As Breast Cancer Awareness Month draws to a close, researchers still struggle against a disease that claims more than 40,000 U.S. lives annually (pdf). Whereas ideas about prevention and treatment may vary from doctor to doctor, early detection is the key to successful treatment—when detected early enough, any cancer has a 90 percent cure rate.

Catching cancer in its early stages is not easy though. Mammographies and other types of imaging tests are hardly foolproof, and neither are painful biopsy procedures, particularly when tumors are small and difficult to find. For these reasons, an interdisciplinary team of geneticists, computer scientists, oncologists and other researchers at The Fred Hutchinson Cancer Research Center in Seattle are working on a way to detect early-stage cancer that is no more invasive or painful than a simple blood test.

Scientific American spoke with geneticist and oncologist Amanda Paulovich, director of the Fred Hutchinson's Early Detection Initiative, about the work she and her team are doing to find biomarkers that can identify cancer in its most formative stages.

[An edited transcript of the interview follows.]


What is the Fred Hutchinson's Early Detection Initiative, and what role is technology playing in this work?
The Early Detection Initiative is an organizational structure within Fred Hutchinson Cancer Research Center to acknowledge and support the many faculty who for years have devoted their research to the discovery of the most effective ways to find and treat cancer at its earliest stages. This is a critically important initiative because early detection is our best chance for a cure. If we can detect tumors early, it's possible to surgically remove them and treat the patient with localized radiation therapy. The members of the initiative hope to improve early detection of cancer by making an impact in a variety of areas, including doing epidemiologic studies on risk factors, developing new and improved methods of diagnostic imaging, discovering and verifying new biomarkers as indicators of cancer, and developing new technologies to aid early detection.

Technology development plays a very significant role in this work. The number of new FDA-approved diagnostic tests for blood has remained static at zero-to-two per year for the past 15 years, indicating that conventional approaches to identify and validate new blood-based diagnostic tests have reached a plateau. For the majority of human proteins, there are currently no methods—that is, assays to quantify the amount in blood samples, so novel assays must be developed for testing of candidate new diagnostics in clinical studies. Unfortunately, this process is prohibitively expensive and time-consuming, creating a severe bottleneck. As a result, few candidate new diagnostics end up undergoing testing. A major focus of my research is on developing new technologies, based on hypothesis-driven, targeted mass spectrometry, for relieving this bottleneck to testing potential new blood-based diagnostics for detection of cancer. Ideally, we need assays to allow us to measure all human proteins in many clinical samples, rather than trying to pick a couple proteins that might work.

Last year, you and your colleagues reported in Nature Biotechnology having demonstrated that new applications of existing proteomic techniques showed promise of greater accuracy in detecting and quantifying proteins biomarkers in bodily fluids. What is the significance of this work and how have you advanced it in the past year? (Scientific American is part of Nature Publishing Group.)
In this work we demonstrated that assays for measuring proteins in blood, based on targeted mass spectrometry, are highly reproducible between different laboratories. This is important because if you are trying to discover or use new diagnostic tests, you need to be able to get the same result regardless of where you send blood samples.

How has the $4.8 million in federal stimulus funding from the National Cancer Institute impacted your research in the past year? What has it enabled you to do that you otherwise would not have been able to do?
The entire biomedical research enterprise is hampered by a lack of methods [assays] for quantifying human proteins. This unmet need slows basic research, development of new drug therapies and development of new diagnostic tests to help us detect or treat diseases. This NCI funding has allowed me to enter into an exciting collaboration with Steve Carr's lab at the Broad Institute in Boston to do a pilot project testing the feasibility of developing targeted mass spectrometry–based assays to all human proteins, starting with about 1 percent of the basic unmodified human proteome.

We will develop multiplexed assays that will allow us to measure multiple proteins at a time in human samples, thus increasing the speed with which we can test candidate biomarkers. This project would serve as a proof of concept to show the time and cost for a much larger effort, which would develop assays to the remaining 99 percent of all human proteins. The scaled project would be comparable to the Human Genome Project in cost and scope and promises to have a comparable impact on the biomedical research enterprise.

What are the challenges of detecting breast cancer in its early stages? What can be done about this?
For current breast cancer screening methods such as mammography, the challenges are access to screening, compliance, detection of some disease that is indolent and need not be treated, detection of noncancerous lesions that undergo biopsy, and failure to detect some disease that does need to be treated. It may be possible to develop blood-based diagnostic tests that complement mammography to address some of these issues associated with screening.

What would it take to create a blood test that could help early screening for cancer?
I envision blood tests and imaging technologies complementing each other—perhaps as a cost-effective panel of tests done in stages. A cheaper but less specific blood-screening test could be done first, followed by imaging. Alternatively, imaging could be used for initial screening, with follow-up blood testing to determine if a biopsy is necessary or if the disease is aggressive and needs to be treated.