Swinging for the fences with prostate cancer tissue bank
The discovery of the HER2/neu
marker was one of the most dramatic advances in breast cancer treatment in the last several decades. It led to the HercepTest and approval of the first novel gene therapy for aggressive breast cancer.
The breakthrough was a home run for patients, but it would not have been possible without the Cooperative Breast Cancer Tissue Resource, funded by the National Cancer Institute. Launched 14 years ago, the breast cancer tissue resource enabled the HER2/neu work because it supplied tissues from hundreds of patients for whom the outcome of treatment was known.
Make way for the breast cancer tissue resource’s cousin: the Cooperative Prostate Cancer Tissue Resource, or CPCTR. Formally funded in 2000, it is now operating at full tilt, and researchers and clinicians hope it will bring equally impressive progress in treating prostate cancer.
Michael J. Becich, MD, PhD, vice chairman and professor of pathology at the University of Pittsburgh Medical Center and one of the leaders in developing the CPCTR, says the resource has what it takes to hit home runs-this time for prostate cancer research. "That’s the purpose of this resource-to provide researchers with hundreds, even thousands, of tissue samples that it would be impossible for any one resource to provide."
Lack of sufficient prostate tissue samples has hampered prostate cancer research, but not just because most academic institutions haven’t shared the wealth. Unlike most other tumors, prostate cancer tumors are difficult to identify through gross examination of fresh tissue; they need high-quality histologic characterization through pathology resources, which are often not available for the job.
"Many scientists were not doing research in the field because they couldn’t get tissues with good clinical annotations," says Jules J. Berman, MD, PhD, CPCTR program director.
The National Cancer Institute decided that a multicenter effort was needed. "The NCI commissioned a number of organ review groups and we had a prostate review group that met toward the end of the 1990s," Dr. Berman says. "Their report listed what they thought needed to be done to enhance progress in the area of prostate cancer research, and one key thing was a tissue resource for scientists."
The NCI organized a cooperative venture including four academic institutions in the form of a "virtual tissue bank," the same structure employed by the breast cancer tissue resource. The cooperative group consists of George Washington University Medical Center in Washington, DC, Medical College of Wisconsin in Milwaukee, New York University School of Medicine, and University of Pittsburgh.
Although the CPCTR has actually been open for business for two and a half years, until now its main job has been acquiring high-quality, well-characterized tissue samples. "It takes at least two years to really get a tissue resource going from scratch," Dr. Berman says.
In 2001, going beyond the breast cancer tissue resource model, the CPCTR started prospective collection of fresh tissue of radical prostatectomy specimens. As of October 2003, the CPCTR had collected and annotated 5,134 cases, and it expects to continue adding about 1,300 cases per year.
The CPCTR invites requests from commercial and academic investigators; typically, researchers have a candidate marker for prostate cancer and are looking at the distribution of the marker in prostate cancer tissues from multiple specimens, so they need tissues from different stages and levels of aggressiveness.
Echoing the experience of the breast cancer tissue resource, requests for tissue microarray slides of prostate cancer specimens are vastly outpacing requests for frozen blocks, Dr. Berman reports. "There is a six-to-one predominance of TMA blocks over paraffin blocks or frozen tissues. It just saves researchers an enormous amount of time if they can do a single experiment, or experiments, on hundreds of pieces of tissue assembled on one slide, rather than design and perform experiments on many different sections they collect and stain."
At the moment, the CPCTR relies on the Medical College of Wisconsin as its tissue microarray laboratory. "We designed the TMAs based on our database; then we collect all the blocks from different sites and they go to one laboratory," Dr. Berman says. But the CPCTR is working on at least one additional TMA laboratory.
Says Dr. Becich: "Microarrays are an unbelievable resource. There’s nothing like them in terms of numbers. Instead of sending 10,000 slides, we may only have to send dozens." But he cites other advantages: All the staining is on one block and in one batch, so there are more uniform and standardized chemical results from each patient. With greater accuracy and higher density of information, there are also cost savings on reagents and antibodies.
Milton Datta, MD, a urologic pathologist who is the principal investigator of the CPCTR at Medical College of Wisconsin and assistant professor of pathology, explains that tissue microarrays help maximize the usefulness of tissue specimens. "In the U.S., we tend to focus on surgical treatment, but radiation treatment has seen improvements in aiming the beam and reducing toxicity, and it has become an extremely important alternative. But what do you do if you want to study the tissues on a patient who gets radiation treatment? You didn’t take out the prostate, but the initial needle biopsy contains cancer. Some of that biopsy tissue is left-not a lot, but it’s there."
"So one of the things we’ve done aggressively is try to collect little tiny bits of leftover tissue from needle biopsies. And Dr. Andre Balla [at the University of Illinois, Chicago] has been collaborating with us on methodology to make tissue microarrays to analyze hundreds of samples from prostate needle biopsies, so we get the most value out of tiny, tiny pieces of tissue."
Despite these economies, the clinical information collected today is much more expansive than it was in the past. The common data elements capturing clinical, pathologic, and inventory data include demographic information, clinical history, clinical pathology values, anatomic grading and pathology characteristics for individual blocks, histologic characteristics, lymph node status, clinical stage and progression, vital status, and information about therapy.
"The specimens we have gain value every year because we refresh our outcome data and they become more useful for research," Dr. Berman says. Exhausting the tissue has not been a problem so far. "For most types of specimens we can give the researchers sections from every block without exhausting them. Dozens of researchers could use a single block, unless they needed many sections from each block."
How does the CPCTR decide
which research projects will be funded? It starts with a research question, Dr. Becich explains. "Prospective recipients of tissues from the resource give us a plan, and we look at the question from three perspectives: Is it important for prostate cancer patients? Is the question scientifically valid and useful? And third, do they have statistical power in terms of the number of samples to be utilized to derive scientifically rigorous results?"
Because there is an iterative process, 90 percent of the proposals will be brought up to the required level. "There are not a lot of proposals that come to us that we won’t supply the resources to," Dr. Becich says.
But the specimens will seldom come from a single site. Dr. Datta explains that under this "virtual data bank," instead of physically moving all of the tissues to one bank, the resource continues to store them at many little banks at various institutions, and they are shipped to investigators from multiple sources.
He compares the setup to business practices at Amazon.com. "If you place an order for several items from Amazon, you’ll get three or four boxes-maybe one from Kentucky, one from Kansas, and so on-because Amazon will ship from whatever distribution house has the item."
Among other advantages, this arrangement helps protect the integrity of the specimens. "We’re always concerned when we’re moving tissues around that they’ll thaw, or get lost, or if they are wax the truck will get hot and they’ll melt and change."
Diversity of sources also benefits the research directly by eliminating some of the potential selection bias. "We’re collecting prostate samples from patients at four different medical centers and each one often has local community hospitals, veterans’ hospitals, and affiliated medical centers as sources, so there is a lot of variation in patients," Dr. Datta says.
He adds: "A lot of large prospective studies you see originate from a single major medical school, with the work usually done by one or two urologists and maybe one pathologist reviewing all the cases. What I like about the material in this resource is it’s ’Everyman’s tissue.’ It’s what happens to the guy being treated in Milwaukee and New York City, in the community hospital and the fancy medical center. So I hope the findings will work for Everyman."
A key area of research for the past four years has been understanding how tumors grow, discovering factors other than testosterone, which has been known since the 1940s to be linked to prostate cancer. Developing a targeted treatment for a prostate cancer growth factor is precisely the kind of purpose the CPCTR was intended for, Dr. Datta says. "Just imagine what we could do if, through combination chemotherapy such as we use for treating lymphoma or AIDS, we could block not only testosterone, but other growth factors as well."
Prostate cancer research has accelerated in the past few years, says Lawrence D. True, MD, associate professor of pathology at the University of Washington, Seattle, and co-director of the Specimen Core of one of the 12 prostate cancer SPOREs in the United States. Sponsored by the National Institutes of Health, SPOREs (Specialized Programs of Research Excellence) usually conduct multiple research projects, all dedicated to cancers of body sites. Leading the push for more research funds was investment banking figure Michael Milken, who was diagnosed with prostate cancer in the early 1990s.
Most of the clinical utility of the research being funded, however, lies down the road. At the University of Washington SPORE, for example, "our major focus is really to understand the biology of progressive metastatic and hormonally resistant prostate cancer," Dr. True says.
"With tissues from more than 1,000 prostate cancer cases treated at the UW, we’ve identified some new androgen-regulated genes that are candidates for genes that may be important in controlling the progression of prostate cancer, so that’s looking promising. We’ve also improved the technology to deal with smaller numbers of tumor cells to characterize transcriptomes and proteomes."
Although the investment for breast cancer tissue came long before the investment for prostate, prostate is making up for lost time, Dr. Becich says. "Now that folks know how successful the model was, we’re on the ascending part of the equation. There’s probably 10 to 20 times as much funding today as 10 years ago. I don’t think it’s peaked yet, but it’s probably maturing. The access to these blocks and microarrays gives an unprecedented infrastructure for prostate cancer research. So the results of all this investment haven’t yet rolled out to patients."
Unfortunately, the potentially most useful specimens can be the most difficult to acquire. Christopher J. Logothetis, MD, chair of genitourinary medical oncology at the University of Texas M.D. Anderson Cancer Center, Houston, describes one of the catches of prostate cancer pathology: "The problem is that if you obtain tissues from patients who regularly get treated with prostatectomies, those are invariably representative of low-stage, low-grade cancers, and there’s neither sufficient cancer, nor do we know the relevance of how threatening all this cancer is for the patient."
"So we end up studying large numbers of patients with small amounts of tissues," he says, "and we require a long followup, because, fortunately for the patient, even if the patient eventually succumbs, it takes forever. So it creates a huge bottleneck in the efficient screening of new therapy targets or new surrogate markers."
When prostate cancer metastasizes to the bone, it becomes difficult to sample and get reliable tissue, he adds. With "warm autopsies," the patient and family members allow malignant tissue to be harvested soon after the death of the patient from prostate cancer. "But the problem with this tissue is it’s end-stage. The good thing is it clearly represents bad disease and often large amounts of cancer, but you don’t have a long serial history to know precisely how it evolved over time, because you don’t have many time points."
Says Dr. Datta: "The most informative prostate cancer patients would be ones where we have both primary and metastatic prostate cancer. But they’re hard to get because the criterion for patients having a radical prostatectomy is that they don’t have metastases at the time of surgery. Patients with known metastases are seldom biopsied, because usually the clinician treats them based on elevated PSA levels, or on radiographic or clinical evidence of tumor recurrence. But when we can get them, we do, and those are the most valuable cases because we have so few of them."
The CPCTR is an attempt to remedy these difficulties by linking the tissues to critical data elements, Dr. Logothetis notes, crediting Dr. Becich with the vision to develop a data-modeling system that makes it possible to inventory tissues of different qualities and to give investigators nationwide access to them.
M.D. Anderson’s prostate cancer SPORE is trying to understand the pathways in the progression of cancer in the bone that create either resistance or sensitivity to therapy. For example, using a novel platform called in vivo phage display, a group of M.D. Anderson investigators, led by Renata Pasqualini, PhD, and Wadih Arap, MD, PhD, has identified molecules that selectively hook up to blood vessels inside prostate cancers that have spread to the bone, and the group is now developing therapeutic agents that target the cancerous cells.
Similarly, an osteoblast regulatory factor (MDA-BF-1) identified by Sue-Hwa Lin, PhD, in the marrow derived from patients with prostate cancer has been implicated in cancer progression. "This exciting finding illustrates the value of unique sources of tissue linked to clinical outcomes," Dr. Logothetis says. "Ultimately we’ll need to find out how prevalent these targets are in large numbers of patients. That’s where the tissue resource will come in."
These lines of research involve basic observations that have become "translational" questions because they probe the connection to human disease, he says. "Translational research is basically research that crosses the bridge between laboratory investigation and clinical investigation."
M.D. Anderson is trying to expand the concept of tissue banking to include knowledge banking, Dr. Logothetis reports. "Every time someone studies a tissue, we obtain a commitment to get any new observation they make into the data bank, so we create an enriching knowledge bank." Because of the expectation that investigators using the tissue will collaborate with M.D. Anderson researchers, "it will be difficult for a purely commercial enterprise to avail itself of our tissue, and commercial use of our specimens is not something that has happened in the past."
A forthcoming report by the
Rand Corp. on best practices in tissue banking confirms that there is much wider appreciation today of the importance of standardized collection and storage, Dr. Logothetis says. The obstacle now is finding the resources to create an infrastructure that places trained tissue procurement and characterization experts near the operating room, and permits them to collect all the critical information with common data elements while remaining compliant with the Health Insurance Portability and Accountability Act of 1996. "Assembling all those pieces is both an expensive and complex exercise," he stresses.
The topic of best practices in handling tissue for research is a timely one, Dr. True says. It was discussed at the January meeting of prostate cancer SPOREs. Since pathologists at different centers have successfully used different approaches in tissue handling, a decision was made to share protocols and do collaborative investigations into which factors most affect the quality of RNA and protein in these specimens.
Despite the challenges, the United States is in the vanguard of prostate cancer research worldwide because of the extent of its funding, Dr. Datta points out. "We have funding not only from NIH and from private foundations like CAPCure, but the Department of Defense’s congressionally mandated disease research program also has money designed for cancers of the breast, prostate, and ovary. What this has done is give scientists a lot of opportunity to think about prostate cancer. A lot of other countries can’t match that."
An even more ambitious project is underway in the effort to accelerate cancer diagnostics and therapeutics: the National Biospecimen Network, or NBN. A partnership of the National Dialogue on Cancer and the National Cancer Institute, the NBN will be the first national network of tissue repositories designed to facilitate genomic and proteomic research.
In December, a three-year demonstration project was proposed to pilot-test the NBN’s plan to standardize the collection and annotating of biospecimens, and allow timely access to tissue based on open peer review. Supplementing the initiative is Rand’s forthcoming report, "Case Studies of Existing Human Tissue Repositories," which will outline best practices in collection, storage, and annotation of tissue specimens.
But the Cooperative Prostate Cancer Tissue Resource has more immediate needs. Its biggest priority now is marketing. "People always think, incorrectly, that when you have a resource and you know many people need it there won’t be any difficulty in getting people to apply," Dr. Berman says. "But actually, like anything else in the world, if you don’t have marketing, you’re not going to get any requests."
"Right now we feel as though we’re at our peak as a resource," he continues. "We have all we need to provide researchers not only with paraffin sections but also frozen tissues and tissue microarrays. We’re getting the word out that we’re ready to take orders. And we have half a dozen requests that are now being reviewed."
The CPCTR is a priority for the National Institutes of Health, Dr. Datta says.
"As far as NIH is concerned, these tissues are worth more than money," he says.
"They’ve invested over $10 million in this project, and they really believe
in it. We have a huge repository of material-one of the largest in the world.
We’re starting to give out tissues for various research projects, and the fun
part is seeing the excitement and interest both from pharmaceutical companies
and from researchers with their own laboratories who are coming in and studying
Anne Paxton is a writer in Seattle. The CPCTR’s Web site is at
www. prostatetissues.org, and the Cooperative Breast Cancer Tissue Resource
site is at www.cbctr.ims.nci.nih.gov. The National Biospecimen Network
blueprint can be found at www.ndoc.org/about_ndc/reports/ pdfs/FINAL_NBNBlueprint.pdf.
For more information on Rand’s “Case Studies of Existing Human Tissue Repositories”
report, visit www.rand.org/publications/ MG/MG120.