Digital vs. Film Mammography in the Digital Mammographic Imaging Screening Trial (DMIST): Questions and Answers
- Posted: September 16, 2005
- The study showed that digital mammography was significantly better that film mammography in screening women who were under age 50, or women of any age who had very dense breasts.
- Digital mammography allows improvement in image storage and transmission because images can be stored and sent electronically.
- Secondary goals measuring the relative cost-effectiveness of both digital and film technologies, and the effect on participant quality of life due to the expected reduction of false positives are still being assessed and will be reported at a later date.
- Death rates from breast cancer have been declining since 1990, and these decreases are believed to be the result, in part, or earlier detection and improved treatment.
- How is digital mammography different from film mammography?
Both digital and film mammography use X-rays to produce an image of the breast.
In film mammography, which has been used for over 35 years, the image is created directly on a film. While standard film mammography is very good, it is less sensitive for women who have dense breasts. Prior studies have suggested that approximately 10 percent to 20 percent of breast cancers that were detected by breast self-examination or physical examination are not visible on film mammography. A major limitation of film mammography is the film itself. Once a film mammogram is obtained, it cannot be significantly altered; if the film is underexposed, for example, contrast is lost and cannot be regained.
Digital mammography takes an electronic image of the breast and stores it directly in a computer. Digital mammography uses less radiation than film mammography. Digital mammography allows improvement in image storage and transmission because images can be stored and sent electronically. Radiologists also can use software to help interpret digital mammograms. One of the obstacles to greater use of digital mammography is its cost, with digital systems currently costing approximately 1.5 to 4 times more than film systems.
- How was DMIST conducted?
The Digital Mammographic Imaging Screening Trial (DMIST), begun in October 2001, enrolled 49,528 women, who had no signs of breast cancer, at 33 sites in the United States. On the appointment day, women provided background health information and filled out brief questionnaires. They also had both digital and film mammograms taken on that day, each with a minimum of two views of each breast. Two different certified radiologists interpreted the conventional and digital mammogram exams for each individual patient. All radiologists who participated read both types of mammograms, and each radiologist read approximately an equal number of mammograms of each type.
Participants were asked to return in one year for their annual mammogram. At that time, a mammogram was performed as part of routine health care. Women who were not able to return to the same site as in year one were asked to submit films from another institution for review by study radiologists.
- Why was DMIST important?
For women, breast cancer is the most common non-skin cancer and the second leading cause of cancer-related death in the United States. Death rates from breast cancer have been declining since 1990, and these decreases are believed to be the result, in part, of earlier detection and improved treatment.
DMIST was performed to measure relatively small, but potentially clinically important, differences in diagnostic accuracy between digital and film mammography. While any differences that were detected might be relatively small, they could improve breast cancer detection for all or some groups of women.
Digital mammography is a newer technology that is becoming more common. Currently, approximately 8 percent of breast imaging units provide digital mammography. Past trials of digital mammography have shown no difference in diagnostic accuracy between digital and film mammography. The U.S. Food and Drug Administration (FDA) trials and three smaller screening trials showed no significant difference in the performance of digital mammography vs. film mammography. These studies were limited, however, because they each included only one type of digital detector and had relatively small numbers of patients, perhaps limiting their ability to detect small differences in diagnostic accuracy.
- Who were the women who enrolled in DMIST?
Over 49,500 women who were requesting their usual breast cancer screening mammogram were recruited at 33 sites in the United States and Canada. The women had no breast cancer symptoms, and they agreed to undergo a follow-up mammogram at the same participating site or provide their mammograms from another institution for review one year from study entry. All women reviewed and signed the study consent form.
The following women were ineligible:
- pregnant women
- women with breast implants
- women who had undergone a screening mammogram in the past 11 months
- women with a focal dominant lump, which is defined as a single lump felt by a woman or her doctor
- women with a bloody or clear nipple discharge
- women with a history of breast cancer treated with lumpectomy
Breast cancer status for DMIST participants was determined through available breast biopsy information within 15 months of study entry or through follow-up mammography ten months or later after study entry.
Who organized the study and how much did it cost?
The American College of Radiology Imaging Network (ACRIN) coordinated the study. ACRIN is a Cooperative Group sponsored by the Division of Cancer Treatment and Diagnosis at the National Cancer Institute (NCI). Enrollment began in October 2001. On Nov. 14, 2003, DMIST reached its targeted 49,500 participant recruitment goal. ACRIN is a network of physicians, scientists, and medical institutions that have joined together to conduct clinical trials of new medical imaging technologies. The total cost of the digital mammography trial was about $26 million.
Which digital mammography equipment was included in DMIST?
General Electric Medical Systems, Fuji Medical Systems, Fischer Imaging, and Hologic digital mammography systems were tested in the trial. Of these, all except for the Fuji system are already FDA-approved and available for clinical use in the United States.
How important is reader training in interpreting digital mammography?
Breast cancer has a very similar appearance on digital and film mammograms, but the display of the images on monitors instead of film requires additional reader (radiologist) training. Under the federal law that governs mammography in the U.S. (the Mammography Quality Standards Act), radiologists who switch from interpreting film to interpreting digital mammography must undergo some additional training.
- What were the main results of DMIST?
DMIST showed that, for the entire population of women studied, digital and film mammography had very similar screening accuracy.
Digital mammography was significantly better in screening women who fit any of these three categories:
- under age 50 (no matter what level of breast tissue density they had)
- of any age with heterogeneously (very dense) or extremely dense breasts
- pre- or perimenopausal women of any age (defined as women who had a last menstrual period within 12 months of their mammograms)
There is no apparent benefit of digital over film mammography for women who fit ALL of the following three categories:
- over age 50
- those who do not have dense or heterogeneously (very dense) breast tissue
- those who are not still menstruating
In addition, there was no statistically significant difference in the accuracy of digital mammography compared to film according to digital mammography machine type, race, or breast cancer risk.
These results suggest that for women who fall into three subgroups (women under age 50, women with heterogeneously dense or extremely dense breasts, and pre- and perimenopausal women), digital mammography may be better at detecting breast cancer than traditional film mammography. Approximately 65 percent of the women in DMIST fit into one of the three subsets that showed a benefit with digital mammography.
Some earlier studies had suggested that digital mammography would result in fewer false positives than film mammography, but the rates of false positives for digital mammography and traditional mammography were the same in DMIST.
- How many of the study participants were diagnosed with cancer?
During the course of the study, including initial screening and follow-up, 335 women were diagnosed with cancer. In general, cancers detected by either film or digital mammography were similar in histology (microscopic structure) and stage (how advanced they were).
However, lesions detected by digital mammography and missed by film in women under age 50, in women with heterogeneously dense or extremely dense breasts, and in pre- and peri-menopausal women, included many invasive cancers and medium and high grade in situ lesions. Many of these cancers were confined to the breast at diagnosis; that is, they had not yet spread to the lymph nodes under the arm. These are precisely the lesions that must be detected early to save more lives through screening. In situ lesions in the breast are those confined to the breast duct without invading the surrounding breast tissue and are known as DCIS, or ductal carcinoma in situ.
Neither digital nor film mammography found all the breast cancers in the study population. Women who develop lumps, breast changes, or symptoms after screening mammography should report them to their physician even if their mammogram showed no signs of breast cancer.
- Was the screening mammography performed in DMIST, both digital and film, accurate?
Both digital and film mammography had sensitivities (the ability to tell if a cancer is present) for breast cancer of 70 percent in the overall study population using the conventional methods for measuring sensitivity in a breast cancer screening trial.
The sensitivity for women with dense breasts was only 55 percent for film mammography while the sensitivity for digital mammography was 70 percent. The overall rate of 70 percent is within the expected rate of detection.
Specificities, or the ability to tell correctly that a cancer is NOT present when the breasts are normal, were high for DMIST, just as would be expected of screening mammography. Using one year follow-up interval, specificities for both digital and film mammography were 92 percent for the overall population.
- Do the trial results suggest that women's lives will be saved if they undergo digital mammograms and they are in one of the three groups that showed a benefit with digital mammography?
DMIST was not designed to study breast cancer mortality, as that would have been a much longer and more costly study. The improved diagnostic accuracy of digital mammography in the subgroups of women found in DMIST may not translate into saved lives. However, the types of lesions that were found by digital mammography and not by film in these subgroups of the women were the types of cancers that can lead to death -- that is, invasive cancers without evidence of metastasis to axillary lymph nodes (lymph nodes under the arm and next to the breast that is affected by cancer) at the time of diagnosis and medium and high-grade in situ lesions (DCIS).
Randomized clinical trials that have studied mortality have shown a reduction in mortality from breast cancer with the use of screening mammography, ranging from 18 percent to 30 percent depending on the age of the women. DMIST results indicate that screening with digital mammography will detect at least as many breast cancers as film mammography over the whole population, and more advanced or serious breast cancers in women in the three subsets of the population. This suggests that at least as many -- and possibly more -- lives will be saved with digital mammography as are now saved by screening with film mammography.
- Do the trial results indicate that ALL women should get digital mammograms instead of film mammograms for breast cancer screening?
No. The study results indicate that only women who fit in ANY of these three categories would benefit from digital mammography instead of film mammography:
- under age 50 (regardless of level of breast tissue density)
- of any age, with heterogeneously (very dense) or extremely dense breast tissue
- or pre- or perimenopausal women of any age (defined as women who had a last menstrual period within 12 months of their mammograms).
According to the results, women who fit ALL of the following three categories would not benefit from digital mammography instead of film mammography:
- over age 50
- those who do not have dense or heterogeneously (very dense) breast tissue
- and those who are no longer menstruating.
At present, only 8 percent of the mammography units in the United States have digital systems, whereas approximately 40 percent of women undergoing screening mammography have dense breasts. It will be impossible for all women who have dense breasts to receive digital mammograms, at least for the near future. As more digital mammography systems become available, more women in the groups who are likely to benefit from digital mammography will have access to this technology.
What were the secondary goals of this trial and what were those results?
Secondary goals included:
- measurement of the relative cost-effectiveness of both digital and film technologies, as digital mammography costs 1.5 to 4 times more than film mammography
- the measurement of the effect on participant quality of life due to the expected reduction of false positives
The results of these parts of the study are still under analysis and will be presented at a later date. In fact, even though a reduction in false positives with digital mammography was expected, none was found in DMIST. The effect of false positive results on quality of life will be reported at a later date.
Reader (radiologist) studies also were conducted using the mammograms obtained in DMIST. What were the studies and what were the results?
Seven controlled reader studies were used to measure the following:
- diagnostic accuracy of softcopy (displayed on a computer monitor) vs. printed film display for digital mammography
- effect of disease prevalence (percent of trial participants who actually had breast cancer) on reader interpretation performance
- effect of breast density on the diagnostic accuracy of digital mammography vs. film mammography
- diagnostic accuracy of each of the four individual digital mammography units vs. film mammography
The analysis of the reader studies has not been completed at this time.
INFORMATION FOR WOMEN
How can women obtain digital mammograms?
Film mammography is still much more common that digital mammography. Women who would like to have digital mammograms can ask their doctors or contact local hospitals or imaging centers to find out if digital mammography is available in their area.
Should women who live in communities that don't have digital mammography facilities, or don't have enough available machines, delay their next mammograms until they can have digital mammograms? Should women in the affected categories try to get digital mammograms before they are scheduled for their next mammogram?
Women should have their next mammogram when they are scheduled for it. It would be better to have a film mammogram when a woman is supposed to have her next mammogram than for her to delay her screening in order to get a digital mammogram. Women should not defer screening with mammography just because of a lack of access to digital mammography. Film mammography has been successfully used as a screening tool for breast cancer for over 35 years.
Women should not receive an extra mammogram because of these trial results. That is, if a woman has had a mammogram in the last year, and she has no breast signs or symptoms, she should undergo her next screening mammogram only when she is due for one, not earlier than she would ordinarily be scheduled.
What should women do who have larger, dense breasts that require multiple digital exposures to obtain an accurate view of a portion of the breast instead of one larger image for each view? Is additional dose an issue?
Only the GE FDA-approved digital mammography system provides an imaging area that is smaller than that used for the usual film mammogram. This means that some women with large breasts will require extra images with digital mammography instead of only two images of each breast with film mammography.
In the DMIST study, 19 percent of all participants required one or more extra views with digital mammography over and above what was required with film mammography in order to include all portions of their breasts. About 25 percent of those women scanned with the GE system, however, required more images with digital mammography than they did for their film mammograms. The women who required the extra images who have dense breasts are typically the women for whom digital mammography was shown to be more accurate than film.
In considering this issue, it is important to realize that increasing the number of images per view does not dramatically increase the dose because not all breast tissue is exposed in each view. For example, taking two digital images of the breast instead of one film mammogram does not double the dose overall, because only a portion of the breast is exposed twice. On average, the larger number of digital images required is more than offset by the lower doses delivered by digital mammography for women with thicker and denser breasts.
How does a postmenopausal woman over age 50 determine if she has extremely dense or heterogeneously dense breasts?
At present, this can only be determined by a prior mammogram. Usually the density rating on mammography should be noted in the written report from the interpreting radiologist who reads the mammogram. If it is not included in the mammography report, it can be determined by a radiologist or qualified mammography technologist by viewing a prior mammogram. Women who are uncertain about their density status should inquire about it at the time of their next mammography visit.
If a woman has dense breasts, will she have dense breasts for the rest of her life?
Breast density can change over time. Most frequently, breast tissue becomes less dense with age. Estrogen replacement therapy, menopause, and weight loss or gain can change a woman's breast density. If a woman has questions about her breast density, she can discuss it with her primary care physician or the staff at the clinic where she receives her mammograms.
Does getting a digital mammogram feel similar to getting a film mammogram?
From a woman's perspective, a digital mammography examination is similar to a traditional mammography examination. Positioning and compression of the breast are identical.
What other breast imaging techniques might be useful for breast cancer screening?
In addition to mammography, ultrasound and magnetic resonance imaging (MRI) are both sometimes used to screen for breast cancer. ACRIN is currently conducting another trial of breast cancer screening, which compares ultrasound vs. mammography in high- risk women. MRI has shown promise for women at high-risk for breast cancer. DMIST did not study either of these other technologies. In fact, women who participated in DMIST were not permitted to participate in other screening trials during the one year immediately before and after their entry into DMIST.
There are no multicenter clinical trials investigating the use of either MRI or ultrasound in place of mammography as screening tools for breast cancer for the general population of women over age 40.
MISCELLANEOUS ADDITIONAL INFORMATION
Are there other possible advantages of digital mammography over film mammography?
Digital mammography offers other advantages over film, including improved ease of image access, transmission, retrieval and storage, and lower average radiation dose without a compromise in diagnostic accuracy. In addition, digital mammograms are less likely than film mammograms to be lost.
What levels of radiation are used for digital mammography vs. film mammography?
In DMIST, digital mammograms required approximately three quarters the radiation dose of film mammography. However, the dose in film mammography is quite low and poses no significant danger to patients.
How many women are screened with mammography annually in the United States?
The FDA reports that there are about 33.5 million mammography procedures performed per year in the United States. Data from 2000-2002 show that about 70 percent of all mammograms that are performed annually are for screening purposes (to detect cancer as opposed to following cancer once it has been diagnosed). This translates to about 23.5 million screening procedures every year.
What are the costs of different types of mammograms?
Reimbursement by Medicare in 2005 for film-screen mammograms is $85.65 and for digital screening mammography (for women with two breasts, as opposed to those who have undergone mastectomy) is $135.29. Actual cost for mammograms will vary by region and the form of reimbursement.
What is telemammography?
Telemammography is the transmission of digital mammograms electronically so that they might be interpreted in a remote location. This can be accomplished through wireless networks (such as through satellites) or through more traditional wire-based networks. This may allow access to experts and second opinions more quickly for digital mammograms, particularly for women in underserved areas. Of course, second opinions are also available with film mammography by shipping mammograms using mail and other delivery services.
Who were the investigators who conducted the study?
The study was led by Etta D. Pisano, M.D., of the Department of Radiology and Biomedical Engineering, the Biomedical Research Imaging Center and the Lineberger Comprehensive Cancer Center of the University of North Carolina at Chapel Hill.
Edward Hendrick, Ph.D., with the Department of Radiology, Northwestern University, Chicago, Ill., was co-principal investigator of the study. Martin Yaffe, Ph.D., Department of Imaging Research and Medical Imaging of the University of Toronto, Toronto, Ontario, Canada, was the lead physicist. The Center for Statistical Sciences at Brown University, Providence, R.I., provided statistical coordination for the study under the direction of Constantine Gatsonis, Ph.D. Dennis Fryback, Ph.D., at the University of Wisconsin at Madison, directed the health-related quality of life analysis and Anna Tosteson, Sc.D., at Dartmouth Medical School, Hanover, N.H., directed the cost-effectiveness evaluation. Data/Image collection and study coordination was performed at the American College of Radiology, ACRIN headquarters, located in Philadelphia, Pa.
The citation for the online publication of the study results in the New England Journal of Medicine is: Pisano E, Gatsonis C, Hendrick E, Yaffe M, Baum J, Acharyya S, Conant E, Fajardo L, Bassett L, D'Orsi C, Jong R, and Rebner M. Diagnostic Performance of Digital versus Film Mammography for Breast Cancer Screening - The Results of the American College of Radiology Imaging Network (ACRIN) Digital Mammographic Imaging Screening Trial (DMIST). NEJM, published online September 15, 2005 and in print on October 27, 2005.
For a press release about this study, please go to http://www.nci.nih.gov/newscenter/pressreleases/DMISTrelease.
For more information about cancer, visit the NCI Web site at http://www.cancer.gov or call NCI's Cancer Information Service at 1-800-4 CANCER (1-800-422-6237).