By Charlie Schmidt
Banking on Tissues
Standardized tissue collection could propel cancer research
By Charlie Schmidt
What's the biggest roadblock to curing cancer? According to a think tank convened in 2002, a shortage of high-quality human tumor and tissue samples is the top barrier to progress in cancer research.
If that surprises you, consider that scientists are pinning their hopes for new diagnostics and cures on molecular biomarkers. Biomarkers are proteins and other biochemical signs that indicate the presence of a disease, or a response to treatment. Found in tumors and other tissues, biomarkers could enable early cancer detection and provide new targets for drugs. But to study them comprehensively, scientists need better access to human samples from large numbers of people.
Tissue Banks Play Growing Role
Emerging to meet that need, a variety of tissue banks—also known as tumor banks, biobanks and biorepositories—are specializing in collecting human biological samples and making them available for research. Tissue collection arguably has a long history—Johns Hopkins University, for instance, began to save tissue from autopsies more than 100 years ago. However, tissues that have been collected and stored using traditional methods have a limited capacity to serve modern research. Most samples are fixed in formalin and embedded in paraffin, which can degrade the structure of proteins and other biological molecules, such as DNA and RNA (a type of genetic material essential to the production of proteins).
Researchers must therefore rely on the more sophisticated facilities that can supply fresh tissue samples for molecular studies. Some of the facilities are connected to large-scale, clinical research centers that supply a steady flow of material. Still others are hosted by privately funded advocacy groups for specific cancers, which collect tumors from around the country to increase sample sizes for research.
Collectively, these tissue facilities comprise an emerging field that is essential to progress in cancer. "Right now, there are only two ways to get live tissues," explains Christopher Corless, a pathology professor at the Oregon Health and Science University (OHSU) Cancer Institute in Portland. "You can either collect them yourself from individual patients, which is difficult because of all the legal issues involved, or you can tap into a bank of pre-existing samples obtained from consenting patients. If properly stored, these samples can be used for virtually any type of analysis: DNA, proteins, or a combination of the two. There's no substitute for them."
Corless directs the Cancer Pathology Shared Resource, a tissue bank within the OHSU Cancer Institute. He says scientists primarily use biological samples to test medical hypotheses: They start with an idea that cancer might be linked to a protein or other molecular signal, and then they investigate that signal in as many samples as possible. Studying a large number of tissue samples is essential because humans vary in their genetics and their responses to diseases, he says. The more samples scientists have, the greater their confidence that study findings will actually apply to the whole population. At Corless' tissue bank, samples obtained from patients during surgery are flash-frozen and stored, with the donor's permission, at -80° Celsius or in liquid nitrogen. The tissues are then supplied to OHSU scientists and their collaborators under the oversight of an institutional review board, which ensures compliance with ethical standards.
Repositories for Rare Cancers
Tissue banks' contributions are especially valuable in the case of rare cancers. Patients with these cancers tend to be separated by both distance and time, and sample shortages pose difficult problems for research. Take, for example, inflammatory breast cancer—a cancer with a poor prognosis that is diagnosed in at most 10,000 American women every year. "We were told by the scientific community that the reason there's so little research on inflammatory breast cancer is because the patient cohorts are too small," says Ginny Mason, the executive director of the Inflammatory Breast Cancer Research Foundation (ibcRF). "So we asked if having better access to tumor samples would make a positive difference, and the scientists said yes, that it would."
Upon learning this, Mason—a survivor of the disease herself—and her colleagues at the ibcRF created a tissue bank specifically to advance research into the disease. Because they were part of a small nonprofit organization, rather than a large academic institution, everyone involved had a lot to learn, Mason says. But the ibcRF persevered, ultimately linking up with advocacy groups from six other diseases to form the Genetic Alliance Biobank (GAB), a cooperative project that launched in October 2004. Today, the GAB provides common protocols for tissue banking that apply to each of its seven affiliate members, which include organizations dedicated to rare disorders such as cardiofaciocutaneous syndrome, Joubert syndrome, and a disease called neurodegeneration with brain iron accumulation. "All the GAB members have a similar goal," says Owen Johnson, the ibcRF's president. "We all want to help researchers find genetic abnormalities that either cause the disease, define it, or provide opportunities for targeted therapies."
The ibcRF's own tissue bank, called the Inflammatory Breast Cancer Biobank, is housed in Wisconsin by a private contract laboratory. Mason says scientists who want patient samples can submit a proposal to the ibcRF. If the request is approved, the tissue bank will release the samples together with sets of coded clinical data, which are presented in a format that protects a patient's privacy. Clinical data are crucial to putting analytical results in the proper context, Corless says. Scientists need to consider variables like a patient's race or ethnicity, sex, age, treatments and other factors when they assess a sample's molecular profile. This information helps confirm that a particular molecular signal derives from the disease, or an experimental treatment under study, rather than a random factor.
Issues at Home and Abroad
Scientists investigating multiple myeloma—an uncommon cancer that originates in bone marrow—also have faced the challenges of tissue sample collection. But today, many of these samples are sent to a large advocacy-based tissue bank known as the Multiple Myeloma Research Consortium (MMRC). Launched in August 2004 by the Multiple Myeloma Research Foundation, the MMRC coordinates with 11 cancer centers to collect bone marrow and blood samples, which wind up in a central repository at the Mayo Clinic in Scottsdale, Ariz.
The consortium was created to knock down barriers to collaboration, says its executive director, Steven Young. Through the organization, scientists can access high-quality bone marrow extracts and matched blood samples taken from the same patients. Both types of samples are crucial for research, Young says. Bone marrow is useful for testing new drugs because it's the main target for multiple myeloma therapies. And blood samples could contain molecular clues that indicate a patient has the disease, which may help in the development of new early detection methods.
The MMRC's administrators have developed a customized system and software that combines laboratory and clinical data. Each sample is assigned a unique identifier, which links it to clinical information. But only the patient's own doctors can trace the sample back to a name, Young emphasizes.
Young stresses this point because patient privacy and related issues pose difficult challenges for tissue banking, particularly for publicly funded facilities supported by the National Institutes of Health. This is especially true because the Health Insurance Portability and Accountability Act (HIPAA) penalizes clinicians and administrators who breach patient confidentiality. "We have significant problems with tissue banking's ethical, legal and policy aspects," acknowledges pathologist Carolyn Compton, who directs the National Cancer Institute's Office of Biorepositories and Biospecimen Research. "There are significant disparities among the federal agencies that regulate the use of biospecimens."
Security breaches could make it hard for offending biobanks to continue collecting new tissues and data. Hoping to minimize the risk, Compton's office is now producing sample templates for patient consent, tissue transfer agreements, and other protocols that ideally will pave the way for smoother research.
Few doubt the value of these efforts, but some in the tissue-banking field say privacy concerns are overblown. "I think there's too much anxiety over this," says Corless. "Our experience shows patients hardly ever object to scientists using their tissues for research. You offer them the opportunity to turn that down, but they virtually never do."
Overseas, some countries—particularly those with highly educated populations and centralized medical systems—have mounted effective national dialogues aimed at resolving such ethical concerns, says Brett Davis, an executive with IBM Healthcare and Life Sciences, which has sponsored four global biobanking summits during the last two years. For instance, in the United Kingdom, an array of scientific, religious and community leaders were instrumental in the formation of the U.K. Biobank, which now proclaims itself the "world's biggest resource for the study of the role of nature and nurture in health and disease."
The Karolinska Institute Biobank in Stockholm, Sweden—hosted by the Karolinska Institute, home of the Nobel Prize in physiology or medicine—is another example of a biobank success story. The facility, which is working with IBM, collects tissue samples from throughout Sweden.
The Singapore Tissue Network is also on the cutting edge of the field, Davis says. Created in 2002 by the Singapore Biomedical Research Council, the country's Ministry of Health and the Genome Institute of Singapore, this tissue bank provides a national repository for tissues, including blood and DNA samples. The facility also promotes numerous research projects, including genetic comparisons of normal and cancerous tissues, and molecular studies of diabetes, high blood pressure and heart disease.
"Other countries are much farther ahead of us in tissue banking," Compton concludes. "They're investing in personalized medicine with centralized, national repositories that collect specimens from every member of the population. And because they have national health care systems, they aren't faced with the sorts of legal problems that make things much more difficult for us. If we don't come to grips with this, we may fall so far behind in terms of personalized medicine that we may never be able to catch up."
Worries over Standardization
Ethical and legal challenges don't comprise the only impediments to tissue banking in the United States. Even more daunting perhaps are the widely divergent collection and storage practices used by tissue banks throughout the country. Proteins and RNA are highly sensitive to environmental changes. A sample left on a surgical tray for 10 minutes before freezing might yield a different molecular pattern than one frozen an hour after surgery, says Elisa Eiseman, a senior natural scientist with RAND Corporation, in Washington, D.C. Likewise, adds Corless, differences in surgical techniques can influence proteins and RNA. "Some surgeons remove a tumor without clamping the vein while others clamp the vein first," he explains. "And this can make a huge impact on the [protein or metabolite] profiles you derive from that tumor—individual surgical differences can really impact on the quality of the material."
What this means, Compton says, is that researchers can't readily compare results obtained from institutions that use different methods to handle tissues. "If labs produce different results, we don't know if that reflects real biological variation," she says. "And that's hindering progress—we need to be able to separate biological truth from experimental and technical artifacts." This same concern was highlighted by RAND in a 2003 report, which concluded that American tissue banks are awash in varying standards, and that these differences are significantly impeding gene- and protein-based research efforts in the biomedical sciences.
According to RAND's Eiseman, who co-authored the 2003 report, the best tissue banks in the U.S. have certain key features: They characterize tumors to the greatest extent possible, and they emphasize close working relationships among all those involved in the process. "They offer a lot of training so everyone is working off the same page," she explains.
"Good tumor banks also collect as much clinical information as they can," Eiseman adds. "That's one of the biggest things people in this field need: They need to know what's going on with the patients, they need information on long-term follow-up, they need to know whether the cancer recurred, and which treatments worked and didn't work."
The NCI's Office of Biorepositories and Biospecimen Research has just released an initial set of guidelines that it describes as a "first step toward unifying policies and procedures for NCI-supported biorepositories." The guidelines, developed through consultations with a wide range of experts, and distributed to NCI-supported facilities' managers, lay out voluntary measures for sample documentation, collection, processing, storage and retrieval. Forthcoming guidelines will cover handling and analytical methods for specific tissue types, such as normal and cancerous tissues, as well as blood, plasma and urine.
As these various issues are resolved, tissue banking will likely play a growing role in advancing cancer research. "Most patients want their disease experience to benefit someone else," says the ibcRF's Ginny Mason. "We're finding that patients want to help and our own patients are very excited about this."