Breast Imaging in 2008
Lawrence W. Bassett, MD, FACR, Director of the Iris Cantor Breast Imaging Center, David Geffen School of Medicine at UCLA, radiologist and researcher
This is a summary of a lecture presented on April 8, 2008.
Breast imaging techniques continue to evolve and are now more effective than ever at detecting cancers of the breast. In addition to improvements of existing technology, there are new techniques with proven benefits that are being used at leading institutions around the world. The three technologies that are considered to be “proven technologies” are mammography, breast ultrasound and breast MRI.
Mammography
One of the first mammograms was done in New York in 1940 by Stafford Warren, who later became the founding Dean of the UCLA School of Medicine. Since that time radiologists have used different types of mammography with increasing levels of clarity and detail. Between 1976 and 2000, screening mammography with film was the most commonly used technique until being replaced by a new standard, digital mammography.
There has been significant debate about whether increased use of mammography has led to a decrease in mortality from breast cancer. In 2002, Duffy and colleagues published a seminal study conducted in Sweden in which breast cancer mortality rates in seven Swedish counties were compared in pre-screening and post-screening eras, and then compared with counties with no screening. Each county utilized the same treatments so decreases in mortality could not be attributed to how patients were treated. The study found that overall about 33% of the population was screened. There was a 30% reduction in breast cancer mortality in those counties where screening had been offered. However, not every woman in these counties actually was screened. When this variable was taken into consideration and they examined the population of women who were actually screened they found a substantial 45% drop in breast cancer deaths when compared to women not screened. These data underscore the value of mammography for decreasing mortality from breast cancer.
Because mammography is successful in identifying tumors at an early stage, another issue has arisen regarding how to classify tumors. For many years physicians have used what is called the TNM staging system for breast cancer. In this system the T (tumor), N (lymph node) and M (distant spread of tumor) are classified. Tumors were classified according to size: T1 represents tumors less than or equal to 2 cm in size, T2 represents tumors 2 cm to 5 cm in size, and T3 represents tumors greater than 5 cm in size. Fortunately, almost all women are now falling within the T1 category which provides very little information about which women at various stages might be doing better or worse. In 2002, Dr. Bassett, along with lead author Dr. Singletary, published a new staging system, which has been adopted by the American Joint Committee on Cancer (AJCC), that articulates the staging standards to be used. Dr. Singletary wrote, “The need for substantial changes in the staging system for breast cancer stemmed from continuing developments in breast cancer diagnosis and management. First, with the widespread use of screening mammography most breast tumors are first detected when they are very small…”
The new TNM staging system for breast cancer includes five new levels less than T2 in size. The first is DCIS, which represents the pre invasive stage or carcinoma in situ. Before mammography screening only about 5% of all breast cancers were found in situ. However, today after the introduction of screening mammography 30% of cancers detected are in situ. T1 continues to represent all invasive tumors that are less than 2 cm in their greatest dimension; however, there are now four sub-levels of T1. T1mic represents all tumors that have only microinvasion and are 0.1 cm or less. T1a represents tumors from 0.1 cm to 0.5 cm. T1b represents tumors great than 0.5 cm up to 1 cm. T1c represents tumors greater than 1 cm up to 2 cm.
Breast tissue that is fattier is less opaque; it is easier to see breast cancer lesions in less opaque tissue. Tissues that are denser are considered less fatty. Sometimes as women age and have less estrogen their breasts become less dense, but this is not always the case. There are four levels of breast density. Type 1 is almost entirely fatty tissue, Type 2 is primarily fatty with scattered fibroglandular densities, Type 3 is heterogeneously dense, and Type 4 is extremely dense. The sensitivity of mammography goes down with increased density. Thus, it is most difficult to find a cancer in Type 4 breast tissue.
Many have wondered if there is a way to increase the sensitivity of mammography for dense breasts. The Digital Mammography Imaging Screening Trial (DMIST) published in 2005 was an NCI sponsored multi-institutional trial with 50,000 screening exams in which film-screen was compared to digital mammography. Digital mammography was more accurate in women with dense breasts (which included women under 50, premenopausal women and women on hormones). Today about 30% of all mammography units are digital. All mammography at UCLA is digital. There are other advantages to digital mammography. Because it creates computerized images, specific areas of concern can be blown up and looked at more closely. There is no need for film libraries, thus reducing lost films. This can reduce the cost of maintaining space-intensive libraries, especially in facilities with limited space. While some might be concerned about digital images being lost this is not a concern because of regular back-ups. For example, UCLA stores back-ups both here in Los Angeles and in a GE storage unit in Chicago. Digital mammography can create greater access for patients in rural areas where there may not be a radiology breast specialist to read them; digital films taken in remote areas can be quickly sent to centers where there are experts to read them. In addition, Centers can get “outside” films in minutes rather than weeks.
There is a new digital technology technique going into trial called “Breast Mammography Tomosynthesis.” This new digital technology offers the advantage of acquiring a three dimensional data set of the entire breast which can be used to reconstruct and look at slices of it. This cross-sectional analysis will help dissect the breast in imaging. Preliminary ues of this technology suggests that it may help eliminate some false positives, in which women are called back for more mammography. UCLA is installing one of these machines and will be using it in trials to evaluate whether women who get called back for additional mammography might not have needed to return had the Tomosynthesis process been used.
There is a significant crisis developing regarding mammography. In the near future there may be insufficient numbers of well trained radiologists with expertise in reading mammograms to read all of the mammograms. In a national survey conducted in 2002, 30% of the medical practices surveyed reported having unfilled positions for radiologists who read mammograms. Facilities with job openings had long appointment waiting times for screening exams. Thirty-five percent reported financial losses in breast imaging due to poor insurance reimbursements. As physician shortages and financial constraints increase so, too, does the population of women 40 years and older who are in need of mammograms. Adding to this problem is the fact that of the 106 radiology training programs, most of which are attached to medical schools, there were more than 570 job vacancies, 70% of them in breast imaging. In addition, young trainees are not pursuing careers in breast imaging. A survey of residency training program in the US found that 65% would not consider a fellowship in breast imaging if it was offered. Some of the reasons for not wanting to interpret mammograms included believing that the technology was not particularly interesting, fear of lawsuits due to missed cancers─which is the leading reason for malpractice lawsuits, the stresses associated with missing a cancer, and low pay (breast imaging specialists are usually paid less than others). Breast imaging fellowships that were offered at 53 programs were only able to recruit 43 individual. These issues were finally recognized by the Institute of Medicine and National Research Council of the National Academies in 2004; they acknowledged that increased access to quality mammography is needed to reduce cancer deaths and the shortage of screening specialists should be addressed to deal with the capacity crisis. Stay tuned on this topic. Dr. Bassett established a program at UCLA for training these specialists; since its inception the program has trained 62 specialists.
Breast Ultrasound: The Promises and Limitations
In 1980 when ultrasound began being used for the breast, there was very limited differentiation that was visible. Ultrasound is now able to distinguish the skin, subcutaneous fat, fibroglandular tissue, pectoral muscles and chest wall. This makes ultrasound a very helpful tool in detecting the presence of breast cancer. The primary accepted use of breast ultrasound is for identifying and further characterizing palpable and non-palpable abnormalities, e.g., solid tumors respond differently to the sound waves than do cysts. Ultrasound of the breast is also good for guiding a needle biopsy procedure. It is recommended as an initial method to evaluate masses in women under 30, lactating and pregnant women. In a woman with a suspicious ultrasound of the breast, a mammogram should be performed even if the woman is under 30. Many people wonder about the use of ultrasound for screening. Unlike mammography, ultrasound is actually better at detecting masses in dense breasts, but not in detecting pre-cancer as ultrasound does not identify calcifications which can be an indication of pre-cancer. There are no data on whether ultrasound as a screening technique will be helpful in reducing the mortality due to breast cancer.
The National Cancer Institute and the Avon Foundation sponsored a study (ACRIN 6666) which evaluated the role of breast ultrasound screening in high risk women. It was run as a multi-center study in 20 different sites with over 2,500 women enrolled in the study. The results indicate the ultrasound screening detected 10% additional cancers in women with dense breasts. The downside is that the ultrasound also reported a large number of false positives. A false positive means that the ultrasound found a questionable mass, a biopsy was done, but it turned out to be a benign mass. One of the questions that this study raises is whether the benefits are worth the risks. Is this an acceptable rate of false positives or unacceptable in biopsies, costs and anxiety?
Breast Magnetic Resonance Imaging (MRI)
There has been increasing interest in the use of Magnetic Resonance imaging (MRI) for breast cancer detection. MRI is used to detail images of the body in any plane. It does not use radiation, but instead uses a powerful magnetic field to align the magnetization of hydrogen atoms in the body. Radio waves are used to systematically alter the alignment of this magnetization, causing the hydrogen atoms to produce a rotating magnetic field detectable by the scanner. This signal is used to reconstruct an image of the body. Since 1993 there have been at least 5 significant studies evaluating the sensitivity and specificity of the MRI technique for breast cancer detection. Sensitivity is the percent of cancers in the breast that were detected by MRI and specificity is the percent of suspicious findings leading to biopsy that were cancer. The range in sensitivity is 91-99%; however, the specificity range has been from 28-83%.
The current thinking regarding the use of breast MRI is that it is very important in certain situations; however, it should never be a substitute for a conventional imaging workup. Breast MRI is important as a pre-surgical evaluation to determine the extent of disease in the breast and/or in the other breast. It can also be used as a post-surgical evaluation to determine if there is any residual cancer. It can be used to evaluate axillary (arm pit) lymph nodes. It can be helpful in patients who get neoadjuvant chemotherapy (chemotherapy before surgery often used to shrink the tumor to a size that allows a lumpectomy rather than a mastectomy). The pre- and post-breast MRI can evaluate the impact of the neoadjuvant chemotherapy.
Recently the American Cancer Society recommended that breast MRI screening be done for women at high risk for breast cancer (over 25-30% increased risk). Breast MRI screening should also be used for women who have the BR-CA1 and/or BR-CA2 gene mutation. Breast MRI needs to be done with contrast in order to be able to be effective in detecting cancers. Breast MRIs should also be done between the 7th and 12th day of a menstruating woman’s menstrual cycle as hormonal changes can affect the readings.
Other Imaging Techniques: Real or Hype
There have been hosts of imaging techniques that have been around through the years; many new techniques being studied have promise. Thermography, however, is not one that has potential even though it has been around for a long time and continues to be reintroduced as a potential screening tool. It usually involves wearing a bra-like device that measures the temperature at different areas of the breast, ostensibly to detect the presence of tumors. Increased blood flow to tumors would theoretically result in higher temperatures. There is no scientific evidence that this works and it has not been evaluated in comparison to the standard techniques. Its advantage is that it has no side effects. Because it has no harmful effects, the FDA approved this device. Once a device is approved by the FDA others can also be approved by demonstrating that they are as good as the existing approved device. If a device has no side effects or harmful effects, it will not be pulled from use. Therefore, many of these thermography devices have FDA approval; however, they do not detect cancers and certainly not small cancers.
Scintimammography is another tool that has received some attention. Initial studies in the late 1990’s with over 2000 patients showed that scintimammography detected tumors with an average size of 2.4 cm but two thirds of the tumors were already palpable. There was only 40%-60% accuracy for lesions less than 1 cm. The limiting factor to this technique is that there is not enough resolution to detect small cancers plus newer devices show more resolution.
There are several other techniques that are under investigation including positron emission tomography/mammography, breast CT scanning and electromagnetic imaging techniques such as electrical impedance (EI) scanning, microwave Imaging (MI) spectroscopy, and near infrared (NIR) spectroscopy. None of these are ready for use on patients. While some show the promise of being effective we need to conduct clinical trials to show their efficacy and value.
Summary
Remember there is a lot of hype about breast cancer treatment and detection. Hype is defined as “extravagant or intensive publicity or promotion” or “a deception carried out for the sake of publicity” It is also defined “to promote or publicize (a product or idea) intensively, often exaggerating its importance or benefits.” Alternative imaging techniques are and will continue to be developed. Some will succeed while others will fail, but it is important to trust scientific studies and turn to reliable resources for medical care to avoid the hype and make the right choices. The future is likely to involve a multi-modality approach to breast cancer detection.
For reprint authorization, contact SimmsMannCenter@mednet.ucla.edu.