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

Learn More →

Detection of Ductal Carcinoma In Situ in Women Undergoing Screening Mammography

Detection of Ductal Carcinoma In Situ in Women Undergoing Screening Mammography Abstract Background: With the large number of women having mammography—an estimated 28.4 million U.S. women aged 40 years and older in 1998—the percentage of cancers detected as ductal carcinoma in situ (DCIS), which has an uncertain prognosis, has increased. We pooled data from seven regional mammography registries to determine the percentage of mammographically detected cancers that are DCIS and the rate of DCIS per 1000 mammograms. Methods: We analyzed data on 653 833 mammograms from 540 738 women between 40 and 84 years of age who underwent screening mammography at facilities participating in the National Cancer Institute’s Breast Cancer Surveillance Consortium (BCSC) throughout 1996 and 1997. Mammography results were linked to population-based cancer and pathology registries. We calculated the percentage of screen-detected breast cancers that were DCIS, the rate of screen-detected DCIS per 1000 mammograms by age and by previous mammography status, and the sensitivity of screening mammography. Statistical tests were two-sided. Results: A total of 3266 cases of breast cancer were identified, 591 DCIS and 2675 invasive breast cancer. The percentage of screen-detected breast cancers that were DCIS decreased with age (from 28.2% [95% confidence interval (CI) = 23.9% to 32.5%] for women aged 40–49 years to 16.0% [95% CI = 13.3% to 18.7%] for women aged 70–84 years). However, the rate of screen-detected DCIS cases per 1000 mammograms increased with age (from 0.56 [95% CI = 0.41 to 0.70] for women aged 40–49 years to 1.07 [95% CI = 0.87 to 1.27] for women aged 70–84 years). Sensitivity of screening mammography in all age groups combined was higher for detecting DCIS (86.0% [95% CI = 83.2% to 88.8%]) than it was for detecting invasive breast cancer (75.1% [95% CI = 73.5% to 76.8%]). Conclusions: Overall, approximately 1 in every 1300 screening mammography examinations leads to a diagnosis of DCIS. Given uncertainty about the natural history of DCIS, the clinical significance of screen-detected DCIS needs further investigation. Incidence rates of ductal carcinoma in situ (DCIS) of the breast, a noninvasive form of breast cancer, have risen dramatically in the United States and elsewhere since the early 1980s (1–5). The increase in DCIS incidence parallels an increase in the use of screening mammography, which makes the detection of DCIS much more likely than in the past. DCIS is a diverse process of the breast epithelium that is, by definition, confined within the breast duct above the basement membrane. The cellular appearance of DCIS varies from low-grade lesions similar to atypical hyperplasia to high-grade or anaplastic lesions. Because DCIS is excised when detected, it is not possible to accurately estimate the fraction of untreated cases that would progress to invasive malignancy. In addition to undergoing surgical treatment (lumpectomy or mastectomy), women with DCIS are frequently treated with radiation or hormone therapy (1,6,7). Moreover, in addition to other treatment, axillary lymph node dissection may also be performed for women with DCIS, although this procedure is not standard of care (6,7). It is unlikely that there will ever be definitive studies to demonstrate whether detection of DCIS by mammography has contributed to the reduction in breast cancer mortality associated with screening mammography. However, it seems reasonable to conclude that some women benefit from having DCIS detected through screening mammography, whereas other women do not, depending on whether the disease was likely to progress. Because it is not possible to screen women for invasive breast cancer alone, the detection of DCIS is part of any screening mammography program. A recent survey of 479 women in the United States found that few women had heard about DCIS or were aware that some forms of breast cancer might not progress (8). When these women were informed, however, 60% of them wanted to take the possibility of DCIS detection into account when deciding about participating in screening mammography. No information is currently available about the rate of screen-detected DCIS. Therefore, we pooled data from mammography registries across the United States to determine the percentage of DCIS among all screen-detected breast cancers and the rate of screen-detected DCIS per 1000 mammograms performed. Methods Study Population We obtained patient history and clinical and radiologic assessment data from screening mammography examinations from the mammography registries of the Breast Cancer Surveillance Consortium (BCSC) performed from January 1996 through December 1997. The mammography registries were located in Colorado, New Hampshire, New Mexico, North Carolina, San Francisco (CA), Vermont, and western Washington State. The BCSC, funded by the National Cancer Institute, and its confidentiality procedures are described elsewhere (9,10). Each mammography registry had Institutional Review Board approval to collect data from mammography facilities for research purposes. All patient identifiers were removed from the collected data before they were transferred to the BCSC’s Statistical Coordinating Center for analysis. The present analysis is limited to the results of screening mammography examinations for women between 40 and 84 years old who had no self-report of previous breast cancer. To be eligible for inclusion, the woman’s mammography examination had to be designated as a screening examination rather than a diagnostic examination by the radiologist. We excluded unilateral screening examinations and examinations that did not have an assessment code indicating whether they had been considered negative or positive for an abnormality indicative of cancer. Assessment codes were based on the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) (11). We also excluded mammographic examinations if the women had had any breast imaging (including ultrasound or another mammogram) in the preceding 9 months, because imaging within this period may indicate that the screening mammographic examination was not a true screening examination but rather a follow-up examination. We used 9 months rather than a longer time period as the cutoff because women sometimes schedule their annual mammographic examination a month or two early. All screening examinations from the same woman over the 2-year study period were included, subject to the 9-month exclusion rule. Mammogram Assessment, Categorization, and Cancer Diagnoses Mammograms were linked to either cancer or pathology registries in the same geographic region as the mammography registry. Not all mammography facilities in a particular geographic region are included in the BCSC for that region, so not all mammograms for that region were ascertained; however, ascertainment of cancers is believed to be almost complete. Statistics by state for each cancer registry are maintained by the North American Association of Central Cancer Registries and show adjusted completion rates based on observed versus expected rates for all cancers in excess of 94.3% in all states included in this study except for Vermont, which did not have statistics available (12). Cancer records from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program1 1 Editor’s note: SEER is a set of geographically defined, population-based, central cancer registries in the United States, operated by local nonprofit organizations under contract to the National Cancer Institute (NCI). Registry data are submitted electronically without personal identifiers to the NCI on a biannual basis, and the NCI makes the data available to the public for scientific research. were used for the three BCSC sites that are located in SEER registry areas—that is, New Mexico, San Francisco, and western Washington State—and a fourth site, Colorado, used a non-SEER cancer registry that uses SEER guidelines for data collection. For the other three sites, where there can be a long delay between cancer diagnosis and appearance of the cancer record in the local cancer registry, pathology registries served as the primary source of case reporting, supplemented by cancer registry data. In this article, we present combined data from all mammography registries participating in the BCSC, recognizing that there may be some variation in cancer ascertainment across registries and, therefore, potential variation in cancer incidence rates. Cases were categorized as either DCIS or invasive breast cancer based on either cancer registry or pathology registry data. Cases of in situ lesions that were specifically coded as lobular carcinoma in situ (LCIS) were excluded, but other cases of in situ lesions were included (for example, in situ lesions that were coded as “not otherwise specified”). Thus, some in situ lesions other than DCIS were categorized as DCIS, but the number of these was small. If a cancer case was coded within 60 days of diagnosis as having both invasive and in situ components, it was categorized as invasive breast cancer rather than as DCIS. For comparison with the DCIS findings, we also provide some findings for invasive breast cancer; however, the data on invasive breast disease were not examined in detail. Throughout this article the term “screen-detected” refers to screening mammography and not to other forms of breast cancer detection. We used the initial mammographic assessment category assigned to each screening mammography examination to determine whether cases of DCIS and invasive breast cancer had been screen-detected. We considered cases of DCIS or invasive breast cancer to be screen-detected (i.e., positive) if the patient’s most recent eligible screening mammography examination in the 365 days prior to diagnosis resulted in any of the following BI-RADS assessment codes: 0 (needs additional evaluation); 3 (probably benign finding), with a recommendation for immediate work-up; 4 (suspicious abnormality); or 5 (highly suggestive of malignancy). Cases of DCIS or invasive breast cancer were considered non-screen-detected (i.e., negative) if the patient’s most recent eligible screening mammography examination in the 365 days prior to diagnosis had resulted in a BI-RADS assessment code of 1 (negative); 2 (benign finding); or 3 (probably benign finding), with no recommendation for immediate work-up or with an unknown recommendation. These case definitions were used to generate the numerators for calculating rates per 1000 mammograms for screen-detected and non-screen-detected DCIS and invasive breast cancer. The denominators used in calculating rates were the number of screening mammography examinations, not the number of women. The follow-up period after a screening examination was 365 days or the next screening examination, whichever came first. This follow-up period ensured that a diagnosis of cancer was allocated to a single screening mammography examination. Data Analysis We first determined the sensitivity of screening mammography for detecting DCIS and invasive breast cancer. Sensitivity was defined as the number of cases with a positive mammographic assessment at screening divided by the total number of women. We then calculated the percentage of all screen-detected and non-screen-detected breast cancers that were DCIS, and the rates of screen-detected and non-screen-detected DCIS per 1000 mammograms by age group (40–49 years, 50–59 years, 60–69 years, and 70–84 years) and for all ages combined. The denominator used to calculate the DCIS rate in each category was the total number of screening examinations in that category. We further cross-classified the percentage and rate variables by previous mammography status as determined by women’s self-reported mammography history (screening or diagnostic mammogram). At the time of screening mammography, women were routinely asked if they had had a prior mammographic examination and the date of or approximate time in years since that examination. Women who did not provide complete information about history of previous mammography were categorized as having no previous mammography; 461 129 (71%) women reported a history of a prior mammographic examination and 192 704 (29%) women did not report having a prior mammographic examination. Data from individual BCSC sites on percentages of cancers that are DCIS and rates of DCIS per 1000 mammograms are not reported, but ranges of these variables over the seven BCSC sites are presented (see Tables 3 and 4). Ninety-five percent confidence intervals (CIs) were calculated for sensitivity, for percentage of cancers that are DCIS, and for the rate of DCIS per 1000 mammograms based on the pooled BCSC data and were not corrected for multiple comparisons. We assumed a normal approximation to the binomial distribution for proportions and a Poisson distribution for each cancer rate. Standard chi-square tests were used as statistical tests for proportions, and Poisson regression was used as the statistical test for comparison of cancer rates. Statistical tests were two-sided. We also plotted the distributions of time-to-diagnosis for cases defined as screen-detected and those defined as non-screen-detected. Results Demographic and Clinical Variables Across the seven BCSC sites, data from 653 833 screening mammography examinations (ranging from 43 624 to 150 387 examinations among 540 738 women were available for evaluation (Table 1). Thus, approximately 20% of women contributed more than one screening mammography examination to the analysis. Based on questionnaire data collected at the time of screening, 29% of all mammographic examinations were classified as first ever: 33% for ages 40–49 years, 27% for ages 50–59 years, 28% for ages 60–69 years, and 29% for ages 70–84 years. Race was self-reported for 86% of the women who were screened; of these women, 79% were white, 5% were African-American, 2% were Asian/Pacific Islander, 2% were Native American, and 12% were other/mixed. Of the 81% of women who responded to a question about whether they were of Hispanic origin, 4% reported being Hispanic. Educational level was either not asked for or not reported for 39% of women. Of those women who did report educational level, 34% completed college, 57% completed high school but not college, and 9% did not complete high school. Because this demographic information was not reported for many women, we did not analyze the results of screening mammography data by race or educational level. Linkage of mammography screening data and cancer case reports resulted in the identification of 3266 cases of breast cancer (2675 invasive and 591 DCIS) occurring within 365 days following screening mammography. The distributions of the number of screening mammography examinations, DCIS cases, and invasive breast cancer cases by age are given in Table 1, where we further distribute by positive (screen-detected) cases. Sensitivity of Screening Mammography for Detecting DCIS The sensitivity of screening mammography for detecting DCIS was higher than that for invasive breast cancer. Of all DCIS cases diagnosed during the study period, 86.0% (508/591, 95% CI = 83.2% to 88.8%) were associated with a positive screening mammography assessment (compared with 75.1% (2010/2675, 95% CI = 73.5% to 76.8%) of all invasive breast cancer cases (Table 1). Sensitivity of screening mammography for detecting DCIS did not differ statistically significantly by age (P = .47). By comparison, sensitivity for detecting invasive breast cancer increased statistically significantly with age (Ptrend<.001), from 66.9% for women aged 40–49 years to 82.7% for women aged 70–84 years (difference in sensitivity = 15.8%, 95% CI = 10.7% to 21.0%). Sensitivity for detecting both types of cancers combined also increased statistically significantly with age (Ptrend<.001) from 71.7% for women aged 40–49 years to 82.8% for women aged 70–84 years (difference in sensitivity = 11.1%, 95% CI = 6.6% to 15.5%), as might be expected, because the invasive breast cancer cases made up 81.9% of all cases. Sensitivity of screening mammography for detecting DCIS was slightly higher for women with no previous mammography (88.6%) than it was for those with previous mammography (84.8%), but these values were not statistically significantly different from one another (difference in sensitivity = 3.8%, 95% CI = –2.0% to 9.6%; P = .25) (Table 2). The highest sensitivity for detecting DCIS was seen in women aged 40–49 years who reported no previous mammography (97.4%, 95% CI = 85.0% to 100%); however, there was no clear association between sensitivity for detecting DCIS and age when the data were stratified by previous mammography status. For invasive breast cancer, the difference in sensitivity between women with no previous mammography (82.3%) and those with previous mammography (71.5%) was greater than that observed for DCIS (difference in sensitivity = 10.8%, 95% CI = 7.5% to 14.0%; P<.001), and sensitivity increased statistically significantly with age for women without and with previous mammography (Ptrend<.001 for both). Percentage of DCIS Among Screen-Detected and Non-Screen-Detected Breast Cancers The percentages of screen-detected and non-screen-detected breast cancer cases that were DCIS across all study sites combined varied by age and by prior screening history, as shown in Table 3. The percentage of cancers that were DCIS differed statistically significantly by age overall (P<.001) and for women with positive mammography screens who had previous mammography (P<.001). However, if the women had positive mammography screens and had not had a previous mammogram, or if they had negative screens, the percentage of cancers that were DCIS did not vary statistically significantly by age (P = .19 and P = .07, respectively). For all age groups combined, 20.2% (95% CI = 18.6% to 21.7%) of screen-detected breast cancers were DCIS, with the percentage being higher among women aged 40–49 years (28.2%, 95% CI = 23.9% to 32.5%) than among women in all older age groups (ranging from 16% [95% CI = 13.3% to 18.7%] to 20.5% [95% CI = 17.5% to 23.6%] among women aged 60–84 years). In every age group, a higher percentage of screen-detected breast cancer was DCIS among women with previous mammography than among women with no previous mammography, with the difference being most pronounced for women aged 40–49 years (31.9% versus 22.6%, respectively). Rates of Screen-Detected DCIS The association of screen-detected DCIS with age differed depending on whether one was examining the percentage of cases that are DCIS or the rate of DCIS per 1000 mammograms (Fig. 1). The percentage of screen-detected breast cancer that was DCIS decreased with age (Fig. 1, A), whereas the DCIS rate per 1000 mammograms increased statistically significantly with age (Fig. 1, B; P<.001). Table 4 shows the rate of screen-detected and non-screen-detected DCIS (for all sites combined) and the ranges of the rates across sites for all DCIS cases. The rate of screen-detected DCIS was 0.56 (95% CI = 0.41 to 0.70) per 1000 mammograms among women aged 40–49 years and increased to 1.07 (95% CI = 0.87 to 1.27) per 1000 mammograms among women aged 70–84 years. The overall rate (i.e., for all age groups combined) of non-screen-detected DCIS (0.13 [95% CI = 0.05 to 0.20] per 1000 mammograms) was only approximately one sixth that of screen-detected DCIS (0.78 [95% CI = 0.60 to 0.95] per 1000 mammograms). Similar to screen-detected DCIS, the rate of non-screen-detected DCIS increased with age from 0.08 (95% CI = 0.02 to 0.13) per 1000 mammograms among women aged 40–49 years to 0.22 (95% CI = 0.13 to 0.31) per 1000 mammograms among women aged 70–84 years (P<.001). Differences in screen-detected DCIS rates between women with and without previous mammography were small, with no consistent direction across age groups. The rate of screen-detected cancers increased more markedly with age for invasive breast cancer than for DCIS, from 1.42 (95% CI = 1.19 to 1.66) per 1000 mammograms among women aged 40–49 years to 5.62 (95% CI = 5.15 to 6.08) per 1000 mammograms among women aged 70–84 years (data not shown). In addition, although in every age group the rates of non-screen-detected invasive breast cancer were lower than those of screen-detected cancers, the rate of non-screen-detected invasive breast cancer was approximately one half that of screen-detected invasive breast cancer among women aged 40–49 years (0.70, 95% CI = 0.54 to 0.87 per 1000 mammograms) and approximately one fifth that of screen-detected invasive breast cancer among women aged 70–84 years (1.17, 95% CI = 0.96 to 1.38 per 1000 mammograms) (data not shown). Time to Diagnosis of DCIS The time to DCIS diagnosis following screening mammography was generally much shorter for women with a positive mammography screen than for women with a negative screen (Fig. 2). Whereas 91% of DCIS cases associated with a positive screen were diagnosed within 3 months after screening, the distribution of DCIS cases diagnosed following a negative screen was spread throughout the subsequent year. The distribution of time to diagnosis following a negative screen had distinct peaks at around 1, 6–7, and 12 months. This distribution is consistent with the occurrence of further follow-up mammograms and subsequent diagnostic evaluation immediately following the screening mammography examination (perhaps due to clinical findings, such as a breast mass, even though the indication for the examination was routine screening), at the 6-month short-interval follow-up and at the repeat annual screening mammography examination occurring just prior to 12 months. Discussion Our study of nearly 654 000 mammography screening examinations among nearly 541 000 women aged 40–84 years from seven mammography registries throughout the United States showed that approximately 20% of all screen-detected breast cancers were DCIS and that the overall rate of DCIS detection by screening mammography was 0.78 per 1000 mammograms; this rate increased with age. Our results suggest that one case of DCIS is detected for approximately every 1300 screening mammography examinations performed. This rate varies by age, ranging from approximately 1 in every 1800 mammograms among women aged 40–49 years to 1 in every 935 mammograms among women aged 70–84 years. Several reports regarding the occurrence of in situ lesions have been published that were based on data from population-based cancer registries in the United States (1,2,4,13) and elsewhere (3,5). Data from cancer registries, however, do not permit determination of the incidence of screen-detected disease specifically, because cancer registries do not routinely collect information on screening mammography practices either for individual cancer cases or for the populations they cover. To the extent that non-screened women are included in cancer registry statistics, estimates of DCIS incidence based on cancer data from population-based cancer registries will underestimate the incidence of DCIS among screened women because DCIS is primarily detected by mammography. Reports from individual mammography programs and from case series have provided additional estimates of the percentage of all breast cancers or of nonpalpable breast cancers that are DCIS (14–16). There are, however, several reports of the occurrence of in situ lesions among women participating in large, organized screening mammography programs. The Breast Cancer Detection Demonstration Project, conducted in the 1970s, reported that 18% of breast cancers detected at initial screening mammography in study participants were intraductal and in situ cases, with a smaller percentage of DCIS being detected in subsequent screenings (17). In the early 1990s, community-based screening mammography programs in New Mexico and British Columbia found that approximately 20% of all breast cancers detected were in situ (18,19). The U.S. National Breast and Cervical Cancer Early Detection Project reported in 1998 (20,21) that 25.3% of all breast cancers detected on the first round of screening mammography and 32.8% of those detected on subsequent rounds of screening were in situ. DCIS detection rates in this study were approximately 1.5 cases per 1000 first-round mammograms and 0.9 cases per 1000 subsequent mammograms. Several international studies (22–24) have also found that the proportion of screen-detected cancers that were in situ ranged from 15% to 22%. Precise comparisons of either the percentage of screen-detected cancers that are DCIS or of the rates of DCIS per 1000 screening mammograms across screened populations are not possible because of differences in calendar period, age distributions, racial/ethnic composition, socioeconomic status, recommended screening intervals, definitions of screen-detected cancer and prior mammography, and type of in situ cases included. Nevertheless, the rates we estimated in this study are similar to those reported elsewhere. In this study, we found that the percentage of all screen-detected breast cancer that was DCIS decreased with increasing age. That is, if a woman had screen-detected breast cancer, it was less likely to be DCIS if she was older. Other investigators (25,26) with similar findings have misinterpreted the percentages of DCIS in their studies as meaning that the incidence of DCIS is higher in younger women than it is in older women. However, as was the case for invasive breast cancer, we found that the rate of DCIS detection actually increased with age, from 0.56 per 1000 mammograms among women aged 40–49 years to 1.07 per 1000 mammograms among women aged 70–84 years. Other studies (20,22,23) have confirmed that the percentage of DCIS among all screen-detected breast cancers does generally decrease with age, but those studies did not always show a consistent increase in the rate of detection of DCIS with age. Similar to another report (27), we found that screening mammography was more sensitive for detecting DCIS (86%) than it was for detecting invasive breast cancer (75%). Invasive carcinoma is usually associated with a mass or density found on mammography, whereas DCIS is usually associated with microcalcifications (25). Missed DCIS is much less likely than missed invasive breast cancer to be diagnosed subsequently as a palpable mass, and it may never become clinically apparent. This phenomenon has the potential effect of overestimating the ability of screening mammography to detect in situ disease, so that sensitivity for detecting DCIS would be biased upward. Furthermore, in dense or heterogeneously dense breast tissue, microcalcifications are easier to detect on mammography than are masses. In our study, sensitivity of screening mammography for detecting both DCIS and invasive breast cancer was higher among women for whom no previous mammography was recorded than it was for women with previous mammography. Earlier work has also shown that the sensitivity of mammography is higher for women without a known previous mammogram, presumably because the cancers available for screen detection include prevalent cases that may be larger and/or slower growing than cancers among women undergoing regular screening (28). Although we found that sensitivity for detecting invasive breast cancer increased with age, both for women with previous mammography and for women with no previous mammography, there was no clear relationship between sensitivity for detecting DCIS and age. Among women with no previous mammography, however, sensitivity for detecting DCIS was higher in women younger than 50 years of age than for older women. Thus, the proportion of DCIS cases in different study populations, as well as age and previous mammography status of the women, will influence reported overall sensitivities of screening mammography for detecting breast cancer. Surprisingly, given the general understanding that DCIS is readily detected by screening mammography, we found that 14% of DCIS cases (83/591) were among women whose most recent screening mammography examination had a negative assessment. It should be noted that the screening examinations of 21 of the 83 non-screen-detected DCIS cases had an assessment code of 3 with a recommendation for short interval follow-up (i.e. a follow-up generally occurring within 3–6 months from a screening mammography examination), which indicates that the radiologist had some concern about the result even though an immediate work-up was not recommended. Over 90% of DCIS diagnoses for women with positive screens occurred within 3 months of mammography, whereas DCIS diagnoses for women with negative screens were spread throughout the year following mammography, suggesting that many of these DCIS cases were detected by other means. We do not know, however, whether these non-screen-detected DCIS cases represent cases that were present but missed at the time of screening mammography or were new (i.e., interval) occurrences of DCIS detected by means other than mammography. For example, some studies (29,30) indicate that some DCIS cases can be detected by clinical breast examination. Our study has several limitations. First, we relied on self-reported questionnaire data to classify women as to whether they had had a previous screening mammography examination, and if there was no information provided on the questionnaire about previous mammography, we assumed that the screening mammogram was the woman’s first ever. To the extent that misclassification occurred, “true” differences between women with and without previous mammography would be diminished. Second, we used the initial mammographic assessment category assigned to each screening mammogram to determine whether the finding was positive or negative. Subsequent imaging mammograms of the study participants may have resulted in a reappraisal of the BI-RADS assessment code and its associated recommendation. However, estimates of sensitivity for detecting cancer are most commonly based on the initial assessment, so we chose to use the initial mammographic assessment to allow comparison to existing literature. Given that the likelihood of detecting DCIS among women undergoing screening mammography has increased and is not small at 20% of all screen-detected breast cancer cases, is there evidence to suggest that detecting DCIS prevents deaths from breast cancer? The rationale for early detection and treatment of DCIS is based on several lines of evidence. First, a small series of untreated women with DCIS diagnosed in the premammography era had higher than expected subsequent occurrences of invasive breast cancer (31,32). In addition, larger, more recent series of women treated for DCIS have also been shown to be at increased risk of recurrence of both in situ and invasive breast cancer (33). The estimates of recurrence rates range fairly broadly across studies and are influenced by patient and tumor characteristics, such as age and tumor grade, and by treatment (34–36). In a large clinical trial, after a mean follow-up of 90 months, the incidence of ipsilateral invasive breast cancer among women with DCIS who had been treated with lumpectomy alone was 13.4% (37). Thus, DCIS can recur as invasive breast cancer. Findings from randomized clinical trials that have shown that the addition of radiation therapy and tamoxifen to lumpectomy treatment of DCIS reduces the chance of future invasive disease recurrence compared with lumpectomy alone also bolster the case for treatment of DCIS (37–39). Finally, data are accumulating to suggest that there are both epidemiologic and genetic similarities between DCIS and invasive breast cancers, further suggesting that treatment for DCIS may yield benefits similar to those of treatment for invasive breast cancer (40–46). In contrast, some lines of evidence temper enthusiasm for detecting DCIS. For example, autopsy studies have shown that DCIS is not uncommon among women who died of causes unrelated to breast cancer; across seven autopsy studies, median prevalence of DCIS was 8.9% (range = 0% to 14.7%), although, presumably, not all such cases of DCIS would have been detectable by mammography (47). This finding suggests that some cases of DCIS do not progress to clinically significant lesions and may never require treatment in the patient’s lifetime. In addition, although their risk of subsequent invasive disease is elevated, most women treated for DCIS do not experience an invasive recurrence, at least within 10–15 years after DCIS diagnosis. Follow-up data from population-based registries show that only about 2% of women who have been diagnosed with and treated for DCIS since the time when screening mammography was widely introduced in the early 1990s died of breast cancer within 10 years following DCIS diagnosis (48,49). These favorable outcomes presumably reflect both the effectiveness of current treatment regimens and the relatively benign nature of most mammographically detected DCIS. Finally, even though over four times as many in situ cases were found among women in the Canadian National Breast Screening Study-2 who had received mammography plus clinical breast examination as were found among women who had received clinical breast examination alone, there was no overall difference in breast cancer mortality between the two groups of women after 13 years of follow-up (30). The use of screening mammography has become widespread. Based on 1998 nationally representative data from the National Health Interview Survey, the estimated number of U.S. women aged 40 years and older who had mammography in the previous year was more than 28.4 million (Breen N, National Cancer Institute: personal communication). Our data suggest that approximately 1 in every 1300 screening mammography examinations results in a DCIS diagnosis. If early detection and treatment of DCIS have contributed to the recent decline in U.S. breast cancer mortality (50), independent of effects attributable to early detection and treatment of invasive disease and independent of recent advances in breast cancer treatment, then some women who have been treated for screen-detected DCIS have benefited. The pathology of DCIS is heterogeneous, and presumably, the likelihood of benefit from treatment is greater for women with larger, higher grade, or multifocal lesions than for those with very small low-grade lesions. The BCSC affords the opportunity to explore both the distribution of grade of DCIS in a population-based mammography program and the long-term sequelae of DCIS identification. More research is needed to understand the biology and clinical significance of DCIS to identify disease that is likely to progress and to better tailor treatment decisions. Table 1. Number of screening mammography examinations and the percentage of women with ductal carcinoma in situ (DCIS) or invasive breast cancer who had a positive screen (sensitivity), by age*     DCIS  Invasive breast cancer  DCIS and invasive breast cancer combined  Age range, y  No. of screening mammography examinations  No. of cases diagnosed  Percentage of cases with positive screen  No. of cases diagnosed  Percentage of cases with positive screen†  Total no. of cases diagnosed  Percentage of cases with positive screen†  *Positive screen is a mammography screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.001).  40–49  211 551  134  88.1  450  66.9  584  71.7  50–59  200 255  155  88.4  792  72.2  947  74.9  60–69  135 376  165  84.2  709  75.9  874  77.5  70–84  106 651  137  83.2  724  82.7  861  82.8      Total  653 833  591  86.0  2675  75.1  3266  77.1      DCIS  Invasive breast cancer  DCIS and invasive breast cancer combined  Age range, y  No. of screening mammography examinations  No. of cases diagnosed  Percentage of cases with positive screen  No. of cases diagnosed  Percentage of cases with positive screen†  Total no. of cases diagnosed  Percentage of cases with positive screen†  *Positive screen is a mammography screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.001).  40–49  211 551  134  88.1  450  66.9  584  71.7  50–59  200 255  155  88.4  792  72.2  947  74.9  60–69  135 376  165  84.2  709  75.9  874  77.5  70–84  106 651  137  83.2  724  82.7  861  82.8      Total  653 833  591  86.0  2675  75.1  3266  77.1  View Large Table 2. Percentage of women with ductal carcinoma in situ (DCIS) or invasive breast cancer who had a positive screen (sensitivity), by age and previous mammography*   DCIS  Invasive breast cancer    No previous mammogram  Previous mammogram  No previous mammogram  Previous mammogram  Age range, y  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen†  No. of cases  Percentage of cases with positive screen†  *Positive screen is a mammography-screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.01).  40–49  39  97.4  95  84.2  172  75.6  278  61.5  50–59  46  87.0  109  89.0  254  79.1  538  69.0  60–69  45  84.4  120  84.2  209  83.3  500  72.8  70–84  46  87.0  91  81.3  263  89.0  461  79.2      Total  176  88.6  415  84.8  898  82.3  1777  71.5    DCIS  Invasive breast cancer    No previous mammogram  Previous mammogram  No previous mammogram  Previous mammogram  Age range, y  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen†  No. of cases  Percentage of cases with positive screen†  *Positive screen is a mammography-screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.01).  40–49  39  97.4  95  84.2  172  75.6  278  61.5  50–59  46  87.0  109  89.0  254  79.1  538  69.0  60–69  45  84.4  120  84.2  209  83.3  500  72.8  70–84  46  87.0  91  81.3  263  89.0  461  79.2      Total  176  88.6  415  84.8  898  82.3  1777  71.5  View Large Table 3. Percentage of ductal carcinoma in situ (DCIS) cases among all breast cancer cases, by age, previous mammography, and whether screening mammography was positive or negative*   Percentage of DCIS among all breast cancer cases with a positive screen  Percentage of DCIS among all cancer cases with a negative screen  Percentage of DCIS among all breast cancer cases  Age range, y  No previous mammogram  Previous mammogram  Total (95% CI)†  Total (95% CI)  Total (95% CI)†  Range over BCSC sites  *Mammography screening was considered positive when it was associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) and considered negative when associated with a negative screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval; BCSC = Breast Cancer Surveillance Consortium.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  22.6  31.9  28.2 (23.9 to 32.5)  9.7 (5.2 to 14.2)  22.9 (19.5 to 26.4)  18.3–30.3  50–59  16.6  20.7  19.3 (16.4 to 22.2)  7.6 (4.2 to 10.9)  16.4 (14.0 to 18.7)  11.1–24.7  60–69  17.9  21.7  20.5 (17.5 to 23.6)  13.2 (8.5 to 17.9)  18.9 (16.3 to 21.5)  16.0–23.9  70–84  14.6  16.9  16.0 (13.3 to 18.7)  15.5 (9.7 to 21.4)  15.9 (13.5 to 18.4)  9.1–21.8      Total  17.4  21.7  20.2 (18.6 to 21.7)  11.1 (8.8 to 13.4)  18.1 (16.8 to 19.4)  14.6–23.8    Percentage of DCIS among all breast cancer cases with a positive screen  Percentage of DCIS among all cancer cases with a negative screen  Percentage of DCIS among all breast cancer cases  Age range, y  No previous mammogram  Previous mammogram  Total (95% CI)†  Total (95% CI)  Total (95% CI)†  Range over BCSC sites  *Mammography screening was considered positive when it was associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) and considered negative when associated with a negative screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval; BCSC = Breast Cancer Surveillance Consortium.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  22.6  31.9  28.2 (23.9 to 32.5)  9.7 (5.2 to 14.2)  22.9 (19.5 to 26.4)  18.3–30.3  50–59  16.6  20.7  19.3 (16.4 to 22.2)  7.6 (4.2 to 10.9)  16.4 (14.0 to 18.7)  11.1–24.7  60–69  17.9  21.7  20.5 (17.5 to 23.6)  13.2 (8.5 to 17.9)  18.9 (16.3 to 21.5)  16.0–23.9  70–84  14.6  16.9  16.0 (13.3 to 18.7)  15.5 (9.7 to 21.4)  15.9 (13.5 to 18.4)  9.1–21.8      Total  17.4  21.7  20.2 (18.6 to 21.7)  11.1 (8.8 to 13.4)  18.1 (16.8 to 19.4)  14.6–23.8  View Large Table 4. Screen-detected and non-screen-detected ductal carcinoma in situ (DCIS) rates per 1000 mammograms, by age and previous mammography*   Screen-detected DCIS rate  Non-screen-detected DCIS rate  DCIS rates, all cases  Age range, y  No previous mammogram†  Previous mammogram†  Total (95% CI)†  Total (95% CI)†  Total (95% CI)†  Range over BCSC sites  *Screen-detected = cases of DCIS associated with a positive mammography-screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to diagnosis. Non-screen-detected = cases of DCIS associated with a negative mammography-screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval.  †Proportions using chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  0.54  0.57  0.56 (0.41 to 0.70)  0.08 (0.02 to 0.13)  0.63 (0.48 to 0.79)  0.44–1.10  50–59  0.74  0.66  0.68 (0.52 to 0.85)  0.09 (0.03 to 0.15)  0.77 (0.60 to 0.95)  0.51–1.38  60–69  1.00  1.04  1.03 (0.83 to 1.23)  0.19 (0.11 to 0.28)  1.22 (1.00 to 1.44)  1.00–2.26  70–84  1.31  0.97  1.07 (0.87 to 1.27)  0.22 (0.13 to 0.31)  1.28 (1.06 to 1.51)  0.50–1.75      Total  0.81  0.76  0.78 (0.60 to 0.95)  0.13 (0.05 to 0.20)  0.90 (0.72 to 1.09)  0.72–1.42    Screen-detected DCIS rate  Non-screen-detected DCIS rate  DCIS rates, all cases  Age range, y  No previous mammogram†  Previous mammogram†  Total (95% CI)†  Total (95% CI)†  Total (95% CI)†  Range over BCSC sites  *Screen-detected = cases of DCIS associated with a positive mammography-screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to diagnosis. Non-screen-detected = cases of DCIS associated with a negative mammography-screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval.  †Proportions using chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  0.54  0.57  0.56 (0.41 to 0.70)  0.08 (0.02 to 0.13)  0.63 (0.48 to 0.79)  0.44–1.10  50–59  0.74  0.66  0.68 (0.52 to 0.85)  0.09 (0.03 to 0.15)  0.77 (0.60 to 0.95)  0.51–1.38  60–69  1.00  1.04  1.03 (0.83 to 1.23)  0.19 (0.11 to 0.28)  1.22 (1.00 to 1.44)  1.00–2.26  70–84  1.31  0.97  1.07 (0.87 to 1.27)  0.22 (0.13 to 0.31)  1.28 (1.06 to 1.51)  0.50–1.75      Total  0.81  0.76  0.78 (0.60 to 0.95)  0.13 (0.05 to 0.20)  0.90 (0.72 to 1.09)  0.72–1.42  View Large Fig. 1. View largeDownload slide Percentage of ductal carcinoma in situ (DCIS) cases among all screen-detected breast cancers and screen-detected DCIS rates per 1000 mammograms. A) Percentage of all screen-detected breast cancers that were diagnosed as DCIS stratified by age group. B) Rate of screen-detected DCIS per 1000 mammograms stratified by age group. Cancers were considered screen-detected if they were associated with a positive screening mammography examination according to the American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment codes of 0 (needs additional evaluation), 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality), or 5 (highly suggestive of malignancy) within 365 days prior to cancer diagnosis. The data in A are in Table 3, and the data in B are in Table 4. Fig. 1. View largeDownload slide Percentage of ductal carcinoma in situ (DCIS) cases among all screen-detected breast cancers and screen-detected DCIS rates per 1000 mammograms. A) Percentage of all screen-detected breast cancers that were diagnosed as DCIS stratified by age group. B) Rate of screen-detected DCIS per 1000 mammograms stratified by age group. Cancers were considered screen-detected if they were associated with a positive screening mammography examination according to the American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment codes of 0 (needs additional evaluation), 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality), or 5 (highly suggestive of malignancy) within 365 days prior to cancer diagnosis. The data in A are in Table 3, and the data in B are in Table 4. Fig. 2. View largeDownload slide Distribution of ductal carcinoma in situ (DCIS) cases by time from screening mammography examination to DCIS diagnosis. A) Cases of DCIS diagnosed after a positive screen. DCIS cases were considered screen-detected if they were associated with a positive screening mammographic examination according to the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) assessment codes of 0 (needs additional evaluation); 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality); or 5 (highly suggestive of malignancy) within the 365 days prior to DCIS diagnosis. B) Cases of DCIS diagnosed after a negative screen. DCIS cases were considered non-screen-detected if they were associated with a negative screening mammographic examination (BI-RADS assessment codes of 1 [negative], 2 [benign finding], or 3 [probably benign finding with no recommendation for immediate work-up or unknown recommendation]) within 365 days prior to DCIS diagnosis. Fig. 2. View largeDownload slide Distribution of ductal carcinoma in situ (DCIS) cases by time from screening mammography examination to DCIS diagnosis. A) Cases of DCIS diagnosed after a positive screen. DCIS cases were considered screen-detected if they were associated with a positive screening mammographic examination according to the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) assessment codes of 0 (needs additional evaluation); 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality); or 5 (highly suggestive of malignancy) within the 365 days prior to DCIS diagnosis. B) Cases of DCIS diagnosed after a negative screen. DCIS cases were considered non-screen-detected if they were associated with a negative screening mammographic examination (BI-RADS assessment codes of 1 [negative], 2 [benign finding], or 3 [probably benign finding with no recommendation for immediate work-up or unknown recommendation]) within 365 days prior to DCIS diagnosis. Supported by cooperative agreements U01CA63731, U01CA63736, U01CA63740, U01CA69976, U01CA70013, U01CA70040, U01CA86076, and U01CA86082, as part of the National Cancer Institute’s (NCI’s) Breast Cancer Surveillance Consortium, and by grant R01CA63146 from the NCI, National Institutes of Health, Department of Health and Human Services. References 1 Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson IC. Incidence and treatment for ductal carcinoma in situ of the breast. JAMA  1996; 275: 913–8. Google Scholar 2 Choi WS, Parker BA, Pierce JP, Greenberg ER. Regional differences in the incidence and treatment of carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev  1996; 5: 317–20. Google Scholar 3 Levi F, Te VC, Randimbison L, La Vecchia C. Trends of in situ carcinoma of the breast in Vaud, Switzerland. Eur J Cancer  1997; 33: 903–6. Google Scholar 4 Adams-Cameron M, Gilliland FD, Hunt WC, Key CR. Trends in incidence and treatment for ductal carcinoma in situ in Hispanic, American Indian, and non-Hispanic white women in New Mexico, 1973–1994. Cancer  1999; 85: 1084–90. Google Scholar 5 Barchielli A, Paci E, Girogi D. Recent trends of in situ carcinoma of the breast and mammographic screening in the Florence area, Italy. Cancer Causes Control  1999; 10: 313–7. Google Scholar 6 Winchester DP, Menck HR, Osteen RT, Kraybill W. Treatment trends for ductal carcinoma in situ of the breast. Ann Surg Oncol  1995; 2: 207–13. Google Scholar 7 Schwartz GF, Solin LJ, Olivotto IA, Ernster VL, Pressman PI, and the Consensus Conference Committee. The Consensus Conference on the Treatment of In Situ Ductal Carcinoma of the Breast, April 22–25, 1999. Hum Pathol  2000; 31: 131–9. Google Scholar 8 Schwartz LM, Woloshin S, Sox HC, Fischhoff B, Welch HG. US women’s attitudes to false positive mammography results and detection of ductal carcinoma in situ: cross sectional survey. BMJ  2000; 320: 1635–40. Google Scholar 9 Ballard-Barbash R, Taplin SH, Yankaskas BC, Ernster VL, Rosenberg RD, Carney PA, et al. Breast Cancer Surveillance Consortium: a national mammography screening and outcomes database. AJR Am J Roentgenol  1997; 169: 1001–8. Google Scholar 10 Carney PA, Geller BM, Moffett H, Ganger M, Sewell M, Barlow WE, et al. Current medicolegal and confidentiality issues in a large multicenter research program. Am J Epidemiol  2000; 152: 371–8. Google Scholar 11 American College of Radiology. Illustrated Breast Imaging Reporting and Data System (BI-RADS TM). 3rd ed. Reston (VA): American College of Radiology; 1998. p. 179–81. Google Scholar 12 Wu XC, Hotes JL, Fulton PJ, Cormier M, Corre CN, McLaughlin CC, et al., editors. Cancer in North America, 1995–1999. Vol. 1. Incidence. Springfield (IL): North American Association of Cancer Registries; April 2002. p. II.33, II.45, II.193, II.271. Google Scholar 13 Ries LA, Eisner MP, Kosary CL, Hankey BF, Miller BA, Clegg L, et al. SEER Cancer Statistics Review, 1973–1998. Bethesda (MD): National Cancer Institute; 2001. Google Scholar 14 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 15 Frykberg ER, Bland KI. In situ breast carcinoma. In: Cameron JL, editor. Advances in surgery. Chicago (IL): Mosby Year Book, Inc.; 1993. p. 29–72. Google Scholar 16 Evans AJ, Pinder SE, Ellis IO, Wilson AR. Screen detected ductal carcinoma in situ (DCIS): overdiagnosis or an obligate precursor of invasive disease? J Med Screen  2001; 8: 149–51. Google Scholar 17 Seidman H, Gelb SK, Silverberg E, LaVerda N, Lubera JA. Survival experience in the Breast Cancer Detection Demonstration Project. CA Cancer J Clin  1987; 37: 258–90. Google Scholar 18 Rosenberg RD, Lando JF, Hunt WC, Darling RR, Williamson MR, Linver MN, et al. The New Mexico Mammography Project: screening mammography performance in Albuquerque, New Mexico, 1991 to 1993. Cancer  1996; 78: 1731–9. Google Scholar 19 Olivotto IA, Kan L, d’Yachkova Y, Burhenne LJ, Hayes M, Hislop TG. Ten years of breast screening in the Screening Mammography Program of British Columbia. J Med Screen  2000; 7: 152–9. Google Scholar 20 May DS, Lee NC, Nadel MR, Henson RM, Miller DS. The National Breast and Cervical Cancer Early Detection Program: report on the first 4 years of mammography provided to medically underserved women. AJR Am J Roentgenol  1998; 170: 97–104. Google Scholar 21 May DS, Lee NC, Richardson LC, Giustozzi AG, Bobo JK. Mammography and breast cancer detection by race and Hispanic ethnicity: results from a national program (United States). Cancer Causes Control  2000; 11: 697–705. Google Scholar 22 Canada H. Organized breast cancer screening programs in Canada: 1996 report: Laboratory Centre for Disease Control, Health Canada. Minister of Public Works and Government Services Canada; 1999. Google Scholar 23 UK Trial of Early Detection of Breast Cancer Group. 16-year mortality from breast cancer in the UK Trial of Early Detection of Breast Cancer. Lancet  1999; 353: 1909–14. Google Scholar 24 Fracheboud J, Groenewoud JH, Boer RI, van Ineveld BM, Verbeek AL, Hendricks JH, et al. Landelijke evaluatie van bevolkingsonderzoek naar borstkanker in Nederland. Rotterdam (The Netherlands): 2000. Google Scholar 25 Wazer DE, Gage I, Homer MJ, Krosnick SH, Schmid C. Age-related differences in patients with nonpalpable breast carcinomas. Cancer  1996; 78: 1432–7. Google Scholar 26 Evans WP, Starr AL, Bennos ES. Comparison of the relative incidence of impalpable invasive breast carcinoma and ductal carcinoma in situ in cancers detected in patients older and younger than 50 years of age. Radiology  1997; 204: 489–91. Google Scholar 27 Kerlikowske K, Grady D, Barclay J, Sickles EA, Ernster VL. Effect of age, breast density, and family history on the sensitivity of first screening mammography. JAMA  1996; 276: 33–8. Google Scholar 28 Rosenberg RD, Yankaskas BC, Hunt WC, Ballard-Barbash R, Urban N, Ernster VL, et al. Effect of variations in operational definitions on performance estimates for screening mammography. Acad Radiol  2000; 7: 1058–68. Google Scholar 29 Bobo JK, Lee NC, Thames SF. Findings from 752,081 clinical breast examinations reported to a national screening program from 1995 through 1998. J Natl Cancer Inst  2000; 92: 971–6. Google Scholar 30 Miller AB, To T, Baines CJ, Wall C. Canadian National Breast Screening Study-2: 13-year results of a randomized trial in women aged 50–59 years. J Natl Cancer Inst  2000; 92: 1490–9. Google Scholar 31 Page DL, Dupont WD, Rogers LW, Jensen RA, Schuyler PA. Continued local recurrence of carcinoma 15–25 years after a diagnosis of low grade ductal carcinoma in situ of the breast treated only by biopsy. Cancer  1995; 76: 1197–200. Google Scholar 32 Eusebi V, Foschini MP, Cook MG, Berrino F, Azzopardi JG. Long-term follow-up of in situ carcinoma of the breast with special emphasis on clinging carcinoma. Semin Diagn Pathol  1989; 6: 165–73. Google Scholar 33 Fonseca R, Hartmann LC, Petersen IA, Donohue JH, Crotty TB, Gisvold JJ. Ductal carcinoma in situ of the breast. Ann Intern Med  1997; 127: 1013–22. Google Scholar 34 Van Zee KJ, Liberman L, Samli B, Tran KN, McCormick B, Petrek JA, et al. Long term follow-up of women with ductal carcinoma in situ treated with breast-conserving surgery: the effect of age. Cancer  1999; 86: 1757–67. Google Scholar 35 Franceschi S, Levi F, La Vecchia C, Randimbison L, Te VC. Second cancers following in situ carcinoma of the breast. Int J Cancer  1998; 77: 392–5. Google Scholar 36 Mirza NQ, Vlastos G, Meric F, Sahin AA, Singletary SE, Newman LA, et al. Ductal carcinoma-in-situ: long-term results of breast-conserving therapy. Ann Surg Oncol  2000; 7: 656–64. Google Scholar 37 Fisher B, Dignam J, Wolmark N, Mamounas E, Castantino J, Poller W, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from the National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol  1998; 16: 441–52. Google Scholar 38 Fisher B, Dignam J, Wolmark N, Wickerham LD, Fisher ER, Mamounas E, et al. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet  1999; 353: 1993–2000. Google Scholar 39 Julien JP, Bijker N, Fentiman IS, Peterse JL, Delledonne V, Rouanet P, et al. Radiotherapy in breast-conserving treatment for ductal carcinoma in situ: first results of the EORTC randomized phase III trial 10853. Lancet  2000; 355: 528–33. Google Scholar 40 Longnecker MP, Bernstein L, Paganini-Hill A, Enger SM, Ross RK. Risk factors for in situ breast cancer. Cancer Epidemiol Biomarkers Prev  1996; 5: 961–5. Google Scholar 41 Kerlikowske K, Barclay J, Grady D, Sickles EA, Ernster VL. Comparison of risk factors for ductal carcinoma in situ and invasive breast cancer. J Natl Cancer Inst  1997; 89: 76–82. Google Scholar 42 Lambe M, Hsieh CC, Tsaih SW, Ekbom A, Trichopoulos D, Adami HO. Parity, age at first birth and the risk of carcinoma in situ of the breast. Int J Cancer  1998; 77: 330–2. Google Scholar 43 Zhuang Z, Merino MJ, Chuaqui R, Liotta LA, Emmert-Buck MR. Identical allelic loss on chromosome 11q13 in microdissected in situ and invasive human breast cancer. Cancer Res  1995; 55: 467–71. Google Scholar 44 James LA, Mitchell EL, Menasce L, Varley JM. Comparative genomic hybridisation of ductal carcinoma of the breast: identification of regions of DNA amplification and deletion in common with invasive breast carcinoma. Oncogene  1997; 14: 1059–65. Google Scholar 45 Marsh KL, Varley JM. Loss of heterozygosity at chromosome 9p in ductal carcinoma in situ and invasive carcinoma of the breast. Br J Cancer  1998; 77: 1439–47. Google Scholar 46 Trentham-Dietz A, Newcomb PA, Storer BE, Remington PL. Risk factors for carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev  2000; 9: 697–703. Google Scholar 47 Welch HG, Black WC. Using autopsy series to estimate the disease “reservoir” for ductal carcinoma in situ of the breast: how much more breast cancer can we find? Ann Intern Med  1997; 127: 1023–8. Google Scholar 48 Ernster VL, Barclay J, Kerlikowske K, Wilkie H, Ballard-Babash R. Mortality among women with ductal carcinoma in situ of the breast in the population-based Surveillance, Epidemiology and End Results program. Arch Intern Med  2000; 160: 953–8. Google Scholar 49 Warnberg F, Bergh J, Holmberg L. Prognosis in women with a carcinoma in situ of the breast: a population-based study in Sweden. Cancer Epidemiol Biomarkers Prev  1999; 8: 769–74. Google Scholar 50 Howe HL, Wingo PA, Thun MJ, Ries LA, Rosenberg HM, Feigal EG, et al. Annual report to the nation on the status of cancer (1973 through 1998), featuring cancers with recent increasing trends. J Natl Cancer Inst  2001; 93: 824–42. Google Scholar © Oxford University Press http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI: Journal of the National Cancer Institute Oxford University Press

Loading next page...
 
/lp/oxford-university-press/detection-of-ductal-carcinoma-in-situ-in-women-undergoing-screening-62Qg2L02d0

References (47)

Publisher
Oxford University Press
Copyright
© Oxford University Press
ISSN
0027-8874
eISSN
1460-2105
DOI
10.1093/jnci/94.20.1546
Publisher site
See Article on Publisher Site

Abstract

Abstract Background: With the large number of women having mammography—an estimated 28.4 million U.S. women aged 40 years and older in 1998—the percentage of cancers detected as ductal carcinoma in situ (DCIS), which has an uncertain prognosis, has increased. We pooled data from seven regional mammography registries to determine the percentage of mammographically detected cancers that are DCIS and the rate of DCIS per 1000 mammograms. Methods: We analyzed data on 653 833 mammograms from 540 738 women between 40 and 84 years of age who underwent screening mammography at facilities participating in the National Cancer Institute’s Breast Cancer Surveillance Consortium (BCSC) throughout 1996 and 1997. Mammography results were linked to population-based cancer and pathology registries. We calculated the percentage of screen-detected breast cancers that were DCIS, the rate of screen-detected DCIS per 1000 mammograms by age and by previous mammography status, and the sensitivity of screening mammography. Statistical tests were two-sided. Results: A total of 3266 cases of breast cancer were identified, 591 DCIS and 2675 invasive breast cancer. The percentage of screen-detected breast cancers that were DCIS decreased with age (from 28.2% [95% confidence interval (CI) = 23.9% to 32.5%] for women aged 40–49 years to 16.0% [95% CI = 13.3% to 18.7%] for women aged 70–84 years). However, the rate of screen-detected DCIS cases per 1000 mammograms increased with age (from 0.56 [95% CI = 0.41 to 0.70] for women aged 40–49 years to 1.07 [95% CI = 0.87 to 1.27] for women aged 70–84 years). Sensitivity of screening mammography in all age groups combined was higher for detecting DCIS (86.0% [95% CI = 83.2% to 88.8%]) than it was for detecting invasive breast cancer (75.1% [95% CI = 73.5% to 76.8%]). Conclusions: Overall, approximately 1 in every 1300 screening mammography examinations leads to a diagnosis of DCIS. Given uncertainty about the natural history of DCIS, the clinical significance of screen-detected DCIS needs further investigation. Incidence rates of ductal carcinoma in situ (DCIS) of the breast, a noninvasive form of breast cancer, have risen dramatically in the United States and elsewhere since the early 1980s (1–5). The increase in DCIS incidence parallels an increase in the use of screening mammography, which makes the detection of DCIS much more likely than in the past. DCIS is a diverse process of the breast epithelium that is, by definition, confined within the breast duct above the basement membrane. The cellular appearance of DCIS varies from low-grade lesions similar to atypical hyperplasia to high-grade or anaplastic lesions. Because DCIS is excised when detected, it is not possible to accurately estimate the fraction of untreated cases that would progress to invasive malignancy. In addition to undergoing surgical treatment (lumpectomy or mastectomy), women with DCIS are frequently treated with radiation or hormone therapy (1,6,7). Moreover, in addition to other treatment, axillary lymph node dissection may also be performed for women with DCIS, although this procedure is not standard of care (6,7). It is unlikely that there will ever be definitive studies to demonstrate whether detection of DCIS by mammography has contributed to the reduction in breast cancer mortality associated with screening mammography. However, it seems reasonable to conclude that some women benefit from having DCIS detected through screening mammography, whereas other women do not, depending on whether the disease was likely to progress. Because it is not possible to screen women for invasive breast cancer alone, the detection of DCIS is part of any screening mammography program. A recent survey of 479 women in the United States found that few women had heard about DCIS or were aware that some forms of breast cancer might not progress (8). When these women were informed, however, 60% of them wanted to take the possibility of DCIS detection into account when deciding about participating in screening mammography. No information is currently available about the rate of screen-detected DCIS. Therefore, we pooled data from mammography registries across the United States to determine the percentage of DCIS among all screen-detected breast cancers and the rate of screen-detected DCIS per 1000 mammograms performed. Methods Study Population We obtained patient history and clinical and radiologic assessment data from screening mammography examinations from the mammography registries of the Breast Cancer Surveillance Consortium (BCSC) performed from January 1996 through December 1997. The mammography registries were located in Colorado, New Hampshire, New Mexico, North Carolina, San Francisco (CA), Vermont, and western Washington State. The BCSC, funded by the National Cancer Institute, and its confidentiality procedures are described elsewhere (9,10). Each mammography registry had Institutional Review Board approval to collect data from mammography facilities for research purposes. All patient identifiers were removed from the collected data before they were transferred to the BCSC’s Statistical Coordinating Center for analysis. The present analysis is limited to the results of screening mammography examinations for women between 40 and 84 years old who had no self-report of previous breast cancer. To be eligible for inclusion, the woman’s mammography examination had to be designated as a screening examination rather than a diagnostic examination by the radiologist. We excluded unilateral screening examinations and examinations that did not have an assessment code indicating whether they had been considered negative or positive for an abnormality indicative of cancer. Assessment codes were based on the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) (11). We also excluded mammographic examinations if the women had had any breast imaging (including ultrasound or another mammogram) in the preceding 9 months, because imaging within this period may indicate that the screening mammographic examination was not a true screening examination but rather a follow-up examination. We used 9 months rather than a longer time period as the cutoff because women sometimes schedule their annual mammographic examination a month or two early. All screening examinations from the same woman over the 2-year study period were included, subject to the 9-month exclusion rule. Mammogram Assessment, Categorization, and Cancer Diagnoses Mammograms were linked to either cancer or pathology registries in the same geographic region as the mammography registry. Not all mammography facilities in a particular geographic region are included in the BCSC for that region, so not all mammograms for that region were ascertained; however, ascertainment of cancers is believed to be almost complete. Statistics by state for each cancer registry are maintained by the North American Association of Central Cancer Registries and show adjusted completion rates based on observed versus expected rates for all cancers in excess of 94.3% in all states included in this study except for Vermont, which did not have statistics available (12). Cancer records from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program1 1 Editor’s note: SEER is a set of geographically defined, population-based, central cancer registries in the United States, operated by local nonprofit organizations under contract to the National Cancer Institute (NCI). Registry data are submitted electronically without personal identifiers to the NCI on a biannual basis, and the NCI makes the data available to the public for scientific research. were used for the three BCSC sites that are located in SEER registry areas—that is, New Mexico, San Francisco, and western Washington State—and a fourth site, Colorado, used a non-SEER cancer registry that uses SEER guidelines for data collection. For the other three sites, where there can be a long delay between cancer diagnosis and appearance of the cancer record in the local cancer registry, pathology registries served as the primary source of case reporting, supplemented by cancer registry data. In this article, we present combined data from all mammography registries participating in the BCSC, recognizing that there may be some variation in cancer ascertainment across registries and, therefore, potential variation in cancer incidence rates. Cases were categorized as either DCIS or invasive breast cancer based on either cancer registry or pathology registry data. Cases of in situ lesions that were specifically coded as lobular carcinoma in situ (LCIS) were excluded, but other cases of in situ lesions were included (for example, in situ lesions that were coded as “not otherwise specified”). Thus, some in situ lesions other than DCIS were categorized as DCIS, but the number of these was small. If a cancer case was coded within 60 days of diagnosis as having both invasive and in situ components, it was categorized as invasive breast cancer rather than as DCIS. For comparison with the DCIS findings, we also provide some findings for invasive breast cancer; however, the data on invasive breast disease were not examined in detail. Throughout this article the term “screen-detected” refers to screening mammography and not to other forms of breast cancer detection. We used the initial mammographic assessment category assigned to each screening mammography examination to determine whether cases of DCIS and invasive breast cancer had been screen-detected. We considered cases of DCIS or invasive breast cancer to be screen-detected (i.e., positive) if the patient’s most recent eligible screening mammography examination in the 365 days prior to diagnosis resulted in any of the following BI-RADS assessment codes: 0 (needs additional evaluation); 3 (probably benign finding), with a recommendation for immediate work-up; 4 (suspicious abnormality); or 5 (highly suggestive of malignancy). Cases of DCIS or invasive breast cancer were considered non-screen-detected (i.e., negative) if the patient’s most recent eligible screening mammography examination in the 365 days prior to diagnosis had resulted in a BI-RADS assessment code of 1 (negative); 2 (benign finding); or 3 (probably benign finding), with no recommendation for immediate work-up or with an unknown recommendation. These case definitions were used to generate the numerators for calculating rates per 1000 mammograms for screen-detected and non-screen-detected DCIS and invasive breast cancer. The denominators used in calculating rates were the number of screening mammography examinations, not the number of women. The follow-up period after a screening examination was 365 days or the next screening examination, whichever came first. This follow-up period ensured that a diagnosis of cancer was allocated to a single screening mammography examination. Data Analysis We first determined the sensitivity of screening mammography for detecting DCIS and invasive breast cancer. Sensitivity was defined as the number of cases with a positive mammographic assessment at screening divided by the total number of women. We then calculated the percentage of all screen-detected and non-screen-detected breast cancers that were DCIS, and the rates of screen-detected and non-screen-detected DCIS per 1000 mammograms by age group (40–49 years, 50–59 years, 60–69 years, and 70–84 years) and for all ages combined. The denominator used to calculate the DCIS rate in each category was the total number of screening examinations in that category. We further cross-classified the percentage and rate variables by previous mammography status as determined by women’s self-reported mammography history (screening or diagnostic mammogram). At the time of screening mammography, women were routinely asked if they had had a prior mammographic examination and the date of or approximate time in years since that examination. Women who did not provide complete information about history of previous mammography were categorized as having no previous mammography; 461 129 (71%) women reported a history of a prior mammographic examination and 192 704 (29%) women did not report having a prior mammographic examination. Data from individual BCSC sites on percentages of cancers that are DCIS and rates of DCIS per 1000 mammograms are not reported, but ranges of these variables over the seven BCSC sites are presented (see Tables 3 and 4). Ninety-five percent confidence intervals (CIs) were calculated for sensitivity, for percentage of cancers that are DCIS, and for the rate of DCIS per 1000 mammograms based on the pooled BCSC data and were not corrected for multiple comparisons. We assumed a normal approximation to the binomial distribution for proportions and a Poisson distribution for each cancer rate. Standard chi-square tests were used as statistical tests for proportions, and Poisson regression was used as the statistical test for comparison of cancer rates. Statistical tests were two-sided. We also plotted the distributions of time-to-diagnosis for cases defined as screen-detected and those defined as non-screen-detected. Results Demographic and Clinical Variables Across the seven BCSC sites, data from 653 833 screening mammography examinations (ranging from 43 624 to 150 387 examinations among 540 738 women were available for evaluation (Table 1). Thus, approximately 20% of women contributed more than one screening mammography examination to the analysis. Based on questionnaire data collected at the time of screening, 29% of all mammographic examinations were classified as first ever: 33% for ages 40–49 years, 27% for ages 50–59 years, 28% for ages 60–69 years, and 29% for ages 70–84 years. Race was self-reported for 86% of the women who were screened; of these women, 79% were white, 5% were African-American, 2% were Asian/Pacific Islander, 2% were Native American, and 12% were other/mixed. Of the 81% of women who responded to a question about whether they were of Hispanic origin, 4% reported being Hispanic. Educational level was either not asked for or not reported for 39% of women. Of those women who did report educational level, 34% completed college, 57% completed high school but not college, and 9% did not complete high school. Because this demographic information was not reported for many women, we did not analyze the results of screening mammography data by race or educational level. Linkage of mammography screening data and cancer case reports resulted in the identification of 3266 cases of breast cancer (2675 invasive and 591 DCIS) occurring within 365 days following screening mammography. The distributions of the number of screening mammography examinations, DCIS cases, and invasive breast cancer cases by age are given in Table 1, where we further distribute by positive (screen-detected) cases. Sensitivity of Screening Mammography for Detecting DCIS The sensitivity of screening mammography for detecting DCIS was higher than that for invasive breast cancer. Of all DCIS cases diagnosed during the study period, 86.0% (508/591, 95% CI = 83.2% to 88.8%) were associated with a positive screening mammography assessment (compared with 75.1% (2010/2675, 95% CI = 73.5% to 76.8%) of all invasive breast cancer cases (Table 1). Sensitivity of screening mammography for detecting DCIS did not differ statistically significantly by age (P = .47). By comparison, sensitivity for detecting invasive breast cancer increased statistically significantly with age (Ptrend<.001), from 66.9% for women aged 40–49 years to 82.7% for women aged 70–84 years (difference in sensitivity = 15.8%, 95% CI = 10.7% to 21.0%). Sensitivity for detecting both types of cancers combined also increased statistically significantly with age (Ptrend<.001) from 71.7% for women aged 40–49 years to 82.8% for women aged 70–84 years (difference in sensitivity = 11.1%, 95% CI = 6.6% to 15.5%), as might be expected, because the invasive breast cancer cases made up 81.9% of all cases. Sensitivity of screening mammography for detecting DCIS was slightly higher for women with no previous mammography (88.6%) than it was for those with previous mammography (84.8%), but these values were not statistically significantly different from one another (difference in sensitivity = 3.8%, 95% CI = –2.0% to 9.6%; P = .25) (Table 2). The highest sensitivity for detecting DCIS was seen in women aged 40–49 years who reported no previous mammography (97.4%, 95% CI = 85.0% to 100%); however, there was no clear association between sensitivity for detecting DCIS and age when the data were stratified by previous mammography status. For invasive breast cancer, the difference in sensitivity between women with no previous mammography (82.3%) and those with previous mammography (71.5%) was greater than that observed for DCIS (difference in sensitivity = 10.8%, 95% CI = 7.5% to 14.0%; P<.001), and sensitivity increased statistically significantly with age for women without and with previous mammography (Ptrend<.001 for both). Percentage of DCIS Among Screen-Detected and Non-Screen-Detected Breast Cancers The percentages of screen-detected and non-screen-detected breast cancer cases that were DCIS across all study sites combined varied by age and by prior screening history, as shown in Table 3. The percentage of cancers that were DCIS differed statistically significantly by age overall (P<.001) and for women with positive mammography screens who had previous mammography (P<.001). However, if the women had positive mammography screens and had not had a previous mammogram, or if they had negative screens, the percentage of cancers that were DCIS did not vary statistically significantly by age (P = .19 and P = .07, respectively). For all age groups combined, 20.2% (95% CI = 18.6% to 21.7%) of screen-detected breast cancers were DCIS, with the percentage being higher among women aged 40–49 years (28.2%, 95% CI = 23.9% to 32.5%) than among women in all older age groups (ranging from 16% [95% CI = 13.3% to 18.7%] to 20.5% [95% CI = 17.5% to 23.6%] among women aged 60–84 years). In every age group, a higher percentage of screen-detected breast cancer was DCIS among women with previous mammography than among women with no previous mammography, with the difference being most pronounced for women aged 40–49 years (31.9% versus 22.6%, respectively). Rates of Screen-Detected DCIS The association of screen-detected DCIS with age differed depending on whether one was examining the percentage of cases that are DCIS or the rate of DCIS per 1000 mammograms (Fig. 1). The percentage of screen-detected breast cancer that was DCIS decreased with age (Fig. 1, A), whereas the DCIS rate per 1000 mammograms increased statistically significantly with age (Fig. 1, B; P<.001). Table 4 shows the rate of screen-detected and non-screen-detected DCIS (for all sites combined) and the ranges of the rates across sites for all DCIS cases. The rate of screen-detected DCIS was 0.56 (95% CI = 0.41 to 0.70) per 1000 mammograms among women aged 40–49 years and increased to 1.07 (95% CI = 0.87 to 1.27) per 1000 mammograms among women aged 70–84 years. The overall rate (i.e., for all age groups combined) of non-screen-detected DCIS (0.13 [95% CI = 0.05 to 0.20] per 1000 mammograms) was only approximately one sixth that of screen-detected DCIS (0.78 [95% CI = 0.60 to 0.95] per 1000 mammograms). Similar to screen-detected DCIS, the rate of non-screen-detected DCIS increased with age from 0.08 (95% CI = 0.02 to 0.13) per 1000 mammograms among women aged 40–49 years to 0.22 (95% CI = 0.13 to 0.31) per 1000 mammograms among women aged 70–84 years (P<.001). Differences in screen-detected DCIS rates between women with and without previous mammography were small, with no consistent direction across age groups. The rate of screen-detected cancers increased more markedly with age for invasive breast cancer than for DCIS, from 1.42 (95% CI = 1.19 to 1.66) per 1000 mammograms among women aged 40–49 years to 5.62 (95% CI = 5.15 to 6.08) per 1000 mammograms among women aged 70–84 years (data not shown). In addition, although in every age group the rates of non-screen-detected invasive breast cancer were lower than those of screen-detected cancers, the rate of non-screen-detected invasive breast cancer was approximately one half that of screen-detected invasive breast cancer among women aged 40–49 years (0.70, 95% CI = 0.54 to 0.87 per 1000 mammograms) and approximately one fifth that of screen-detected invasive breast cancer among women aged 70–84 years (1.17, 95% CI = 0.96 to 1.38 per 1000 mammograms) (data not shown). Time to Diagnosis of DCIS The time to DCIS diagnosis following screening mammography was generally much shorter for women with a positive mammography screen than for women with a negative screen (Fig. 2). Whereas 91% of DCIS cases associated with a positive screen were diagnosed within 3 months after screening, the distribution of DCIS cases diagnosed following a negative screen was spread throughout the subsequent year. The distribution of time to diagnosis following a negative screen had distinct peaks at around 1, 6–7, and 12 months. This distribution is consistent with the occurrence of further follow-up mammograms and subsequent diagnostic evaluation immediately following the screening mammography examination (perhaps due to clinical findings, such as a breast mass, even though the indication for the examination was routine screening), at the 6-month short-interval follow-up and at the repeat annual screening mammography examination occurring just prior to 12 months. Discussion Our study of nearly 654 000 mammography screening examinations among nearly 541 000 women aged 40–84 years from seven mammography registries throughout the United States showed that approximately 20% of all screen-detected breast cancers were DCIS and that the overall rate of DCIS detection by screening mammography was 0.78 per 1000 mammograms; this rate increased with age. Our results suggest that one case of DCIS is detected for approximately every 1300 screening mammography examinations performed. This rate varies by age, ranging from approximately 1 in every 1800 mammograms among women aged 40–49 years to 1 in every 935 mammograms among women aged 70–84 years. Several reports regarding the occurrence of in situ lesions have been published that were based on data from population-based cancer registries in the United States (1,2,4,13) and elsewhere (3,5). Data from cancer registries, however, do not permit determination of the incidence of screen-detected disease specifically, because cancer registries do not routinely collect information on screening mammography practices either for individual cancer cases or for the populations they cover. To the extent that non-screened women are included in cancer registry statistics, estimates of DCIS incidence based on cancer data from population-based cancer registries will underestimate the incidence of DCIS among screened women because DCIS is primarily detected by mammography. Reports from individual mammography programs and from case series have provided additional estimates of the percentage of all breast cancers or of nonpalpable breast cancers that are DCIS (14–16). There are, however, several reports of the occurrence of in situ lesions among women participating in large, organized screening mammography programs. The Breast Cancer Detection Demonstration Project, conducted in the 1970s, reported that 18% of breast cancers detected at initial screening mammography in study participants were intraductal and in situ cases, with a smaller percentage of DCIS being detected in subsequent screenings (17). In the early 1990s, community-based screening mammography programs in New Mexico and British Columbia found that approximately 20% of all breast cancers detected were in situ (18,19). The U.S. National Breast and Cervical Cancer Early Detection Project reported in 1998 (20,21) that 25.3% of all breast cancers detected on the first round of screening mammography and 32.8% of those detected on subsequent rounds of screening were in situ. DCIS detection rates in this study were approximately 1.5 cases per 1000 first-round mammograms and 0.9 cases per 1000 subsequent mammograms. Several international studies (22–24) have also found that the proportion of screen-detected cancers that were in situ ranged from 15% to 22%. Precise comparisons of either the percentage of screen-detected cancers that are DCIS or of the rates of DCIS per 1000 screening mammograms across screened populations are not possible because of differences in calendar period, age distributions, racial/ethnic composition, socioeconomic status, recommended screening intervals, definitions of screen-detected cancer and prior mammography, and type of in situ cases included. Nevertheless, the rates we estimated in this study are similar to those reported elsewhere. In this study, we found that the percentage of all screen-detected breast cancer that was DCIS decreased with increasing age. That is, if a woman had screen-detected breast cancer, it was less likely to be DCIS if she was older. Other investigators (25,26) with similar findings have misinterpreted the percentages of DCIS in their studies as meaning that the incidence of DCIS is higher in younger women than it is in older women. However, as was the case for invasive breast cancer, we found that the rate of DCIS detection actually increased with age, from 0.56 per 1000 mammograms among women aged 40–49 years to 1.07 per 1000 mammograms among women aged 70–84 years. Other studies (20,22,23) have confirmed that the percentage of DCIS among all screen-detected breast cancers does generally decrease with age, but those studies did not always show a consistent increase in the rate of detection of DCIS with age. Similar to another report (27), we found that screening mammography was more sensitive for detecting DCIS (86%) than it was for detecting invasive breast cancer (75%). Invasive carcinoma is usually associated with a mass or density found on mammography, whereas DCIS is usually associated with microcalcifications (25). Missed DCIS is much less likely than missed invasive breast cancer to be diagnosed subsequently as a palpable mass, and it may never become clinically apparent. This phenomenon has the potential effect of overestimating the ability of screening mammography to detect in situ disease, so that sensitivity for detecting DCIS would be biased upward. Furthermore, in dense or heterogeneously dense breast tissue, microcalcifications are easier to detect on mammography than are masses. In our study, sensitivity of screening mammography for detecting both DCIS and invasive breast cancer was higher among women for whom no previous mammography was recorded than it was for women with previous mammography. Earlier work has also shown that the sensitivity of mammography is higher for women without a known previous mammogram, presumably because the cancers available for screen detection include prevalent cases that may be larger and/or slower growing than cancers among women undergoing regular screening (28). Although we found that sensitivity for detecting invasive breast cancer increased with age, both for women with previous mammography and for women with no previous mammography, there was no clear relationship between sensitivity for detecting DCIS and age. Among women with no previous mammography, however, sensitivity for detecting DCIS was higher in women younger than 50 years of age than for older women. Thus, the proportion of DCIS cases in different study populations, as well as age and previous mammography status of the women, will influence reported overall sensitivities of screening mammography for detecting breast cancer. Surprisingly, given the general understanding that DCIS is readily detected by screening mammography, we found that 14% of DCIS cases (83/591) were among women whose most recent screening mammography examination had a negative assessment. It should be noted that the screening examinations of 21 of the 83 non-screen-detected DCIS cases had an assessment code of 3 with a recommendation for short interval follow-up (i.e. a follow-up generally occurring within 3–6 months from a screening mammography examination), which indicates that the radiologist had some concern about the result even though an immediate work-up was not recommended. Over 90% of DCIS diagnoses for women with positive screens occurred within 3 months of mammography, whereas DCIS diagnoses for women with negative screens were spread throughout the year following mammography, suggesting that many of these DCIS cases were detected by other means. We do not know, however, whether these non-screen-detected DCIS cases represent cases that were present but missed at the time of screening mammography or were new (i.e., interval) occurrences of DCIS detected by means other than mammography. For example, some studies (29,30) indicate that some DCIS cases can be detected by clinical breast examination. Our study has several limitations. First, we relied on self-reported questionnaire data to classify women as to whether they had had a previous screening mammography examination, and if there was no information provided on the questionnaire about previous mammography, we assumed that the screening mammogram was the woman’s first ever. To the extent that misclassification occurred, “true” differences between women with and without previous mammography would be diminished. Second, we used the initial mammographic assessment category assigned to each screening mammogram to determine whether the finding was positive or negative. Subsequent imaging mammograms of the study participants may have resulted in a reappraisal of the BI-RADS assessment code and its associated recommendation. However, estimates of sensitivity for detecting cancer are most commonly based on the initial assessment, so we chose to use the initial mammographic assessment to allow comparison to existing literature. Given that the likelihood of detecting DCIS among women undergoing screening mammography has increased and is not small at 20% of all screen-detected breast cancer cases, is there evidence to suggest that detecting DCIS prevents deaths from breast cancer? The rationale for early detection and treatment of DCIS is based on several lines of evidence. First, a small series of untreated women with DCIS diagnosed in the premammography era had higher than expected subsequent occurrences of invasive breast cancer (31,32). In addition, larger, more recent series of women treated for DCIS have also been shown to be at increased risk of recurrence of both in situ and invasive breast cancer (33). The estimates of recurrence rates range fairly broadly across studies and are influenced by patient and tumor characteristics, such as age and tumor grade, and by treatment (34–36). In a large clinical trial, after a mean follow-up of 90 months, the incidence of ipsilateral invasive breast cancer among women with DCIS who had been treated with lumpectomy alone was 13.4% (37). Thus, DCIS can recur as invasive breast cancer. Findings from randomized clinical trials that have shown that the addition of radiation therapy and tamoxifen to lumpectomy treatment of DCIS reduces the chance of future invasive disease recurrence compared with lumpectomy alone also bolster the case for treatment of DCIS (37–39). Finally, data are accumulating to suggest that there are both epidemiologic and genetic similarities between DCIS and invasive breast cancers, further suggesting that treatment for DCIS may yield benefits similar to those of treatment for invasive breast cancer (40–46). In contrast, some lines of evidence temper enthusiasm for detecting DCIS. For example, autopsy studies have shown that DCIS is not uncommon among women who died of causes unrelated to breast cancer; across seven autopsy studies, median prevalence of DCIS was 8.9% (range = 0% to 14.7%), although, presumably, not all such cases of DCIS would have been detectable by mammography (47). This finding suggests that some cases of DCIS do not progress to clinically significant lesions and may never require treatment in the patient’s lifetime. In addition, although their risk of subsequent invasive disease is elevated, most women treated for DCIS do not experience an invasive recurrence, at least within 10–15 years after DCIS diagnosis. Follow-up data from population-based registries show that only about 2% of women who have been diagnosed with and treated for DCIS since the time when screening mammography was widely introduced in the early 1990s died of breast cancer within 10 years following DCIS diagnosis (48,49). These favorable outcomes presumably reflect both the effectiveness of current treatment regimens and the relatively benign nature of most mammographically detected DCIS. Finally, even though over four times as many in situ cases were found among women in the Canadian National Breast Screening Study-2 who had received mammography plus clinical breast examination as were found among women who had received clinical breast examination alone, there was no overall difference in breast cancer mortality between the two groups of women after 13 years of follow-up (30). The use of screening mammography has become widespread. Based on 1998 nationally representative data from the National Health Interview Survey, the estimated number of U.S. women aged 40 years and older who had mammography in the previous year was more than 28.4 million (Breen N, National Cancer Institute: personal communication). Our data suggest that approximately 1 in every 1300 screening mammography examinations results in a DCIS diagnosis. If early detection and treatment of DCIS have contributed to the recent decline in U.S. breast cancer mortality (50), independent of effects attributable to early detection and treatment of invasive disease and independent of recent advances in breast cancer treatment, then some women who have been treated for screen-detected DCIS have benefited. The pathology of DCIS is heterogeneous, and presumably, the likelihood of benefit from treatment is greater for women with larger, higher grade, or multifocal lesions than for those with very small low-grade lesions. The BCSC affords the opportunity to explore both the distribution of grade of DCIS in a population-based mammography program and the long-term sequelae of DCIS identification. More research is needed to understand the biology and clinical significance of DCIS to identify disease that is likely to progress and to better tailor treatment decisions. Table 1. Number of screening mammography examinations and the percentage of women with ductal carcinoma in situ (DCIS) or invasive breast cancer who had a positive screen (sensitivity), by age*     DCIS  Invasive breast cancer  DCIS and invasive breast cancer combined  Age range, y  No. of screening mammography examinations  No. of cases diagnosed  Percentage of cases with positive screen  No. of cases diagnosed  Percentage of cases with positive screen†  Total no. of cases diagnosed  Percentage of cases with positive screen†  *Positive screen is a mammography screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.001).  40–49  211 551  134  88.1  450  66.9  584  71.7  50–59  200 255  155  88.4  792  72.2  947  74.9  60–69  135 376  165  84.2  709  75.9  874  77.5  70–84  106 651  137  83.2  724  82.7  861  82.8      Total  653 833  591  86.0  2675  75.1  3266  77.1      DCIS  Invasive breast cancer  DCIS and invasive breast cancer combined  Age range, y  No. of screening mammography examinations  No. of cases diagnosed  Percentage of cases with positive screen  No. of cases diagnosed  Percentage of cases with positive screen†  Total no. of cases diagnosed  Percentage of cases with positive screen†  *Positive screen is a mammography screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.001).  40–49  211 551  134  88.1  450  66.9  584  71.7  50–59  200 255  155  88.4  792  72.2  947  74.9  60–69  135 376  165  84.2  709  75.9  874  77.5  70–84  106 651  137  83.2  724  82.7  861  82.8      Total  653 833  591  86.0  2675  75.1  3266  77.1  View Large Table 2. Percentage of women with ductal carcinoma in situ (DCIS) or invasive breast cancer who had a positive screen (sensitivity), by age and previous mammography*   DCIS  Invasive breast cancer    No previous mammogram  Previous mammogram  No previous mammogram  Previous mammogram  Age range, y  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen†  No. of cases  Percentage of cases with positive screen†  *Positive screen is a mammography-screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.01).  40–49  39  97.4  95  84.2  172  75.6  278  61.5  50–59  46  87.0  109  89.0  254  79.1  538  69.0  60–69  45  84.4  120  84.2  209  83.3  500  72.8  70–84  46  87.0  91  81.3  263  89.0  461  79.2      Total  176  88.6  415  84.8  898  82.3  1777  71.5    DCIS  Invasive breast cancer    No previous mammogram  Previous mammogram  No previous mammogram  Previous mammogram  Age range, y  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen  No. of cases  Percentage of cases with positive screen†  No. of cases  Percentage of cases with positive screen†  *Positive screen is a mammography-screening examination associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to cancer diagnosis. Percentage of cases with a positive screen is sensitivity.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (Ptrend<.01).  40–49  39  97.4  95  84.2  172  75.6  278  61.5  50–59  46  87.0  109  89.0  254  79.1  538  69.0  60–69  45  84.4  120  84.2  209  83.3  500  72.8  70–84  46  87.0  91  81.3  263  89.0  461  79.2      Total  176  88.6  415  84.8  898  82.3  1777  71.5  View Large Table 3. Percentage of ductal carcinoma in situ (DCIS) cases among all breast cancer cases, by age, previous mammography, and whether screening mammography was positive or negative*   Percentage of DCIS among all breast cancer cases with a positive screen  Percentage of DCIS among all cancer cases with a negative screen  Percentage of DCIS among all breast cancer cases  Age range, y  No previous mammogram  Previous mammogram  Total (95% CI)†  Total (95% CI)  Total (95% CI)†  Range over BCSC sites  *Mammography screening was considered positive when it was associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) and considered negative when associated with a negative screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval; BCSC = Breast Cancer Surveillance Consortium.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  22.6  31.9  28.2 (23.9 to 32.5)  9.7 (5.2 to 14.2)  22.9 (19.5 to 26.4)  18.3–30.3  50–59  16.6  20.7  19.3 (16.4 to 22.2)  7.6 (4.2 to 10.9)  16.4 (14.0 to 18.7)  11.1–24.7  60–69  17.9  21.7  20.5 (17.5 to 23.6)  13.2 (8.5 to 17.9)  18.9 (16.3 to 21.5)  16.0–23.9  70–84  14.6  16.9  16.0 (13.3 to 18.7)  15.5 (9.7 to 21.4)  15.9 (13.5 to 18.4)  9.1–21.8      Total  17.4  21.7  20.2 (18.6 to 21.7)  11.1 (8.8 to 13.4)  18.1 (16.8 to 19.4)  14.6–23.8    Percentage of DCIS among all breast cancer cases with a positive screen  Percentage of DCIS among all cancer cases with a negative screen  Percentage of DCIS among all breast cancer cases  Age range, y  No previous mammogram  Previous mammogram  Total (95% CI)†  Total (95% CI)  Total (95% CI)†  Range over BCSC sites  *Mammography screening was considered positive when it was associated with a positive screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) and considered negative when associated with a negative screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval; BCSC = Breast Cancer Surveillance Consortium.  †Proportions using the chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  22.6  31.9  28.2 (23.9 to 32.5)  9.7 (5.2 to 14.2)  22.9 (19.5 to 26.4)  18.3–30.3  50–59  16.6  20.7  19.3 (16.4 to 22.2)  7.6 (4.2 to 10.9)  16.4 (14.0 to 18.7)  11.1–24.7  60–69  17.9  21.7  20.5 (17.5 to 23.6)  13.2 (8.5 to 17.9)  18.9 (16.3 to 21.5)  16.0–23.9  70–84  14.6  16.9  16.0 (13.3 to 18.7)  15.5 (9.7 to 21.4)  15.9 (13.5 to 18.4)  9.1–21.8      Total  17.4  21.7  20.2 (18.6 to 21.7)  11.1 (8.8 to 13.4)  18.1 (16.8 to 19.4)  14.6–23.8  View Large Table 4. Screen-detected and non-screen-detected ductal carcinoma in situ (DCIS) rates per 1000 mammograms, by age and previous mammography*   Screen-detected DCIS rate  Non-screen-detected DCIS rate  DCIS rates, all cases  Age range, y  No previous mammogram†  Previous mammogram†  Total (95% CI)†  Total (95% CI)†  Total (95% CI)†  Range over BCSC sites  *Screen-detected = cases of DCIS associated with a positive mammography-screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to diagnosis. Non-screen-detected = cases of DCIS associated with a negative mammography-screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval.  †Proportions using chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  0.54  0.57  0.56 (0.41 to 0.70)  0.08 (0.02 to 0.13)  0.63 (0.48 to 0.79)  0.44–1.10  50–59  0.74  0.66  0.68 (0.52 to 0.85)  0.09 (0.03 to 0.15)  0.77 (0.60 to 0.95)  0.51–1.38  60–69  1.00  1.04  1.03 (0.83 to 1.23)  0.19 (0.11 to 0.28)  1.22 (1.00 to 1.44)  1.00–2.26  70–84  1.31  0.97  1.07 (0.87 to 1.27)  0.22 (0.13 to 0.31)  1.28 (1.06 to 1.51)  0.50–1.75      Total  0.81  0.76  0.78 (0.60 to 0.95)  0.13 (0.05 to 0.20)  0.90 (0.72 to 1.09)  0.72–1.42    Screen-detected DCIS rate  Non-screen-detected DCIS rate  DCIS rates, all cases  Age range, y  No previous mammogram†  Previous mammogram†  Total (95% CI)†  Total (95% CI)†  Total (95% CI)†  Range over BCSC sites  *Screen-detected = cases of DCIS associated with a positive mammography-screening assessment (American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment code of 0 - needs additional imaging, 3 - probably benign finding with a recommendation for immediate work-up, 4 - suspicious abnormality, or 5 - highly suggestive of malignancy) within 365 days prior to diagnosis. Non-screen-detected = cases of DCIS associated with a negative mammography-screening assessment (BI-RADS assessment code of 1 - negative finding, 2 - benign finding, or 3 - probably benign finding with no recommendation for immediate work-up or an unknown recommendation) within 365 days prior to cancer diagnosis. CI = confidence interval.  †Proportions using chi-square (χ2) test are statistically significantly different across age categories (P<.001).  40–49  0.54  0.57  0.56 (0.41 to 0.70)  0.08 (0.02 to 0.13)  0.63 (0.48 to 0.79)  0.44–1.10  50–59  0.74  0.66  0.68 (0.52 to 0.85)  0.09 (0.03 to 0.15)  0.77 (0.60 to 0.95)  0.51–1.38  60–69  1.00  1.04  1.03 (0.83 to 1.23)  0.19 (0.11 to 0.28)  1.22 (1.00 to 1.44)  1.00–2.26  70–84  1.31  0.97  1.07 (0.87 to 1.27)  0.22 (0.13 to 0.31)  1.28 (1.06 to 1.51)  0.50–1.75      Total  0.81  0.76  0.78 (0.60 to 0.95)  0.13 (0.05 to 0.20)  0.90 (0.72 to 1.09)  0.72–1.42  View Large Fig. 1. View largeDownload slide Percentage of ductal carcinoma in situ (DCIS) cases among all screen-detected breast cancers and screen-detected DCIS rates per 1000 mammograms. A) Percentage of all screen-detected breast cancers that were diagnosed as DCIS stratified by age group. B) Rate of screen-detected DCIS per 1000 mammograms stratified by age group. Cancers were considered screen-detected if they were associated with a positive screening mammography examination according to the American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment codes of 0 (needs additional evaluation), 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality), or 5 (highly suggestive of malignancy) within 365 days prior to cancer diagnosis. The data in A are in Table 3, and the data in B are in Table 4. Fig. 1. View largeDownload slide Percentage of ductal carcinoma in situ (DCIS) cases among all screen-detected breast cancers and screen-detected DCIS rates per 1000 mammograms. A) Percentage of all screen-detected breast cancers that were diagnosed as DCIS stratified by age group. B) Rate of screen-detected DCIS per 1000 mammograms stratified by age group. Cancers were considered screen-detected if they were associated with a positive screening mammography examination according to the American College of Radiology’s Breast Imaging Reporting and Data System [BI-RADS] assessment codes of 0 (needs additional evaluation), 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality), or 5 (highly suggestive of malignancy) within 365 days prior to cancer diagnosis. The data in A are in Table 3, and the data in B are in Table 4. Fig. 2. View largeDownload slide Distribution of ductal carcinoma in situ (DCIS) cases by time from screening mammography examination to DCIS diagnosis. A) Cases of DCIS diagnosed after a positive screen. DCIS cases were considered screen-detected if they were associated with a positive screening mammographic examination according to the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) assessment codes of 0 (needs additional evaluation); 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality); or 5 (highly suggestive of malignancy) within the 365 days prior to DCIS diagnosis. B) Cases of DCIS diagnosed after a negative screen. DCIS cases were considered non-screen-detected if they were associated with a negative screening mammographic examination (BI-RADS assessment codes of 1 [negative], 2 [benign finding], or 3 [probably benign finding with no recommendation for immediate work-up or unknown recommendation]) within 365 days prior to DCIS diagnosis. Fig. 2. View largeDownload slide Distribution of ductal carcinoma in situ (DCIS) cases by time from screening mammography examination to DCIS diagnosis. A) Cases of DCIS diagnosed after a positive screen. DCIS cases were considered screen-detected if they were associated with a positive screening mammographic examination according to the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) assessment codes of 0 (needs additional evaluation); 3 (probably benign finding, with a recommendation for immediate work-up); 4 (suspicious abnormality); or 5 (highly suggestive of malignancy) within the 365 days prior to DCIS diagnosis. B) Cases of DCIS diagnosed after a negative screen. DCIS cases were considered non-screen-detected if they were associated with a negative screening mammographic examination (BI-RADS assessment codes of 1 [negative], 2 [benign finding], or 3 [probably benign finding with no recommendation for immediate work-up or unknown recommendation]) within 365 days prior to DCIS diagnosis. Supported by cooperative agreements U01CA63731, U01CA63736, U01CA63740, U01CA69976, U01CA70013, U01CA70040, U01CA86076, and U01CA86082, as part of the National Cancer Institute’s (NCI’s) Breast Cancer Surveillance Consortium, and by grant R01CA63146 from the NCI, National Institutes of Health, Department of Health and Human Services. References 1 Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson IC. Incidence and treatment for ductal carcinoma in situ of the breast. JAMA  1996; 275: 913–8. Google Scholar 2 Choi WS, Parker BA, Pierce JP, Greenberg ER. Regional differences in the incidence and treatment of carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev  1996; 5: 317–20. Google Scholar 3 Levi F, Te VC, Randimbison L, La Vecchia C. Trends of in situ carcinoma of the breast in Vaud, Switzerland. Eur J Cancer  1997; 33: 903–6. Google Scholar 4 Adams-Cameron M, Gilliland FD, Hunt WC, Key CR. Trends in incidence and treatment for ductal carcinoma in situ in Hispanic, American Indian, and non-Hispanic white women in New Mexico, 1973–1994. Cancer  1999; 85: 1084–90. Google Scholar 5 Barchielli A, Paci E, Girogi D. Recent trends of in situ carcinoma of the breast and mammographic screening in the Florence area, Italy. Cancer Causes Control  1999; 10: 313–7. Google Scholar 6 Winchester DP, Menck HR, Osteen RT, Kraybill W. Treatment trends for ductal carcinoma in situ of the breast. Ann Surg Oncol  1995; 2: 207–13. Google Scholar 7 Schwartz GF, Solin LJ, Olivotto IA, Ernster VL, Pressman PI, and the Consensus Conference Committee. The Consensus Conference on the Treatment of In Situ Ductal Carcinoma of the Breast, April 22–25, 1999. Hum Pathol  2000; 31: 131–9. Google Scholar 8 Schwartz LM, Woloshin S, Sox HC, Fischhoff B, Welch HG. US women’s attitudes to false positive mammography results and detection of ductal carcinoma in situ: cross sectional survey. BMJ  2000; 320: 1635–40. Google Scholar 9 Ballard-Barbash R, Taplin SH, Yankaskas BC, Ernster VL, Rosenberg RD, Carney PA, et al. Breast Cancer Surveillance Consortium: a national mammography screening and outcomes database. AJR Am J Roentgenol  1997; 169: 1001–8. Google Scholar 10 Carney PA, Geller BM, Moffett H, Ganger M, Sewell M, Barlow WE, et al. Current medicolegal and confidentiality issues in a large multicenter research program. Am J Epidemiol  2000; 152: 371–8. Google Scholar 11 American College of Radiology. Illustrated Breast Imaging Reporting and Data System (BI-RADS TM). 3rd ed. Reston (VA): American College of Radiology; 1998. p. 179–81. Google Scholar 12 Wu XC, Hotes JL, Fulton PJ, Cormier M, Corre CN, McLaughlin CC, et al., editors. Cancer in North America, 1995–1999. Vol. 1. Incidence. Springfield (IL): North American Association of Cancer Registries; April 2002. p. II.33, II.45, II.193, II.271. Google Scholar 13 Ries LA, Eisner MP, Kosary CL, Hankey BF, Miller BA, Clegg L, et al. SEER Cancer Statistics Review, 1973–1998. Bethesda (MD): National Cancer Institute; 2001. Google Scholar 14 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 15 Frykberg ER, Bland KI. In situ breast carcinoma. In: Cameron JL, editor. Advances in surgery. Chicago (IL): Mosby Year Book, Inc.; 1993. p. 29–72. Google Scholar 16 Evans AJ, Pinder SE, Ellis IO, Wilson AR. Screen detected ductal carcinoma in situ (DCIS): overdiagnosis or an obligate precursor of invasive disease? J Med Screen  2001; 8: 149–51. Google Scholar 17 Seidman H, Gelb SK, Silverberg E, LaVerda N, Lubera JA. Survival experience in the Breast Cancer Detection Demonstration Project. CA Cancer J Clin  1987; 37: 258–90. Google Scholar 18 Rosenberg RD, Lando JF, Hunt WC, Darling RR, Williamson MR, Linver MN, et al. The New Mexico Mammography Project: screening mammography performance in Albuquerque, New Mexico, 1991 to 1993. Cancer  1996; 78: 1731–9. Google Scholar 19 Olivotto IA, Kan L, d’Yachkova Y, Burhenne LJ, Hayes M, Hislop TG. Ten years of breast screening in the Screening Mammography Program of British Columbia. J Med Screen  2000; 7: 152–9. Google Scholar 20 May DS, Lee NC, Nadel MR, Henson RM, Miller DS. The National Breast and Cervical Cancer Early Detection Program: report on the first 4 years of mammography provided to medically underserved women. AJR Am J Roentgenol  1998; 170: 97–104. Google Scholar 21 May DS, Lee NC, Richardson LC, Giustozzi AG, Bobo JK. Mammography and breast cancer detection by race and Hispanic ethnicity: results from a national program (United States). Cancer Causes Control  2000; 11: 697–705. Google Scholar 22 Canada H. Organized breast cancer screening programs in Canada: 1996 report: Laboratory Centre for Disease Control, Health Canada. Minister of Public Works and Government Services Canada; 1999. Google Scholar 23 UK Trial of Early Detection of Breast Cancer Group. 16-year mortality from breast cancer in the UK Trial of Early Detection of Breast Cancer. Lancet  1999; 353: 1909–14. Google Scholar 24 Fracheboud J, Groenewoud JH, Boer RI, van Ineveld BM, Verbeek AL, Hendricks JH, et al. Landelijke evaluatie van bevolkingsonderzoek naar borstkanker in Nederland. Rotterdam (The Netherlands): 2000. Google Scholar 25 Wazer DE, Gage I, Homer MJ, Krosnick SH, Schmid C. Age-related differences in patients with nonpalpable breast carcinomas. Cancer  1996; 78: 1432–7. Google Scholar 26 Evans WP, Starr AL, Bennos ES. Comparison of the relative incidence of impalpable invasive breast carcinoma and ductal carcinoma in situ in cancers detected in patients older and younger than 50 years of age. Radiology  1997; 204: 489–91. Google Scholar 27 Kerlikowske K, Grady D, Barclay J, Sickles EA, Ernster VL. Effect of age, breast density, and family history on the sensitivity of first screening mammography. JAMA  1996; 276: 33–8. Google Scholar 28 Rosenberg RD, Yankaskas BC, Hunt WC, Ballard-Barbash R, Urban N, Ernster VL, et al. Effect of variations in operational definitions on performance estimates for screening mammography. Acad Radiol  2000; 7: 1058–68. Google Scholar 29 Bobo JK, Lee NC, Thames SF. Findings from 752,081 clinical breast examinations reported to a national screening program from 1995 through 1998. J Natl Cancer Inst  2000; 92: 971–6. Google Scholar 30 Miller AB, To T, Baines CJ, Wall C. Canadian National Breast Screening Study-2: 13-year results of a randomized trial in women aged 50–59 years. J Natl Cancer Inst  2000; 92: 1490–9. Google Scholar 31 Page DL, Dupont WD, Rogers LW, Jensen RA, Schuyler PA. Continued local recurrence of carcinoma 15–25 years after a diagnosis of low grade ductal carcinoma in situ of the breast treated only by biopsy. Cancer  1995; 76: 1197–200. Google Scholar 32 Eusebi V, Foschini MP, Cook MG, Berrino F, Azzopardi JG. Long-term follow-up of in situ carcinoma of the breast with special emphasis on clinging carcinoma. Semin Diagn Pathol  1989; 6: 165–73. Google Scholar 33 Fonseca R, Hartmann LC, Petersen IA, Donohue JH, Crotty TB, Gisvold JJ. Ductal carcinoma in situ of the breast. Ann Intern Med  1997; 127: 1013–22. Google Scholar 34 Van Zee KJ, Liberman L, Samli B, Tran KN, McCormick B, Petrek JA, et al. Long term follow-up of women with ductal carcinoma in situ treated with breast-conserving surgery: the effect of age. Cancer  1999; 86: 1757–67. Google Scholar 35 Franceschi S, Levi F, La Vecchia C, Randimbison L, Te VC. Second cancers following in situ carcinoma of the breast. Int J Cancer  1998; 77: 392–5. Google Scholar 36 Mirza NQ, Vlastos G, Meric F, Sahin AA, Singletary SE, Newman LA, et al. Ductal carcinoma-in-situ: long-term results of breast-conserving therapy. Ann Surg Oncol  2000; 7: 656–64. Google Scholar 37 Fisher B, Dignam J, Wolmark N, Mamounas E, Castantino J, Poller W, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from the National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol  1998; 16: 441–52. Google Scholar 38 Fisher B, Dignam J, Wolmark N, Wickerham LD, Fisher ER, Mamounas E, et al. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet  1999; 353: 1993–2000. Google Scholar 39 Julien JP, Bijker N, Fentiman IS, Peterse JL, Delledonne V, Rouanet P, et al. Radiotherapy in breast-conserving treatment for ductal carcinoma in situ: first results of the EORTC randomized phase III trial 10853. Lancet  2000; 355: 528–33. Google Scholar 40 Longnecker MP, Bernstein L, Paganini-Hill A, Enger SM, Ross RK. Risk factors for in situ breast cancer. Cancer Epidemiol Biomarkers Prev  1996; 5: 961–5. Google Scholar 41 Kerlikowske K, Barclay J, Grady D, Sickles EA, Ernster VL. Comparison of risk factors for ductal carcinoma in situ and invasive breast cancer. J Natl Cancer Inst  1997; 89: 76–82. Google Scholar 42 Lambe M, Hsieh CC, Tsaih SW, Ekbom A, Trichopoulos D, Adami HO. Parity, age at first birth and the risk of carcinoma in situ of the breast. Int J Cancer  1998; 77: 330–2. Google Scholar 43 Zhuang Z, Merino MJ, Chuaqui R, Liotta LA, Emmert-Buck MR. Identical allelic loss on chromosome 11q13 in microdissected in situ and invasive human breast cancer. Cancer Res  1995; 55: 467–71. Google Scholar 44 James LA, Mitchell EL, Menasce L, Varley JM. Comparative genomic hybridisation of ductal carcinoma of the breast: identification of regions of DNA amplification and deletion in common with invasive breast carcinoma. Oncogene  1997; 14: 1059–65. Google Scholar 45 Marsh KL, Varley JM. Loss of heterozygosity at chromosome 9p in ductal carcinoma in situ and invasive carcinoma of the breast. Br J Cancer  1998; 77: 1439–47. Google Scholar 46 Trentham-Dietz A, Newcomb PA, Storer BE, Remington PL. Risk factors for carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev  2000; 9: 697–703. Google Scholar 47 Welch HG, Black WC. Using autopsy series to estimate the disease “reservoir” for ductal carcinoma in situ of the breast: how much more breast cancer can we find? Ann Intern Med  1997; 127: 1023–8. Google Scholar 48 Ernster VL, Barclay J, Kerlikowske K, Wilkie H, Ballard-Babash R. Mortality among women with ductal carcinoma in situ of the breast in the population-based Surveillance, Epidemiology and End Results program. Arch Intern Med  2000; 160: 953–8. Google Scholar 49 Warnberg F, Bergh J, Holmberg L. Prognosis in women with a carcinoma in situ of the breast: a population-based study in Sweden. Cancer Epidemiol Biomarkers Prev  1999; 8: 769–74. Google Scholar 50 Howe HL, Wingo PA, Thun MJ, Ries LA, Rosenberg HM, Feigal EG, et al. Annual report to the nation on the status of cancer (1973 through 1998), featuring cancers with recent increasing trends. J Natl Cancer Inst  2001; 93: 824–42. Google Scholar © Oxford University Press

Journal

JNCI: Journal of the National Cancer InstituteOxford University Press

Published: Oct 16, 2002

There are no references for this article.