In this model, which is operated by many pathology departments, the raw tissue samples are collected by clinicians and sent to the pathology laboratory for processing or analysis. The pathologist examines, selects, and processes the specimens into formalin-fixed, paraffin-embedded (FFPE) tissue samples and the FFPE samples are archived for future diagnosis and research. Following approval by the hospital's institutional review board (IRB), the pathologist supplies the requested human tissue samples to the researcher (Fig. 1A, Table 1).

The advantages of this model are that it does not require the use of a certified bioresource bank, and management of the pathology FFPE archive for research purposes requires no additional maintenance costs (Table 2). However, this model has many limitations. First, the pathologist can only supply researchers with FFPE tissue samples; the absence of a formal bioresource bank makes it impossible to store fresh or frozen tissue, cell lines, or high quality DNA or RNA samples, which can only be obtained from fresh tissue (Table 2). Second, the use of archived FFPE tissue samples is ethically questionable because informed consent is typically not obtained by the clinician. In most cases, use of archived FFPE tissue samples for research purposes would require the researcher to obtain the informed consent of the specific patient in compliance with the Bioethics and Safety Act.

In this model, the clinician obtains informed consent from the patient to allow storage of the tissue in the hospital's government-certified bioresource bank. The sample is then sent to the pathology laboratory, where a portion of the fresh tissue is processed into FFPE samples and archived, and another portion is sent to the bioresource bank by the pathologist (Fig. 1B, Table 1). A technician at the bioresource bank extracts DNA, RNA, and protein, or cultivates cell lines from the fresh tissue sample. In this model, the use of specimens stored in the hospital's bioresource bank for research purposes does not require IRB approval; therefore, researchers can easily obtain high quality DNA, RNA, and protein samples, as well as isolated cell lines (Table 2). However, as in model 1, if a researcher requires access to the FFPE tissue archive, the informed consent of the patient and IRB approval must be obtained.

It should be noted that the revised Bioethics and Safety Act will allow the use of fresh tissue samples to be exempted from an IRB review (10). From a research perspective, model 2 has an advantage over model 1 because it provides researchers with convenient access to high quality DNA and RNA samples that are essential to modern medical research. On the other hand, model 2 would require hospitals to establish government-certified bioresource banks, which may not be economically or logistically feasible for smaller hospitals or those without sufficient resources (Table 2).

This model is similar to model 2 in that the clinician obtains informed consent from the patient to allow storage of the tissue in government-certified bioresource banks, and fresh tissue samples are processed into cell lines, nucleic acid and protein samples at the hospital's bioresource bank. However, in this model, the pathology department operates its own post-diagnosis FFPE tissue bank; only tissue samples that have already been used for diagnosis may be stored in the FFPE tissue bank (Fig. 1C, Table 1). The remaining tissue is archived so that it may be used for future diagnosis. Access to FFPE tissue samples does not require IRB approval and researchers can simply submit a request to the pathology department itself. However, distribution of tissue from the archive to the FFPE tissue bank remains a controversial ethical issue. For example, archived tissue samples may be required in cases of cancer recurrence and when new medical breakthroughs occur; the type and number of samples that should be archived is a question faced by pathologists on a daily basis.

Since informed consent has been obtained for the samples in both bioresource banks, this model allows researchers to access a large number of human tissue specimens without the need for IRB approval. However, the costs associated with establishing and operating multiple bioresource banks may be a burden that only a few hospitals can afford (Table 2).

Model 4 combines the functions of the two bioresource banks described in model 3 into a single bank that stores and processes both FFPE and fresh tissue samples and is managed by the pathology department (Fig. 1D, Table 1). Implementation of this model would streamline the storage process, reduce operational costs, and simplify the application process when accessing samples for research purposes (Table 2).

Although this model is the most efficient method of managing tissue samples for research, physicians are likely to feel uncomfortable with the idea of assigning complete control of the samples to the pathology department. To alleviate these concerns, a system could be established in which the physician who obtained informed consent from the patient is given control of the samples. Although the initial opposition to model 4 by physicians may be strong, serious consideration should be given to ideas that can accommodate a single bioresource bank.

This model is an alternative to model 4; instead of providing complete control of the FFPE and fresh tissue samples to the pathology department, a hospital bioresource bank would be established and its operation would be supervised by a BUSC (Fig. 1E, Table 1). The advantage of this model is similar to that of model 4, in that access to clinical samples for research purposes would be simplified (Table 2). The disadvantages of model 5 include the costs and resources associated with establishment of the BUSC, and pathologists' concerns over loss of control of the samples (Table 2). However, if the BUSC is composed of multiple individuals that balance the viewpoints of physicians, pathologists, and researchers, these groups would be more inclined to share their tissue sample archives. This balance would be essential for generating the trust that is necessary for pathology departments to actively participate in the system. Whereas the goal of model 4 was to alleviate the concerns of physicians, the main objective of model 5 is to alleviate the concerns of the pathology department. A system in which the pathologists are given ample opportunities to collaborate with various researchers could be established. Because most pathology departments control over 90% of human clinical samples, the success of model 5 is critically dependent on the enthusiastic support of these departments.