Developing the Ideal Breast Cancer Screening Test

Thirty-five years after mammography screening demonstrated a 30 percent mortality reduction,1 researchers are still searching for an ideal breast cancer screening tool. At the recent American Society of Clinical Oncology (ASCO) meeting in San Francisco, members of an Institute of Medicine panel described their recommendations to spur development of such a test. Earlier this year, the panel issued a report titled Mammography and Beyond: Developing New Technologies for the Early Detection of Breast Cancer,2 which was fundamentally a technology assessment of early breast cancer detection technologies. A premise of the report is that mammography is the gold standard to which all emerging detection technologies should be compared. Because nothing yet has proven more effective than mammography, "wider use of mammography should be encouraged," Craig Henderson, IOM panel vice chair and adjunct professor of medicine at the University of California, San Francisco, told ASCO attendees. Yet even though the panel considered mammography as the only test to prove effective for population-wide screening (i.e., testing healthy asymptomatic individuals for early signs of breast cancer), the panel nonetheless drew attention to its limitations. "Despite advances in mammography, it has never proven better than it was 35 years ago when it was relatively crude," said Henderson. Additionally, mammography has important limitations. Among Henderson's concerns were the 15 percent of interval tumors [that] develop between screening exams and the difficulty imaging dense breasts and breasts of women on hormone replacement therapy. Another important drawback is that it "creates a situation of overdiagnosis," he said. Up to three-fourths of suspected cancerous lesions prove to be false positives, "leaving emotional scars, and fear of breast cancer," he observed. These observations alone challenged fundamental tenets that back early cancer detection. Perhaps it is no surprise then, that policymakers, the public, and the cancer community have been slow to embrace this report. Not everyone, however, still thinks that on balance, the benefits of screening mammography are as compelling as the panel suggested. Indeed, Steven Narod, professor of public health sciences at the University of Toronto (and not a panelist) says that the false positive problem with mammography should "prompt us to re-examine the axiom that early detection promises cure." Rather than invest more resources in refining mammography and ensuring its use in women who don't get regular mammograms, Narod thinks that limited funds for breast cancer research should be allocated to studying the biologic processes that control early metastases (a top priority of the report). The IOM panel steered a middle ground on further development of mammography, neither agreeing nor disagreeing with Narod that refining imaging studies would offer an edge in early detection and cure. In preparing the report, the 16-member interdisciplinary panel heard experts describe the status of emerging technologies and reviewed the peer-reviewed literature, holding two public workshops and several meetings. The heart of the report addresses how new government policies could enhance technology development, research, and ultimately adoption and broad use of the most effective breast cancer screening technologies. The report strikes a balance between advocating measures that will expedite research, development, and adoption of effective breast cancer screening tools, but at the same time, making sure that technologies not proven effective are not unleashed too soon. The panel examined the potential for a variety of novel imaging techniques to improve on full-field mammography in the screening setting. Digital mammography, computer aided detection, ultrasound, elastography, magnetic resonance imaging, optical imaging, thermography, and electric impedance imaging, were among the imaging techniques that the committee concluded showed the most promise. At an even earlier stage of development, they reviewed thermoacoustic computed tomography, microwave imaging, Hall effect imaging, magnetomammography, and three-dimensional interactive visualization. An ideal screening test should be able to discriminate between the lethal and nonlethal cancers, according to Henderson, but such a test does not yet exist. He warns against early adoption of tests in the pipeline. The report underscores this point: "If the information generated by new technologies cannot be acted upon appropriately to improve outcomes, then women are not likely to benefit from the technological advances. Furthermore, as imaging methods become better and better at finding very small, early lesions such as carcinoma in situ, treatment decisions can be difficult to make because so little is known about the malignant potential of these premalignant cells. As a result, some women may face the diagnosis of breast cancer and the subsequent therapy for a lesion that may never have become a lethal, metastatic cancer." Basic Breast Cancer Biology Comprehending the basic biology of the breast is the most pressing area to pursue, the panel concludes. "We still don't know the really bad actors," explains panelist Jean J. Latimer, an investigator at Magee Women's Research Institute, and assistant professor of obstetrics, gynecology, and reproductive sciences at the University of Pittsburgh. "We need to understand normal human breast, but we are relying on rat breast in much research." Latimer points out that because breast cancer is a very heterogeneous disease, only by studying human breast tissue, will that heterogeneity become better understood. According to the panel, the top priority should be given to funding research into molecular markers, translational research, building specimen banks, purchase and operation of high-throughput technologies to study genetic and protein markers, and new bioinformatics approaches. To expedite this work, the panel calls for expansion of breast cancer specimen banks, as well as shoring up researcher access to patient samples. Latimer points out that bench researchers can tap many resources to pursue biological studies of breast cancer. Researchers will need more than advances in biology, however, to develop a better screening test. The committee recommended that the Food and Drug Administration (FDA) provide early approval for diagnostic imaging devices conditional on the manufacturer eventually demonstrating effectiveness in a screening situation. With FDA approval contingent on manufacturers agreeing to do definitive studies, the IOM panel acknowledged that such studies would require thousands of participants in screening studies. One strategy for expediting such research would be to have Medicare and private insurers conditionally cover the test for screening only when enrollees participate in these screening studies, the report recommends. Industry sponsors and the National Cancer Institute (NCI) would pick up the other trial expenses. Even though the up-front costs for insurers would be high, in the long run, covering these costs in screening studies would prove far less costly than permitting broad use before the test's effectiveness for screening has been demonstrated. Detection of early invasive cancer would be an acceptable endpoint to meet muster for FDA approval for screening simply because the ideal endpoint, disease-specific mortality, would take too long to get in most cases, and because mammography screening has already shown a mortality reduction, the panel claimed. But the panel took pains to be certain that not too low a bar was recommended before the FDA approval for screening was permitted. Given the current lack of understanding of the biology of these lesions, the panel considered evidence of premalignant or preinvasive breast lesions an unsatisfactory outcome measure. To further expedite getting answers about emerging detection technologies, the panel urged NCI to sponsor large randomized trials every 10 to 15 years to reevaluate the effects of accepted screening techniques on clinical outcomes. Peter Greenwald, director of NCI's division of cancer prevention, applauded many aspects of the report, but he questions the feasibility of conducting screening clinical trials every 10 to 15 years, citing expense. Furthering this agenda would involve NCI establishing a "permanent infrastructure for testing the efficacy and clinical effectiveness of new technologies as they emerge," wrote the panel. Yet NCI questions the need for such an entity, which many spokespersons view as "redundant." Not so, contends Henderson: "We have hit a plateau. We need to be sure that we set up and reaffirm systems that can move the field forward." Laura Newman (newman@nasw.org) is a freelance science writer in New York. References 1. S. Shapiro, "Periodic screening for breast cancer: the HIP randomized controlled trial Health Insurance Plan," Journal of the National Cancer Institute Monograph, 1997:27-30. 2. National Cancer Policy Board, Institute of Medicine and Commission on Life Sciences, National Research Council, Mammography and Beyond: Developing Technologies for the Early Detection of Breast Cancer, Washington, D.C.: National Academy Press, 2001. Resources on the Internet Cancer Genome Anatomy Project cgap.nci.nih.gov Early Detection Research Network edrn.nci.nih.gov National Cancer Institute Initiatives www.nci.nih.gov National Cancer Institute Technology Funding Opportunities otir.nci.nih.gov/tech/funding.html

Developing the Ideal Breast Cancer Screening Test

Interested in reading more?

Become a Member of