A Review of International Biobanks and Networks: Success Factors and Key Benchmarks

Introduction

Biobanking involves the collection, processing, storage, and dissemination of biological samples and their associated clinical data and information, organized in a systematic way. A well-managed biobank is a critical prerequisite for high-quality biomedical research. Recent advances in the tools and technology of molecular biology and genetics have increased the demand for well-annotated, properly preserved specimens. In response to that demand, biobanks have been established on several continents within the past dozen years, and more are in development. This review outlines the approaches that 16 of the largest biobanks and biobanking networks in Europe, North America, Australia, and Asia have taken to collecting and distributing human research specimens and managing scientific initiatives while covering operating costs. This information may be of interest to groups considering, or currently establishing, biospecimen repositories. Several different operational and business approaches have been taken, with varying degrees of success in terms of long-term sustainability and meeting scientific goals. “Success” depends on many technical, ethical and legal, and financial factors. The trend toward international collaboration in biobanking and biospecimen utilization makes it important to understand these factors.

The biobanks covered in this review do not consider biospecimens to be tradable commodities; therefore, they all operate as not-for-profit entities. Further, to avoid exorbitant biospecimen fees that might inhibit research, the banks discussed do not attempt to recover their full operating costs from investigators who request samples. Rather, they depend upon government and/or charitable funds along with reduced fees for researchers to enable biospecimen collection, storage, processing, and dissemination. Most of these repositories accrue specimens from cancer patients and handle them so as to enable a variety of downstream analyses. Additionally, the biobanks reviewed tend to have similar access policies, requiring information about the experiments to be performed and the funding thereof, justification for the specimens requested, and ethical handling of the human materials. Beyond these commonalities, there is little consensus between the featured biobanks regarding whether to store samples in central or distributed locations; whether to incorporate nongovernmental donations into the economic model; whether to offer derivative samples such as purified nucleic acids in addition to tissue and blood products; and whether to require that unused biospecimens or research results be returned to the banking institution.

Biospecimen Collection Characteristics

The 16 biobanks and networks chosen for this review represent a geographically diverse set of organizations with varying approaches to collection and dissemination. Table 1 summarizes the general characteristics for the 16 biobanks in this review. With respect to the population from which biospecimens are collected, two of the repositories obtain biospecimens from the general population and one centers on patients with rare diseases. However, most of the biobanks under consideration—13 of 16—are disease-based and focus on cancer. This is a natural outcome of the fact that cancer treatment frequently entails surgical excision of the diseased tissue, and generates material upon which, under proper consent procedures, research may be conducted without physically endangering the patient. The majority of these biobanks perform general collection, that is, they procure and process biospecimens in a manner that produces samples suited to a variety of technological research platforms. We refer to these collections of tissues with no specific research projects in mind as “general” in Table 1. Two of the repositories collect biospecimens exclusively for predefined research endeavors. We refer these to as “project-driven” collections. Another repository typically procures biospecimens for general use, but may be contracted to perform project-driven collections.

Most of the biobanks under consideration include numerous sites at which samples are stored. In these cases of “federated” collections, researchers generally may query a single database encompassing all of the participating sites. Five of the 16 sites rely on a centralized physical repository. For example, the US National Cancer Institute (NCI)'s Cooperative Human Tissue Network (CHTN) is a virtual biobank that functions under a federated system on a demand and supply model. Investigators request particular samples and the NCI coordinators at the CHTN meet the request from a variety of sources. The banks are evenly divided as to whether they offer researchers access to purified nucleic acids extracted from biospecimens they have accrued; half do and half do not.

Economic Models

Each of the biobanks discussed adheres to the principle that human specimens should not be commercialized; thus, each is a nonprofit entity. The consensus among these banks is that operations require funding beyond what is gained from cost-recovery systems, that is, the fees charged to researchers to access biospecimen samples and data. Table 2 describes for each biobank the funding sources and price per sample (if available). None has implemented a cost-recovery program that fully covers all operating expenditures at this time. However, reliance on governmental and charitable funding varies, as does the degree of cost recovery. The overwhelming majority of these biobanking programs receive support from local and/or national governments. More than half—9 of 16—also depend upon private contributions. Only one does not receive any government monies to support its operations, but relies solely on charitable contributions and cost recovery. The approaches to cost recovery generally aim to defray a portion of the price of biobanking while avoiding fees so expensive as to inhibit research. These range from asking solely for researchers to pay sample transportation costs, to assessing transportation costs plus a nominal fee to defray biospecimen collection and handling, to models in which researchers compensate a greater portion of the burden the biobank incurred to collect, process, store, and transport the samples they request. Four repositories impose different cost-recovery rates for nonprofit research vs. commercial endeavors. Additional sample-processing services are offered by 4 of the 16 biobanks at a cost fully borne by the requesting researcher. It is worth noting that except for the UK Biobank, which is a population-based initiative, all of these biobanks are disease-based. Although data are not available to quantify costs over time, in general a disease-based biobank can expect to incur much higher costs during its initial development period, due to the necessity to invest large amounts of personnel effort, materials and equipment in establishing the collection. In contrast, population-based collections generally require smaller up-front investments but incur higher long-term costs due to the need for years of storage and other personnel and maintenance costs to preserve and analyze specimens until study goals have been completed.

Biospecimen-Access Policies

One of the major criticisms of biobanks is that specimens are collected and stored for lengthy periods before being used, delaying their productive use by researchers, and possibly delaying new discoveries and treatments for patients.

As summarized in Table 3, of the institutions evaluated, 14 have biospecimen-access polices or recommendations in place; 2, onCore UK and UK Biobank, are not yet disbursing biospecimens. EuroBioBank-access policies are member-bank-specific and not summarized here. Several policies were consistent throughout the 13 whose polices were evaluated: all require applicants to describe the proposed experimentation, specifying the desired biospecimens and accompanying annotation. All require that applicants have the expertise to perform the proposed experiments, which must use an established methodology; and all except Tubafrost indicate that applicants should verify that they have sufficient funding to complete the proposed studies. In addition, all the 13 institutions require requestors to provide evidence that the project has obtained ethical clearance and that applicants respect participant privacy by not attempting to identify individual bio-specimen donors. They further specify that the biospecimens provided are to be used only for the proposed purpose and are not to be redistributed to others without written consent.

Several access policy models have been developed among the evaluated institutions. While most have internal panels to review applications, 4 indicate that applications undergo external review. Three national biobanks will supply bio-specimens only to researchers in their own countries or to collaborators thereof. Three do not specify that statistical justification for the requested sample size be included with the application. Two biobanks require authorships on publications resulting from research conducted on biospecimens they have provided; 9 others request acknowledgment. Three institutions specify that they should be notified of any clinically relevant data so that patients may be informed; 6 indicate that experimental data either be deposited with the biobank or made publicly available. More than half of the repositories—9 of 13—anticipate that publications resulting from experimentation on the supplied biospecimens be lodged with them. Seven institutions request that applicants provide reports on the experiments being conducted; the timing of this varies, although the most common requests are for annual reports or a single report after completion of the study. Six of the 13 biobanks request that unused samples be returned to the bank.

Individual Biorepositories and Biobanking Networks

Discussion

The biobanks and networks reviewed in this article operate under a variety of economic and governance models, and although generally successful, some challenges are worth noting to inform new initiatives.

Issues of tissue access and custodianship are often serious challenges to the long-term sustainability of a biobank. For disease-based biobanks, much depends on the location of the biobank in relation to the diagnostic workflow. For example, biobanks can readily be established in hospital pathology departments, because the needs of diagnosis dictate that samples flow through these departments; it is the best practice that the pathologist must examine resected specimens first and take samples for diagnosis, before samples may be obtained for research. However, the establishment of a centralized collection, remote from the biospecimen procurement site, presents a more difficult problem. For such a centralized collection to be successful, clear incentives must be in place, for example, financial reimbursement, to encourage those responsible for the local hospital procurement to provide their samples to a central biobank, otherwise the work of specimen collection and annotation simply distracts hospital staff from their primary clinical and/or research responsibilities.

Financial support is one incentive to encourage contribution of biospecimens to a centralized disease-based biobank. This may include payment for staff salaries and equipment at the local collection site, which may retain some portion of the samples for local research use. Biobanking is an expensive activity requiring dedicated staff to obtain patients' consent and collect samples; the success of this incentive depends on the level of funding available. Experience has shown that the costs of establishing and maintaining a biobank or biobanking system are often not well understood at the beginning of a project. The incentive used in project-driven collections is generally research collaboration, whereby clinician-investigators at the local procurement site are actively involved in research projects using the large sample collections at the central biobank. This approach has recently been adopted by the Singapore Tissue Network, after several years of general collection.

The federated network model (see Table 1 for examples) provides a popular alternative to the centralized biobank model. In a federated network, staff members at the collection site are the custodians of the samples they collect. Membership in a federated network includes the condition that some portion of the local collection is made available to researchers at other sites. Incentives for membership in the network may include financial support and the advantage of access to collections at other sites. This model requires that perceived “ownership” rights to specimen collections are overcome for the benefit of the network.

A national biobank, for example UK Biobank, that obtains public funding requires public support, and needs to have strong leadership that is engaged in assuring the public that specimens are being collected for the common good and are being used in an ethical manner for productive scientific endeavors. In general, public confidence is ensured by the existence of ethics committees or institutional review boards that monitor sample collection and use, together with privacy laws that protect the confidentiality of patients and donors.

The biobanks outlined in this review, although they operate under a variety of models, share many of the following characteristics, which in most cases are detailed in their Web sites:

Governance models with clearly stated technical standards, ethical guidelines, access policies and procedures, scientific rationale, and long-term custodianship plans, that is, assuring that the program is sustainable from technical and economic perspectives.

As the biobanks and networks described here mature and others are established, the factors that contribute to their success will become clearer. Regular communication between biobanks helps administrators identify these success factors and stay abreast of technical advances and policy changes in the field. ISBER provides a forum for such communication through annual meetings, working groups, and listserv discussions. Recently, in recognition of the increasingly international nature of biospecimen resources, ISBER created an Asian chapter; additional international chapters are under consideration. The Public Population Project in Genomics ¹⁸ (P3G) provides another forum, with special emphasis on population biobanking. P3G maintains an extensive set of Internet reference materials (the “Observatory”) of international biobanking guidelines and other materials that facilitate international coordination and collaboration.