How to Efficiently Obtain Human Tissues to Support Specific Biomedical Research Projects

Many investigators have difficulty in obtaining high quality, well characterized, human tissues to support their research. The experience of the Tissue Collection and Banking Facility (TCBF) at University of Alabama (UAB) in providing more than 200,000 tissue specimens to researchers indicates many of the problems result from inappropriate requests, incorrect expectations, as well as limitations of tissue repositories (1-3). Our tissue resource cannot provide a whole “normal” beating heart or 10 cm of normal ascending aorta from surgical procedures, as have been requested. We can, however, provide extensive numbers of well characterized human tissues to support a wide range of biomedical research projects if the requests for specific tissues are reasonable. Our goal is to explain the role of the major tissue resources; why the more demanding an investigator's request, the fewer tissues a repository can supply; and how researchers can work with their tissue resource to improve tissue availability.

The TCBF at UAB is composed of multiple components including the Southern Division of the Cooperative Human Tissue Network (CHTN), the Tissue Procurement Service of the UAB Comprehensive Cancer Center, the Tissue Repository of the Pancreatic Specialized Program of Research Excellence (SPORE) at UAB, the Collection and Preparation Facilities of the Pulmonary Hypertension Breakthrough Initiative, and the Tissue Resource of the Skin Disease Research Core at UAB. Prior facilities have included the Tissue Resource Centers of the Ovarian and Breast SPORES at UAB and the Banking Demonstration Project of the Early Detection Research Network (EDRN).

Over the past 20 years the TCBF at UAB has received more than 4,000 requests for human tissues and provided more than 200,000 tissue specimens to researchers. Many initial requests are found to be difficult or impossible to meet for a variety of reasons.

Basically, tissue resources work under one of two models: prospective collection or tissue banking. The investigator must understand the model under which a given tissue resource operates. For example, the Cooperative Human Tissue Network (CHTN) collects tissues prospectively on the basis of the requirements of a specific investigator and does not have a large bank of previously collected tissues. Using this “model of prospective collection,” an investigator may establish specific collection requirements such as requesting that fresh lung carcinoma be minced and supplied in RPMI 1640 and such tissues are collected and processed to meet the investigators needs (1-3). One major disadvantage of this model is that clinical information is usually limited; however, if permitted, clinical information may be obtained from medical records. Because samples have just been collected, it may be years before clinical outcomes are available. Large numbers of specimens are not immediately available because samples are collected prospectively (1).

In contrast, each SPORE has a tissue bank focused on an organ system. SPOREs exist for all major cancers with more than 50 spore center grants currently funded (http://spores.nci.nih.gov/current/index_current.html). In this “tissue banking model,” samples are collected and processed according to a standard protocol. For example, all solid tissues might be divided into three aliquots and frozen using the same or different preparations. The banking model has the advantage that samples may be associated with both detailed clinical information and with clinical outcome. Also, relatively large numbers of specimens may be available initially; however, tissue processing may not meet investigator needs.

Tissues that are unavailable at one resource may be obtained from a different resource (1-3). Because of the controversial nature of some specimens (e.g., fetal or HIV tissues), some tissue resources elect not to collect such tissues; however, other tissue resources do collect them (e.g., the AIDS and Cancer Specimen Bank). Investigators should be aware of the major resources available and the model under which each resource works (Table 1).¹

Investigators should check with their local Institutional Review Board (IRB) and Health Insurance Portability and Accountability Act of 1996 (HIPAA) privacy board about what is required to accept and use human tissues received from an outside source. Such tissues may be supplied as deidentified so that HIPAA requirements do not apply (http://www.hhs.gov/ohrp/humansubjects/guidance/cdebiol.pdf).

Tissue resources typically rely on remnants of tissue that are not needed for diagnosis and that are available from clinical activities. Procedures that may furnish excess tissues can be surgical, diagnostic, clinical research, transplantation, and/or autopsy. When developing a request it is important to understand that some tissues and/or medical information may not be readily available. Thus, requests for tissues or information that are not available may either cause the rejection of the request and/or a delay in processing the request.

Problems with requests fall under specific categories: (1) size and number of specimens, (2) types (rarity and availability), (3) time constraints, (4) demand versus supply, (5) limitations and goals of the resource, and (6) time and resources that can be devoted to a specific request. Even when informed of specific limitations that prevent the investigator's requests from being met, the investigator's refusal to compromise on their requirements may result in disappointment and delay in the investigator's ability to perform his/her research (1-5).

Before tissues can be utilized in research, all diagnostic requirements of the tissue must be fulfilled. For breast cancer, diagnosis may require about 1 g of cancer to ensure that the histopathologic description is adequate and biomarkers such as estrogen receptor, progesterone receptor, and Her2-neu status may be evaluated. Thus, a 1-cm breast tumor would just fulfill diagnostic requirements. Also, preinvasive neoplastic lesions such as ductal carcinoma in situ (DCIS) are submitted completely to exclude invasion. If a resource receives 1 g of remnant breast tissue, an investigator who requests 1 g of breast cancer is unlikely to receive a specimen from this case, in that rather than providing 1 g of breast cancer to a single investigator, the tissue source is likely to choose to supply 10 investigators with 0.1-g specimens; thus meeting the needs of a greater number of researchers. Requesting a gram or more of high demand tissues is destined to lead to receiving few specimens. Investigators should adapt their experimental methods to small specimens if breast, prostate, and pancreatic cancers or other small, rare, or high demand tissues are requested. Similarly, brain, pancreas, and small cell undifferentiated (oat cell) lung cancers may be treated following needle biopsies, so no fresh or frozen tissues may be available except from autopsies unless the patient is consented to obtain “extra” biopsies for research. Such consent and extra biopsies are being obtained by some SPOREs. In contrast, obtaining grams of tumor for one investigator is possible from some types of more common, bulky tumors for which resections are done (e.g., ovarian carcinomas, renal cell carcinoma).

Some tumors are relatively rare (e.g., paragangliomas), so that requests for multiple samples of such tumors or for many tumors in high demand (breast, prostate, pancreatic cancers) may be impracticable unless a special collection of such tumors is available. Another limitation to the availability of some tumors is treatment prior to surgery (e.g., sarcoma, breast cancer). Neoadjuvant radiation or chemotherapy together with successful early screening for neoplastic diseases can result in small overall tumor size at the time of resection. These approaches have begun to severely limit access to many tumors and to large tumors.

Similarly, some types of tissues are unavailable from diagnostic procedures, especially “truly normal tissues” or certain diseased tissues not treated by surgery. Normal tissues are not removed unless the surgery removes normal adjacent tissues. “Uninvolved matching” tissues may be available (e.g., colonic epithelium from a patient with colon cancer); however, this is not truly normal tissue because molecular changes may have occurred and be present in the normal appearing tissue (5). In contrast, from operations for colon cancer, normal smooth muscle can be collected because colon cancer does not affect muscles unless it invades or the cancer is hereditary.

When “normal” tissue is requested, it is important to specify what is acceptable as “normal.” Is uninvolved tissue acceptable, or does the investigator need only tissues from noncancer cases? Is diseased tissue acceptable? What types of diseases or abnormal conditions are acceptable? An age range should be specified in requests for “normal” prostate because many prostates in men ≥50 years of age may have involvement by benign prostatic hyperplasia.

Some investigators request normal tissues from trauma cases. Tissues from surgical trauma are very difficult to obtain because these surgeries are rarely scheduled, and often are done after working hours. Similarly, repositories may not obtain tissues from forensic autopsies because these may be done without consent.

Unreasonable time constraints reduce the availability of tissues. Some specimens can be obtained within 15 minutes of surgical removal; however this requires the tissue organization to devote special resources to accomplish very rapid processing. Also, only a few aliquots can be processed rapidly because of the time required for dissecting, processing, and labeling of large specimens. For example, obtaining 10 samples from one large, complex specimen may take more than an hour to collect and process, and processing may be slowed if multiple specimens from this or other cases must be processed simultaneously. Rapid collection may not permit critical diagnostic review and could compromise a specimen's diagnostic usefulness, and diagnosis always must take precedence over research. It is important for investigators to have a basic understanding of the types of changes that tissue may undergo and how these changes may impact their results.

The tissue collection timeline can be broken down into several phases including in vivo, ex vivo, and storage phases. In vivo warm ischemia changes occur during the surgical procedure following induction of anesthesia and compromise of the vascular supply. This phase may be prolonged in some procedures such as recently developed robotic procedures for prostatectomy. A similar ischemic process may be seen in vivo as a result of tumor growth exceeding vascular supply, particularly in high grade tumors resulting in degradation of biomolecules. Studies of these changes and their time course have not been done. Although human tissues to support research should be collected and processed rapidly following the removal of a specimen from the patient, the scientific importance of rapid collection following lengthy warm ischemia remains controversial. Several studies, based upon observations of intact ribosomal bands of RNA and lack of smudging of mRNA on gel analysis, have indicated that the rRNA does not change rapidly after tissues are removed from the body (4, 6-8). This finding may be due to a more rapid degradation of RNA during warm in vivo ischemia after organs have their vascular supply compromised; however, relying on intact ribosomal bands considers only degradation, not alteration in expression levels of various genes which may occur.

Huang and colleagues (6) reported modest changes in gene expression in colonic epithelium during ex vivo warm ischemia in tissue held at room temperature (5 minutes versus 60 minutes postoperative). Type I genes (63.8% of all genes) had an average increase in expression of 27% over 60 minutes; 67% of type II genes (17.8% of all genes) had an average increase in expression of 12%; and 33% of type II genes had an average decrease in expression of 12%. For type III genes (13.4% of all genes), half increased on an average of 50% and half decreased on an average of 50%.

Dash and colleagues (7) reported that less than 0.6% of the genes they evaluated in prostate cancer showed increased expression at 1 hour at room temperature. Spruessel and colleagues (8) using Affymetrix chips, reverse transcriptase-PCR, and surface-enhanced laser desorption/ionization-time-of-flight-mass spectrometry reported no changes in the 28S:18S ribosomal bands of RNA at 30 minutes; however, 15% of genes and proteins changed by at least twofold within 5 minutes after removal and 20% by twofold within 30 minutes of removal.

The above studies indicate that after tissues are surgically removed from humans, ribosomal bands usually do not degrade rapidly after tissues have experienced in vivo warm ischemia, that most genes are likely to increase in expression rather than to degrade, and that the expressions of some genes return to baseline within 1 hour. In addition, newer methods of real-time quantitative PCR have been developed that use short amplicons and hence can utilize partially degraded mRNA (9). Also, some proteins such as phosphoproteins have been reported to degrade rapidly, but such degradation is likely to vary with specific tissues (10, 11). The effects of anesthesia and in vivo ischemia on all these genes are unknown and may be greater than the changes observed after removal of the tissue from the patient. Thus, requiring rapid removal and processing of tissues may not always be warranted, especially because this may reduce tissue availability. Each investigator must match the requirement and time restraint needed for their specific research with effects of these on tissue availability.

Although breast tumors are frequently removed, their small sizes and the numerous requests for them may make many requests difficult to fulfill. To prioritize samples, most academically based institutions provide tissues using a priority system. Investigators located at the home of the tissue resource may have first priority, academically based second priority, and commercially based users the lowest priority. Subgroups in priority may be based on the scientific impact, peer reviewed extramural or intramural funding, and aid to new investigators. In some cases, a Tissue Utilization Committee decides which tissues are supplied to investigators (1-3, 12).

Commercial companies may have difficulty obtaining some samples; however, samples of readily available normal tissues and tissues of tumors that are typically large (e.g., renal cell carcinoma, ovarian carcinomas, and sarcomas) can be supplied. Alternatively, tissue resources that cater to commercial users may be accessed (1, 2, 12).

A tissue resource may have requirements or agreements that must be documented prior to the receipt of tissues. These requirements are usually rigid, carefully thought out, and unlikely to change. The tissue resource may require education of personnel in biohazards (12, 13) and signing of an indemnification or material transfer agreement (2, 12). Many repositories require that the IRB and privacy board of the investigator's institution must agree that the investigator has met their requirements (14). The tissue resource may require that it be recognized in publications. Issues related to commercial use of the specimens also may have to be satisfied. Some resources also may require a data use agreement to satisfy HIPAA requirements if a limited data set is provided. It is important to understand the policies of tissue resources before requesting tissues (12).

Most constraints increase the difficulty of obtaining tissue. For example, a sample of breast cancer may be requested from a patient who has type I (juvenile onset) diabetes. The history of type I diabetes may be in the medical chart, but most tissue organizations do not have the time or resources to review charts on all patients with breast cancer prior to surgery to fulfill such unusual and demanding requests. Similarly, some repositories may not have permission to review a patient's medical history or to obtain clinical outcome. When patients are a part of an interventional study, consent for use of tissue or medical information in research can be included in the patient consent form; however, it is more difficult and time consuming to obtain informed consent or HIPAA authorization from the majority of patients undergoing surgery or admitted to a medical facility (12, 14). Thus, investigator requirements that tissues provided to investigators be from consented patients may limit the availability of tissues unless anonymized specimens can be accepted.

The more restrictive and numerous are constraints, the greater the difficulty in meeting them. When a size constraint is added for a specimen that is already difficult to collect, it frequently makes the specimen impossible to obtain. Thus, a request for 1 gram of breast cancer from a patient less than 30 years of age would be almost impossible to meet because the sample requested is large, breast cancers are in high demand, and breast cancers are uncommon in patients this young.

Extensive services (e.g., isolation of cells or microdissection of specific cell populations) usually cannot be provided by tissue resources; however, if provided, the investigator should expect to pay all costs of such services. If a tissue repository includes or has access to a histology facility, histology services may be available for research. Investigators should check with the tissue resource about availability and cost of special services.

By understanding the types of resources available and the constraints under which various tissue resources operate, an investigator can select an appropriate tissue resource for the project and work with the resource to tailor tissue requests to maximize yield. Investigators should review protocols to minimize the amount of tissue required and be aware of the difficulty in obtaining large or numerous samples of tissues in high demand or of rare tissues. Difficult or complex tissue requests may limit tissue availability. To ensure that a requester of tissues obtains needed tissues as efficiently as possible, it is important for the requester to understand the constraints under which tissue resources operate.

No potential conflicts of interest were disclosed.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.