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

Learn More →

The association of mammographic density with ductal carcinoma in situ of the breast: the Multiethnic Cohort

The association of mammographic density with ductal carcinoma in situ of the breast: the... Introduction It is well established that women with high Results Mammographic density was associated with invasive mammographic density are at greater risk for breast cancer than breast cancer and breast DCIS. For the highest category of are women with low breast density. However, little research has percentage breast density (≥50%) as compared with the lowest been done on mammographic density and ductal carcinoma in (<10%), the OR was 3.58 (95% CI 2.26–5.66) for invasive situ (DCIS) of the breast, which is thought to be a precursor breast cancer and 2.86 (1.38–5.94) for breast DCIS. Similarly, lesion to some invasive breast cancers. for the highest category of dense area (≥45 cm ) as compared with the lowest (<15 cm ), the OR was 2.92 (95% CI 2.01– Method We conducted a nested case-control study within the 4.25) for invasive breast cancer and 2.59 (1.39–4.82) for breast Multiethnic Cohort, and compared the mammographic densities DCIS. Trend tests were significant for invasive breast cancer (P of 482 patients with invasive breast cancer and 119 with breast for trend < 0.0001) and breast DCIS (P for trend < 0.001) for DCIS cases versus those of 667 cancer-free control subjects. A both percentage density and dense area. reader blinded to disease status performed computer-assisted density assessment. For women with more than one mammogram, mean density values were computed. Polytomous logistic regression models were used to compute adjusted odds Conclusion The similar strength of association for ratios (ORs) and 95% confidence intervals (CIs) for two mammographic density with breast DCIS and invasive breast measurements of mammographic density: percentage density cancer supports the hypothesis that both diseases may have a and dense area. common etiology. cases [5] or combined invasive and CIS cases [6-10]. One study investigating breast CIS observed that DCIS was more Introduction likely to occur in mammographically dense areas [11], and As a result of increasing early detection efforts, breast carci- another study reported an increase in breast hyperplasia or noma in situ (CIS) constitutes more than 20% of newly diag- atypia/CIS in women with greater than 50% breast density nosed breast cancer cases in the USA [1]. Although breast [12]. This analysis examines the association between mammo- CIS shares a number of risk factors with invasive breast can- graphic density and risk for breast DCIS in comparison with cer, and ductal carcinoma in situ (DCIS) of the breast is con- invasive breast cancer and breast cancer-free control sidered to be a precursor to some invasive breast cancers [2], subjects. it is not clear to what extent breast CIS and invasive breast cancer have the same etiology. Several case-control studies Materials and methods have confirmed that mammographic density is associated with Study population risk for breast cancer [3]; women with high breast density have The data for this analysis were collected using a nested case- a three-fold to six-fold greater risk for developing breast cancer control study within the Hawaii component of the Multiethnic than women with low breast density [3,4]. Most studies of Cohort, which was established between 1993 and 1996 [13]. mammographic density included only invasive breast cancer As described in detail elsewhere [4], all female members BHQ = Breast Health Questionnaire; BMI = body mass index; CI = confidence interval; CIS = carcinoma in situ; DCIS = ductal carcinoma in situ; HRT = hormone replacement therapy; OR = odds ratio; ROC = receiver operating characteristic. Page 1 of 6 (page number not for citation purposes) Breast Cancer Research Vol 8 No 3 Gill et al. Table 1 Characteristics of the study population by disease status Covariate Invasive breast cancer Breast DCIS Controls P value Sample size 482 119 667 - Ethnicity (%) 0.19 Hawaiian 14.3 8.4 24.3 Japanese 46.5 55.5 43.8 Caucasian 32.6 28.6 28.0 Other 6.6 7.6 3.9 Age at diagnosis (years) 63.5 62.9 N/A 0.50 Age at recruitment (years) 60.0 59.6 56.7 0.58 Mean age at all mammograms (years) 59.8 59.1 59.7 0.70 Body mass index (kg/m ) 25.2 24.5 25.7 0.07 Time from first mammogram to diagnosis (years) 6.2 6.7 N/A 0.17 Family history of breast cancer (%) 16.8 16.8 12.0 0.05 Age at first birth (years) 25.0 25.1 24.7 0.65 Age at menarche (years) 13.1 12.9 13.1 0.47 Parous (%) 84.5 83.9 88.5 0.10 Number of children 2.4 2.2 2.6 0.02 Postmenopausal (%) 86.3 89.9 77.7 <0.0001 Ever HRT use (%) 67.2 73.9 69.3 0.01 c c c Number of mammograms 3.2 3.4 2.4 <0.0001 Breast percent density 36.5 38.2 29.4 <0.0001 d 2 Breast dense area (cm ) 36.7 34.9 28.7 <0.0001 d 2 Total breast area (cm ) 114.5 106.9 118.9 0.08 a 2 b Unless otherwise indicated, mean values are given. P ascertained from t test or χ test, as appropriate. Compares only invasive breast cancer and breast DCIS cases. All other comparisons are of invasive breast cancer cases, breast DCIS cases, and controls. The range for number of mammograms is 11, 10, and 7 for invasive breast cancer, breast DCIS, and controls, respectively. Adjusted for age at recruitment. DCIS, ductal carcinoma in situ; HRT, hormone replacement therapy. diagnosed with primary breast cancer between cohort entry All participants provided informed consent to participate in and December 2000 were identified as potential cases (n = both studies. 1,587). A similar number of randomly selected control sub- jects (n = 1,584) who were not known to have breast cancer Data collection were frequency matched to the distribution of ethnicity and 5- Details of the study procedures were reported previously [4]. year age groups of the cases. Cases and controls with a pre- In brief, information on demographics, medical history, repro- vious diagnosis of breast cancer, a history of breast augmen- ductive behavior, hormone replacement therapy (HRT) use, tation or reduction, and no mammogram were excluded. and body mass index (BMI) were collected with an extensive Approximately 13% of breast cancer cases and 4% of con- questionnaire at entry into the cohort during the period from trols were ineligible primarily because of death or pre-existent 1993 to 1996 [13]. As part of the nested case-control study, breast cancer. Of the 1,396 cases eligible to participate, a one-page Breast Health Questionnaire (BHQ) was com- 52.6% responded to the mailings and gave full consent. Of the pleted to elicit information on menopausal status, previous 1,500 eligible controls, 48.7% responded to the mailings and breast surgery, mammography, and HRT medications [4]. gave full consent. After removing women who did not have Mammograms suitable mammograms, the final sample consisted of 607 breast cancer cases and 667 control subjects. The original The mammographic films were retrieved from clinics located cohort and the nested case-control study were approved by throughout the State of Hawaii using the authorization forms the Committee on Human Studies at the University of Hawaii. signed by the study participants. The original cohort study had Page 2 of 6 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/8/3/R30 no records on mammography use except for one item in the mixed ancestry – regardless of admixture – were assigned to baseline questionnaire. At that time, 90% of Caucasian and one ethnic category according to the following priority ranking: Japanese women and 75% of Native Hawaiian women native Hawaiian, Japanese, Caucasian, and, finally, other [13]. reported previous mammography [14]. Only craniocaudal We created an HRT variable using the responses from the views were digitized using a Kodak LS 85 Film Digitizer with a questionnaire at cohort entry and from the BHQ at enrollment pixel size of 260 µm. If available, mammograms for every sec- into the breast density study. A comparison of the HRT infor- ond or third year were scanned with the goal being to cover as mation from the two questionnaires exhibited good agreement wide a time period as possible for each woman. For cases, for overlapping years when both questionnaires reported HRT only mammograms taken before treatment for breast cancer use. If a woman indicated that she had used HRT at any time were selected. However, the image of the contralateral breast but the write-in field in the BHQ was empty, we assigned the taken at the time of diagnosis was used for five cases. The type of HRT from the cohort questionnaire completed at base- scanned images for both breasts were assessed for densities line. For the women with missing HRT type information (5.4%), using Cumulus108 software [15] by one reader (GM), who we imputed the type based on hysterectomy status: estrogen was blinded to case status and time sequence of the mammo- only for women with a hysterectomy and combined therapy grams. After the reader determined a threshold for the edge of otherwise. the breast and for the edge of the dense tissue [15], the com- puter calculated the total number of pixels in the digitized All models were adjusted for the following covariates that are image that constituted the total area and the dense area and known to be associated with breast cancer and mammo- computed the ratio between the two values as percentage graphic density: mean age of all mammograms (continuous), density. Because readings for the right and left breast were ethnicity, BMI (<22.5, 22.5 to <25, 25 to <30, or ≥30 kg/m ), very similar (correlation coefficient >0.90), we averaged the parity (0–1, 2–3, or ≥4), age at menarche (<13, 13–14, or ≥15 values for both to obtain one measure of total breast area, years), age at first live birth (<21, 21–30, >30 years, or no chil- dense area and percentage density. dren), menopausal status (pre- or postmenopausal), family his- tory of breast cancer (breast cancer in a first-degree relative or A random sample of 410 mammograms was read in duplicate no history), and HRT use (never, estrogen only, or estrogen + to assess the reliability of the mammographic readings. The progestin). Tests for trend were performed by fitting a variable intraclass correlation coefficients [16] were 0.96 (95% confi- representing ordinal categories (described above) of percent- dence interval (CI) 0.95–0.97) for the size of dense area and age density or dense area. 0.996 (95% CI 0.995–0.997) for the total breast area, result- ing in an intraclass correlation coefficient for percentage den- We were also interested in comparing how well percentage sity of 0.974 (95% CI 0.968–0.978). breast density and size of dense area predicted invasive breast cancer and breast DCIS. We performed unconditional Statistical analysis binary logistic regression and examined the area under the To test for differences in covariate values across breast cancer receiver operating characteristic (ROC) curve, which is a cases and control subjects, we performed t tests for continu- method used in sensitivity-specificity analyses. The ROC ous variables and χ tests for categorical variables. We used curve assesses the ability of the model to distinguish between unconditional polytomous logistic regression modeling with two groups (for example, diseased and disease free). If the the SAS software [17] to compute odds ratios (ORs) and model is able to separate the two groups perfectly, then the 95% CIs for the risks for DCIS and invasive cancer associated area under the ROC curve is equal to 1; if the model performs with breast density [18]. All P values reported are two sided. no better than chance, then the area will be 0.5. Breast cancer cases were divided into CIS and invasive based on information provided by the state-wide Hawaii Tumor Reg- Results istry, a member of the National Cancer Institute's Surveillance, Of all breast cancer cases, 119 (19.8%) were breast DCIS Epidemiology and End Results program. Of the 125 breast (Table 1). Japanese women had a greater proportion of breast CIS cases, 119 were classified as having DCIS. DCIS than invasive breast cancer, and native Hawaiian women had almost twice the proportion of invasive cases compared We chose two measures of mammographic breast density as with DCIS cases (P = 0.19). The mean age at diagnosis was our exposure variables: size of the dense area and percentage similar for breast DCIS and invasive cases (P = 0.50). How- density. Percentage density was grouped into four commonly ever, the breast DCIS cases had a greater proportion of post- used categories: <10%, 10–24.9%, 25–49.9%, and ≥50%. menopausal women (89.9% versus 86.3% and 77.7% (for The size of the dense areas was classified as follows: <15 DCIS, invasive cases, and controls, respectively)) and had 2 2 2 2 cm , 15–29.9 cm , 30.0–44.9 cm , and ≥45 cm . Study par- fewer children (2.2 versus 2.4 and 2.6 (for DCIS, invasive ticipants were grouped into four categories: Japanese, Cauca- cases and controls, respectively)). Women with invasive sians, native Hawaiians and others (mostly Filipinos). To breast cancer or breast DCIS had more mammograms than maximize the number of participants per group, women of did controls (P < 0.0001). However, mean unadjusted age at Page 3 of 6 (page number not for citation purposes) Breast Cancer Research Vol 8 No 3 Gill et al. Table 2 Mammographic breast density and risk estimates for breast DCIS and invasive breast cancer Exposure variable Invasive breast cancer (n)Breast DCIS (n) Controls (n) Invasive versus controls DCIS versus controls DCIS versus invasive Mean percentage density (%) <10 63 18 158 1 (reference) 1 (reference) 1 (reference) 10–24.9 110 23 170 1.81 (1.21–2.70) 1.15 (0.57–2.30) 0.65 (0.31–1.35) 25–49.9 174 34 212 2.53 (1.69–3.78) 1.29 (0.64–2.59) 0.56 (0.27–1.17) ≥50 135 44 127 3.58 (2.26–5.66) 2.86 (1.38–5.94) 0.89 (0.41–1.91) P value for trend <0.0001 0.0010 Mean breast dense area (cm ) <15 95 26 209 1 (reference) 1 (reference) 1 (reference) 15–29.9 136 29 200 1.58 (1.12–2.23) 1.05 (0.58–1.92) 0.75 (0.40–1.40) 30–44.9 109 28 133 1.93 (1.32–2.81) 1.70 (0.90–3.22) 0.92 (0.48–1.76) ≥45 142 36 125 2.92 (2.01–4.25) 2.59 (1.39–4.82) 0.99 (0.53–1.86) P value for trend <0.0001 0.0026 Values are expressed as odds ratios (95% confidence intervals), which were estimated using polytomous logistic regression and adjusted for ethnicity, mean age of all mammograms, body mass index, age at first live birth, number of children, age at menarche, menopausal status, use of hormone replacement therapy, and family history of breast cancer. DCIS, ductal carcinoma in situ. first mammogram and last mammogram were not different Because of sample size limitations, we were only able to ana- across breast DCIS cases, invasive breast cancer cases, and lyze mammographic density and risk for breast DCIS by eth- controls (P = 0.32 and P = 0.66, respectively; results not nicity for Japanese and Caucasian women (results not shown). shown). Although invasive breast cancer cases had a greater Both ethnic groups exhibited similar trends; risk for breast age-adjusted mean dense breast area than did breast DCIS DCIS increased with increasing percentage density and 2 2 cases (36.7 cm and 34.9 cm , respectively), the breast DCIS dense area. However, the CIs were wide and, except for the cases had higher percentage breast densities (38.2% and highest category of percentage density and dense area in 36.5%, respectively) than did the invasive breast cancer cases Caucasian women, all intervals included 1. because of their smaller mean total breast area. The area under the ROC curve was similar for percentage Both percentage density and the size of the dense area were density and dense area. For the adjusted invasive breast can- associated with breast DCIS and invasive breast cancer in our cer model, both values were 0.74, whereas for the adjusted study (Table 2). For each cancer, trend tests were highly sig- breast DCIS model the areas under the ROC curve were 0.67. nificant. Women with at least 50% percentage density had a When modeled without adjustment for covariates, the area 3.5-fold greater risk for invasive breast cancer than women under the ROC curve for invasive breast cancer was 0.59 for with less than 10% density (OR 3.58, 95% CI 2.26–5.66) both percentage density and dense area. A similar unadjusted when compared with controls. The risk for DCIS was almost model for DCIS yielded values of 0.59 for dense area and 0.61 threefold greater in women with 50% or more percentage den- for percentage density. sity than in women with less than 10% density (OR 2.86, 95% CI 1.38–5.94). For breast DCIS, the 95% CIs for the second Discussion and third categories of density included the null value. The Our analyses revealed that mammographic breast density, as comparison of breast DCIS with invasive breast cancer indi- measured by percentage density and the size of dense area, is cated a lower, although not statistically significant, risk for associated with breast DCIS as well as invasive breast cancer. breast DCIS than for invasive cancer given the same level of The association was slightly weaker for breast DCIS than for percentage density. The risk estimates for the size of dense invasive breast cancer. Comparisons of areas under the ROC area in the breast were not as strong as they were for percent- curve indicated that breast density and dense area performed age breast density. For women with more than 45 cm of equally well in distinguishing breast DCIS and invasive breast dense breast area, the ORs of invasive breast cancer and cancer cases from controls. The ORs for breast DCIS in the breast DCIS were 2.92 (95% CI 2.01–4.25) and 2.86 (95% present study were not as large as those reported in a study CI 1.57–5.20), respectively, compared with controls. The of a cohort of Canadian women [12], which estimated relative comparison of breast DCIS versus invasive breast cancer risks between 7 and 12 for detecting atypia/CIS in biopsy cases showed little difference. specimens from women with more than 50% density. How- ever, the cell sizes within the strata of density were small in that Page 4 of 6 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/8/3/R30 study and cases were selected from a group of women with A limitation of the present study is the lack of data on the fre- biopsies. A British case-control study [19] found much lower quency of mammography use. Although this population has risks for breast CIS, and the estimates were similar in strength high screening rates [14], we cannot rule out bias toward the to those for invasive breast cancer. Women with Wolfe paren- null in estimates of DCIS risk because of the possibility of chymal patterns P2 or DY had 70% greater risk (95% CI 1.1– undetected breast DCIS among controls. We also had rela- 2.6) of screening detected in situ breast cancer compared tively low participation rates, although a comparison of partic- with controls and a 30% greater risk (95% CI 1.2–1.8) of ipating and eligible women revealed that they had very similar screening detected invasive breast cancer compared with characteristics [4]. Our assessments of HRT and BMI are lim- controls. ited because we had to rely on self-report, and assumed that their values remained constant between the time they were Other studies of breast DCIS support the idea that there is a reported and when the study was completed, but an examina- relation between in situ lesions and mammographic density. tion of BMI from a follow-up questionnaire five years after An investigation of 28 mammograms confirmed that breast cohort entry showed that the mean BMI changes by only 0.50 DCIS occurs to a greater extent in areas of the breast that kg/m during that time. Therefore, differences in BMI are exhibit high mammographic density [11]. The relative risk for a unlikely to change the results materially. The use of BMI in cat- second breast cancer was found to be three times higher egories is unlikely to have confounded our results because among women with primary DCIS who had mammographic analyses with continuous values of BMI gave nearly identical densities of 75% and greater compared with women with less results. We had limited power to estimate the risk for DCIS; than 25% density [20]. there were nearly four times as many invasive breast cancer cases as breast DCIS cases. Nevertheless, the cohort design, Although it is known that some breast DCIS lesions will the multiple mammograms, and the frequent mammography develop into invasive breast cancers [21], and that DCIS is use in this population must be considered strengths of this often found near invasive breast cancer lesions [22], it remains project. The high rate of mammography participation problematic to distinguish DCIS lesions that may progress to decreases the probability that a large number of cancers were invasive breast cancer from those that may not. Given the missed in this population. growing number of breast CIS cases with increasing mam- mography screening [1], this question has important implica- Conclusion tions when decisions must be made regarding the In the present study similar patterns of association for mammo- aggressiveness of treatment [23]. A comparison of risk factors graphic density with DCIS and invasive breast cancers add to for CIS (ductal and/or lobular) and invasive breast cancer the growing body of evidence that certain breast CIS and inva- found mixed results. Established invasive breast cancer risk sive breast cancers may share etiologic factors. At the same factors shown to be consistently associated with breast CIS time, it appears likely that factors that have not yet been iden- include family history of breast cancer [24-30], low BMI tified may influence the effects of breast density and other among premenopausal women [24,29,30], and nulliparity known breast cancer risk factors in the progression of breast [24,28,30,31]. However, other invasive breast cancer risk fac- CIS to invasive breast cancer. tors such as early age at menarche, late age at menopause, increased endogenous estrogen levels, and alcohol intake Competing interests were associated with breast CIS in some studies The authors declare that they have no competing interests. [26,27,29,32], but not in others [24-27,30,33]. Authors' contributions Our study and other reports mentioned above indicate that JG carried out the main statistical analysis, created variables mammographic density is associated with breast CIS and with used in this analysis, and drafted the manuscript. GM con- invasive breast cancer. However, because some studies ceived of the nested case-control study, performed the mam- describe differential effects for some risk factors, more mographic density assessment, and helped with the draft and research needs to be done. Studies with larger samples of revisions of the manuscript. IP created a data set and variables breast CIS must be performed to assess whether factors such used in analysis, assisted in writing of SAS programs, and as postmenopausal obesity, growth factors (insulin-like growth advised on statistical analysis techniques. LK conceived of the factor-I), exogenous hormones, and cell proliferation biomark- Multiethnic Cohort and helped in revision of the manuscript. All ers can help in elucidating the association between breast authors read and approved the final manuscript. density, breast CIS, and progression to invasive breast cancer. New molecular techniques may have the ability to identify fac- Acknowledgements The case-control study was funded by a grant from the National Cancer tors that are responsible for the progression from DCIS to Institute (R01 CA 85265). The Multiethnic Cohort Study has been sup- invasive cancer [2]. ported by USPHS (National Cancer Insitute) grant R37 CA 54281. Dr Gill was supported by R25 CA 90956. We are very grateful to Jihae Noh for her outstanding work in mammogram retrieval and scanning, to Page 5 of 6 (page number not for citation purposes) Breast Cancer Research Vol 8 No 3 Gill et al. Andrew Williams for the excellent database, and to Maj Earle for provid- 22. Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson IC: Inci- dence of and treatment for ductal carcinoma in situ of the ing information from the Multiethnic Cohort Study. breast. JAMA 1996, 275:913-918. 23. Leonard GD, Swain SM: Ductal carcinoma in situ, complexities References and challenges. J Natl Cancer Inst 2004, 96:906-920. 1. American Cancer Society SR2: Cancer Facts and Figures 2005 24. Kerlikowske K, Barclay J, Grady D, Sickles EA, Ernster V: Compar- Atlanta, GA: American Cancer Society, Inc.; 2005. ison of risk factors for ductal carcinoma in situ and invasive 2. Bernstein L: The epidemiology of breast carcinoma in situ. In breast cancer. J Natl Cancer Inst 1997, 89:76-82. 25. Gapstur SM, Morrow M, Sellers TA: Hormone replacement ther- Ductal Carcinoma In Situ of the Breast 2nd edition. Edited by: Sil- verstein MJ, Recht A, Lagios MD. Philadelphia: Lippincott, Williams apy and risk of breast cancer with a favorable histology: and Wilkins; 2002. results of the Iowa Women's Health Study. JAMA 1999, 3. Boyd NF, Rommens JM, Vogt K, Lee V, Hopper JL, Yaffe MJ, Pater- 281:2091-2097. son AD: Mammographic breast density as an intermediate 26. Trentham-Dietz A, Newcomb PA, Storer BE, Remington PL: Risk phenotype for breast cancer. Lancet Oncol 2005, 6:798-808. factors for carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev 2000, 9:697-703. 4. Maskarinec G, Pagano I, Lurie G, Wilkens LR, Kolonel LN: Mam- mographic density and breast cancer risk: the Multiethnic 27. Claus EB, Stowe M, Carter D: Breast carcinoma in situ: risk fac- Cohort. Am J Epidemiol 2005, 162:743-752. tors and screening patterns. J Natl Cancer Inst 2001, 5. Ursin G, Ma H, Wu AH, Bernstein L, Salane M, Parisky YR, Astra- 93:1811-1817. han M, Siozon CC, Pike MC: Mammographic density and breast 28. Brinton LA, Hoover R, Fraumeni JF Jr: Epidemiology of minimal cancer in three ethnic groups. Cancer Epidemiol Biomarkers breast cancer. JAMA 1983, 249:483-487. 29. Longnecker MP, Bernstein L, Paganini-Hill A, Enger SM, Ross RK: Prev 2003, 12:332-338. 6. Torres-Mejia G, De Stavola B, Allen DS, Perez-Gavilan JJ, Ferreira Risk factors for in situ breast cancer. Cancer Epidemiol Biomar- JM, Fentiman IS, Dos Santos Silva I: Mammographic features kers Prev 1996, 5:961-965. and subsequent risk of breast cancer: a comparison of quali- 30. Weiss HA, Brinton LA, Brogan D, Coates RJ, Gammon MD, tative and quantitative evaluations in the Guernsey prospec- Malone KE, Schoenberg JB, Swanson CA: Epidemiology of in tive studies. Cancer Epidemiol Biomarkers Prev 2005, situ and invasive breast cancer in women aged under 45. Br J Cancer 1996, 73:1298-1305. 14:1052-1059. 7. Maskarinec G, Meng L: A case-control study of mammographic 31. Lambe M, Hsieh CC, Tsaih SW, Ekbom A, Trichopoulos D, Adami densities in Hawaii. Breast Cancer Res Treat 2000, 63:153-161. HO: Parity, age at first birth and the risk of carcinoma in situ of 8. Byrne C, Schairer C, Wolfe J, Parekh N, Salane M, Brinton LA, the breast. Int J Cancer 1998, 77:330-332. 32. Missmer SA, Eliassen AH, Barbieri RL, Hankinson SE: Endog- Hoover R, Haile R: Mammographic features and breast cancer risk: effects with time, age, and menopause status. J Natl Can- enous estrogen, androgen, and progesterone concentrations and breast cancer risk among postmenopausal women. J Natl cer Inst 1995, 87:1622-1629. 9. Boyd NF, Byng JW, Jong RA, Fishell EK, Little LE, Miller AB, Lock- Cancer Inst 2004, 96:1856-1865. wood GA, Tritchler DL, Yaffe MJ: Quantitative classification of 33. Zeleniuch-Jacquotte A, Gu Y, Shore RE, Koenig KL, Arslan AA, mammographic densities and breast cancer risk: results from Kato I, Rinaldi S, Kaaks R, Toniolo P: Postmenopausal levels of sex hormones and risk of breast carcinoma in situ: results of the Canadian National Breast Screening Study. J Natl Cancer Inst 1995, 87:670-675. a prospective study. Int J Cancer 2005, 114:323-327. 10. Saftlas AF, Hoover RN, Brinton LA, Szklo M, Olson DR, Salane M, Wolfe JN: Mammographic densities and risk of breast cancer. Cancer 1991, 67:2833-2838. 11. Ursin G, Hovanessian-Larsen L, Parisky YR, Pike MC, Wu AH: Greatly increased occurrence of breast cancers in areas of mammographically dense tissue. Breast Cancer Res 2005, 7:R605-R608. 12. Boyd NF, Jensen HM, Cooke G, Lee Han H: Relationship between mammographic and histological risk factors for breast cancer. J Natl Cancer Inst 1992, 15:1170-1179. 13. Kolonel LN, Henderson BE, Hankin JH, Nomura AMY, Wilkens LR, Pike MC, Stram DO, Monroe KR, Earle ME, Nagamine FS: A Mul- tiethnic Cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol 2000, 151:346-357. 14. Hawaii State Department of Health: Hawaii Behavioral Risk Factor Surveillance Report 1994–2000 Honolulu: State of Hawaii; 2002. 15. Byng JW, Boyd NF, Fishell E, Jong RA, Yaffe MJ: The quantitative analysis of mammographic densities. Phys Med Biol 1994, 39:1629-1638. 16. Shrout P, Fleiss J: Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979, 86:420-428. 17. SAS Institute Inc: SAS OnlineDoc 9.1.2 Cary, NC: SAS Institute Inc; 2004. 18. Hosmer DW, Lemeshow S: Applied Logistic Regression 2nd edi- tion. New York: John Wiley & SOns Inc; 2000. 19. Sala E, Solomon L, Warren R, McCann J, Duffy S, Luben R, Day N: Size, node status and grade of breast tumours: association with mammographic parenchymal patterns. Eur Radiol 2000, 10:157-161. 20. Habel LA, Dignam JJ, Land SR, Salane M, Capra AM, Julian TB: Mammographic density and breast cancer after ductal carci- noma in situ. J Natl Cancer Inst 2004, 96:1467-1472. 21. Page DL, Dupont WD, Rogers LW, Jensen RA, Schuyler PA: Con- tinued local recurrence of carcinoma 15–25 years after a diag- nosis of low grade ductal carcinoma in situ of the breast treated only by biopsy. Cancer 1995, 76:1197-1200. Page 6 of 6 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Breast Cancer Research Springer Journals

The association of mammographic density with ductal carcinoma in situ of the breast: the Multiethnic Cohort

Loading next page...
 
/lp/springer-journals/the-association-of-mammographic-density-with-ductal-carcinoma-in-situ-6jWzwn6V06

References (37)

Publisher
Springer Journals
Copyright
2006 Gill et al.; licensee BioMed Central Ltd.
eISSN
1465-542X
DOI
10.1186/bcr1507
Publisher site
See Article on Publisher Site

Abstract

Introduction It is well established that women with high Results Mammographic density was associated with invasive mammographic density are at greater risk for breast cancer than breast cancer and breast DCIS. For the highest category of are women with low breast density. However, little research has percentage breast density (≥50%) as compared with the lowest been done on mammographic density and ductal carcinoma in (<10%), the OR was 3.58 (95% CI 2.26–5.66) for invasive situ (DCIS) of the breast, which is thought to be a precursor breast cancer and 2.86 (1.38–5.94) for breast DCIS. Similarly, lesion to some invasive breast cancers. for the highest category of dense area (≥45 cm ) as compared with the lowest (<15 cm ), the OR was 2.92 (95% CI 2.01– Method We conducted a nested case-control study within the 4.25) for invasive breast cancer and 2.59 (1.39–4.82) for breast Multiethnic Cohort, and compared the mammographic densities DCIS. Trend tests were significant for invasive breast cancer (P of 482 patients with invasive breast cancer and 119 with breast for trend < 0.0001) and breast DCIS (P for trend < 0.001) for DCIS cases versus those of 667 cancer-free control subjects. A both percentage density and dense area. reader blinded to disease status performed computer-assisted density assessment. For women with more than one mammogram, mean density values were computed. Polytomous logistic regression models were used to compute adjusted odds Conclusion The similar strength of association for ratios (ORs) and 95% confidence intervals (CIs) for two mammographic density with breast DCIS and invasive breast measurements of mammographic density: percentage density cancer supports the hypothesis that both diseases may have a and dense area. common etiology. cases [5] or combined invasive and CIS cases [6-10]. One study investigating breast CIS observed that DCIS was more Introduction likely to occur in mammographically dense areas [11], and As a result of increasing early detection efforts, breast carci- another study reported an increase in breast hyperplasia or noma in situ (CIS) constitutes more than 20% of newly diag- atypia/CIS in women with greater than 50% breast density nosed breast cancer cases in the USA [1]. Although breast [12]. This analysis examines the association between mammo- CIS shares a number of risk factors with invasive breast can- graphic density and risk for breast DCIS in comparison with cer, and ductal carcinoma in situ (DCIS) of the breast is con- invasive breast cancer and breast cancer-free control sidered to be a precursor to some invasive breast cancers [2], subjects. it is not clear to what extent breast CIS and invasive breast cancer have the same etiology. Several case-control studies Materials and methods have confirmed that mammographic density is associated with Study population risk for breast cancer [3]; women with high breast density have The data for this analysis were collected using a nested case- a three-fold to six-fold greater risk for developing breast cancer control study within the Hawaii component of the Multiethnic than women with low breast density [3,4]. Most studies of Cohort, which was established between 1993 and 1996 [13]. mammographic density included only invasive breast cancer As described in detail elsewhere [4], all female members BHQ = Breast Health Questionnaire; BMI = body mass index; CI = confidence interval; CIS = carcinoma in situ; DCIS = ductal carcinoma in situ; HRT = hormone replacement therapy; OR = odds ratio; ROC = receiver operating characteristic. Page 1 of 6 (page number not for citation purposes) Breast Cancer Research Vol 8 No 3 Gill et al. Table 1 Characteristics of the study population by disease status Covariate Invasive breast cancer Breast DCIS Controls P value Sample size 482 119 667 - Ethnicity (%) 0.19 Hawaiian 14.3 8.4 24.3 Japanese 46.5 55.5 43.8 Caucasian 32.6 28.6 28.0 Other 6.6 7.6 3.9 Age at diagnosis (years) 63.5 62.9 N/A 0.50 Age at recruitment (years) 60.0 59.6 56.7 0.58 Mean age at all mammograms (years) 59.8 59.1 59.7 0.70 Body mass index (kg/m ) 25.2 24.5 25.7 0.07 Time from first mammogram to diagnosis (years) 6.2 6.7 N/A 0.17 Family history of breast cancer (%) 16.8 16.8 12.0 0.05 Age at first birth (years) 25.0 25.1 24.7 0.65 Age at menarche (years) 13.1 12.9 13.1 0.47 Parous (%) 84.5 83.9 88.5 0.10 Number of children 2.4 2.2 2.6 0.02 Postmenopausal (%) 86.3 89.9 77.7 <0.0001 Ever HRT use (%) 67.2 73.9 69.3 0.01 c c c Number of mammograms 3.2 3.4 2.4 <0.0001 Breast percent density 36.5 38.2 29.4 <0.0001 d 2 Breast dense area (cm ) 36.7 34.9 28.7 <0.0001 d 2 Total breast area (cm ) 114.5 106.9 118.9 0.08 a 2 b Unless otherwise indicated, mean values are given. P ascertained from t test or χ test, as appropriate. Compares only invasive breast cancer and breast DCIS cases. All other comparisons are of invasive breast cancer cases, breast DCIS cases, and controls. The range for number of mammograms is 11, 10, and 7 for invasive breast cancer, breast DCIS, and controls, respectively. Adjusted for age at recruitment. DCIS, ductal carcinoma in situ; HRT, hormone replacement therapy. diagnosed with primary breast cancer between cohort entry All participants provided informed consent to participate in and December 2000 were identified as potential cases (n = both studies. 1,587). A similar number of randomly selected control sub- jects (n = 1,584) who were not known to have breast cancer Data collection were frequency matched to the distribution of ethnicity and 5- Details of the study procedures were reported previously [4]. year age groups of the cases. Cases and controls with a pre- In brief, information on demographics, medical history, repro- vious diagnosis of breast cancer, a history of breast augmen- ductive behavior, hormone replacement therapy (HRT) use, tation or reduction, and no mammogram were excluded. and body mass index (BMI) were collected with an extensive Approximately 13% of breast cancer cases and 4% of con- questionnaire at entry into the cohort during the period from trols were ineligible primarily because of death or pre-existent 1993 to 1996 [13]. As part of the nested case-control study, breast cancer. Of the 1,396 cases eligible to participate, a one-page Breast Health Questionnaire (BHQ) was com- 52.6% responded to the mailings and gave full consent. Of the pleted to elicit information on menopausal status, previous 1,500 eligible controls, 48.7% responded to the mailings and breast surgery, mammography, and HRT medications [4]. gave full consent. After removing women who did not have Mammograms suitable mammograms, the final sample consisted of 607 breast cancer cases and 667 control subjects. The original The mammographic films were retrieved from clinics located cohort and the nested case-control study were approved by throughout the State of Hawaii using the authorization forms the Committee on Human Studies at the University of Hawaii. signed by the study participants. The original cohort study had Page 2 of 6 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/8/3/R30 no records on mammography use except for one item in the mixed ancestry – regardless of admixture – were assigned to baseline questionnaire. At that time, 90% of Caucasian and one ethnic category according to the following priority ranking: Japanese women and 75% of Native Hawaiian women native Hawaiian, Japanese, Caucasian, and, finally, other [13]. reported previous mammography [14]. Only craniocaudal We created an HRT variable using the responses from the views were digitized using a Kodak LS 85 Film Digitizer with a questionnaire at cohort entry and from the BHQ at enrollment pixel size of 260 µm. If available, mammograms for every sec- into the breast density study. A comparison of the HRT infor- ond or third year were scanned with the goal being to cover as mation from the two questionnaires exhibited good agreement wide a time period as possible for each woman. For cases, for overlapping years when both questionnaires reported HRT only mammograms taken before treatment for breast cancer use. If a woman indicated that she had used HRT at any time were selected. However, the image of the contralateral breast but the write-in field in the BHQ was empty, we assigned the taken at the time of diagnosis was used for five cases. The type of HRT from the cohort questionnaire completed at base- scanned images for both breasts were assessed for densities line. For the women with missing HRT type information (5.4%), using Cumulus108 software [15] by one reader (GM), who we imputed the type based on hysterectomy status: estrogen was blinded to case status and time sequence of the mammo- only for women with a hysterectomy and combined therapy grams. After the reader determined a threshold for the edge of otherwise. the breast and for the edge of the dense tissue [15], the com- puter calculated the total number of pixels in the digitized All models were adjusted for the following covariates that are image that constituted the total area and the dense area and known to be associated with breast cancer and mammo- computed the ratio between the two values as percentage graphic density: mean age of all mammograms (continuous), density. Because readings for the right and left breast were ethnicity, BMI (<22.5, 22.5 to <25, 25 to <30, or ≥30 kg/m ), very similar (correlation coefficient >0.90), we averaged the parity (0–1, 2–3, or ≥4), age at menarche (<13, 13–14, or ≥15 values for both to obtain one measure of total breast area, years), age at first live birth (<21, 21–30, >30 years, or no chil- dense area and percentage density. dren), menopausal status (pre- or postmenopausal), family his- tory of breast cancer (breast cancer in a first-degree relative or A random sample of 410 mammograms was read in duplicate no history), and HRT use (never, estrogen only, or estrogen + to assess the reliability of the mammographic readings. The progestin). Tests for trend were performed by fitting a variable intraclass correlation coefficients [16] were 0.96 (95% confi- representing ordinal categories (described above) of percent- dence interval (CI) 0.95–0.97) for the size of dense area and age density or dense area. 0.996 (95% CI 0.995–0.997) for the total breast area, result- ing in an intraclass correlation coefficient for percentage den- We were also interested in comparing how well percentage sity of 0.974 (95% CI 0.968–0.978). breast density and size of dense area predicted invasive breast cancer and breast DCIS. We performed unconditional Statistical analysis binary logistic regression and examined the area under the To test for differences in covariate values across breast cancer receiver operating characteristic (ROC) curve, which is a cases and control subjects, we performed t tests for continu- method used in sensitivity-specificity analyses. The ROC ous variables and χ tests for categorical variables. We used curve assesses the ability of the model to distinguish between unconditional polytomous logistic regression modeling with two groups (for example, diseased and disease free). If the the SAS software [17] to compute odds ratios (ORs) and model is able to separate the two groups perfectly, then the 95% CIs for the risks for DCIS and invasive cancer associated area under the ROC curve is equal to 1; if the model performs with breast density [18]. All P values reported are two sided. no better than chance, then the area will be 0.5. Breast cancer cases were divided into CIS and invasive based on information provided by the state-wide Hawaii Tumor Reg- Results istry, a member of the National Cancer Institute's Surveillance, Of all breast cancer cases, 119 (19.8%) were breast DCIS Epidemiology and End Results program. Of the 125 breast (Table 1). Japanese women had a greater proportion of breast CIS cases, 119 were classified as having DCIS. DCIS than invasive breast cancer, and native Hawaiian women had almost twice the proportion of invasive cases compared We chose two measures of mammographic breast density as with DCIS cases (P = 0.19). The mean age at diagnosis was our exposure variables: size of the dense area and percentage similar for breast DCIS and invasive cases (P = 0.50). How- density. Percentage density was grouped into four commonly ever, the breast DCIS cases had a greater proportion of post- used categories: <10%, 10–24.9%, 25–49.9%, and ≥50%. menopausal women (89.9% versus 86.3% and 77.7% (for The size of the dense areas was classified as follows: <15 DCIS, invasive cases, and controls, respectively)) and had 2 2 2 2 cm , 15–29.9 cm , 30.0–44.9 cm , and ≥45 cm . Study par- fewer children (2.2 versus 2.4 and 2.6 (for DCIS, invasive ticipants were grouped into four categories: Japanese, Cauca- cases and controls, respectively)). Women with invasive sians, native Hawaiians and others (mostly Filipinos). To breast cancer or breast DCIS had more mammograms than maximize the number of participants per group, women of did controls (P < 0.0001). However, mean unadjusted age at Page 3 of 6 (page number not for citation purposes) Breast Cancer Research Vol 8 No 3 Gill et al. Table 2 Mammographic breast density and risk estimates for breast DCIS and invasive breast cancer Exposure variable Invasive breast cancer (n)Breast DCIS (n) Controls (n) Invasive versus controls DCIS versus controls DCIS versus invasive Mean percentage density (%) <10 63 18 158 1 (reference) 1 (reference) 1 (reference) 10–24.9 110 23 170 1.81 (1.21–2.70) 1.15 (0.57–2.30) 0.65 (0.31–1.35) 25–49.9 174 34 212 2.53 (1.69–3.78) 1.29 (0.64–2.59) 0.56 (0.27–1.17) ≥50 135 44 127 3.58 (2.26–5.66) 2.86 (1.38–5.94) 0.89 (0.41–1.91) P value for trend <0.0001 0.0010 Mean breast dense area (cm ) <15 95 26 209 1 (reference) 1 (reference) 1 (reference) 15–29.9 136 29 200 1.58 (1.12–2.23) 1.05 (0.58–1.92) 0.75 (0.40–1.40) 30–44.9 109 28 133 1.93 (1.32–2.81) 1.70 (0.90–3.22) 0.92 (0.48–1.76) ≥45 142 36 125 2.92 (2.01–4.25) 2.59 (1.39–4.82) 0.99 (0.53–1.86) P value for trend <0.0001 0.0026 Values are expressed as odds ratios (95% confidence intervals), which were estimated using polytomous logistic regression and adjusted for ethnicity, mean age of all mammograms, body mass index, age at first live birth, number of children, age at menarche, menopausal status, use of hormone replacement therapy, and family history of breast cancer. DCIS, ductal carcinoma in situ. first mammogram and last mammogram were not different Because of sample size limitations, we were only able to ana- across breast DCIS cases, invasive breast cancer cases, and lyze mammographic density and risk for breast DCIS by eth- controls (P = 0.32 and P = 0.66, respectively; results not nicity for Japanese and Caucasian women (results not shown). shown). Although invasive breast cancer cases had a greater Both ethnic groups exhibited similar trends; risk for breast age-adjusted mean dense breast area than did breast DCIS DCIS increased with increasing percentage density and 2 2 cases (36.7 cm and 34.9 cm , respectively), the breast DCIS dense area. However, the CIs were wide and, except for the cases had higher percentage breast densities (38.2% and highest category of percentage density and dense area in 36.5%, respectively) than did the invasive breast cancer cases Caucasian women, all intervals included 1. because of their smaller mean total breast area. The area under the ROC curve was similar for percentage Both percentage density and the size of the dense area were density and dense area. For the adjusted invasive breast can- associated with breast DCIS and invasive breast cancer in our cer model, both values were 0.74, whereas for the adjusted study (Table 2). For each cancer, trend tests were highly sig- breast DCIS model the areas under the ROC curve were 0.67. nificant. Women with at least 50% percentage density had a When modeled without adjustment for covariates, the area 3.5-fold greater risk for invasive breast cancer than women under the ROC curve for invasive breast cancer was 0.59 for with less than 10% density (OR 3.58, 95% CI 2.26–5.66) both percentage density and dense area. A similar unadjusted when compared with controls. The risk for DCIS was almost model for DCIS yielded values of 0.59 for dense area and 0.61 threefold greater in women with 50% or more percentage den- for percentage density. sity than in women with less than 10% density (OR 2.86, 95% CI 1.38–5.94). For breast DCIS, the 95% CIs for the second Discussion and third categories of density included the null value. The Our analyses revealed that mammographic breast density, as comparison of breast DCIS with invasive breast cancer indi- measured by percentage density and the size of dense area, is cated a lower, although not statistically significant, risk for associated with breast DCIS as well as invasive breast cancer. breast DCIS than for invasive cancer given the same level of The association was slightly weaker for breast DCIS than for percentage density. The risk estimates for the size of dense invasive breast cancer. Comparisons of areas under the ROC area in the breast were not as strong as they were for percent- curve indicated that breast density and dense area performed age breast density. For women with more than 45 cm of equally well in distinguishing breast DCIS and invasive breast dense breast area, the ORs of invasive breast cancer and cancer cases from controls. The ORs for breast DCIS in the breast DCIS were 2.92 (95% CI 2.01–4.25) and 2.86 (95% present study were not as large as those reported in a study CI 1.57–5.20), respectively, compared with controls. The of a cohort of Canadian women [12], which estimated relative comparison of breast DCIS versus invasive breast cancer risks between 7 and 12 for detecting atypia/CIS in biopsy cases showed little difference. specimens from women with more than 50% density. How- ever, the cell sizes within the strata of density were small in that Page 4 of 6 (page number not for citation purposes) Available online http://breast-cancer-research.com/content/8/3/R30 study and cases were selected from a group of women with A limitation of the present study is the lack of data on the fre- biopsies. A British case-control study [19] found much lower quency of mammography use. Although this population has risks for breast CIS, and the estimates were similar in strength high screening rates [14], we cannot rule out bias toward the to those for invasive breast cancer. Women with Wolfe paren- null in estimates of DCIS risk because of the possibility of chymal patterns P2 or DY had 70% greater risk (95% CI 1.1– undetected breast DCIS among controls. We also had rela- 2.6) of screening detected in situ breast cancer compared tively low participation rates, although a comparison of partic- with controls and a 30% greater risk (95% CI 1.2–1.8) of ipating and eligible women revealed that they had very similar screening detected invasive breast cancer compared with characteristics [4]. Our assessments of HRT and BMI are lim- controls. ited because we had to rely on self-report, and assumed that their values remained constant between the time they were Other studies of breast DCIS support the idea that there is a reported and when the study was completed, but an examina- relation between in situ lesions and mammographic density. tion of BMI from a follow-up questionnaire five years after An investigation of 28 mammograms confirmed that breast cohort entry showed that the mean BMI changes by only 0.50 DCIS occurs to a greater extent in areas of the breast that kg/m during that time. Therefore, differences in BMI are exhibit high mammographic density [11]. The relative risk for a unlikely to change the results materially. The use of BMI in cat- second breast cancer was found to be three times higher egories is unlikely to have confounded our results because among women with primary DCIS who had mammographic analyses with continuous values of BMI gave nearly identical densities of 75% and greater compared with women with less results. We had limited power to estimate the risk for DCIS; than 25% density [20]. there were nearly four times as many invasive breast cancer cases as breast DCIS cases. Nevertheless, the cohort design, Although it is known that some breast DCIS lesions will the multiple mammograms, and the frequent mammography develop into invasive breast cancers [21], and that DCIS is use in this population must be considered strengths of this often found near invasive breast cancer lesions [22], it remains project. The high rate of mammography participation problematic to distinguish DCIS lesions that may progress to decreases the probability that a large number of cancers were invasive breast cancer from those that may not. Given the missed in this population. growing number of breast CIS cases with increasing mam- mography screening [1], this question has important implica- Conclusion tions when decisions must be made regarding the In the present study similar patterns of association for mammo- aggressiveness of treatment [23]. A comparison of risk factors graphic density with DCIS and invasive breast cancers add to for CIS (ductal and/or lobular) and invasive breast cancer the growing body of evidence that certain breast CIS and inva- found mixed results. Established invasive breast cancer risk sive breast cancers may share etiologic factors. At the same factors shown to be consistently associated with breast CIS time, it appears likely that factors that have not yet been iden- include family history of breast cancer [24-30], low BMI tified may influence the effects of breast density and other among premenopausal women [24,29,30], and nulliparity known breast cancer risk factors in the progression of breast [24,28,30,31]. However, other invasive breast cancer risk fac- CIS to invasive breast cancer. tors such as early age at menarche, late age at menopause, increased endogenous estrogen levels, and alcohol intake Competing interests were associated with breast CIS in some studies The authors declare that they have no competing interests. [26,27,29,32], but not in others [24-27,30,33]. Authors' contributions Our study and other reports mentioned above indicate that JG carried out the main statistical analysis, created variables mammographic density is associated with breast CIS and with used in this analysis, and drafted the manuscript. GM con- invasive breast cancer. However, because some studies ceived of the nested case-control study, performed the mam- describe differential effects for some risk factors, more mographic density assessment, and helped with the draft and research needs to be done. Studies with larger samples of revisions of the manuscript. IP created a data set and variables breast CIS must be performed to assess whether factors such used in analysis, assisted in writing of SAS programs, and as postmenopausal obesity, growth factors (insulin-like growth advised on statistical analysis techniques. LK conceived of the factor-I), exogenous hormones, and cell proliferation biomark- Multiethnic Cohort and helped in revision of the manuscript. All ers can help in elucidating the association between breast authors read and approved the final manuscript. density, breast CIS, and progression to invasive breast cancer. New molecular techniques may have the ability to identify fac- Acknowledgements The case-control study was funded by a grant from the National Cancer tors that are responsible for the progression from DCIS to Institute (R01 CA 85265). The Multiethnic Cohort Study has been sup- invasive cancer [2]. ported by USPHS (National Cancer Insitute) grant R37 CA 54281. Dr Gill was supported by R25 CA 90956. We are very grateful to Jihae Noh for her outstanding work in mammogram retrieval and scanning, to Page 5 of 6 (page number not for citation purposes) Breast Cancer Research Vol 8 No 3 Gill et al. Andrew Williams for the excellent database, and to Maj Earle for provid- 22. Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson IC: Inci- dence of and treatment for ductal carcinoma in situ of the ing information from the Multiethnic Cohort Study. breast. JAMA 1996, 275:913-918. 23. Leonard GD, Swain SM: Ductal carcinoma in situ, complexities References and challenges. J Natl Cancer Inst 2004, 96:906-920. 1. American Cancer Society SR2: Cancer Facts and Figures 2005 24. Kerlikowske K, Barclay J, Grady D, Sickles EA, Ernster V: Compar- Atlanta, GA: American Cancer Society, Inc.; 2005. ison of risk factors for ductal carcinoma in situ and invasive 2. Bernstein L: The epidemiology of breast carcinoma in situ. In breast cancer. J Natl Cancer Inst 1997, 89:76-82. 25. Gapstur SM, Morrow M, Sellers TA: Hormone replacement ther- Ductal Carcinoma In Situ of the Breast 2nd edition. Edited by: Sil- verstein MJ, Recht A, Lagios MD. Philadelphia: Lippincott, Williams apy and risk of breast cancer with a favorable histology: and Wilkins; 2002. results of the Iowa Women's Health Study. JAMA 1999, 3. Boyd NF, Rommens JM, Vogt K, Lee V, Hopper JL, Yaffe MJ, Pater- 281:2091-2097. son AD: Mammographic breast density as an intermediate 26. Trentham-Dietz A, Newcomb PA, Storer BE, Remington PL: Risk phenotype for breast cancer. Lancet Oncol 2005, 6:798-808. factors for carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev 2000, 9:697-703. 4. Maskarinec G, Pagano I, Lurie G, Wilkens LR, Kolonel LN: Mam- mographic density and breast cancer risk: the Multiethnic 27. Claus EB, Stowe M, Carter D: Breast carcinoma in situ: risk fac- Cohort. Am J Epidemiol 2005, 162:743-752. tors and screening patterns. J Natl Cancer Inst 2001, 5. Ursin G, Ma H, Wu AH, Bernstein L, Salane M, Parisky YR, Astra- 93:1811-1817. han M, Siozon CC, Pike MC: Mammographic density and breast 28. Brinton LA, Hoover R, Fraumeni JF Jr: Epidemiology of minimal cancer in three ethnic groups. Cancer Epidemiol Biomarkers breast cancer. JAMA 1983, 249:483-487. 29. Longnecker MP, Bernstein L, Paganini-Hill A, Enger SM, Ross RK: Prev 2003, 12:332-338. 6. Torres-Mejia G, De Stavola B, Allen DS, Perez-Gavilan JJ, Ferreira Risk factors for in situ breast cancer. Cancer Epidemiol Biomar- JM, Fentiman IS, Dos Santos Silva I: Mammographic features kers Prev 1996, 5:961-965. and subsequent risk of breast cancer: a comparison of quali- 30. Weiss HA, Brinton LA, Brogan D, Coates RJ, Gammon MD, tative and quantitative evaluations in the Guernsey prospec- Malone KE, Schoenberg JB, Swanson CA: Epidemiology of in tive studies. Cancer Epidemiol Biomarkers Prev 2005, situ and invasive breast cancer in women aged under 45. Br J Cancer 1996, 73:1298-1305. 14:1052-1059. 7. Maskarinec G, Meng L: A case-control study of mammographic 31. Lambe M, Hsieh CC, Tsaih SW, Ekbom A, Trichopoulos D, Adami densities in Hawaii. Breast Cancer Res Treat 2000, 63:153-161. HO: Parity, age at first birth and the risk of carcinoma in situ of 8. Byrne C, Schairer C, Wolfe J, Parekh N, Salane M, Brinton LA, the breast. Int J Cancer 1998, 77:330-332. 32. Missmer SA, Eliassen AH, Barbieri RL, Hankinson SE: Endog- Hoover R, Haile R: Mammographic features and breast cancer risk: effects with time, age, and menopause status. J Natl Can- enous estrogen, androgen, and progesterone concentrations and breast cancer risk among postmenopausal women. J Natl cer Inst 1995, 87:1622-1629. 9. Boyd NF, Byng JW, Jong RA, Fishell EK, Little LE, Miller AB, Lock- Cancer Inst 2004, 96:1856-1865. wood GA, Tritchler DL, Yaffe MJ: Quantitative classification of 33. Zeleniuch-Jacquotte A, Gu Y, Shore RE, Koenig KL, Arslan AA, mammographic densities and breast cancer risk: results from Kato I, Rinaldi S, Kaaks R, Toniolo P: Postmenopausal levels of sex hormones and risk of breast carcinoma in situ: results of the Canadian National Breast Screening Study. J Natl Cancer Inst 1995, 87:670-675. a prospective study. Int J Cancer 2005, 114:323-327. 10. Saftlas AF, Hoover RN, Brinton LA, Szklo M, Olson DR, Salane M, Wolfe JN: Mammographic densities and risk of breast cancer. Cancer 1991, 67:2833-2838. 11. Ursin G, Hovanessian-Larsen L, Parisky YR, Pike MC, Wu AH: Greatly increased occurrence of breast cancers in areas of mammographically dense tissue. Breast Cancer Res 2005, 7:R605-R608. 12. Boyd NF, Jensen HM, Cooke G, Lee Han H: Relationship between mammographic and histological risk factors for breast cancer. J Natl Cancer Inst 1992, 15:1170-1179. 13. Kolonel LN, Henderson BE, Hankin JH, Nomura AMY, Wilkens LR, Pike MC, Stram DO, Monroe KR, Earle ME, Nagamine FS: A Mul- tiethnic Cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol 2000, 151:346-357. 14. Hawaii State Department of Health: Hawaii Behavioral Risk Factor Surveillance Report 1994–2000 Honolulu: State of Hawaii; 2002. 15. Byng JW, Boyd NF, Fishell E, Jong RA, Yaffe MJ: The quantitative analysis of mammographic densities. Phys Med Biol 1994, 39:1629-1638. 16. Shrout P, Fleiss J: Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979, 86:420-428. 17. SAS Institute Inc: SAS OnlineDoc 9.1.2 Cary, NC: SAS Institute Inc; 2004. 18. Hosmer DW, Lemeshow S: Applied Logistic Regression 2nd edi- tion. New York: John Wiley & SOns Inc; 2000. 19. Sala E, Solomon L, Warren R, McCann J, Duffy S, Luben R, Day N: Size, node status and grade of breast tumours: association with mammographic parenchymal patterns. Eur Radiol 2000, 10:157-161. 20. Habel LA, Dignam JJ, Land SR, Salane M, Capra AM, Julian TB: Mammographic density and breast cancer after ductal carci- noma in situ. J Natl Cancer Inst 2004, 96:1467-1472. 21. Page DL, Dupont WD, Rogers LW, Jensen RA, Schuyler PA: Con- tinued local recurrence of carcinoma 15–25 years after a diag- nosis of low grade ductal carcinoma in situ of the breast treated only by biopsy. Cancer 1995, 76:1197-1200. Page 6 of 6 (page number not for citation purposes)

Journal

Breast Cancer ResearchSpringer Journals

Published: Jun 1, 2006

Keywords: Cancer Research; Oncology; Surgical Oncology

There are no references for this article.