Paget�s Disease of the Breast: There Is a Role forBreast-Conserving Therapy

Kazumi Kawase, MD,1 Dominick J. DiMaio, MD,1 Susan L. Tucker, PhD,2

Thomas A. Buchholz, MD,3 Merrick I. Ross, MD,1 Barry W. Feig, MD,1

Henry M. Kuerer, MD, PhD,1 Funda Meric-Bernstam, MD,1 Gildy Babiera, MD,1

Frederick C. Ames, MD,1 S. Eva Singletary, MD,1 and Kelly K. Hunt, MD, FACS1

1Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Unit 444,Houston, Texas 77030

2Department of Biomathematics, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Unit 237,Houston, Texas 77030

3Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard,Unit 1202, Houston, Texas 77030

Background: The optimal surgical management of Paget�s disease of the breast remains tobe defined. Mastectomy has been the standard of care, but several institutions have recentlyadvocated breast-conserving surgery, particularly for patients with minimal disease. In aneffort to develop rational treatment guidelines, we examined our institutional experience withPaget�s disease of the breast.Methods: Patients with Paget�s disease of the breast who had surgical therapy at our

institution between 1949 and 1993 were reviewed. In addition to patient and tumor charac-teristics, charts were reviewed for treatment modalities, locoregional recurrence patterns, andsurvival. Subgroups were compared for differences in survival in both univariate and multi-variate analyses.Results: A total of 104 patients met the study criteria. The most common presenting

symptoms were nipple discharge and eczematous changes of the nipple/areola complex.Ninety-seven patients (93.2%) had an underlying invasive or noninvasive cancer associatedwith Paget�s disease. Ninety-two patients (88.5%) underwent mastectomy, and 12 (11.5%) hada breast-conserving procedure. On univariate analysis, patients with age <60 years at diag-nosis, stage II disease, positive lymph nodes, invasive disease, or a palpable mass had sig-nificantly lower 10-year disease-specific and recurrence-free survival. There were fourlocoregional recurrences (three after mastectomy and one after breast conservation). Therewere no significant differences in overall, disease-specific, or recurrence-free survival accordingto the type of surgery.Conclusions: Paget�s disease of the breast is almost always associated with an underlying

breast cancer. Breast-conserving approaches result in local control and survival rates similar tothose achieved with mastectomy.Key Words: Paget�s disease—Breast conservation—Breast cancer—Survival.

Paget�s disease of the breast is an uncommon dis-order that accounts for 1% to 3% of all mammarytumors.1�3 Clinically, patients with Paget�s diseasehave erythema and scaling or eczematous changesin the nipple and areola that they often describeas pruritic or burning. These nipple changes laterprogress to ulcerations and erosion and may occur

Received May 22, 2004; accepted January 5, 2005; publishedonline j.Presented in part at the 26th Annual San Antonio Breast Cancer

Symposium, San Antonio, Texas, December 2�6, 2003.Address correspondence and reprint requests to: Kelly K. Hunt,

MD, FACS; E-mail: khunt@mdanderson.org.

Published by Springer Science+Business Media, Inc. � 2005The Society of Surgical Oncology, Inc.

Annals of Surgical Oncology, 12(5): 1)7DOI: 10.1245/ASO.2005.05.026

1

alone or in conjunction with an underlying palpablemass. Paget�s disease is often mistaken for dermatitisor a benign dermatological condition involving thenipple. It is not unusual for the diagnosis to be de-layed for 6 to 12 months before a biopsy is per-formed.There are many theories regarding the origin of

Paget�s disease: Paget�s cells may have an intraepi-dermal origin (in situ transformation theory),4 orthey may originate from an intraductal cancer andspread by upward migration to the nipple (epi-dermotropic theory).5,6 Most researchers agree,however, that regardless of their origin, Paget�s cellsare a sign of underlying breast cancer.1,2,7,8 Pagetsuggested in his original description that an under-lying cancer succeeded the nipple changes in mostcases within 1 year and in all cases within 2 years.9

However, cases of Paget�s disease of the nipple alone,with no underlying cancer, have been reported.10

Published treatment recommendations for Paget�sdisease have varied from simple excision or radio-therapy to total mastectomy. Mastectomy has longbeen considered as a definitive treatment.1,2,11�14

However, the surgical management of breast cancerhas changed significantly over the last three decadesas a result of several large randomized trials dem-onstrating that breast-conservation therapy results insurvival equal to that after mastectomy in patientswith early-stage breast cancer.15�18 Thus, althoughthe rarity of Paget�s disease does not allow for ran-domized trials comparing treatment modalities, withthe aid of improved mammographic and radiother-apy techniques, the breast-conserving approach hasbeen considered as a treatment option in selectedpatients. 19�22

We initiated this study to examine our institutionalexperience with Paget�s disease in an attempt to de-velop rational treatment guidelines.

PATIENTS AND METHODS

We conducted a retrospective study of patientsdiagnosed with Paget�s disease of the breast andtreated with primary surgical therapy at the Univer-sity of Texas M. D. Anderson Cancer Center between1949 and 1993. Patients were identified through asearch of a prospectively compiled institutionaldatabase. Patients who had involvement of the nippleas part of a locally advanced cancer were not in-cluded. Patient records were reviewed to determineclinical presentation, histopathologic features, treat-ment modalities, locoregional recurrence patterns,

and survival. Patients were restaged according to the5th American Joint Committee for Cancer tumor-node-metastasis staging manual.23 Survival data wereanalyzed by using the method of Kaplan and Meier.24

The log-rank test was used to compare disease-spe-cific survival (DSS) and recurrence-free survival(RFS). The v2 test was used to correlate prognosticfactors significant in univariate but not multivariateanalysis with nodal status. Statistical analysis wasperformed with Stata statistical software, release 8(Stata Corp., College Station, TX).

RESULTS

Patient Characteristics

We identified 113 patients with a diagnosis of Paget�sdisease of the breast treated at the M. D. AndersonCancer Center between 1949 and 1993. Nine patientswith stage III or IV disease were excluded from ouranalysis because nipple involvement in these patientswas believed to be a late manifestation of locally ad-vanced or recurrent disease. The remaining 104 pa-tients formed the basis of this study. Of these 104patients, 2 weremen and 102were women. Themedianage was 57 years (range, 24�90 years).

Tumor Characteristics

Presenting symptoms are listed in Table 1. Mostpatients had more than one symptom at presentation.The most common presenting symptoms were nippledischarge and eczema or scaling of the nipple (Fig. 1).In addition to nipple changes, 36 patients (35%) had apalpable mass. Twelve (33%) of these masses wereadjacent or deep to the nipple, and 24 (67%) werelocated distant from the nipple. Ten patients had apalpable breast mass without obvious nipple changes.Mammography was not commonly performed before1970. Mammography was performed at presentationin 81 patients, 60 (74%) of whom had abnormalfindings.The histopathologic findings at definitive surgical

resection are listed in Table 2. Overall, 63 patients(60%) also had invasive carcinoma, 34 patients (33%)had ductal carcinoma-in-situ, and the lesion wasconfined within the nipple without evidence of inva-sive or noninvasive disease in 7 patients (7%). Mul-tifocal disease was identified in the breast in 37patients (36%). Of the 91 patients who had lymphnodes examined, 26 (29%) had metastatic disease onpathologic examination. Forty-one patients (39%)

K. KAWASE ET AL.2

Ann. Surg. Oncol. Vol. 12, No. 5, 2005

had stage 0 disease, 23 (22%) had stage I disease, and40 (38%) had stage II disease.

Treatment Modalities

All patients underwent surgical excision as part oftheir local therapy. Patients were selected for nipple/areola complex excision alone when the treatingsurgeon believed that all disease could be encom-passed in a single excision. In addition, patients witha mammographic abnormality or palpable mass inaddition to the nipple changes were considered can-didates for breast-conserving surgery if all diseasecould be resected with negative margins and withoutpoor cosmesis.Ninety-two patients (88%) were treated with simple

(3 patients), extended simple (13 patients), modifiedradical (45 patients), or radical (31 patients) mastec-tomy. Twelve patients were treated with primarybreast-conserving therapy in addition to completeexcision of the nipple/areola complex: 11 (92%) hadsegmental mastectomy alone, and 1 (8%) had seg-mental mastectomy with axillary dissection. Radio-therapy was used in 30 patients (29%), including all12 patients treated with breast-conserving surgery.The median dose to the breast was 50 Gy. Eighteenpatients (17%) were treated with adjuvant chemo-therapy—in most cases, a doxorubicin-based combi-nation chemotherapy regimen. Nine patients weretreated with both chemotherapy and radiotherapy inaddition to surgical resection.

Survival Analysis

At the time of our analysis, 52 patients were alive,20 had died of breast cancer, and 32 had died of othercauses. Among the surviving patients, the medianfollow-up time was 7 years (range, 10 months to 29years). The 5-year DSS rate for all patients was84% ± 8%, and the 10-year DSS rate was79% ± 9%. There were four locoregional recurrences

(three in the mastectomy group and one in the breast-conservation group). RFS rates were calculated fromthe date of surgery to the date of recurrence or lastfollow-up. The 5-year RFS rate was 79% ± 8%, andthe 10-year RFS rate was 75% ± 9%. Subgroups ofpatients were compared for differences in DSS andRFS as shown in Table 3. For both DSS and RFS,the only adverse factor selected by multivariateanalysis at a significance level of p £ .05 was positivenodal status. The type of surgical treatment did notresult in differences in DSS and RFS (p = .697 and.953, respectively). For DSS, the relative risk associ-ated with positive nodal status was 9.1 (95% confi-dence interval [CI], 2.8�23.0). For RFS, the relativerisk associated with positive nodal status was 9.8(95% CI, 3.5�27.2). After correction for nodal status,the only factor close to statistical significance wasage, with a relative risk in younger patients of 4.0(95% CI, .9�17.5; p = .067) for DSS and 3.3 (95%CI, .97-11.4; p = .056) for RFS. All of the prognosticfactors that predicted decreased DSS and RFS onunivariate analysis, except for age, were correlatedwith nodal status (Table 4).

DISCUSSION

Over 44 years (1949 to 1993), we identified 113patients treated at our institution for Paget�s diseaseof the breast. We excluded 9 patients who had locallyadvanced or metastatic disease, leaving a total of 104patients. The median age was 57 years. This finding issimilar to those of other series—it has previouslybeen noted that Paget�s disease often presents at aslightly older age than breast carcinoma withoutPaget�s disease. There were 2 men in our series,adding to the approximately 40 cases in men reportedin the literature to date. The most frequently reportedsymptoms in our patients were nipple discharge andscaling or eczematous changes. Most patients hadtwo or more symptoms at presentation. In the liter-ature, approximately 40% to 50% of patients withPaget�s disease present with an associated palpablemass,1,7�9 and 20% to 60% of these masses are re-mote from the nipple.2,8 In our series, one third of thepatients had a palpable mass at presentation, andclinically, 67% of these palpable masses were distantfrom the nipple/areola complex.In our study, mammography was performed on 81

patients at presentation. Of these patients, 60 (74%)had abnormalities recognized by the mammographer.A mammographic abnormality has been reported in40% to 50% of cases described in the literature. One

TABLE 1. Presenting symptoms

Symptom No. of patients (%)a

Nipple discharge 62 (60)Eczema or scaling 54 (52)Nipple erythema or edema 38 (37)Abnormal nipple sensation 38 (37)Nipple inversion 23 (22)

a Sixteen patients (15%) had 1 symptom, 41 patients (39%) had 2symptoms, 21 patients (20%) had 3 symptoms, 11 patients (11%)had 4 symptoms, and 2 patients (2%) had 5 symptoms. Fourteenpatients (13%) had no presenting symptoms.

SURGICAL MANAGEMENT OF PAGET’S DISEASE OF THE BREAST 3

Ann. Surg. Oncol. Vol. 12, No. 5, 2005

possible explanation for the higher rate of mammo-graphic abnormalities in our study is the fact thatdiagnostic mammography was used: this may resultin a higher yield than standard screening mammog-raphy. Breast ultrasonography and magnetic reso-nance imaging have been proposed as potentialmodalities for screening and diagnosis.25,26 Thesemodalities can be used to confirm mammographicallydetected abnormalities or to detect mammographi-cally occult lesions in patients with Paget�s disease.The incidence of underlying carcinoma associated

with Paget�s disease has been reported in 82% to100% of cases.1,6�8,27,28 The finding of underlyingcarcinoma reaches almost 100% when a palpablemass is also present.6,8,27 Multifocal or multicentriccarcinoma associated with Paget�s disease has beenreported in 32% to 41% of patients.8,28 In our study,93% of patients had an associated underlying carci-noma; of these, 33% were ductal carcinoma-in-situonly, and 36% had multifocal disease. These resultsare consistent with other studies and confirm the highfrequency of underlying carcinoma associated withPaget�s disease.The surgical treatment of Paget�s disease has been

controversial. Historically, mastectomy with or with-out axillary lymph node dissection was the standardtherapy, and this approach was supported by theobservation that there was almost always an under-lying invasive or noninvasive carcinoma. Several re-

ports have shown that 20% to 40% of mastectomyspecimens from patients with Paget�s disease havemultifocal or multicentric disease, and this disease wasoften underestimated on mammography.2,3,27,28 Inthe literature, almost all patients treated with localexcision, radiotherapy, or both had the clinical pre-sentation of Paget�s disease without a palpable mass.Dixon et al.3 reported a high local recurrence rate

of 40% with a median follow-up time of 56 months in10 patients treated with local excision alone whosemammograms suggested in situ changes confined tothe immediate nipple area. Polgar et al.29 reported alocal recurrence rate of 33% with a median follow-uptime of 6 years in 33 patients treated with localexcision alone with or without underlying ductalcarcinoma-in-situ confined to the nipple/areolacomplex. They concluded that local excision alonewas not an appropriate treatment for patients withPaget�s disease of the nipple.However, increasing numbers of clinicians suggest

that breast-conserving therapy is effective for selectedpatients. Stockdale et al.19 reviewed 28 patients withPaget�s disease of the nipple treated with radiother-apy alone. Sixteen (84%) of 19 patients without apalpable mass and no abnormality seen on mam-mography remained free of disease at a median fol-low-up time of 63 months. Six patients had nopalpable mass but either had a mammographicallydetected underlying tumor or did not undergoscreening mammography. Of these six patients, five(83%) had relapsed at the time of their report. Of thethree patients with a palpable underlying tumor, onehad concomitant systemic metastasis, and two hadlocal recurrences. Fourquet et al.20 reported on 20patients with Paget�s disease of the breast confined tothe nipple without clinical or radiological signs of anassociated intraductal or infiltrating breast tumorwho were treated conservatively with radiotherapy

TABLE 2. Histopathologic findings at surgical resection

Finding No. of patients (%)

Invasive ductal carcinoma 43 (41)Ductal carcinoma-in-situ 34 (33)Adenocarcinoma not otherwise specified 19 (18)Invasive lobular carcinoma 1 (1)Paget�s disease of the nipple only 7 (7)

FIG. 1. Clinical presentation of Paget�s disease. (A) Inthis patient, the erythema and scaling changes areconfined to the nipple. (B) In this patient, changesconsistent with Paget�s disease are more extensive, withinvolvement of the nipple and areola.

K. KAWASE ET AL.4

Ann. Surg. Oncol. Vol. 12, No. 5, 2005

alone or limited surgery and radiotherapy. Theactuarial 7-year probability of freedom from diseasein these patients was 81%.Bijker et al.21 reported the results of a prospective

study conducted by the European Organization forResearch and Treatment of Cancer. In this study, 61patients with Paget�s disease of the breast were trea-ted with complete excision of the nipple/areola com-plex, including the underlying breast tissue, withtumor-free margins, followed by irradiation of theentire breast. Ninety-seven percent of patients pre-sented without an associated palpable mass, and 84%had no abnormality on mammography. At a medianfollow-up time of 6.4 years, four patients (7%) had alocal recurrence; three of those were invasive cancer.The authors concluded that breast-conserving ther-apy was a feasible alternative for patients with Paget�sdisease and a limited extent of underlying ductalcarcinoma-in-situ.

Marshall et al.22 reported the long-term follow-upresults of breast-conserving surgery with radiother-apy in a study of 38 patients with Paget�s disease ofthe breast without a palpable mass or mammographicdensity. Of these patients, 94% underwent completeor partial excision of the nipple/areola complex,100% underwent irradiation of the breast, and 97%also received a boost to the remaining nipple or tu-mor bed. The actuarial local control rate for breastrecurrence as a component of first failure was 91%,83%, and 76% at 5, 10, and 15 years, respectively.A treatment algorithm based on the clinical and

pathologic features in patients with Paget�s disease ofthe breast is outlined in Fig. 2. At our institution,patients were considered for nipple/areola complexexcision alone when they presented without a palpablemass or mammographic abnormality. In addition,patients with a mammographic abnormality or pal-pable mass were considered candidates for breast-

TABLE 4. Prognostic factors identified by univariate analysis and their relationship to nodal status

Factor Group No. of node-negative patients No. of node-positive patients P valuea

Age at diagnosis <60 y 38 18 .336‡60 y 27 8

Stage 0 30 0 <.001I 22 0II 13 26

Histology Adenocarcinoma 12 6 <.001In situ 28 0Invasive ductal 23 19Invasive lobular 0 1Other 2 0

Palpable mass No 49 7 <.001Yes 16 19

a P value from v2 test.

TABLE 3. Analyses of factors possibly influencing disease-specific survival (DSS) and recurrence-free survival (RFS)

Factor Group No. of patients10-y DSS,

% (95% CIb)UnivariateP valuea

10-y RFS, %(95% CIb)

UnivariateP value

Age at diagnosis <60 y 65 70 ± 13 .008 68 ± 2 .060‡60 y 39 94 ± 9 86 ± 11

Stage 0 41 92 ± 12 .0011 90 ± 12 .001I 23 90 ± 14 80 ± 8II 40 61 ± 16 57 ± 16

Nodal status Positive 26 47 ± 22 <.0001 41 ± 21 <.0001Negative 65 93 ± 7 91 ± 8

Histology Adenocarcinoma 19 78 ± 20 .021 71 ± 22 .007Invasive ductal 43 71 ± 15 65 ± 15In situ ductal 34 96 ± 8 96 ± 8

Palpable mass Yes 36 69 ± 16 .036 65 ± 16 .024No 68 84 ± 10 80 ± 11

Multifocal disease Yes 37 74 ± 17 .499 71 ± 17 .485No 63 85 ± 10 81 ± 10

Type of surgery Mastectomy 92 79 ± 9 .697 75 ± 9 .953BCT 12 67 ± 54 61 ± 51

CI, confidence interval; BCT, breast-conserving therapy.a Log-rank test.b ± 2 SD.

SURGICAL MANAGEMENT OF PAGET’S DISEASE OF THE BREAST 5

Ann. Surg. Oncol. Vol. 12, No. 5, 2005

conserving surgery if all disease could be resected withnegative margins and without poor cosmesis. All pa-tients who had breast-conserving surgery underwentadjuvant radiotherapy with a median dose of 50 Gy.In our study, at a median follow-up time of 7 years,locoregional recurrence was observed in only 1 (8%)of 12 patients treated with breast-conserving surgery.In addition, there was no difference in DSS or RFSbetween patients selected for breast-conserving sur-gery and those selected for mastectomy. However,because the 2 treatment groups (12 breast-conservingsurgery vs. 92 mastectomy) were highly biased towardmastectomy, this study may not have had the statis-tical power to detect a significant difference.Recent reports from several large randomized

studies with long-term follow-up have shown thatbreast-conserving surgery is equivalent to mastec-tomy in terms of overall and disease-free survival inpatients with breast cancer.11�14 Although the num-ber of patients in our study is very small, our resultsare similar to these randomized trial results in termsof local recurrence rates, DSS, and RFS. This wouldsuggest that for selected patients, breast-conservingsurgery is an appropriate treatment for Paget�s dis-ease of the breast. Once multicentric disease has beenruled out by careful examination and imaging studies,surgical treatment of Paget�s disease of the breast canbe chosen according to the stage and histologicalsubtype of the underlying carcinoma.Traditionally, axillary lymph node dissection has

been used as the preferred approach to evaluate thestatus of the regional nodes; however, sentinel lymphnode dissection has emerged as a reliable strategy topredict axillary lymph node status for patients with

invasive breast carcinoma.30 The treatment of pa-tients with Paget�s disease of the breast shouldincorporate these recent developments.It is clear that prognosis is different in patients with

a palpable mass and those without a palpable mass.Patients with a palpable mass almost always have anassociated underlying invasive carcinoma (75%-100%) and a high rate of axillary lymph nodemetastasis (45%-65%), whereas patients without apalpable mass have an associated invasive carcinomain only 20% to 30% of cases and positive axillarylymph nodes in 10% to 20% of cases,1,7�9,15,16,27

Overall survival has been shown to be affected bylymph node status and is reported to be 75% to 95%in patients with negative lymph nodes and as low as20% to 25% in those with positive lymph nodes.1,8

The survival analysis in our study demonstrated a 10-year DSS rate of 47% in patients with positive nodesand 93% in those with negative nodes, and this isconsistent with other studies (Table 3). Univariateanalysis confirmed that tumor stage, invasiveness,and palpability were related to nodal status (Table 4),as has been suggested in the literature.1,6,7,27 There-fore, adjuvant treatment (radiotherapy, chemother-apy, or hormonal therapy) is recommended on thebasis of lymph node status and features of the pri-mary tumor as determined by final pathologicexamination.In summary, Paget�s disease of the nipple is almost

always associated with an underlying invasive ornoninvasive breast cancer. Breast-conserving therapyresults in local control and survival rates similar tothose achieved with mastectomy. The surgical treat-ment plan must be chosen on the basis of carefulclinical and imaging assessment of each patient. Be-cause prognosis is determined largely by nodal status,postoperative adjuvant therapies should be based onthe final tumor-node-metastasis stage.

ACKNOWLEDGMENTS

The authors thank Stephanie Deming for her crit-ical reading of the manuscript.

REFERENCES

1. Ashikari R, Park K, Huvos AG, Urban JA. Paget�s disease ofthe breast. Cancer 1970;26:680�5.

2. Chaudary MA, Millis RR, Lane EB, Miller NA. Paget�s dis-ease of the nipple: a ten-year review including clinical, patho-logical, and immunohistochemical findings. Breast Cancer ResTreat 1986;8:139�46.

FIG. 2. A treatment algorithm based on the clinical and pathologicfeatures of patients presenting with Paget�s disease of the breast.

K. KAWASE ET AL.6

Ann. Surg. Oncol. Vol. 12, No. 5, 2005

3. Dixon AR, Galea MH, Ellis IO, Elston CW, Blamey RW.Paget�s disease of the nipple. Br J Surg 1991;78:722�3.

4. Muir R. Pathogenesis of Paget�s disease of the nipple andassociated lesions. Br J Surg 1935;22:728�37.

5. Inglis K. Paget�s disease of the nipple, with special reference tochanges in the ducts. Am J Pathol 1946;22:1�33.

6. Yim JH, Wick MR, Philpott GW, Norton JA, Doherty GM.Underlying pathology in mammary Paget�s disease. Ann SurgOncol 1997;4:287�92.

7. Kollmorgen DR, Varanasi JS, Edge SB, Carson WE III. Pa-get�s disease of the breast: a 33-year experience. J Am Coll Surg1998;187:171�7.

8. Fu W, Mittel VK, Young SC. Paget disease of the breast:analysis of 41 patients. Am J Clin Oncol 2001;24:397�400.

9. Paget J. On the disease of the mammary areola precedingcancer of the mammary gland. St Bartholomew Hosp Rep1874;10:87�9.

10. Lagios MD, Westdahl PR, Rose MR, Concannon S. Paget�sdisease of the nipple. Alternative management in cases withoutor with minimal extent of underlying breast carcinoma. Can-cer1984;54:545�51.

11. Kay S. Paget�s disease of the nipple. Surg Gynecol Obstet1966;123:1010�4.

12. Rissanen P, Holsti P. Paget�s disease of the breast: the influ-ence of the presence or absence of an underlying palpable tu-mor on the prognosis and on the choice of treatment. Oncology1969;23:209�16.

13. Freund H, Maydovnik M, Laufer N, Durst A. Paget�s diseaseof the breast. J Surg Oncol 1977;9:93�8.

14. Paone JF, Baker RR. Pathogenesis and treatment of Paget�sdisease of the breast. Cancer 1981;48:825�9.

15. van Dongen JA, Voogd AC, Fentiman IS, et al. Long-termresults of a randomized trial comparing breast-conservingtherapy with mastectomy: European Organization for Re-search and Treatment of Cancer 10801 trial. J Natl Cancer Inst2000;92:1143�50.

16. Veronesi U, Cascinelli N, Mariani L, et al. Twenty-year fol-low-up of a randomized study comparing breast-conservingsurgery with radical mastectomy for early breast cancer. NEngl J Med 2002;347:1227�32.

17. Fisher B, Anderson S, Bryant J, et al. Twenty-year followup of a randomized trial comparing total mastectomy,lumpectomy, and lumpectomy plus irradiation for the treat-ment of invasive breast cancer. N Engl J Med 2002;347:1233�41.

18. Poggi MM, Danforth DN, Sciuto LC, et al. Eighteen-yearresults in the treatment of early breast carcinoma with mas-tectomy versus breast conservation therapy. The NationalCancer Institute Randomized Trial. Cancer 2003;98:697�702.

19. Stockdale AD, Brierley JD, White WF, Folkes A, Rostom AY.Radiotherapy for Paget�s disease of the nipple: a conservativealternative. Lancet 1989;2:664�6.

20. Fourquet A, Campana F, Vielh P, Schlienger P, JullienD, Vilcoq JR. Paget�s disease of the nipple withoutdetectable breast tumor: conservative management withradiation therapy. Int J Radiat Oncol Biol Phys 1987;13:1463�5.

21. Bijker N, Rutgers EJ, Duchateau L, et al. Breast-conservingtherapy for Paget disease of the nipple. A prospective Euro-pean Organization for Research and Treatment of Cancerstudy of 61 patients. Cancer 2001;91:472�7.

22. Marshall JK, Griffith KA, Haffty BG, et al. Conservativemanagement of Paget disease of the breast with radiotherapy.10- and 15-year results. Cancer 2003;97:2142�9.

23. Fleming ID, Cooper JS, Henson DE, et al. (eds.). AJCCCancer Staging Manual 5th ed. Philadelphia, Lippincott-

Raven, 1997;171�80.24. Kaplan E, Meier P. Nonparametric estimation from incom-

plete observations. J Am Stat Assoc 1958;53:457�81.25. LaTrenta LR, Menell JH, Morris EA, Abramson AF,

Dershaw DD, Liberman L. Breast lesions detected with MRimaging: utility and histopathologic importance of identifica-tion with US. Radiology 2003;227:856�61.

26. Friedman EP, Hall-Craggs MA, Mumtaz H, Schneidau A.Breast MR and the appearance of the normal and abnormalnipple. Clin Radiol 1997;52:854�61.

27. Ikeda DM, Helvie MA, Frank TS, Chapel KL, Andersson IT.Paget disease of the nipple: radiographic-pathologic correla-tion. Radiology 1993;189:89�94.

28. Kothari AS, Beechey-Newman N, Hamed H, et al. Pagetdisease of the nipple: a multifocal manifestation of higher-riskdisease. Cancer 2002;95:1�7.

29. Polgar C, Orosz Z, Kovacs T, Fodor J. Breast-conservingtherapy for Paget disease of the nipple: a prospective EuropeanOrganization for Research and Treatment of Cancer Study of61 patients. Cancer 2002;94:1904�5.

30. Veronesi U, Paganelli G, Viale G, et al. A randomized com-parison of sentinel-node biopsy with routine axillary dissectionin breast cancer. N Engl J Med 2003;349:546�53.

SURGICAL MANAGEMENT OF PAGET’S DISEASE OF THE BREAST 7

Ann. Surg. Oncol. Vol. 12, No. 5, 2005