Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Breast Cancer Screening Among Women in Their Forties: An Overview of the Issues

Breast Cancer Screening Among Women in Their Forties: An Overview of the Issues Abstract This article summarizes the issues prompting a recent NIH Consensus Conference on mammography screening for women in their forties. To date, eight randomized controlled trials of breast cancer screening have been conducted, and a reduction in breast cancer mortality has emerged after 10 to 15 years of follow-up among women offered screening in their forties. No effect appears for at least eight years, and the reason for the delay, compared to that seen in women aged 50-69, is not clear. Two possibilities include cancer-stage shift due to screening in younger women and the aging of women into their fifties during the course of screening. Possible adverse effects of screening include radiation risk, although this is low, false-negative and false-positive screening tests, and overdiagnosis due to detection of ductal carcinoma in situ (DCIS). In order to make appropriate decisions regarding mammography, women need age-related information about both the benefits and potential risks of screening. Although 85% of breast cancers occur in women after they reach the age of 50, breast cancer is the number one cause of cancer death for American women aged 40-49. In 1993, it is estimated that 30,940 American women in this age group developed breast cancer and 4,843 died of it (Harras, A; personal communication). Each year, for every 100,000 women in their forties, 163 are diagnosed with breast cancer and 30 die of the disease (1). Women in their forties need information to understand their risk for breast cancer. Data from SEER statistics indicate that for every 1,000 American women turning 40 years old, approximately 16 will develop breast cancer at some time before their fiftieth birthday (1). How many of these women will survive the cancer? SEER statistics show that nationally, 52% of women under 50 years of age who were diagnosed with invasive breast cancer in 1973 were still living 18 years later (1). Few, if any, of these women were likely screened. In the Health Insurance Plan (HIP) study, the only randomized controlled trial (RCT) with 18-year follow-up data, 58% of women in the group not offered screening survived to 18 years (2). With the advent of improved therapies over the past two decades, the percentage of women surviving breast cancer is improving (3). Thus, of the 16 women out of a thousand who will develop breast cancer in their forties, at least eight, and probably more, will survive the cancer regardless of screening. Therefore, breast cancer screening for women in their forties is primarily directed at the eight or fewer women in every 1,000 who might be saved by earlier detection of the cancer. If screening decreases mortality by as much as 25%, it would save one or two of the 16 women in a thousand who develop breast cancer in their forties. Any potentially fatal illness striking persons in the prime of life is a terrible occurrence, but breast cancer is doubly so because it not only threatens a woman's life, but an emotionally and sexually important part of her body as well. Black and colleagues found that fear of breast cancer is so great that women in their forties overestimated their risk of dying of breast cancer 20-fold and their risk of developing breast cancer sixfold (4). With such a terrifying disease, it is important to find better ways to cure and prevent it. What can screening, especially screening with mammography, contribute to the control of breast cancer in women in their forties? When considering screening for a particular medical disease, usually three questions are asked relating to the burden of disease, the characteristics of the screening test, and the effectiveness of early treatment (Table 1).1 In particular, it is important to examine the mortality benefits that accrue from the intervention, its adverse effects, and its costs. My task, then, is to present an overview of these issues as they pertain to breast cancer screening in women aged 40 to 49. Breast Cancer Mortality Reduction Most attention in breast cancer control has been directed towards determining the effect of screening on breast cancer mortality. Eight RCTs of mammography, with or without clinical breast examination, have been conducted in four countries: the HIP study from the United States; the Kopparberg, Östergötland, Malmö, Stockholm, and Gothenburg studies from Sweden; the Edinburgh study from the United Kingdom; and the National Breast Screening Study (NBSS-I) from Canada. At the National Cancer Institute International Workshop on Breast Cancer Screening in 1993, all seven trials published at that time found mortality reductions among women aged 50-69, two with statistically significant results (5). A meta-analysis presented at the Workshop found a statistically significant 34% reduction in breast cancer mortality after seven years of follow-up among women aged 50-69, with a relative risk ratio for screened to nonscreened women of 0.66 (95% confidence interval [CI]: 0.55-0.79) (6). However, the findings among younger women were less clear. The meta-analysis showed no effect at seven years of follow-up, with a relative risk of 0.99 (95% CI: 0.74-1.32) or 1.08 (95% CI: 0.85-1.39), depending on whether or not results from the Canadian study were included. A new overview of all five Swedish studies was also presented at the Workshop (7), and it showed a statistically insignificant 10% to 13% mortality reduction at 12 years of follow-up among women aged 40-49 (Fig. 1).1 This overview was more current than the meta-analysis because it included results from the Gothenburg trial that had not been previously published and because all cases in the other studies were re-reviewed, leading to some reassignment of subjects and outcomes. Nevertheless, the new Swedish analysis did not alter the conclusion from the meta-analysis that by seven years of follow-up, no beneficial effect is seen in younger women. With the data presented from the eight randomized trials, the Report of the International Workshop concluded, “For [women aged 40-49 years] it is clear that in the first 5-7 years after study entry, there is no reduction in mortality from breast cancer that can be attributed to screening. There is an uncertain and, if present, marginal reduction in mortality at about 10-12 years. Only one study (HIP) provides information on long-term effects beyond 12 years, and more information is needed.” In March 1996, an updated meta-analysis of the studies' results was reported in Falun, Sweden, for women aged 40-49 (Fig. 2).2 Five of the eight showed mortality reductions after 10-15 years of follow-up, and three showed no benefit. Pooled results demonstrated mortality reductions, with relative risk ratios of 0.77 (95% CI: 0.59-1.10), 0.76 (95% CI: 0.62-0.93), or 0.85 (95% CI: 0.71-1.01), depending on which trials were included. As results of the trials continue to accrue, it appears that the time required to demonstrate beneficial screening effect, and the size of the effect, vary by age. Whereas mortality differences between screened and control groups began to emerge after only a few years of follow-up for women aged 50-69, the studies showed effects more slowly for women in their forties. In the combined Swedish studies (Fig. 11), mortality rates were similar in the invited and control groups during the first eight years of follow-up, after which a beneficial effect of screening began to emerge. This effect has continued to grow and is nearly statistically significant with three more years of follow-up (see Fig. 22, all Swedish trials). The same trend occurred in the HIP and Edinburgh studies. In most studies, breast cancer mortality reduction in younger women was less than in older women. The cause of the time difference in effect by age is not yet clear. It has been suggested that screening picks up such early cancers in younger women that it takes longer for mortality reductions to occur (8). Indeed, randomized trials have shown that screening shifts diagnosis of breast cancer to earlier stages. Also, detection of ductal carcinomas is more common with the use of mammography. However, for stage shift to cause the time difference, screening would have to shift the stage of cancer differently according to age. Analysis of the Kopparberg trial showed shifts to earlier-stage cancers among women in both the forties and fifties (9). Data from all trials, however, should be examined to determine whether and to what degree stage shift could explain the delayed effect of screening in younger women. The slower appearance of mortality reduction in younger women could also be partly due to “age creep.” Because screening studies occur over several years, some women who enter trials during their forties age into their fifties over the course of screening. It has been suggested that as women move into their fifties, when breast cancer screening is known to work, a benefit becomes apparent (10,11). Analyses of the RCTs are reported according to the age of women at entry into the trial, not their age at the time of diagnosis of breast cancer. This approach is necessary to preserve the comparability of the screened and control groups. Nevertheless, when there is the possibility that the effect of breast cancer screening varies by age, information about the age at diagnosis is needed. Most trials have not yet provided this information. The issue is especially important in the two trials in which only women 45 and older at entry were included. Two groups have reported data about this issue. In the HIP study, 32% of cancers in women aged 40-49 at entry were detected after the women had turned 50 (2). Shapiro et al. demonstrated that screened women aged 45-49 at entry benefited when their cancers were diagnosed after age 50 but not when their cancers were diagnosed earlier. The numbers of women in each subgroup, however, were small. On the other hand, Tabar and Duffy did not find any effect of age creep in the Swedish two-county trial in which 36% of cancers were diagnosed after women in their forties turned 50 (12). The relative mortality was 0.95 (95% CI: 0.44-2.03) for women in whom breast cancer was diagnosed after they turned age 50 and 0.85 (95% CI: 0.49-1.45) for women in whom breast cancer was diagnosed before age 50. Information from the other trials is needed. Ultimately, the degree to which age creep influences mortality effects for women in their forties will be best addressed by the results of the National Health Service Breast Screening Programme underway in the United Kingdom (13). In this large RCT, women aged 40 and 41 at entry are being screened annually for five years, which means that all breast cancers diagnosed will be in women who are still in their forties. Why would a screening test for breast cancer have differential effects by age? Part of the explanation may be the lower accuracy (both sensitivity and specificity) of screening tests in younger women (5). Also, breast cancer growth rates may differ by age of the woman. Tabar et al. found that the mean sojourn time (time in the preclinical detectable state) was 1.25 years for women in their forties and 3.03 years for women in their fifties (14). Whether and how estrogen levels and menopause, rather than age per se, influence effectiveness of breast cancer screening remains unclear and needs to be determined. The question as to when to start breast cancer screening should not be arbitrarily linked to a particular decade of a woman's life. It is important to determine the effect of screening in groups at high risk for breast cancer, especially in women aged 40-49. To date, no reports from the randomized trials have examined screening effects according to risk status of the participants. Adverse Effects Important possible adverse effects of breast cancer screening include radiation risk from mammography, adverse physical and psychosocial sequelae of false-positive and false-negative tests, and overdiagnosis. Radiation risk from modern mammography appears to be very low. Feig and Ehrlich reviewed recent estimates for lifetime radiation risk to the breast and reported that a single mammogram exposure is estimated to cause between 2.9 and 8.8 excess breast cancers per million women screened during their forties (15). The estimates varied according to the age of the woman and the method of calculation, with more recent estimates being lower than older ones. If women in their forties are screened annually, radiation risk might cause between 29 and 88 additional breast cancers per million women screened for an entire decade. Most previous work has analyzed the degree of accuracy of screening tests and the related problems of false-positive and false-negative results. The International Workshop Report summarized data from the randomized trials showing breast cancer screening in younger women is not as sensitive as in older women (5). Sensitivity by the detection method (ratio of screen-detected cancers to screen-detected plus interval cancers) among women aged 40-49 at study entry varied from 53% to 81% in the Stockholm, Swedish two-county, and Canada NBSS-I trials, while comparably defined sensitivity among women aged 50-59 varied from 73% to 88%. Thus, the risk of false-negative screening tests is greater in younger women. Recent data suggest that false-positive mammograms—those requiring further evaluation to rule out cancer diagnosis—are a substantial problem in the United States. In a national survey of community mammography facilities conducted in 1992 and 1993, Brown et al. found that 11% (95% CI: 9%-13%) of screening mammograms were read as abnormal; 10.6% were false-positive readings (16). At a state-of-the-art program in northern California, Kerlikowske et al. reported that 6.3% of first-screen mammograms among women aged 40-49 were abnormal, and 5.9% proved to be false-positive readings (17). Table 22 presents the percentage of women in each of these two studies who underwent additional procedures, including biopsies, following abnormal mammogram readings. On average, between one and two additional procedures were performed for each abnormal reading. Approximately 10% to 15% of women in each study underwent an invasive procedure. Lidbrink et al. have reported comparable results from the Stockholm trial (18). The psychological effect of false-positive mammograms has been studied by Lerman et al. (19). They found that among women with high-suspicion mammograms that subsequently proved to be false-positive, three months later 47% reported being quite anxious about mammography, 41% were quite anxious about breast cancer, 26% reported that worry affected their mood, and 17% reported that it adversely affected their daily function. Women with low-suspicion mammograms reported less concern, but even among these women anxiety about mammography and breast cancer was relatively high (29% and 40% respectively, compared to 24% and 29% in women with normal mammograms). In a study from Norway, 18 months after screening mammography, 29% of 126 women with false-positive mammograms reported anxiety about breast cancer, compared to 13% of 152 randomly selected women with negative mammograms (p = 0.001) (20). In Britain, Ellman et al. found that 25% of women with normal mammograms, 30% of women with false-positive mammograms, and 35% of women with breast symptoms but with normal mammograms had General Health Questionnaire scores indicating probable psychiatric morbidity (21). Because breast cancer screening is periodically repeated, it is important to determine the percentage of women who will experience a false-positive mammogram or clinical breast examination over an extended period of time. In the studies to date, it is clear that the percentage of false-positive mammograms decreases as women are rescreened. Nevertheless, it has been estimated that over a 10-year period of annual mammograms, as many as 30% of women in their forties will experience a false-positive mammogram or clinical breast examination (22). It is important to determine the actual number. Efforts to decrease the number of false-positive screening tests and their resultant adverse effects are also urgently needed. As screening for breast cancer has increased, detection of ductal carcinomas in situ (DCISs) has risen—328% from 1983 to 1992 (23). An increasing percentage of breast cancers detected by screening are DCIS. In the northern Californian study discussed above, 43% of cancers detected among women in their forties were DCIS (17). Some experts are concerned that early lesions such as DCIS have led to overdiagnosis of breast cancer and is partly responsible for the recent increased incidence of breast cancer (23). Finding breast cancer before any invasion, even microinvasion, has occurred should help save lives. However, there are a number of questions about DCIS. Pathologically, it appears difficult to diagnose. One study, for instance, asked six experienced pathologists to interpret 24 slides; there was complete agreement among the six in only two of the 10 cases in which at least one pathologist diagnosed DCIS (24). The prevalence and natural history of the condition are not clear and are important in determining if DCIS detection is leading to overdiagnosis. Finally, Ernster and colleagues have demonstrated that there is a wide range of treatment approaches for DCIS, not all of which may be appropriate (23). There is an urgent need for studies of all these issues. In sum, progress has been made in better understanding the breast cancer mortality reduction that might occur with a screening program for women in their forties. Randomized trials have demonstrated that for approximately a decade, no benefit occurs, but after 10 to 15 years, a 15% to 25% mortality reduction appears. This translates into one or two women per 1,000 who might be saved. The reasons for the delay in mortality benefits and the degree to which these benefits could be achieved by beginning screening later remain to be determined. Also less clear are other benefits that might occur from earlier screening, such as more limited surgery or less debilitating adjuvant therapy. To achieve these benefits, however, substantial numbers of women will experience adverse effects, especially those caused by false-positive mammograms and possible overdiagnosis because of DCIS. Finally, costs of screening programs and the resultant procedures carried out because of abnormal readings cannot be ignored. Women need information about all these issues. They rightly demand to be involved in an important decision about their lives and their bodies. Ultimately, it is the job of medical science to search for new and better ways to promote health, and along the way, to share with the public the very complicated facts as we understand them. Armed with facts, women can then apply their own set of values to cope with the important problem of breast cancer. Table 1. Criteria for deciding whether a medical condition should be included in periodic health examinations* 1. How great is the burden of suffering caused by the condition in terms of:    Death  Discomfort    Disease  Dissatisfaction    Disability  Destitution  2. How good is the screening test, if one is to be performed, in terms of:    Sensitivity  Cost  Labeling Effects    Specificity  Safety    Simplicity  Acceptability  3. a. For primary prevention, how effective is the intervention?   or   b. For secondary prevention, if the condition is found, how effective is the ensuing treatment in terms of:    Efficacy    Patient Compliance    Early treatment being more effective than later treatment  1. How great is the burden of suffering caused by the condition in terms of:    Death  Discomfort    Disease  Dissatisfaction    Disability  Destitution  2. How good is the screening test, if one is to be performed, in terms of:    Sensitivity  Cost  Labeling Effects    Specificity  Safety    Simplicity  Acceptability  3. a. For primary prevention, how effective is the intervention?   or   b. For secondary prevention, if the condition is found, how effective is the ensuing treatment in terms of:    Efficacy    Patient Compliance    Early treatment being more effective than later treatment  * Reprinted by permission from Fletcher R, Fletcher S, Wagner E. Clinical epidemiology—the essentials. Baltimore, Williams & Wilkins, 1996 View Large Table 2. Follow-up procedures for abnormal screening mammograms   NSMF study* All ages, %   Northern CA study** 40-49 years, first screen %   Clinical breast examination  6.8  3.2  Repeat mammogram  37.0  —  Additional mammogram  41.2  56.1  Ultrasonography  20.0  10.9  Fine needle aspiration  3.0  5.8  Needle biopsy  2.9  —  Excisional biopsy  10.5  13.0  Needle localization  —  11.0    NSMF study* All ages, %   Northern CA study** 40-49 years, first screen %   Clinical breast examination  6.8  3.2  Repeat mammogram  37.0  —  Additional mammogram  41.2  56.1  Ultrasonography  20.0  10.9  Fine needle aspiration  3.0  5.8  Needle biopsy  2.9  —  Excisional biopsy  10.5  13.0  Needle localization  —  11.0  * National Survey of Mammography Facilities. Data from (16). ** Data from (17). View Large Fig. 1. View largeDownload slide Overview of Swedish randomized trials. Cumulative breast cancer mortality (per 1000) up to 12 years after randomization by age at randomization. Solid line = invited group and dotted line = control group. Adapted with permission from Nystrom et al. Breast cancer screening with mammography: overview of Swedish randomized trials. Lancet 1993;341:973-978. Fig. 1. View largeDownload slide Overview of Swedish randomized trials. Cumulative breast cancer mortality (per 1000) up to 12 years after randomization by age at randomization. Solid line = invited group and dotted line = control group. Adapted with permission from Nystrom et al. Breast cancer screening with mammography: overview of Swedish randomized trials. Lancet 1993;341:973-978. Fig. 2. View largeDownload slide Meta-analysis of randomized controlled trials of breast cancer screening for women in their forties, presented at Falun, Sweden; March, 1996. Relative risks are presented for each study, all Swedish trials, all population-based trials, and all trials. Adapted with permission from Breast cancer screening with mammography in women aged 40-49 years. Int J Cancer 1996;68:693-699. © Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc., 1996. Fig. 2. View largeDownload slide Meta-analysis of randomized controlled trials of breast cancer screening for women in their forties, presented at Falun, Sweden; March, 1996. Relative risks are presented for each study, all Swedish trials, all population-based trials, and all trials. Adapted with permission from Breast cancer screening with mammography in women aged 40-49 years. Int J Cancer 1996;68:693-699. © Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc., 1996. References (1) Kosary CL, Ries LAG, Miller BA, Hankey BF, Harras A, Edwards BK. SEER Cancer Statistics Review, 1973-1992: Tables and graphs. Bethesda (MD): National Cancer Institute; 1995: DHHS Publ No. (NIH)96-2789. Google Scholar (2) Shapiro S, Venet W, Strax P, Venet L. Periodic screening for breast cancer: The Health Insurance Plan project and its sequelae; 1963-1986. Baltimore (MD): Johns Hopkins University Press, 1988. Google Scholar (3) Chu KC, Tarone RE, Kessler LG, Ries LA, Hankey B, Miller BA, et al. Recent trends in US breast cancer incidence, survival and mortality rates. J Natl Cancer Inst  1996; 88: 1571-1579. Google Scholar (4) Black WC, Nease RF Jr, Tosteson AN. Perceptions of breast cancer risk and screening effectiveness in women younger than 50 years of age. J Natl Cancer Inst  1995; 87: 720-31. Google Scholar (5) Fletcher SW, Black W, Harris R, Rimer BK, Shapiro S. Report of the International Workshop on Screening for Breast Cancer. J Natl Cancer Inst  1993; 85: 1644-56. Google Scholar (6) Elwood JM, Cox B, Richardson AK. The effectiveness of breast cancer screening by mammography in younger women [published errata appear in Online J Curr Clin Trials 1993; Doc No. 34 and 1994; Doc No. 121]. Online J Curr Clin Trials 1993; Doc No. 32. Google Scholar (7) Nystrom L, Rutqvist LE, Wall S, Lindgren A, Lindqvist M, Ryden S, et al. Breast cancer screening with mammography: overview of Swedish randomized trials [published erratum appears in Lancet 1993;342:1372]. Lancet  1993; 341: 973-8. Google Scholar (8) Kopans DB. Mammography screening and the controversy concerning women aged 40 to 49. Radiol Clin North Am  1995; 33: 1273-90. Google Scholar (9) Tabar L, Gad A, Holmberg L, Ljungquist U. Significant reduction in advanced breast cancer. Results of the first seven years of mammography screening in Kopparberg, Sweden. Diagn Imag Clin Med  1985; 54: 158-164. Google Scholar (10) Kerlikowske K, Grady D, Rubin SM, Sandrock C, Ernster VL. Efficacy of screening mammography. A meta-analysis. JAMA  1995; 273: 149-54. Google Scholar (11) Fletcher SW. Why question screening mammography for women in their forties? Radiol Clin North Am  1995; 33: 1259-1271. Google Scholar (12) Tabar L, Duffy SW, Chen HH. Re: Quantitative interpretation of age-specific mortality reductions from the Swedish breast cancer screening trials. J Natl Cancer Inst  1996; 88: 52-5. Google Scholar (13) Moss SM, Michel M, Patnick J, Johns L, Blanks R, Chamberlain J. Results from the NHS breast screening programme 1990-1993. J Med Screen  1995; 4: 186-90. Google Scholar (14) Tabar L, Fagerberg G, Duffy SW, Day NE, Gad A, Grontoft O. Update of the Swedish two-county program of mammographic screening for breast cancer. Radiol Clin North Am  1992; 30: 187-210. Google Scholar (15) Feig SA, Ehrlich SM. Estimation of radiation risk from screening mammography: recent trends and comparison with expected benefits. Radiology  1990; 174: 638-7. Google Scholar (16) Brown ML, Houn F, Sickles EA, Kessler LG. Screening mammography in community practice: positive predictive value of abnormal findings and yield of follow-up diagnostic procedures. AJR Am J Roentgenol  1995; 165: 1373-7. Google Scholar (17) Kerlikowske K, Grady D, Barclay J, Sickles EA, Eaton A, Ernster V. Positive predictive value of screening mammography by age and family history of breast cancer. JAMA  1993; 270: 2444-50. Google Scholar (18) Lidbrink E, Elfving J, Frisell J, Jonsson E. Neglected aspects of false positive findings of mammography in breast cancer screening: analysis of false positive cases from the Stockholm trial. BMJ  1996; 312: 273-6. Google Scholar (19) Lerman C, Trock B, Rimer BK, Boyce A, Jepson C, Engstrom PF. Psychological and behavioral implications of abnormal mammograms. Ann Intern Med  1991; 114: 657-61. Google Scholar (20) Graham IT, Lund E, Slenker SE. Quality of life following a false positive mammogram. Br J Cancer  1990; 62: 1018-22. Google Scholar (21) Ellman R, Angeli N, Christians A, Moss A, Chamberlain J, Macquire P. Psychiatric morbidity associated with screening for breast cancer. Br J Cancer  1989; 60: 781-4. Google Scholar (22) Eddy DM. Screening for breast cancer. Ann Intern Med  1989; 111: 389-99. Google Scholar (23) Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson C. Incidence of and treatment for ductal carcinoma in situ of the breast. JAMA  1996; 275: 913-8. Google Scholar (24) Schnitt SJ, Connolly JL, Tavassoli FA, Fechner RE, Kempson RL, Gelman R, Page DL. Interobserver reproducibility in the diagnosis of ductal proliferative breast lesions using standardized criteria. Am J Surg Pathol.  1992; 16: 1133-43. Google Scholar Oxford University Press http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI Monographs Oxford University Press

Breast Cancer Screening Among Women in Their Forties: An Overview of the Issues

Fletcher, Suzanne W.
JNCI Monographs , Volume 1997 (22) – Jan 1, 1997
Download PDF
7 pages

Loading next page...
 
/lp/oxford-university-press/breast-cancer-screening-among-women-in-their-forties-an-overview-of-pOPheCABCj

References (25)

Publisher
Oxford University Press
Copyright
Oxford University Press
ISSN
1052-6773
eISSN
1745-6614
DOI
10.1093/jncimono/1997.22.5
Publisher site
See Article on Publisher Site

Abstract

Abstract This article summarizes the issues prompting a recent NIH Consensus Conference on mammography screening for women in their forties. To date, eight randomized controlled trials of breast cancer screening have been conducted, and a reduction in breast cancer mortality has emerged after 10 to 15 years of follow-up among women offered screening in their forties. No effect appears for at least eight years, and the reason for the delay, compared to that seen in women aged 50-69, is not clear. Two possibilities include cancer-stage shift due to screening in younger women and the aging of women into their fifties during the course of screening. Possible adverse effects of screening include radiation risk, although this is low, false-negative and false-positive screening tests, and overdiagnosis due to detection of ductal carcinoma in situ (DCIS). In order to make appropriate decisions regarding mammography, women need age-related information about both the benefits and potential risks of screening. Although 85% of breast cancers occur in women after they reach the age of 50, breast cancer is the number one cause of cancer death for American women aged 40-49. In 1993, it is estimated that 30,940 American women in this age group developed breast cancer and 4,843 died of it (Harras, A; personal communication). Each year, for every 100,000 women in their forties, 163 are diagnosed with breast cancer and 30 die of the disease (1). Women in their forties need information to understand their risk for breast cancer. Data from SEER statistics indicate that for every 1,000 American women turning 40 years old, approximately 16 will develop breast cancer at some time before their fiftieth birthday (1). How many of these women will survive the cancer? SEER statistics show that nationally, 52% of women under 50 years of age who were diagnosed with invasive breast cancer in 1973 were still living 18 years later (1). Few, if any, of these women were likely screened. In the Health Insurance Plan (HIP) study, the only randomized controlled trial (RCT) with 18-year follow-up data, 58% of women in the group not offered screening survived to 18 years (2). With the advent of improved therapies over the past two decades, the percentage of women surviving breast cancer is improving (3). Thus, of the 16 women out of a thousand who will develop breast cancer in their forties, at least eight, and probably more, will survive the cancer regardless of screening. Therefore, breast cancer screening for women in their forties is primarily directed at the eight or fewer women in every 1,000 who might be saved by earlier detection of the cancer. If screening decreases mortality by as much as 25%, it would save one or two of the 16 women in a thousand who develop breast cancer in their forties. Any potentially fatal illness striking persons in the prime of life is a terrible occurrence, but breast cancer is doubly so because it not only threatens a woman's life, but an emotionally and sexually important part of her body as well. Black and colleagues found that fear of breast cancer is so great that women in their forties overestimated their risk of dying of breast cancer 20-fold and their risk of developing breast cancer sixfold (4). With such a terrifying disease, it is important to find better ways to cure and prevent it. What can screening, especially screening with mammography, contribute to the control of breast cancer in women in their forties? When considering screening for a particular medical disease, usually three questions are asked relating to the burden of disease, the characteristics of the screening test, and the effectiveness of early treatment (Table 1).1 In particular, it is important to examine the mortality benefits that accrue from the intervention, its adverse effects, and its costs. My task, then, is to present an overview of these issues as they pertain to breast cancer screening in women aged 40 to 49. Breast Cancer Mortality Reduction Most attention in breast cancer control has been directed towards determining the effect of screening on breast cancer mortality. Eight RCTs of mammography, with or without clinical breast examination, have been conducted in four countries: the HIP study from the United States; the Kopparberg, Östergötland, Malmö, Stockholm, and Gothenburg studies from Sweden; the Edinburgh study from the United Kingdom; and the National Breast Screening Study (NBSS-I) from Canada. At the National Cancer Institute International Workshop on Breast Cancer Screening in 1993, all seven trials published at that time found mortality reductions among women aged 50-69, two with statistically significant results (5). A meta-analysis presented at the Workshop found a statistically significant 34% reduction in breast cancer mortality after seven years of follow-up among women aged 50-69, with a relative risk ratio for screened to nonscreened women of 0.66 (95% confidence interval [CI]: 0.55-0.79) (6). However, the findings among younger women were less clear. The meta-analysis showed no effect at seven years of follow-up, with a relative risk of 0.99 (95% CI: 0.74-1.32) or 1.08 (95% CI: 0.85-1.39), depending on whether or not results from the Canadian study were included. A new overview of all five Swedish studies was also presented at the Workshop (7), and it showed a statistically insignificant 10% to 13% mortality reduction at 12 years of follow-up among women aged 40-49 (Fig. 1).1 This overview was more current than the meta-analysis because it included results from the Gothenburg trial that had not been previously published and because all cases in the other studies were re-reviewed, leading to some reassignment of subjects and outcomes. Nevertheless, the new Swedish analysis did not alter the conclusion from the meta-analysis that by seven years of follow-up, no beneficial effect is seen in younger women. With the data presented from the eight randomized trials, the Report of the International Workshop concluded, “For [women aged 40-49 years] it is clear that in the first 5-7 years after study entry, there is no reduction in mortality from breast cancer that can be attributed to screening. There is an uncertain and, if present, marginal reduction in mortality at about 10-12 years. Only one study (HIP) provides information on long-term effects beyond 12 years, and more information is needed.” In March 1996, an updated meta-analysis of the studies' results was reported in Falun, Sweden, for women aged 40-49 (Fig. 2).2 Five of the eight showed mortality reductions after 10-15 years of follow-up, and three showed no benefit. Pooled results demonstrated mortality reductions, with relative risk ratios of 0.77 (95% CI: 0.59-1.10), 0.76 (95% CI: 0.62-0.93), or 0.85 (95% CI: 0.71-1.01), depending on which trials were included. As results of the trials continue to accrue, it appears that the time required to demonstrate beneficial screening effect, and the size of the effect, vary by age. Whereas mortality differences between screened and control groups began to emerge after only a few years of follow-up for women aged 50-69, the studies showed effects more slowly for women in their forties. In the combined Swedish studies (Fig. 11), mortality rates were similar in the invited and control groups during the first eight years of follow-up, after which a beneficial effect of screening began to emerge. This effect has continued to grow and is nearly statistically significant with three more years of follow-up (see Fig. 22, all Swedish trials). The same trend occurred in the HIP and Edinburgh studies. In most studies, breast cancer mortality reduction in younger women was less than in older women. The cause of the time difference in effect by age is not yet clear. It has been suggested that screening picks up such early cancers in younger women that it takes longer for mortality reductions to occur (8). Indeed, randomized trials have shown that screening shifts diagnosis of breast cancer to earlier stages. Also, detection of ductal carcinomas is more common with the use of mammography. However, for stage shift to cause the time difference, screening would have to shift the stage of cancer differently according to age. Analysis of the Kopparberg trial showed shifts to earlier-stage cancers among women in both the forties and fifties (9). Data from all trials, however, should be examined to determine whether and to what degree stage shift could explain the delayed effect of screening in younger women. The slower appearance of mortality reduction in younger women could also be partly due to “age creep.” Because screening studies occur over several years, some women who enter trials during their forties age into their fifties over the course of screening. It has been suggested that as women move into their fifties, when breast cancer screening is known to work, a benefit becomes apparent (10,11). Analyses of the RCTs are reported according to the age of women at entry into the trial, not their age at the time of diagnosis of breast cancer. This approach is necessary to preserve the comparability of the screened and control groups. Nevertheless, when there is the possibility that the effect of breast cancer screening varies by age, information about the age at diagnosis is needed. Most trials have not yet provided this information. The issue is especially important in the two trials in which only women 45 and older at entry were included. Two groups have reported data about this issue. In the HIP study, 32% of cancers in women aged 40-49 at entry were detected after the women had turned 50 (2). Shapiro et al. demonstrated that screened women aged 45-49 at entry benefited when their cancers were diagnosed after age 50 but not when their cancers were diagnosed earlier. The numbers of women in each subgroup, however, were small. On the other hand, Tabar and Duffy did not find any effect of age creep in the Swedish two-county trial in which 36% of cancers were diagnosed after women in their forties turned 50 (12). The relative mortality was 0.95 (95% CI: 0.44-2.03) for women in whom breast cancer was diagnosed after they turned age 50 and 0.85 (95% CI: 0.49-1.45) for women in whom breast cancer was diagnosed before age 50. Information from the other trials is needed. Ultimately, the degree to which age creep influences mortality effects for women in their forties will be best addressed by the results of the National Health Service Breast Screening Programme underway in the United Kingdom (13). In this large RCT, women aged 40 and 41 at entry are being screened annually for five years, which means that all breast cancers diagnosed will be in women who are still in their forties. Why would a screening test for breast cancer have differential effects by age? Part of the explanation may be the lower accuracy (both sensitivity and specificity) of screening tests in younger women (5). Also, breast cancer growth rates may differ by age of the woman. Tabar et al. found that the mean sojourn time (time in the preclinical detectable state) was 1.25 years for women in their forties and 3.03 years for women in their fifties (14). Whether and how estrogen levels and menopause, rather than age per se, influence effectiveness of breast cancer screening remains unclear and needs to be determined. The question as to when to start breast cancer screening should not be arbitrarily linked to a particular decade of a woman's life. It is important to determine the effect of screening in groups at high risk for breast cancer, especially in women aged 40-49. To date, no reports from the randomized trials have examined screening effects according to risk status of the participants. Adverse Effects Important possible adverse effects of breast cancer screening include radiation risk from mammography, adverse physical and psychosocial sequelae of false-positive and false-negative tests, and overdiagnosis. Radiation risk from modern mammography appears to be very low. Feig and Ehrlich reviewed recent estimates for lifetime radiation risk to the breast and reported that a single mammogram exposure is estimated to cause between 2.9 and 8.8 excess breast cancers per million women screened during their forties (15). The estimates varied according to the age of the woman and the method of calculation, with more recent estimates being lower than older ones. If women in their forties are screened annually, radiation risk might cause between 29 and 88 additional breast cancers per million women screened for an entire decade. Most previous work has analyzed the degree of accuracy of screening tests and the related problems of false-positive and false-negative results. The International Workshop Report summarized data from the randomized trials showing breast cancer screening in younger women is not as sensitive as in older women (5). Sensitivity by the detection method (ratio of screen-detected cancers to screen-detected plus interval cancers) among women aged 40-49 at study entry varied from 53% to 81% in the Stockholm, Swedish two-county, and Canada NBSS-I trials, while comparably defined sensitivity among women aged 50-59 varied from 73% to 88%. Thus, the risk of false-negative screening tests is greater in younger women. Recent data suggest that false-positive mammograms—those requiring further evaluation to rule out cancer diagnosis—are a substantial problem in the United States. In a national survey of community mammography facilities conducted in 1992 and 1993, Brown et al. found that 11% (95% CI: 9%-13%) of screening mammograms were read as abnormal; 10.6% were false-positive readings (16). At a state-of-the-art program in northern California, Kerlikowske et al. reported that 6.3% of first-screen mammograms among women aged 40-49 were abnormal, and 5.9% proved to be false-positive readings (17). Table 22 presents the percentage of women in each of these two studies who underwent additional procedures, including biopsies, following abnormal mammogram readings. On average, between one and two additional procedures were performed for each abnormal reading. Approximately 10% to 15% of women in each study underwent an invasive procedure. Lidbrink et al. have reported comparable results from the Stockholm trial (18). The psychological effect of false-positive mammograms has been studied by Lerman et al. (19). They found that among women with high-suspicion mammograms that subsequently proved to be false-positive, three months later 47% reported being quite anxious about mammography, 41% were quite anxious about breast cancer, 26% reported that worry affected their mood, and 17% reported that it adversely affected their daily function. Women with low-suspicion mammograms reported less concern, but even among these women anxiety about mammography and breast cancer was relatively high (29% and 40% respectively, compared to 24% and 29% in women with normal mammograms). In a study from Norway, 18 months after screening mammography, 29% of 126 women with false-positive mammograms reported anxiety about breast cancer, compared to 13% of 152 randomly selected women with negative mammograms (p = 0.001) (20). In Britain, Ellman et al. found that 25% of women with normal mammograms, 30% of women with false-positive mammograms, and 35% of women with breast symptoms but with normal mammograms had General Health Questionnaire scores indicating probable psychiatric morbidity (21). Because breast cancer screening is periodically repeated, it is important to determine the percentage of women who will experience a false-positive mammogram or clinical breast examination over an extended period of time. In the studies to date, it is clear that the percentage of false-positive mammograms decreases as women are rescreened. Nevertheless, it has been estimated that over a 10-year period of annual mammograms, as many as 30% of women in their forties will experience a false-positive mammogram or clinical breast examination (22). It is important to determine the actual number. Efforts to decrease the number of false-positive screening tests and their resultant adverse effects are also urgently needed. As screening for breast cancer has increased, detection of ductal carcinomas in situ (DCISs) has risen—328% from 1983 to 1992 (23). An increasing percentage of breast cancers detected by screening are DCIS. In the northern Californian study discussed above, 43% of cancers detected among women in their forties were DCIS (17). Some experts are concerned that early lesions such as DCIS have led to overdiagnosis of breast cancer and is partly responsible for the recent increased incidence of breast cancer (23). Finding breast cancer before any invasion, even microinvasion, has occurred should help save lives. However, there are a number of questions about DCIS. Pathologically, it appears difficult to diagnose. One study, for instance, asked six experienced pathologists to interpret 24 slides; there was complete agreement among the six in only two of the 10 cases in which at least one pathologist diagnosed DCIS (24). The prevalence and natural history of the condition are not clear and are important in determining if DCIS detection is leading to overdiagnosis. Finally, Ernster and colleagues have demonstrated that there is a wide range of treatment approaches for DCIS, not all of which may be appropriate (23). There is an urgent need for studies of all these issues. In sum, progress has been made in better understanding the breast cancer mortality reduction that might occur with a screening program for women in their forties. Randomized trials have demonstrated that for approximately a decade, no benefit occurs, but after 10 to 15 years, a 15% to 25% mortality reduction appears. This translates into one or two women per 1,000 who might be saved. The reasons for the delay in mortality benefits and the degree to which these benefits could be achieved by beginning screening later remain to be determined. Also less clear are other benefits that might occur from earlier screening, such as more limited surgery or less debilitating adjuvant therapy. To achieve these benefits, however, substantial numbers of women will experience adverse effects, especially those caused by false-positive mammograms and possible overdiagnosis because of DCIS. Finally, costs of screening programs and the resultant procedures carried out because of abnormal readings cannot be ignored. Women need information about all these issues. They rightly demand to be involved in an important decision about their lives and their bodies. Ultimately, it is the job of medical science to search for new and better ways to promote health, and along the way, to share with the public the very complicated facts as we understand them. Armed with facts, women can then apply their own set of values to cope with the important problem of breast cancer. Table 1. Criteria for deciding whether a medical condition should be included in periodic health examinations* 1. How great is the burden of suffering caused by the condition in terms of:    Death  Discomfort    Disease  Dissatisfaction    Disability  Destitution  2. How good is the screening test, if one is to be performed, in terms of:    Sensitivity  Cost  Labeling Effects    Specificity  Safety    Simplicity  Acceptability  3. a. For primary prevention, how effective is the intervention?   or   b. For secondary prevention, if the condition is found, how effective is the ensuing treatment in terms of:    Efficacy    Patient Compliance    Early treatment being more effective than later treatment  1. How great is the burden of suffering caused by the condition in terms of:    Death  Discomfort    Disease  Dissatisfaction    Disability  Destitution  2. How good is the screening test, if one is to be performed, in terms of:    Sensitivity  Cost  Labeling Effects    Specificity  Safety    Simplicity  Acceptability  3. a. For primary prevention, how effective is the intervention?   or   b. For secondary prevention, if the condition is found, how effective is the ensuing treatment in terms of:    Efficacy    Patient Compliance    Early treatment being more effective than later treatment  * Reprinted by permission from Fletcher R, Fletcher S, Wagner E. Clinical epidemiology—the essentials. Baltimore, Williams & Wilkins, 1996 View Large Table 2. Follow-up procedures for abnormal screening mammograms   NSMF study* All ages, %   Northern CA study** 40-49 years, first screen %   Clinical breast examination  6.8  3.2  Repeat mammogram  37.0  —  Additional mammogram  41.2  56.1  Ultrasonography  20.0  10.9  Fine needle aspiration  3.0  5.8  Needle biopsy  2.9  —  Excisional biopsy  10.5  13.0  Needle localization  —  11.0    NSMF study* All ages, %   Northern CA study** 40-49 years, first screen %   Clinical breast examination  6.8  3.2  Repeat mammogram  37.0  —  Additional mammogram  41.2  56.1  Ultrasonography  20.0  10.9  Fine needle aspiration  3.0  5.8  Needle biopsy  2.9  —  Excisional biopsy  10.5  13.0  Needle localization  —  11.0  * National Survey of Mammography Facilities. Data from (16). ** Data from (17). View Large Fig. 1. View largeDownload slide Overview of Swedish randomized trials. Cumulative breast cancer mortality (per 1000) up to 12 years after randomization by age at randomization. Solid line = invited group and dotted line = control group. Adapted with permission from Nystrom et al. Breast cancer screening with mammography: overview of Swedish randomized trials. Lancet 1993;341:973-978. Fig. 1. View largeDownload slide Overview of Swedish randomized trials. Cumulative breast cancer mortality (per 1000) up to 12 years after randomization by age at randomization. Solid line = invited group and dotted line = control group. Adapted with permission from Nystrom et al. Breast cancer screening with mammography: overview of Swedish randomized trials. Lancet 1993;341:973-978. Fig. 2. View largeDownload slide Meta-analysis of randomized controlled trials of breast cancer screening for women in their forties, presented at Falun, Sweden; March, 1996. Relative risks are presented for each study, all Swedish trials, all population-based trials, and all trials. Adapted with permission from Breast cancer screening with mammography in women aged 40-49 years. Int J Cancer 1996;68:693-699. © Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc., 1996. Fig. 2. View largeDownload slide Meta-analysis of randomized controlled trials of breast cancer screening for women in their forties, presented at Falun, Sweden; March, 1996. Relative risks are presented for each study, all Swedish trials, all population-based trials, and all trials. Adapted with permission from Breast cancer screening with mammography in women aged 40-49 years. Int J Cancer 1996;68:693-699. © Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc., 1996. References (1) Kosary CL, Ries LAG, Miller BA, Hankey BF, Harras A, Edwards BK. SEER Cancer Statistics Review, 1973-1992: Tables and graphs. Bethesda (MD): National Cancer Institute; 1995: DHHS Publ No. (NIH)96-2789. Google Scholar (2) Shapiro S, Venet W, Strax P, Venet L. Periodic screening for breast cancer: The Health Insurance Plan project and its sequelae; 1963-1986. Baltimore (MD): Johns Hopkins University Press, 1988. Google Scholar (3) Chu KC, Tarone RE, Kessler LG, Ries LA, Hankey B, Miller BA, et al. Recent trends in US breast cancer incidence, survival and mortality rates. J Natl Cancer Inst  1996; 88: 1571-1579. Google Scholar (4) Black WC, Nease RF Jr, Tosteson AN. Perceptions of breast cancer risk and screening effectiveness in women younger than 50 years of age. J Natl Cancer Inst  1995; 87: 720-31. Google Scholar (5) Fletcher SW, Black W, Harris R, Rimer BK, Shapiro S. Report of the International Workshop on Screening for Breast Cancer. J Natl Cancer Inst  1993; 85: 1644-56. Google Scholar (6) Elwood JM, Cox B, Richardson AK. The effectiveness of breast cancer screening by mammography in younger women [published errata appear in Online J Curr Clin Trials 1993; Doc No. 34 and 1994; Doc No. 121]. Online J Curr Clin Trials 1993; Doc No. 32. Google Scholar (7) Nystrom L, Rutqvist LE, Wall S, Lindgren A, Lindqvist M, Ryden S, et al. Breast cancer screening with mammography: overview of Swedish randomized trials [published erratum appears in Lancet 1993;342:1372]. Lancet  1993; 341: 973-8. Google Scholar (8) Kopans DB. Mammography screening and the controversy concerning women aged 40 to 49. Radiol Clin North Am  1995; 33: 1273-90. Google Scholar (9) Tabar L, Gad A, Holmberg L, Ljungquist U. Significant reduction in advanced breast cancer. Results of the first seven years of mammography screening in Kopparberg, Sweden. Diagn Imag Clin Med  1985; 54: 158-164. Google Scholar (10) Kerlikowske K, Grady D, Rubin SM, Sandrock C, Ernster VL. Efficacy of screening mammography. A meta-analysis. JAMA  1995; 273: 149-54. Google Scholar (11) Fletcher SW. Why question screening mammography for women in their forties? Radiol Clin North Am  1995; 33: 1259-1271. Google Scholar (12) Tabar L, Duffy SW, Chen HH. Re: Quantitative interpretation of age-specific mortality reductions from the Swedish breast cancer screening trials. J Natl Cancer Inst  1996; 88: 52-5. Google Scholar (13) Moss SM, Michel M, Patnick J, Johns L, Blanks R, Chamberlain J. Results from the NHS breast screening programme 1990-1993. J Med Screen  1995; 4: 186-90. Google Scholar (14) Tabar L, Fagerberg G, Duffy SW, Day NE, Gad A, Grontoft O. Update of the Swedish two-county program of mammographic screening for breast cancer. Radiol Clin North Am  1992; 30: 187-210. Google Scholar (15) Feig SA, Ehrlich SM. Estimation of radiation risk from screening mammography: recent trends and comparison with expected benefits. Radiology  1990; 174: 638-7. Google Scholar (16) Brown ML, Houn F, Sickles EA, Kessler LG. Screening mammography in community practice: positive predictive value of abnormal findings and yield of follow-up diagnostic procedures. AJR Am J Roentgenol  1995; 165: 1373-7. Google Scholar (17) Kerlikowske K, Grady D, Barclay J, Sickles EA, Eaton A, Ernster V. Positive predictive value of screening mammography by age and family history of breast cancer. JAMA  1993; 270: 2444-50. Google Scholar (18) Lidbrink E, Elfving J, Frisell J, Jonsson E. Neglected aspects of false positive findings of mammography in breast cancer screening: analysis of false positive cases from the Stockholm trial. BMJ  1996; 312: 273-6. Google Scholar (19) Lerman C, Trock B, Rimer BK, Boyce A, Jepson C, Engstrom PF. Psychological and behavioral implications of abnormal mammograms. Ann Intern Med  1991; 114: 657-61. Google Scholar (20) Graham IT, Lund E, Slenker SE. Quality of life following a false positive mammogram. Br J Cancer  1990; 62: 1018-22. Google Scholar (21) Ellman R, Angeli N, Christians A, Moss A, Chamberlain J, Macquire P. Psychiatric morbidity associated with screening for breast cancer. Br J Cancer  1989; 60: 781-4. Google Scholar (22) Eddy DM. Screening for breast cancer. Ann Intern Med  1989; 111: 389-99. Google Scholar (23) Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson C. Incidence of and treatment for ductal carcinoma in situ of the breast. JAMA  1996; 275: 913-8. Google Scholar (24) Schnitt SJ, Connolly JL, Tavassoli FA, Fechner RE, Kempson RL, Gelman R, Page DL. Interobserver reproducibility in the diagnosis of ductal proliferative breast lesions using standardized criteria. Am J Surg Pathol.  1992; 16: 1133-43. Google Scholar Oxford University Press

Journal

JNCI Monographs – Oxford University Press

Published: Jan 1, 1997

There are no references for this article.