Page 1

Executive Summary

Breast cancer takes a tremendous toll in the United States. After lung cancer, breast cancer is the second leading cause of death from cancer among women in the United States and is the most common non-skin-related malignancy among U.S. women. Each year, more than 180,000 new cases of invasive breast cancer are diagnosed and more than 40,000 women die from the disease. Until research uncovers a way to prevent breast cancer or to cure all women regardless of when their tumors are found, early detection will be looked upon as the best hope for reducing the burden of this disease. The hope is that early detection of breast cancer by screening could be as effective at saving lives as the Papanicolaou smear (Pap smear) used for cervical cancer screening.

Early detection is widely believed to reduce breast cancer mortality by allowing intervention at an earlier stage of cancer progression. Clinical data show that women diagnosed with early-stage breast cancers are less likely to die of the disease than those diagnosed with more advanced stages of breast cancer. A thorough annual physical breast examination and monthly breast self-examination can often detect tumors that are smaller than those found in the absence of such examinations, but data on the ability of physical examinations alone to reduce breast cancer mortality are limited. X-ray mammography, with or without a clinical examination, has been shown in randomized clinical trials both to detect cancer at an earlier stage and to reduce disease-specific mortality. As a result, screening mammography has secured a place as part of routine health maintenance procedures for women in the United States. The mortality



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
Page 1 Executive Summary Breast cancer takes a tremendous toll in the United States. After lung cancer, breast cancer is the second leading cause of death from cancer among women in the United States and is the most common non-skin-related malignancy among U.S. women. Each year, more than 180,000 new cases of invasive breast cancer are diagnosed and more than 40,000 women die from the disease. Until research uncovers a way to prevent breast cancer or to cure all women regardless of when their tumors are found, early detection will be looked upon as the best hope for reducing the burden of this disease. The hope is that early detection of breast cancer by screening could be as effective at saving lives as the Papanicolaou smear (Pap smear) used for cervical cancer screening. Early detection is widely believed to reduce breast cancer mortality by allowing intervention at an earlier stage of cancer progression. Clinical data show that women diagnosed with early-stage breast cancers are less likely to die of the disease than those diagnosed with more advanced stages of breast cancer. A thorough annual physical breast examination and monthly breast self-examination can often detect tumors that are smaller than those found in the absence of such examinations, but data on the ability of physical examinations alone to reduce breast cancer mortality are limited. X-ray mammography, with or without a clinical examination, has been shown in randomized clinical trials both to detect cancer at an earlier stage and to reduce disease-specific mortality. As a result, screening mammography has secured a place as part of routine health maintenance procedures for women in the United States. The mortality

OCR for page 1
Page 2 rate from breast cancer has been decreasing in the United States by about 2 percent per year over the last decade, suggesting that early detection and improved therapy are both having an impact on the disease. Mammography is not perfect, however. Routine screening in clinical trials resulted in a 25 to 30 percent decrease in breast cancer mortality among women between the ages of 50 and 70. A lesser benefit was seen among women ages 40 to 49. The benefit of screening mammography for women over age 70 is more difficult to assess because of a lack of data for this age group from randomized clinical trials. Screening mammography cannot eliminate all deaths from breast cancer because it does not detect all cancers, including some that are detected by physical examination. Some tumors may also develop too quickly to be identified at an early, “curable” stage using the standard screening intervals. Furthermore, it is technically difficult to consistently produce mammograms of high quality, and interpretation is subjective and can be variable among radiologists. Mammograms are particularly difficult to interpret for women with dense breast tissue, which is especially common in young women. The dense tissue interferes with the identification of abnormalities associated with tumors, leading to a higher rate of false-positive and false-negative test results among these women. These difficulties associated with dense tissue are especially problematic for young women with heritable mutations who wish to begin screening at a younger age than what is recommended for the general population. Mammography can also have deleterious effects on some women, in the form of false-positive results and overdiagnosis and overtreatment. As many as three-quarters of all breast lesions that are biopsied as a result of suspicious findings on a mammogram, turn out to be benign; that is, the mammographic findings were falsely positive. (Many tissue biopsies performed on lumps found by physical examination are also benign, but the false-positive rate for physical examination has not been carefully studied.) “Overdiagnosis” is the labeling of small lesions as cancer or precancer when in fact the lesions may never have progressed to a life-threatening disease if they had been left undetected and untreated. In such cases, some of the “cures” that occur after early detection may not be real, and thus, such women are unnecessarily “overtreated.” Technical improvements in breast imaging techniques have led to an increase in the rate of detection of these small abnormalities, such as carcinoma in situ, the biology of which is not well understood. Currently, the methods for classification of such lesions detected by mammography are based on the appearance of the tissue structure, and the ability to determine the lethal potential of breast abnormalities from this classification is crude at best. The immense burden of breast cancer, combined with the inherent limitations of mammography and other detection modalities, have been the driving forces behind the enormous efforts that have been and that

OCR for page 1
Page 3 continue to be devoted to the development and refining of technologies for the early detection of breast cancer. The purpose of the study described in this report was to review the breast cancer detection technologies in development and to examine the many steps in medical technology development as they specifically apply to methods for the early detection of breast cancer. The study committee was charged with surveying existing technologies and identifying promising new technologies for early detection, and assessing the technical and scientific opportunities. The committee was further charged with examining the policies that influence the development, adoption, and use of technologies. Funding for the study was provided by seven independent foundations and individuals, including the Breast Cancer Research Foundation, the Carl J. Herzog Foundation, Mr. John K. Castle, the Jewish Healthcare Foundation, the Josiah Macy, Jr., Foundation, the Kansas Health Foundation, and the New York Community Trust. TECHNOLOGIES IN DEVELOPMENT Most of the progress thus far in the field of breast cancer detection has resulted in incremental improvements in traditional imaging technologies. These technical advances have likely led to more consistent detection of early lesions, but clinical trials have not been undertaken to determine whether their use has also resulted in a greater reduction in breast cancer mortality compared with that of older technologies. Many technical improvements have been made to mammography since its initial introduction. One recent example is full-field digital mammography (FFDM). FFDM systems are identical to traditional film-screen mammography (FSM) systems except for the electronic detectors that capture and display the X-ray signals on a computer rather than directly on film. This digital process provides the opportunity to adjust the contrast, brightness, and magnification of the image without additional exposures. Many consider FFDM to be a major technical advance over traditional mammography, but studies to date have not demonstrated a meaningful improvement in screening accuracy. Although one could argue that studies thus far have not directly tested the full potential of FFDM through the use of “soft-copy” image analysis (on a computer screen as opposed to film), difficulties remain with regard to the limited resolution and brightness of the soft-copy display. The technology could potentially improve the practice of screening mammography in other ways, for example, by facilitating electronic storage, retrieval, and transmission of mammograms. Computer-aided detection, through the use of sophisticated computer programs designed to recognize patterns in images, has also shown potential for improving the accuracy of screening mammography, at least among less experienced readers. However, questions remain as to how

OCR for page 1
Page 4 this technology will ultimately be used and whether it will have a beneficial effect on current screening practices. Other breast imaging technologies approved by the Food and Drug Administration (FDA) include ultrasound, magnetic resonance imaging (MRI), scintimammography, thermography, and electrical impedance imaging (Table 1). Ideal detection performance may ultimately depend on TABLE 1 Current Status of Imaging and Related Technologies Under Development for Breast Cancer Detection Current Status Technology Screening Diagnosis FDA approved for breast imaging/detection Film-screen mammography (FSM) +++ +++ Yes Full-field digital mammography (FFDM) ++ ++ Yes Computer-assisted detection (CAD) ++ o Yes Ultrasound (US) + +++ Yes Novel US methods (compound, three-dimensional, Doppler, harmonic) o o No Elastography (MR and US) o o No Magnetic resonance imaging (MRI) + ++ Yes Magnetic resonance spectroscopy (MRS) −/o a +/o a No Scintimammography o + Yes Positron emission tomography (PET) o o Yes Optical imaging o + No Optical spectroscopy − o No Thermography o + Yes Electrical potential measurements o + No Electrical impedance imaging o + Yes Electronic palpation o NA No Thermoacoustic computed tomography, microwave imaging, Hall effect imaging, magnetomammography NA NA No NOTE: This table is an attempt to classify a very diverse set of technologies in a rapidly changing field and thus is subject to change in the near future. aEx vivo analysis of biopsy material/in vivo MRS. Current Status Explanation of Scale - Technology is not useful for the given application NA Data are not available regarding use of the techonology for given application o Preclinical data are suggestive that the technology might be useful for breast cancer detection, but clinical data are absent or very sparse for the given application. +Clinical data suggest the technology could play a role in breast cancer detection, but more study is needed to define a role in relation to existing technologies ++Data suggest that technology could be useful in selected situations because it adds (or is equivalent) to existing technologies, but not currently recommended for routine use +++Technology is routinely used to make clinical decisions for the given application

OCR for page 1
Page 5 multimodality imaging, as no single imaging technology to date can accurately detect all significant lesions. Ultrasound and MRI in particular have shown potential as adjuncts to mammography for diagnostic and screening purposes, especially for women in whom the accuracy of mammography is not optimal, such as those with dense breasts. MRI and ultrasound imaging may also facilitate new minimally invasive methods for the treatment of early lesions that are under investigation, but clinical trials are needed to assess the value of the procedures. Many additional technologies are at earlier stages of development, but to date, it appears that no quantum steps forward have been taken in this area. Furthermore, improved imaging technologies that allow detection of more lesions at an earlier, precancer stage may or may not lead to reduced breast cancer mortality and may lead to more overtreatment of women. The dilemma of overtreatment could potentially be overcome by coupling imaging technologies with biologically based technologies, such as functional imaging, that can determine which lesions are likely to become lethal. The benefit of discovering early lesions could also be enhanced by developing new and effective preventive and therapeutic interventions that are minimally invasive and more acceptable to women. Thus, a great deal of work remains to be done to optimize the benefits and minimize the risks of breast cancer screening. A number of technologies that may help to define the biological nature of breast lesions are being developed, including culture of breast cancer cells in the laboratory, measurement of protein expression in cancer cells, identification of markers of cancer cells or the proteins that they secrete in blood or breast fluid, or identification of genetic changes in tumors ( Table 2). Further progress in this field will depend on the establishment, maintenance, and accessibility of tissue specimen banks, as well as access to new high-throughput technologies and bioinformatics. Technologies based on biology could potentially contribute to improved patient outcomes in several ways. For example, they could distinguish between early lesions that require treatment because they are highly likely to become lethal and those that are not. In many instances, these technologies could also potentially identify fundamental changes in the breast that appear before a lesion can be detected by current imaging methods. Thus, they may identify women at high risk of developing breast cancer or, more importantly, women at high risk of dying from breast cancer. Such women could then undergo more frequent screening or would perhaps benefit from newer imaging technologies. Some women might also choose to explore a “risk reduction strategy” that would affect all breast cells (e.g., bilateral prophylactic mastectomy), although current strategies for risk reduction are less than ideal. Improved understanding of the biology and etiology of breast cancer could also lead to better prevention strategies, which would further increase the benefits of early detection.

OCR for page 1
Page 6 TABLE 2 Technologies Under Development for Biological Characterization and Detection of Breast Cancer Detection Objective Techniques Stage of Development Current Limitations Identify germ-line mutations associated with cancer risk Various genetic tests for BRCA1, BRCA2, p53, and AT In clinical use ≤10% of women with breast cancer affected Survival benefit not proven DNA arrays Preclinical stage Identify polymorphisms associated with cancer risk Various sequencing techniques Used as research tools Value for predicting cancer risk poorly defined Identify and characterize somatic mutations and epigenetic changes Fluorescent In Situ Hybridization (FISH), Comparative Genomic Hybridization, Polymerase Chain Reaction (PCR) for loss of heterozygosity (LOH), DNA arrays, DNA methylation assays Used as research tools Diagnostic and prognostic values of mutations are poorly defined; tumors are heterogeneous Measure changes in RNA expression in cancer cells a Northern analysis Reverse transcription-PCR, cDNA array Used as research tools Same as directly above; difficult to assess small samples; Need computer algorithms for array data

OCR for page 1
Page 7 Measure changes in levels of protein expression in cancer cellsa Immunohistochemistry (IHC) Same as directly above Some IHC in clinical use Two-dimensional gel electrophoresis, Mass spectroscopy All used as research tools Identify markers of cancer in breast fluids Nipple aspiration Breast lavage Used as research tools, early stage of clinical testing Sensitivity and specificity are low; appropriate markers to be examined are not clear Identify markers of cancer or risk in serum or blood Various tests for protein markers Some in clinical use for monitoring disease Others are research tools Same as directly above Isolation and analysis of cancer cells in circulation Early stage of development Culture of breast cells Various tissue culture methods Used as research tools Primary cultures have been difficult to grow Prognostic value of newer methods unproven aPotential targets of functional imaging.

OCR for page 1
Page 8 TECHNOLOGY DEVELOPMENT PROCESS The pathway from technical innovation to accepted clinical practice is long, arduous, and costly. Although the activity and investment in research aimed at developing new technologies for early breast cancer detection have increased substantially over the last decade, biomedical research has also become more complex and capital intensive. Moreover, in addition to the developers of new technologies, many groups participate in the process, including FDA, health care insurers and managed care organizations, and other technology assessment institutions. These public and private organizations and policy makers play a role in evaluating medical technologies at various points along the way, making decisions about FDA approval, insurance coverage, and reimbursement that ultimately determine whether new technologies will be adopted and disseminated. Those who evaluate the potential of new technologies consider many factors, including clinical need, technical performance, clinical performance, economic issues, and patient and societal perspectives. Government funding of research in the health care sector has traditionally focused primarily on basic scientific discovery, but recently, a new emphasis on the translation of science into practice through the development of technology has received considerable attention, including the creation of joint public- and private-sector initiatives. The private sector has made considerable investment in this area as well, although private investment in breast imaging technologies appears to be less attractive than investment in other areas of the health care industry. A variety of factors may contribute to this phenomenon, but it is likely due to the perception that there is a high degree of economic risk in this field, including considerations of the time and resources needed to develop technologies, the size of the potential market, and the remuneration possible. The end results of research are always unpredictable, but for medical devices, the requirements for FDA approval and insurance coverage have been variable and unpredictable, adding additional levels of risk to the development process. Furthermore, because technical innovations are often first introduced into the system in rather crude form, it can be difficult and problematic to judge them solely on the basis of their early versions. ASSESSMENT OF NEW TECHNOLOGIES The dominant framework for medical technology regulation and evaluation has historically been based on therapeutics, whereas early detection relies on screening and diagnostic methods. The evaluation of therapeutic and detection technologies, however, may be intrinsically different. The stages of development for drugs are more standardized, and

OCR for page 1
Page 9 therapeutic interventions generate direct outcomes that can be observed in patients. In contrast, most patient-level effects of screening and diagnostic tests are mediated by subsequent therapeutic decisions. Screening and diagnostic tests also generate information that is subject to interpretation. Furthermore, this information is only one of the inputs into the decision-making process. Hence, the evaluation of detection technologies is fundamentally an assessment of the value of information. The development process for devices also tends to be iterative, and thus, assessment at early stages of development may not recognize the full potential of a new medical device. That is, most technologies that ultimately achieve widespread use go through successive stages of development, variation, and appraisal of the actual experience in the market. With the exception of mammography, new breast cancer detection technologies have been evaluated by diagnostic studies that primarily measure sensitivity (the proportion of people with the disease who test positive) and specificity (the proportion of people without the disease who test negative). Even if the technologies ultimately are intended to be used for screening, they are generally not evaluated through screening studies that measure health outcomes. Adoption of new detection technologies for screening purposes before assessment of their effects on clinical outcome has been common and quite problematic for technologies used to screen for other diseases because data on detection accuracy are not adequate to assess the potential value of new technologies for screening. The ideal end points for assessment of screening technologies are reductions in disease-specific mortality or morbidity, or both, but the clinical trials needed to measure those end points are quite large, lengthy, and costly. Surrogate end points for morbidity and mortality are difficult to define because the net effect of new detection technologies could be either positive (more accurate detection, leading to lower breast cancer mortality) or negative (capable of identifying more lesions but not changing disease-specific mortality and thus leading to greater morbidity and higher screening costs). TECHNOLOGY DISSEMINATION After the hurdles of FDA approval, insurance coverage, and reimbursement have been cleared, the adoption and dissemination of new breast cancer detection technologies will ultimately depend on whether women and their health care providers find them acceptable. Much is already known about the adoption and dissemination of screening mammography, and this knowledge may prove instructive for other developing technologies. Experience from current mammography programs suggests that outreach to women, education of women and providers, and

OCR for page 1
Page 10 access to facilities and services are all essential components of successful dissemination. The use of screening mammography has increased greatly in the last decade, but a significant number of women still do not get screened, and many others do not undergo screening at the recommended intervals. Women often express concerns about discomfort from the procedure, the inconvenience of scheduling an annual test, lack of access to screening facilities, and fear of what could be found (including false-positive results). Studies indicate that physician recommendation is the single most influential factor in determining whether women are screened. Access to screening facilities may be particularly difficult for women who lack health insurance. The National Breast and Cervical Cancer Early Detection Program was established through the Centers for Disease Control and Prevention with the goal of providing screening examinations for uninsured women. The program has grown considerably since it was launched 10 years ago, but it still only reaches about 12 to 15 percent of eligible women nationwide. New federal legislation that would allow Medicaid coverage for treatment of breast cancer detected through the program was recently passed, but adoption of this program by the states is pending. As more women adopt the practice of routine screening and the number of women eligible for screening mammography increases (because of the aging U.S. population), there will be increased demands for trained mammographers and certified screening facilities. There are anecdotal reports that inadequate numbers of mammographers and mammography technologists are being trained to fulfill current and future needs, but quantitative data to support these assertions are not available. Concerns have also been expressed among radiologists and health care administrators that the reimbursement rate for mammography is too low to cover the procedure's actual costs (including the costs of complying with federally mandated quality standards, which are unique to mammography) and that this situation could lead to a reduction in the availability of screening services. Quantitative data are unavailable to confirm or refute these concerns. If the rate of reimbursement for mammography truly is artificially low, then cost comparisons with new technologies may also unfairly favor mammography. When mammography was introduced, it was a “void-filling” technology and thus had no competition during the dissemination process. New technologies face a much different scenario. Evaluation will likely include comparison with mammography, and adoption of a new technology will require competition with other detection technologies that are currently available. A goal of new technologies is to provide additional choices for women and their physicians, allowing an individualized approach to screening and diagnosis depending on a woman's specific needs

OCR for page 1
Page 11 and characteristics. At the same time, new technologies may add layers of complexity to the decision-making processes associated with screening and diagnosis, making it more challenging to establish practice guidelines and to define a standard of care. RECOMMENDATIONS The committee's recommendations fall into two general categories: those that aim to improve the development and adoption processes for new technologies (Recommendations 1 to 5) and those that aim to make the most of the technologies currently available for breast cancer detection (Recommendations 6 to 10). 1. Government support for the development of new breast cancer detection technologies should continue to emphasize research on the basic biology and etiology of breast cancer and on the creation of classification schemes for breast lesions based on molecular biology. A major goal of this research should be to determine which lesions identified by screening are likely to become lethal and thus require treatment. This approach would increase the potential benefits of screening while reducing the potential risk of screening programs. Funding should focus on the development of biological markers and translational research to determine the appropriate uses and applications of the markers, including functional imaging. Research on cancer markers should focus on screening as well as on downstream decisions associated with diagnosis and treatment. Funding priorities should include specimen banks (including specimens of early lesions), purchase and operation of high-throughput technologies for the study and assessment of genetic and protein markers, and new bioinformatics approaches to the analysis of biological data. 2. Breast cancer specimen banks should be expanded and researcher access to patient samples should be enhanced. Health care professionals and breast cancer advocacy groups should educate women about the importance of building tumor banks and encourage women to provide consent for research on patient samples. Stronger protective legislation should be enacted at the national level to prevent genetic discrimination and ensure the confidentiality of genetic test results. The National Cancer Institute (NCI) should devise and enforce strategies to facilitate researcher access to the patient samples in specimen banks. For example, the costs associated with the sharing of samples with

OCR for page 1
Page 12 collaborators should be included in the funding for the establishment and maintenance of the specimen banks, and specimen banks supported by government funds should not place excessive restrictions on the use of the specimens with regard to intellectual property issues. 3. Consistent criteria should be developed and applied by the Food and Drug Administration (FDA) for the approval of screening and diagnostic devices and tests. Guidance documents for determination of “safety and effectiveness,” especially with regard to clinical data, should be articulated more clearly and applied more uniformly. Given the complexity of assessing new technologies, the FDA advisory panels could be improved by including more experts in biostatistics, technology assessment, and epidemiology. 4. For new screening technologies, approval by the Food and Drug Administration (FDA) and coverage decisions by the Health Care Financing Administration (HCFA) and private insurers should depend on evidence of improved clinical outcome. This pursuit should be streamlined by coordinating oversight and support from all relevant participants (FDA, NCI, HCFA, private insurers, and breast cancer advocacy organizations) at a very early stage in the process. Such an approach should prevent technologies that have been approved for diagnostic use from being used prematurely for screening in the absence of evidence of benefit. Technology sponsors generally lack the resources and incentive to undertake large, long-lasting, and expensive screening studies, but a coordinated approach would make it easier to conduct clinical trials to gather the necessary outcome data. The proposed process should provide for the following: FDA should approve new cancer detection technologies for diagnostic use in the traditional fashion, based on evidence of the accuracy (sensitivity and specificity) of new devices or tests in the diagnostic setting. In the case of “next-generation” devices (in which technical improvements have been made to a predicate device already on the market), technical advantages such as patient comfort or ease of data acquisition and storage could be considered in the determination of approval. If a new device that has been approved for diagnostic use shows potential for use as a screening tool (based on evidence of accuracy) and the developers wish to pursue a screening use, an investigational device exemption should be granted for this use and conditional coverage should be provided for the purpose of conducting large-scale screening trials to assess clinical outcomes.

OCR for page 1
Page 13 Trials should be designed and conducted with input from FDA, NCI, HCFA, the Agency for Healthcare Research and Quality, and breast cancer advocacy organizations. Informed consent acknowledging the specific risks of participating in a screening trial would be necessary. HCFA and other payers should agree to conditionally cover the cost of performing the test in the approved clinical trials, whereas NCI and the technology's sponsors should take responsibility for other trial expenses. Participation by private insurers would be particularly important for the assessment of new technologies intended for use in younger women who are not yet eligible for Medicare coverage. Although this expense may initially seem burdensome to private insurers, the cost of providing tests within a clinical trial would be much less than the costs associated with broad adoption by the public (and the associated pressure to provide coverage) in the absence of experimental evidence for improved clinical outcome. Trial data should be reviewed at appropriate intervals, and the results should determine whether FDA approval should be granted (for those deemed sufficiently effective) and coverage should be extended to use outside of the trials. (A prior approval for diagnosis would remain in place regardless of the decision for screening applications.) The ideal end point for clinical outcome is decreased disease-specific mortality. However, given the length of time required to assess that end point and the fact that early detection by screening mammography has already been proven to reduce breast cancer mortality, a surrogate end point for breast cancer detection is appropriate in some cases. As a general rule, a screening technology that consistently detects early invasive breast cancer could be presumed efficacious for the purposes of FDA approval. Detection of premalignant or preinvasive breast lesions, however, cannot be assumed to reduce breast cancer mortality or increase benefits to women, and it is not an appropriate surrogate end point for FDA approval, given the current lack of understanding of the biology of these lesions. 5. The National Cancer Institute should create a permanent infrastructure for testing the efficacy and clinical effectiveness of new technologies for early cancer detection as they emerge. The NCI Breast Cancer Surveillance Consortium and the American College of Radiology Imaging Network may provide novel platforms for this purpose through the creation of databases and archives of clinical samples from thousands of study participants. 6. The Health Care Financing Administration should analyze the current Medicare and Medicaid reimbursement rates for mammography, including a comparison with other radiological techniques, to de-

OCR for page 1
Page 14 termine whether they adequately cover the total costs of providing the procedure. The cost analysis of mammography should include the costs associated with meeting the requirements of the Mammography Quality Standards Act. A panel of external and independent experts should be involved in the analysis. 7. The Health Resources and Services Administration (HRSA) should undertake or fund a study that analyzes trends in specialty training for breast cancer screening among radiologists and radiologic technologists and that examines the factors that affect practitioners' decisions to enter or remain in the field. If the trend suggests an impending shortage of trained experts, HRSA should seek input from professional societies such as the American College of Radiology and the Society of Breast Imaging in making recommendations to reverse the trend. 8. Until health insurance becomes more universally available, the U.S. Congress should expand the Centers for Disease Control and Prevention screening program to reach a much larger fraction of eligible women, and state legislatures should participate in the federal Breast and Cervical Treatment Act by providing funds for cancer treatment for eligible women. The Centers for Disease Control and Prevention should be expected to reach 70 percent of eligible women (as opposed to the current 15 percent). This objective is based on the stated goals of the U.S. Department of Health and Human Services' Healthy People 2010 report, which by the year 2010 expects 70 percent of women over age 40 to have had a recent (within the last 2 years) screening mammogram. 9. The National Cancer Institute should sponsor large randomized trials every 10 to 15 years to reassess the effects of accepted screening modalities on clinical outcome. These trials would compare two currently used technologies that are known to have different sensitivities. Breast cancer-specific mortality would be the principal outcome under evaluation. Such studies are needed because detection technologies and treatments are both continually evolving. Hence, the benefit of a screening method may change over time. 10. The National Cancer Institute, through the American College of Radiology Imaging Network or the Breast Cancer Surveillance Consortium, should sponsor further studies to define more accurately the benefits and risks of screening mammography in women over age 70. As the age distribution of the U.S. population continues to shift toward older ages, the question of whether these women benefit from screening mammography will become increasingly important.