Downstaging of Breast Carcinomas in Older Women Associated with Mammographic Screening

© 1999 Blackwell Science Inc., 1075-122X/99/$14.00/0The Breast Journal, Volume 5, Number 2, 1999 94–100

Downstaging of Breast Carcinomas in Older Women

Associated with

Mammographic Screening

Lawrence J. Solin, MD,* Delray J. Schultz, PhD,

†

Howard B. Kessler, MD,

‡

and Nicholas A. Hanchak, MD

§

*

Department of Radiation Oncology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania

†

Department of Mathematics, Millersville University, Millersville, Pennsylvania andthe University of Pennsylvania Cancer Center, Philadelphia, Pennsylvania

‡

Department of Radiology, Fox Chase Cancer Center, Philadelphia, Pennsylvania

§

U.S. Quality Algorithms,

®

Inc., Aetna U.S. Healthcare,™ Blue Bell, Pennsylvania

graphic screening (p

,

0.0001). Clinical Tis and T1 tumorswere found in 7% (8/107) and 74% (79/107), respectively, ofthe breast cancers detected in women who had undergonemammographic screening compared to 0% (0/23) and 22% (5/23), respectively, of the breast cancers detected in womenwho had not undergone mammographic screening (p

,

0.0001). Of the 102 AJCC clinical stage I–II breast cancers withknown pathologic axillary lymph node staging, lymph nodestage was N0 for 76% (66/87) of the women who had under-gone mammographic screening compared to 53% (8/15) ofthe women who had not undergone mammographic screen-ing (p

5

0.019). The results of downstaging reported in thisstudy are important because earlier staged lesions are associ-ated with an improved prognosis and an increased potentialfor breast-conservation treatment. These findings have shownthat mammographic screening is effective in downstagingbreast carcinomas found in older women.

j

Key Words:

breast carcinoma, breast cancer screening,health maintenance organization (HMO), mammography,older women

M

ammographic screening for the early detection ofbreast cancer has been shown in multiple pro-

spective, randomized trials to achieve downstaging ofbreast carcinomas and to improve survival (1–6). How-

j

Abstract:

The efficacy of mammographic screening forolder women has not been well studied. The present studywas designed to evaluate for downstaging of breast carcino-mas associated with mammographic screening in olderwomen. The study population consisted of 130 women age

>

65 years with newly diagnosed breast carcinoma in 1993–1994 and was obtained from women enrolled in a largehealth maintenance organization. Mammographic screening,if done, was performed by a network of predominantly com-munity-based radiologists. Significant downstaging was foundfor the breast cancers detected in women who had under-gone mammographic screening compared to the breast can-cers detected in women who had not undergone mammo-graphic screening. The American Joint Committee on Cancer(AJCC) clinical stage was stage 0 and stage I in 7% (8/107) and73% (78/107), respectively, of the breast cancers detected inwomen who had undergone mammographic screening com-pared to 0% (0/23) and 22% (5/23), respectively, of the breastcancers detected in women who had not undergone mammo-

Address correspondence and reprint requests to: Lawrence J. Solin,MD, Department of Radiation Oncology, Hospital of the University ofPennsylvania, 3400 Spruce St., Philadelphia, PA 19104, U.S.A., or e-mail:[email protected]


•

95

ever, randomized trials have not examined the efficacyof mammographic screening specifically for olderwomen. There is no theoretical reason to preclude usingmammographic screening for older women without sig-nificant comorbid medical conditions and with a normallife expectancy. The relative absence of older women inmammographic screening trials may be related to a per-ceived inability to obtain long-term follow-up informa-tion rather than a perceived lack of effectiveness ofscreening mammography. In the Swedish two-countystudy, 13% (7,307/55,985) of the women were age 70–74 years (5). Current American Cancer Society guide-lines recommend routine screening mammograms be-ginning at age 40 years, without specifying an upper agelimit (7). For women age 65 years or older, Burns et al.(8) reported that only 15% of the women underwent amammogram during a 1-year period and that the use ofmammography decreased with increasing age.

Breast cancer is a major cause of cancer morbidityand mortality in older women, and the incidence ofbreast cancer is known to rise with increasing age. How-ever, the life expectancy of the older patient can be rela-tively prolonged. The average life expectancy for fe-males in the United States is approximately 18 years atage 65, 15 years at age 70, and 9 years at age 80 (9).

The present study has evaluated downstaging ofbreast cancers associated with mammographic screeningfor women age

>

65 years. Downstaging of breast can-cers is one end point to measure the efficacy of mammo-graphic screening for the early detection of breast can-cer. The study population of older patients was obtainedfrom women enrolled in a large health maintenance or-ganization (HMO).

METHODS

Patients for the present study were 130 women age

>

65 years with a new diagnosis of breast cancer. Thesepatients were obtained from women enrolled in U.S.Healthcare

®

, an independent practice association (IPA)model HMO. All patients included in the present studyhad newly diagnosed breast cancer during the study pe-riod of January 1, 1993, to December 31, 1994. At themidpoint of the study period (January 1, 1994), an esti-mated 613,940 women age

>

18 years and 24,415women age

>

65 years were enrolled in the HMO.All of the 130 patients in the study population pre-

sented with unilateral carcinoma of the breast. At thetime of diagnosis of carcinoma of the breast, the medianage for the 130 women was 72 years (mean 72 years;range 65–91 years).

A patient was considered eligible for the presentstudy if the diagnosis of breast cancer was made duringthe study period of January 1, 1993, to December 31,1994, even if definitive treatment was not begun untilafter the study period. A patient was scored as ineligiblefor the present study if the diagnosis of breast cancerwas made prior to the study period, even if definitivetreatment was begun during the study period.

The study population was identified through an anal-ysis of the HMO claims database. To identify this studypopulation, a previously published algorithm was usedto identify women age

>

65 years potentially with thenew diagnosis of breast cancer (10,11). Medical recordswere then requested and reviewed to confirm that carci-noma of the breast had been newly diagnosed during thestudy period. Complete information for analysis wasobtained from the medical records for all of the 130 pa-tients included in the present study. Not included in thepresent study were: 13 patients without complete infor-mation available; 3 patients with lobular carcinoma insitu (see below); 1 patient with clinically suspicious, butnot pathologically confirmed, findings of bilateral carci-noma of the breast; and 1 patient who refused a biopsyfor a suspicious breast mass.

Before reviewing the medical records for the presentstudy, the decision was made to score the new diagnosisof breast cancer as correct for a patient with any of thefollowing: invasive carcinoma of the breast of any histo-logic subtype; ductal carcinoma in situ (DCIS; intraductalcarcinoma) of the breast; or Paget’s disease of the breast(12,13). Before reviewing the medical records, the deci-sion was also made to exclude patients diagnosed withlobular carcinoma in situ, cystosarcoma phylloides, orother sarcomas of the breast. Patients diagnosed withDCIS or Paget’s disease of the breast were included in thepresent study because the usual treatment for DCIS orPaget’s disease of the breast is either mastectomy orbreast-conservation treatment (14–17). Patients diag-nosed with lobular carcinoma in situ were excluded fromthe present study as the usual treatment for patients withlobular carcinoma in situ is careful observation or bilat-eral mastectomy (14).

The medical records for all of the 130 cases were re-viewed by one physician (L.J.S.) experienced in breastcancer research and treatment. Each patient’s medicalrecord was reviewed to determine the American JointCommittee on Cancer (AJCC) clinical stage (18) and todetermine whether or not the patient had undergonemammographic screening for breast cancer. The AJCCclinical stage was scored from the medical record, if re-

96

•

solin et al.

corded, or determined retrospectively from review of themedical record, if not recorded. When known, patho-logic stage was also recorded. Pathologic stage could notbe determined for all patients for various reasons (e.g.,preoperative chemotherapy, simple mastectomy, patientrefusal to undergo definitive treatment, patient treat-ment based on clinical stage because of older patient ageor significant comorbid medical conditions).

A patient was scored as having undergone mammo-graphic screening if a mammogram had been performedprior to the diagnosis of carcinoma of the breast. A pa-tient was scored as not having undergone mammo-graphic screening if a mammogram had been ordered toevaluate signs or symptoms of carcinoma of the breastat the time of diagnosis of carcinoma of the breast with-out a prior screening mammogram. If a patient had un-dergone a negative mammographic screening, but wassubsequently found to have a breast carcinoma that wasdetected by the patient or her physician in the course ofevaluating signs or symptoms of carcinoma (e.g., abreast mass), the patient was scored both as having un-dergone mammographic screening and as having breastcancer detected after a negative mammographic screen-ing. Mammographic screening was performed by a net-work of predominantly community-based radiologygroups (13). The frequency of mammographic screeningfor each individual patient prior to the diagnosis ofbreast cancer could not be reliably determined from therecords obtained for the present study.

Subgroups chosen for analysis were women age 65–69 years, women age 70–79 years, and women age

>

80years. These subgroups were chosen for analysis in anattempt to identify any potential effect of increasing pa-tient age relative to the end points studied.

Statistical comparisons among groups were performedusing the Pearson chi-square test or Fisher’s exact test.

RESULTS

Table 1 shows the mammographic and physical exam-ination findings at presentation for the 130 breast carci-nomas. Breast carcinoma presented with mammographicfindings alone in 39% (51/130) of the overall group andin 48% (51/107) of the cases detected in women who hadundergone mammographic screening.

Of the overall group, 82% (107/130) of the breastcancers were detected in women who had undergonemammographic screening. Of the 107 breast cancers de-tected in women who had undergone mammographicscreening, 8 (7%) were detected after a negative screen-ing mammogram. The median interval from last nega-tive screening mammogram to the detection of theseeight breast cancers was 8 months (mean 10 months;range 1–29 months).

Table 2 shows the AJCC clinical stage at presentationfor the overall group of 130 breast carcinomas. Signifi-cant downstaging was found for the breast cancers de-tected in women who had undergone mammographicscreening compared to the breast cancers detected inwomen who had not undergone mammographic screen-ing. For the overall group of 130 breast cancers, theAJCC clinical stage was stage 0 and stage I in 7% (8/107)and 73% (78/107), respectively, of the breast cancersdetected in women who had undergone mammographicscreening compared to 0% (0/23) and 22% (5/23), re-spectively, of the breast cancers detected in women whohad not undergone mammographic screening (p

,

0.0001). Downstaging was also suggested for the sub-groups of women age 65–69 years, age 70–79 years, andage

>

80 years (all p

<

0.016), although the number ofpatients in some of the subgroups was small.

The AJCC clinical T stage at presentation for theoverall group of 130 breast carcinomas is shown in Ta-ble 3, and significant downstaging was associated with

Table 1. Clinical Presentation for 130 Breast Carcinomas

Mammographic screening performed for women age 65–69 years


Mammographic screening performed for women age

>

80 years


>

65 years

Clinical presentation Yes No Yes No Yes No Yes No Total

Mammography 20 (50)* 0 29 (50) 0 2 (22) 0 51 (48) 0 51 (39)Physical examination 3 (8) 3 (60) 7 (12) 3 (21) 0 2 (50) 10 (9) 8 (35) 18 (14)Both 17 (42) 2 (40) 21 (36) 10 (71) 7 (78) 2 (50) 45 (42) 14 (61) 59 (45)Other 0 0 1

†

(2) 1

‡

(7) 0 0 1 (1) 1 (4) 2 (2)Total 40 (100) 5 (100) 58 (100) 14 (100) 9 (100) 4 (100) 107 (100) 23 (100) 130 (100)

*Numbers in parentheses are percentages.

†

One patient presented with an axillary lymph node on physical examination, a negative mammogram, and a primary breast carcinoma on magnetic resonance imaging (MRI).

‡

One patient presented with an axillary lymph node on physical examination and a negative mammogram.


•

97

the use of screening mammography. Clinical Tis and T1tumors were found in 7% (8/107) and 74% (79/107),respectively, of the breast cancers detected in womenwho had undergone mammographic screening com-pared to 0% (0/23) and 22% (5/23), respectively, of thebreast cancers detected in women who had not under-gone mammographic screening (p

,

0.0001). Down-staging was also suggested for the subgroups of womenage 65–69 years, age 70–79 years, and age

>

80 years(all p

<

0.043), although the number of patients in someof the subgroups was small.

Of the 112 AJCC clinical stage I–II breast cancers,102 (91%) had undergone pathologic axillary lymphnode staging (Table 4). For these 102 cases, pathologicaxillary lymph node stage was N0 for 76% (66/87) ofthe women who had undergone mammographic screen-ing compared to 53% (8/15) of the women who had notundergone mammographic screening (p

5

0.019). For

the subgroups of women age 65–69 years and age 70–79 years, the percentage of patients with node-negativedisease was higher for women who had undergonemammographic screening compared to women who hadnot undergone mammographic screening, but neither ofthe comparisons was statistically significant (both p

>

0.068). For the subgroup of women age

>

80 years,there were too few patients for meaningful statisticaltesting. Three of the eight patients with DCIS underwentpathologic axillary lymph node staging, and all threewere pathologically node negative.

DISCUSSION

The present study has documented significant down-staging of breast carcinomas associated with mammo-graphic screening in older women age

>

65 years. Signifi-cant downstaging was found for AJCC clinical stage(Table 2), AJCC clinical T stage (Table 3), and pathologic

Table 2. AJCC Clinical Stage at Presentation for 130 Breast Carcinomas




>

80 years


>

65 years

AJCC clinical stage Yes No Yes No Yes No Yes No Total

0 1 (2)* 0 7 (12) 0 0 0 8 (7) 0 8 (6)I 35 (88) 0 36 (62) 5 (36) 7 (78) 0 78 (73) 5 (22) 83 (64)II 2 (5) 3 (60) 13 (22) 6 (43) 1 (11) 4 (100) 16 (15) 13 (57) 29 (22)III 0 2 (40) 2 (3) 3 (21) 0 0 2 (2) 5 (22) 7 (5)IV 2 (5) 0 0 0 1 (11) 0 3 (3) 0 3 (2)Total 40 (100) 5 (100) 58 (100) 14 (100) 9 (100) 4 (100) 107 (100) 23 (100) 130 (100)

*Numbers in parentheses are percentages.For age 65–69 years, p

5

0.0001.For age 70–79 years, p

5

0.016.For age

>

80 years, p

5

0.0070.For age

>

65 years, p

,

0.0001.

Table 3. AJCC Clinical T Stage at Presentation for 130 Breast Carcinomas




>

80 years


>

65 years

AJCC clinical T stage Yes No Yes No Yes No Yes No Total

T0* 0 0 0 1 (7) 0 0 0 1 (4) 1 (1)Tis 1 (2)

†

0 7 (12) 0 0 0 8 (7) 0 8 (6)T1 35 (88) 0 36 (62) 5 (36) 8 (89) 0 79 (74) 5 (22) 84 (65)T2 4 (10) 3 (60) 11 (19) 4 (29) 1 (11) 3 (75) 16 (15) 10 (43) 26 (20)T3 0 0 2 (3) 1 (7) 0 1 (25) 2 (2) 2 (9) 4 (3)T4 0 2 (40) 2 (3) 3 (21) 0 0 2 (2) 5 (22) 7 (5)Total 40 (100) 5 (100) 58 (100) 14 (100) 9 (100) 4 (100) 107 (100) 23 (100) 130 (100)

*One patient with an axillary lymph node presentation and no detected primary breast carcinoma.

†

Numbers in parentheses are percentages.

‡

Excluding the one T0 case.For age 65–69 years, p

5


5

0.043.

‡

For age

>

80 years, p

5

0.0070.For age

>

65 years, p

,

0.0001.

‡

98

•

solin et al.

axillary lymph node stage for the clinical stage I–II lesions(Table 4). The findings in this report are from a studypopulation of patients from a large HMO.

A number of prospective, randomized studies ofscreening mammography have demonstrated significantdownstaging of breast cancers associated with mammo-graphic screening by analyzing various end points suchas clinical stage, percentage of noninvasive tumors, tu-mor size, or pathologic axillary lymph node stage (1–6).Women age

>

65 years have been included in at leasttwo of the randomized studies of screening mammogra-phy (1,4,5) and one nonrandomized study of screeningmammography (19,20). Tabar et al. (4) reported thatthe relative risk (RR) of death from breast cancer was0.79 for women age 70–74 years who were randomizedto undergo mammographic screening compared towomen who were not randomized to undergo mammo-graphic screening; this RR was similar to the risk forwomen age 60–69 years (RR

5

0.60) and women age50–59 years (RR

5

0.66). Van Dijck et al. (19) reporteda reduction in breast cancer mortality associated withmammographic screening for women age 65–74 years,but no reduction for women age

>

75 years in a nonran-domized study. In the Breast Cancer Detection Demon-stration Project (BCDDP), breast cancer screening con-sisted of mammography plus physical examination;women age 65–74 years in the BCDDP had a higher rateof breast cancers detected at the first screening com-pared to younger women, but a similar rate of detectionat the second screening (21). In a retrospective study ofan HMO population of women of all ages, Solin et al.(13) found significant downstaging associated withmammographic screening as measured by AJCC clinicalstage and AJCC clinical T stage.

One difference between the randomized studies andthe present study is that the downstaging reported for the

present study is for women who had undergone mammo-graphic screening, whereas the downstaging reported inthe randomized studies is for women who were random-ized to undergo mammographic screening. One limita-tion of the present study is that it was not possible toidentify women who were invited to undergo, but whochose not to undergo, mammographic screening. In ran-domized studies of screening mammography, the com-parison is generally between women randomized versuswomen not randomized to undergo mammographicscreening. Some of these studies also report separatelytwo subsets of women: women who were randomized toundergo and who actually underwent mammographicscreening; and women who were randomized to undergobut who did not undergo mammographic screening.

Three studies have demonstrated significant down-staging of pathologic axillary lymph nodes associatedwith mammographic screening (3,4,6). For women age50–74 years, Tabar et al. (4) found that the incidence ofpathologically negative axillary lymph nodes was 70.4%for breast cancers detected in women who were random-ized to undergo mammographic screening compared to61.5% for breast cancers detected in women who werenot randomized to undergo mammographic screening.For women age 45–64 years with clinical stage I–II dis-ease and for whom the pathologic axillary lymph nodestage was known, Roberts et al. (3) reported that the inci-dence of pathologically negative nodes for breast cancersdetected in women who were randomized to undergomammographic screening was 69% compared to 60%for breast cancers detected in women who were not ran-domized to undergo mammographic screening. Robertset al. (3) also reported that there were more pathologi-cally negative axillary lymph node cases for the breastcancers detected in women who were randomized to un-dergo mammographic screening when the tumors were

Table 4. Pathologic Axillary Lymph Node Stage for 102 Patients with AJCC Clinical Stage I–II Breast Carcinomas

Pathologic axillary lymph node stage

Mammographic screening performed for women age 65–69

years



>

80 years


>

65 years

Yes No Yes No Yes No Yes No Total

N0 29 (78)* 1 (50) 34 (77) 7 (64) 3 (50) 0 66 (76) 8 (53) 74 (73)N1

1–3 positive 7 (19) 1 (50) 8 (18) 1 (9) 2 (33) 1 (50) 17 (20) 3 (20) 20 (20)

>

4 positive 1 (3) 0 2 (5) 3 (27) 1 (17) 1 (50) 4 (5) 4 (27) 8 (8)Total 37 (100) 2 (100) 44 (100) 11 (100) 6 (100) 2 (100) 87 (100) 15 (100) 102 (100)

*Numbers in parentheses are percentages.For age 65–69 years, p

5


5

0.068.For age

>

80 years, insufficient numbers of patients for meaningful statistical testing.For age

>

65 years, p

5

0.019.


•

99

stratified by size. Bjurstam et al. (6) found that the RR forthe detection of lymph node-positive breast cancer was0.65 for women invited compared to women not invitedto undergo mammographic screening. In the presentstudy of women age

>

65 years, the incidence of patho-logically node-negative disease was 76% for women whohad undergone mammographic screening and 53% forwomen who had not undergone mammographic screen-ing, and the incidence of four or more positive axillarylymph nodes was higher in women who had not under-gone mammographic screening (27% versus 5%, respec-tively; p

5

0.019) (Table 4).Analyzing pathologic lymph node downstaging asso-

ciated with screening mammography is problematic be-cause pathologic lymph node stage usually cannot beobtained for all patients identified as having carcinomaof the breast. Tabar et al. (4) reported pathologic axil-lary lymph node stage or presence of distant metastaticdisease for 87% (2,146/2,467) of the overall group. Inthe study of Roberts et al. (3), pathologic axillary lymphnode stage was known in 92% (410/448) of the clinicalstage I–II cases. Pathologic axillary lymph node stagewas known in 91% (102/112) of the clinical stage I–IIcases in the present study. Clinical stage, in contrast topathologic stage, generally can be obtained in essentiallyall cases. Clinical stage was known for 99% (657/663)of the cases reported by Roberts et al. (3) and for all ofthe cases in the present study (Table 2).

Downstaging of breast cancers, with the associatedimprovement in prognosis and survival, is one of two im-portant benefits of mammographic screening for the earlydetection of breast cancer. The second important benefitof screening mammography is the increased potential forbreast conservation treatment. In a previous analysis ofan HMO population of women of all ages, Solin et al.(12) confirmed a greater eligibility for breast conserva-tion treatment as an alternative to mastectomy for thebreast cancers detected in women who had undergonemammographic screening compared to the breast cancersdetected in women who had not undergone mammo-graphic screening (88% versus 60%, respectively; p

,

0.0001). For the subset of women age

>

50 years, the ac-tual local treatment of the breast received by the patientshowed a significantly increased use of breast conserva-tion treatment for the breast cancers detected in womenwho had undergone mammographic screening comparedto the breast cancers detected in women who had not un-dergone mammographic screening (49% versus 21%, re-spectively; p

5

0.016). Andersson et al. (1) and Roberts etal. (3) have also reported an increase in the use of breast-

conserving surgery associated with mammographicscreening; however, neither of these two studies analyzedolder women separately.

In summary, the present report has demonstrated sig-nificant downstaging of breast cancers associated withmammographic screening for women age

>

65 years.Downstaging was documented for AJCC clinical stage,AJCC clinical T stage, and pathologic axillary lymphnode stage for the clinical stage I–II tumors. The signifi-cant downstaging of breast cancers associated withmammographic screening, which has previously beenfound in prospective, randomized trials, has been docu-mented in the present study for an HMO population ofwomen age

>

65 years.

REFERENCES

1. Andersson I, Aspegren K, Janzon L,

et al.

Mammo-graphic screening and mortality from breast cancer: the Malmomammographic screening trial.

Br Med J

1988;297:943–48.2. Chu KC, Smart CR, Tarone RE. Analysis of breast can-

cer mortality and stage distribution by age for the Health Insur-ance Plan clinical trial.

J Natl Cancer Inst

1988;80:1125–32.3. Roberts MM, Alexander FE, Anderson TJ,

et al.

Edin-burgh trial of screening for breast cancer: mortality at sevenyears.

Lancet

1990;335:241–46.4. Tabar L, Fagerberg G, Chen H-H,

et al.

Efficacy ofbreast cancer screening by age: new results from the Swedishtwo-county trial.

Cancer

1995;75:2507–17.5. Tabar L, Fagerberg G, Duffy SW,

et al.

Update of theSwedish two-county program of mammographic screening forbreast cancer.

Radiol Clin N Am

1992;30:187–210.6. Bjurstam N, Bjorneld L, Duffy SW,

et al.

The Gothen-burg breast screening trial: first results on mortality, incidence,and mode of detection for women ages 39–49 years at random-ization.

Cancer

1997;80:2091–99.7. Leitch AM, Dodd GD, Costanza M,

et al.

AmericanCancer Society guidelines for the early detection of breast can-cer: update 1997.

CA Cancer J Clin

1997;47:150–53.8. Burns RB, McCarthy EP, Freund KM,

et al.

Variabilityin mammography use among older women.

J Am Geriatr Soc

1996;44:922–26.9. U.S. Bureau of the Census.

Statistical abstract of theUnited States: 1996

, 116th ed. Washington, D.C.: U.S. Gov-ernment Printing Office, 1996:89.

10. Solin LJ, Legorreta A, Schultz DJ,

et al.

Analysis of aclaims database for the identification of patients with carci-noma of the breast.

J Med Syst

1994;18:23–32.11. Solin LJ, MacPherson S, Schultz DJ, Hanchak NA.

Evaluation of an algorithm to identify women with carcinomaof the breast.

J Med Syst

1997;21:189–99.12. Solin LJ, Legorreta A, Schultz DJ,

et al.

The importance

100

•

solin et al.

of mammographic screening relative to the treatment ofwomen with carcinoma of the breast.

Arch Intern Med

1994;154:745–52.

13. Solin LJ, Schultz DJ, Legorreta AP, Goodman RL.Downstaging of breast carcinomas associated with mammo-graphic screening.

Breast Dis

1995;8:45–56.14. Fowble B. In situ breast cancer. In: Fowble B, Good-

man RL, Glick JH, Rosato EF, eds. Breast cancer treatment: acomprehensive guide to management. St. Louis: Mosby YearBook, 1991:325–44.

15. Morrow M, Schnitt SJ, Harris JR. Ductal carcinoma insitu. In: Harris JR, Lippman ME, Morrow M, Hellman S, eds.Diseases of the breast. Philadelphia: Lippincott-Raven, 1996:355–68.

16. Pierce LJ, Haffty BG, Solin LJ, et al. The conservativemanagement of Paget’s disease of the breast with radiotherapy.Cancer 1997;80:1065–72.

17. Solin LJ, Kurtz J, Fourquet A, et al. Fifteen-year results

of breast-conserving surgery and definitive breast irradiationfor the treatment of ductal carcinoma in situ of the breast. JClin Oncol 1996;14:754–63.

18. American Joint Committee on Cancer. AJCC cancerstaging manual, 5th ed. Philadelphia: Lippincott-Raven, 1997:171–78.

19. Van Dijck JAAM, Verbeek ALM, Beex LVAM, et al.Mammographic screening after the age of 65 years: evidencefor a reduction in breast cancer mortality. Int J Cancer 1996;66:727–31.

20. Van Dijck JAAM, Verbeek ALM, Beex LVAM, et al.Breast-cancer mortality in a non-randomized trial on mammo-graphic screening in women over age 65. Int J Cancer 1997;70:164–68.

21. Smart CR, Byrne C, Smith RA, et al. Twenty-year fol-low-up of the breast cancers diagnosed during the Breast Can-cer Detection Demonstration Project. CA Cancer J Clin 1997;47:134–49.

Documents

Downstaging of Breast Carcinomas in Older Women Associated with Mammographic Screening