospecimens for Use in Biomedical Research

Elisa Eiseman, Ph.D.

Senior Natural Scientist
RAND Infrastructure,
Safety, and Environment

High-quality human biospecimens annotated with essential clinical data and collected with robust informed consent are essential if we are to ultimately realize the promise of genomics and proteomics for preventing and curing diseases such as cancer, HIV/AIDS, and heart disease. Although human biospecimens and associated demographic and clinical data are collected and stored for research purposes at hundreds of biorepositories throughout the United States, researchers are often unaware of the wealth of resources available to them because of the lack of a single guide that provides information about the availability of biospecimens. In addition, techniques for tissue collection, processing, and storage – the core functions of a biorepository – vary depending on the purpose for which the biorepository was established. Likewise, the quality and extent of information collected with the biospecimens vary depending on the purpose for which the tissue was originally collected. Therefore, it is not enough for biorepositories to have extensive biospecimen collections; information about biospecimen availability must be readily accessible to researchers, and the quality of biospecimens must be optimized.

Human biospecimens have been collected and stored at institutions in the United States for more than 100 years. A 1999 RAND study conservatively estimated that there were more than 307 million human biospecimens from more than 178 million cases stored in the United States, accumulating at a rate of more than 20 million specimens per year. Biospecimens are stored at military facilities, the National Institutes of Health (NIH) and its sponsored facilities, other federal agencies, state collection agencies (e.g., newborn screening laboratories), diagnostic pathology and cytology laboratories, university- and hospital-based research laboratories, commercial enterprises, and non-profit organizations. Collections of biospecimens vary considerably, ranging from formal repositories to the informal storage of blood or tissue specimens in a researcher’s freezer, and range in size from fewer than 200 to more than 92 million specimens.

These valuable biospecimens allow researchers to study the molecular characteristics of disease by providing information about the physiologic or pathologic condition of the person from whom they were derived. Before a biospecimen is removed from a person and becomes a biospecimen, it first exists in situ within a specific biologic context, which is reflected in its molecular profile (i.e., the pattern of expression of genes and proteins). At any point during the acquisition, processing, storage, and distribution of a biospecimen, environmental and biological variables (i.e., preanalytical variables) may be introduced that can alter the molecular profile of the biospecimen. For example, medical and/or surgical procedures conducted during the removal of the biospecimen from the patient (e.g., administration of antibiotics and anesthesia, disruption of blood supply to the tissue), the way the biospecimen is processed (e.g., time at room temperature, method and rate of freezing), or the way it is stored (e.g., storage temperature, duration of storage) may have an effect on the molecular profile of the biospecimen.

Once it is removed from a person, a biospecimen reflects the state of the tissue at the time it was collected – i.e., the expression pattern of genes and proteins will depend on both the biological state of the tissue (e.g., whether it is lung or colon tissue; whether it is diseased or normal tissue) and the environmental and biological stresses the tissue experiences during the processes of acquisition, processing, storage, and distribution (i.e., preanalytical variables). Since the value of a biospecimen to a researcher is the information it contains about the actual biological state of the tissue as it exists in the person from whom it was derived, determining which changes are disease-related and which are artifacts is of utmost importance.

Complicating matters is the fact that the way one biorepository collects, processes, stores, and distributes its specimens may be very different from the way another biorepository performs these operations. These differences occur primarily because there is no standardization or harmonization among biorepositories. The repositories that exist today were each established to fulfill a specific set of objectives and the design and operations of each repository is integrally linked to those objectives. Therefore, techniques for biospecimen collection, processing, and storage vary depending on the purpose of the biorepository. For example, pathology laboratories that collect tissue from patients undergoing surgical or diagnostic procedures routinely store specimens in paraffin blocks, while some biospecimen collections established as part of research protocols are composed entirely of snap-frozen specimens. In addition, biospecimens can be processed and stored in other formats, such as thin sections mounted on slides, cell cultures, or extracted DNA or RNA, depending on their intended use. Similarly, the quality and extent of information collected with the specimens can vary depending on the purpose for which the tissue was originally collected. Some repositories collect only pathologic characterizations of collected tissues; others collect extensive family histories, medical histories, and longitudinal data. The type of informed consent – whether general surgical consent or specific informed consent for the use of tissue for research purposes – also varies from repository to repository. Therefore, biospecimens currently stored at biorepositories may be of limited use for certain types of genomics- and proteomics-based research due to the method in which they were collected, processed and/or stored (e.g., paraffin embedded instead of snap frozen), a lack of sufficient clinical data, and/or the type of informed consent.

In addition, the lack of nationally agreed-upon quality control and standard operating procedures (SOPs) for the collection, processing, storage, and distribution of biospecimens may limit the usefulness of existing collections for research requiring highly standardized specimen collection and preparation, and may complicate comparisons of research results obtained using biospecimens from different biorepositories. Some guidelines and best practices have been issued, including those from the National Cancer Institute (NCI), the International Society for Biological and Environmental Repositories (ISBER), and RAND. However, adherence to these guidelines and best practices is on a voluntary basis.

RAND Infrastructure, Safety and Environment is currently conducting two studies on the availability and quality of human biospecimens used for research purposes. The first project, funded by a contract from the NIH Office of Rare Diseases (ORD), is being conducted to facilitate research leading to treatments and cures for rare diseases, as well as more common diseases, by improving researchers’ access to needed human biospecimens. Other than the 1999 RAND report on biorepositories, there is no comprehensive source for researchers to use to identify the resources they need to conduct their research. RAND and ORD are creating the first ever publicly available, searchable, Web-based database that contains detailed information about biorepositories that collect, store, and distribute human biospecimens in the United States for research use. The database should be available on the Internet by the end of 2006. This project will also help to identify unmet needs of researchers in obtaining human biospecimens for research, and provide information to researchers, the wider research community and other interested parties about biorepositories and biospecimens.

The purpose of the second study, which is being conducted under contract with the National Cancer Institute (NCI) Office of Biorepositories and Biospecimen Research (OBBR), is to maximize the quality and utility of human biospecimens for cancer research by identifying and analyzing existing data on how biospecimens are affected by preanalytical variables. Specifically, this project will focus on the effects of preanalytical variables on biospecimens used to study genomic and proteomic changes in cancer. The information generated by this project will provide OBBR with insight into the molecular impacts of different preanalytical variables on different biospecimen types, research questions, and analysis methods. OBBR can use this information to develop detailed, data-driven best practices that will follow from its “First-Generation Guidelines for NCI-Supported Biorepositories.” In addition, RAND and OBBR are creating a publicly available, searchable Web site that scientists, pathologists, repositories, and others can visit to learn about data, methods, and other relevant information on how biospecimens are affected by preanalytical variables. The Web site should be available by the middle of 2007.

The sequencing of the human genome, advances in genomic and proteomic research, and a focus on pharmacogenomics (i.e., personalized medicine) have placed a renewed emphasis on the need for biospecimens to advance our understanding of the genetic basis of many diseases. To speed the translation of research findings from the bench to the bedside, it is necessary for researchers to have ready access to high quality biospecimens accompanied by detailed medical information and collected with robust informed consent. RAND’s research to create a Web-based, searchable database of human biospecimens available for research use and to determine the effects of preanalytical variables on biospecimen quality will support the development of guidelines and evidence-based standards for biospecimens and biorepositories that will optimize the accessibility and quality of biospecimens for the biomedical research community.