Strategies for Improving Breast Cancer Detection and Diagnosis


Copyright © 2005 National Academy of Sciences
All right reserved.

ISBN: 978-0-309-09213-5

Chapter One

Executive Summary

The outlook for women with breast cancer has improved significantly since 1989 as the mortality rate has declined steadily, a decline attributed both to earlier detection through wider use of mammography screening and to improved treatments. Yet breast cancer remains a major problem, second only to lung cancer as a leading cause of death from cancer for women. This year over 200,000 new cases will be diagnosed and about 40,000 women-most diagnosed in earlier years-will die from the disease.

As their basic understanding has improved, researchers have discovered that breast cancer is far from simple. The disease has many forms that follow many pathways. Some are swift and lethal while others may never progress. Unfortunately, the tools available today cannot distinguish between the small pre-invasive lesions that will become lethal and those that will not. Consequently, most breast cancers are treated as if they were destined to be lethal and many women undergo difficult treatments, such as mastectomy, radiation, and chemotherapy, that might never have been needed.

Current treatments for breast cancer range from the relatively simple, but daunting, procedure known as lumpectomy, which removes cancerous and surrounding breast tissues, to the modified radical mastectomy in which an entire breast and the adjacent lymph nodes are excised. Both may be accompanied by chemotherapy and/or radiation therapy. None of these treatments, however, is guaranteed to save a woman's life, and, because so little is understood about the cellular mechanisms and processes that govern cancer progression, no one can predict with certainty which patients will be "cancer survivors" after treatment.

To date, no way to prevent breast cancer has been discovered and experience has shown that treatments are most effective when a cancer is detected early, while still small and contained and before it has spread to other tissues. Those two facts suggest that, at the present time, improving early detection and diagnosis is the most effective way to continue reducing the toll from breast cancer.

Several years ago an Institute of Medicine (IOM) and National Research Council (NRC) committee examined the array of promising detection and diagnostic technologies then in various stages of development, and concluded that mammography, while far from perfect, was still the best choice for screening the general population to detect breast cancer at early and treatable stages. Their findings and recommendations were published in 2001 in Mammography and Beyond: Developing Technologies for Early Detection of Breast Cancer.

For a variety of reasons, many women do not undergo regular screening. These reasons include limited availability of screening in some areas, inadequate insurance coverage, and misunderstanding of the value of screening. Also, some women are so afraid of breast cancer they choose not to be screened. Others find the procedure painful. The fact that mammography does not work equally well for all women, especially those with dense breast tissue, is a further complication.

In addition, the potential for false-positive and false-negative results remains high. Studies suggest that, due to a lack of sensitivity leading to false-negative findings, mammography screening may miss as many as 1 in 6 tumors. At the other extreme, the risk of a false-positive result is about 1 in 10, meaning that about 1 in 10 suspicious findings on a screening mammogram are false alarms. About three-quarters of suspicious areas biopsied as a result of a mammogram turn out to be benign-though only after a woman has endured the fear that she has breast cancer and borne the costs and discomfort of additional medical procedures.

In 2002, the IOM and NRC named a second committee to examine which of the approaches identified in Mammography and Beyond held the greatest promise for improving early detection and diagnosis. In addition, this group was asked to both identify and recommend ways to overcome and/or circumvent barriers to the development, evaluation, and, finally, incorporation into clinical practice of those strategies with the greatest potential.

Charged with developing a rational and workable framework for the early detection and diagnosis of breast cancer, the committee was also given the broader, and in some ways more formidable, challenge of improving the understanding of both the media and the general public of the public health issues that both underlie and impede the development of new approaches, including the role of regulatory policies and insurance coverage.

With Mammography and Beyond as a starting point, the committee identified several potential approaches: broader access to and use of mammography, better quality mammography, or the development of new technologies. They concluded that for the immediate future, broader and better use of mammography holds the greatest potential to save lives.

Even the most promising of the new technologies, committee members determined, will probably lead only to incremental improvements in existing technologies, and will not replace them. Indeed, finding ways to ensure those incremental advances are integrated into existing systems holds more immediate promise for improving outcomes for breast cancer patients than attempts to isolate a single new technology that might replace mammography. Important avenues of research and development for exciting technologies, such as biological markers of cancer and molecular profiling, although still in their infancy, are especially promising as diagnostic tools.

Simply identifying promising technologies, even those proven through extensive clinical trials, would have no value unless those technologies are suitable for and adopted in clinical practice so they become available to the women who might benefit. Because most clinical trials for cancer detection are designed to evaluate a single technology and do not provide information that might help physicians choose which competing approaches would most benefit patients, the questions asked of new technologies should be which should be used and when, not which is best. As the committee reminded, breast cancer is a complex disease that passes through numerous critical stages, each requiring different tools for detection and diagnosis, and demanding different sets of decisions.

The first decision, of course, is whether a woman decides to be screened for breast cancer, a decision that depends, in part, on a woman's perception of her own breast cancer risk, which is often distorted. For many women, the very topic of breast cancer provokes confusion and dread. Many women overestimate their risk of getting and dying of breast cancer before the age of 50, a finding mirrored in the many magazine articles that suggest a significant risk of breast cancer in younger women. Much of their information comes from news reports and advertisements in the mass media, and more recently the Internet, which tend to emphasize dramatic, unusual, and extreme examples rather than balanced and factual presentations.

Extensive, and sometimes inaccurate, media coverage of recent controversies about the effectiveness of screening mammography has contributed to public confusion about the value of mammography, its role in breast cancer detection, and the ages at which it is most likely to be beneficial. Also, glowing reports of "medical breakthroughs" and "promising" technologies that have not been submitted for approval or even tested in patients add another layer of confusion and uncertainty.

Physicians face different kinds of decisions. When confronted with an abnormal mammogram, they must decide which technology will provide the most expedient and reliable result and, then, how much faith to put in that result. At present, they receive little research-based guidance about emerging technologies, which combinations of technologies, and which approaches would be most effective for certain groups of patients.

The committee included clinicians involved in breast cancer screening, detection, and treatment; experts in cancer and molecular biology; those with expertise in clinical studies, as well as those involved with the development, evaluation, and adoption of medical technology and with experience in health care administration.

To supplement their own considerable expertise, members held a number of background workshops and heard from a range of technology developers, researchers, and leaders of clinical studies designed to improve systems for early detection and diagnosis. They also discussed the many issues involved in assessing new medical technologies with senior staff at the federal agencies and with representatives of private insurance groups, all the groups that act as gatekeepers for medical technology.

Based on this information and their lengthy deliberations, the committee identified four major categories for recommendations aimed at improving early detection and diagnosis of breast cancer: improve current application of screening mammography; integrate biology, technology, and risk models to develop new screening strategies; improve the environment for research and development; and improve the implementation and use of new technologies. The detailed rationale and supporting data for each category are in the body of the report. A brief summary of pertinent findings, together with the recommendations, follows (recommendations are also listed separately in the box at the end of this summary).


A growing shortage of radiologists who specialize in reading mammograms, coupled with an imbalance between the closures and openings of screening facilities, has created unacceptable delays in some parts of the country. At the same time the number of false-positive readings appears to be increasing, possibly due to increasing defensive medicine in reaction to the frequency of malpractice litigation.

Improving screening practices to reduce the number of false positives could reduce the costs of additional testing by an estimated $100 million per year, in addition to eliminating the mental anguish and the possible need for a biopsy for thousands of women and also cutting unnecessary waiting time. Though no one knows the actual costs of settling malpractice suits, since so many are settled out of court, these settlements are thought to contribute to the ever-escalating costs of malpractice insurance for radiologists who read mammograms, a trend that discourages physicians from entering the profession.

Given these, and other, factors, the committee sought ways to optimize the productivity of radiologists who interpret mammograms and, at the same time, improve their accuracy. They looked toward the experience of other countries, notably the United Kingdom, and their organization of screening services. Although differences in the number of "excess" biopsies due to false-positive readings were difficult to assess, for even within the United States significant regional variations exist, committee members did identify elements in the programs of some European countries, as well as Canada and Australia, that could be useful in the United States, which has limited national or regional standards or programs for breast cancer screening. For instance, in the United Kingdom radiologic technologists, who are not physicians, are trained to meet national certification standards, and have proven comparable in accuracy and speed to radiologists.

Also, the British National Breast Cancer Screening Program invites every woman for a screening mammogram, which is paid for through the National Health Service-but only at three-year intervals. In the United States, the recommended screening interval is one year, which is likely to detect more cancers, but women do not get screened unless they are referred by health care providers or refer themselves. Many women are never screened because they lack adequate, if any, insurance coverage. That group tends to include underserved women in lower socioeconomic groups in whom breast cancer may not be detected at an early stage when still treatable.

A program that might be adapted by health care providers in the United States is the European Code Against Cancer which stresses that screening should be done within integrated breast care centers that have quality assurance programs. Another model is Britain's National Health Service Breast Screening Program, which has developed national quality assurance standards and a quality assurance network though which programs are regularly monitored, with results measured against established targets. In the United States no organization collects or monitors data to promote high performance levels and guidelines are only voluntary. (The Mammography Quality Standards Act [MQSA] requires facilities in the United States to collect quality data for internal use, but does not require the facilities to use the data in any specific or documented approach for quality improvement.)

In Sweden and the Netherlands, which both report low rates of false positives, screening takes place in outlying centers and diagnosis and workup takes place in centralized facilities. Great Britain has developed a quality assurance self-assessment program, the only one of its kind in the world, which, while voluntary, is used by 90 percent of that nation's radiologists to identify weaknesses and improve interpretive skills.

By contrast, in the United States screening services are rarely integrated within a comprehensive breast cancer center, and typically separated from treatment, counseling, and support services. The MQSA addressed the technical quality of mammograms, but does not require standards to improve delivery of services and quality of interpretation, or quality assurance and a continuing education program intended to enhance the accuracy of interpretation.

To improve services in the United States, the committee recommended:

Health care providers and payers should consider adopting elements of successful breast cancer screening programs from other countries. Such programs involve centralized expert interpretation in regionalized programs, outcome analysis, and benchmarking. (Recommendation A1)

At this time, one of the few regulations directly relating to the quality of interpretation in the United States requires physicians who interpret mammograms to read a minimum of 960 exams in a 24-month period, which averages out to 480 per year. By comparison, breast imaging specialists in the United Kingdom are required to read at least 5,000 each year.

A number of technologies under development have potential to improve the quality and accuracy of mammography interpretation. These include such technologies as computer-aided detection (CAD), which does not replace interpretation by a radiologist but can highlight areas of concern for further review by the radiologist. The greatest value of CAD may prove to be its potential to increase the performance level of general radiologists to that of those who specialize in breast imaging.

Too, the shortage of mammography personnel may actually impede the kinds of innovation that would improve their efficiency for experts are needed to both assess and properly use these new technologies.

To address these issues the committee recommended:

Breast imagers and technology developers should work in collaboration with health care providers and payers to improve the overall quality of mammographic interpretation by: (Recommendation A2)

adopting and further developing practices that promote self-improvement of breast imagers, but that do not jeopardize the workforce; and

developing technologies, such as CAD, that have the potential to improve quality, and expanding their use once they have been validated.


Excerpted from SAVING WOMEN'S LIVES Copyright © 2005 by National Academy of Sciences. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.