lable at ScienceDirect

The Breast 23 (2014) 97e103

Contents lists avai

The Breast

journal homepage: www.elsevier .com/brst

Review

Extensive pure ductal carcinoma in situ of the breast: Identificationof predictors of associated infiltrating carcinoma and lymph nodemetastasis before immediate reconstructive surgeryq

E. Guillot a,*, C. Vaysse a, J. Goetgeluck a, M.C. Falcou b, B. Couturaud a, A. Fitoussi a,V. Fourchotte a, F. Laki a, C. Malhaire c, B. Sigal-Zafrani d, X. Sastre-Garau d, M.A. Bollet e,V. Mosseri b, F. Reyal a

aDepartment of Surgery, Institut Curie, 26 rue d’Ulm, 75005 Paris, FrancebDepartment of Biostatistics, Institut Curie, Paris, FrancecDepartment of Radiology, Institut Curie, Paris, FrancedDepartment of Tumors’ Biology, Institut Curie, Paris, FranceeDepartment of Radiotherapy, Institut Curie, Paris, France

a r t i c l e i n f o

Article history:Received 25 February 2013Received in revised form28 August 2013Accepted 1 December 2013

Keywords:Breast cancerDuctal carcinoma in situInvasive cancerAxillary lymph node metastasisPredictors

q This is an open-access article distributed undeCommons Attribution-NonCommercial-No Derivativemits non-commercial use, distribution, and reproductthe original author and source are credited.* Corresponding author. Tel.: þ33 0680706516.

E-mail addresses: Eugenie.guillot@curie.net, eug_g

0960-9776/$ e see front matter � 2014 The Authors.http://dx.doi.org/10.1016/j.breast.2013.12.002

a b s t r a c t

Aim: To identify predictors for infiltrating carcinoma and lymph node involvement, before immediatebreast reconstructive surgery, in patients with an initial diagnosis of extensive pure ductal carcinoma insitu of the breast (DCIS).Patients and methods: Between January 2000 and December 2009, 241 patients with pure extensive DCISin preoperative biopsy had underwent mastectomy. Axillary staging (sentinel node and/or axillarydissection) was performed in 92% (n ¼ 221) of patients. Patients with micro-invasive lesions at initialdiagnosis, recurrence or contralateral breast cancer were excluded.Results: Respectively 14% and 21% of patients had a final diagnosis of micro-invasive carcinoma (MIC) andinvasive ductal carcinoma (IDC). Univariate analysis showed that the following variables at diagnosis weresignificantly correlatedwith thepresence of eitherMIC or IDC in themastectomyspecimen: palpable tumor(p¼ 0.002), high grade DCIS (p¼ 0.002) and detection of an opacity bymammography (p¼ 0.019). Axillarylymph node (ALN) involvement was reported in 9% of patients. Univariate analysis suggested that a bodymass index higher than 25 (p ¼ 0.007), a palpable tumor (p ¼ 0.012) and the detection of an opacity bymammography (p ¼ 0.044) were associated with an increased rate of ALN involvement.Conclusion: Skin-sparing mastectomy and immediate breast reconstruction (IBRS) has become increas-ingly popular, especially for patients with extended DCIS of the breast. This study confirmed thatextended DCIS is associated with a substantial risk of finding MIC or IDC on the surgical specimen butalso ALN involvement. Adjuvant systemic treatment and/or radiotherapy could be indicated for some ofthese patients after the surgery. Patients should be informed of the rate of 1) complications associated toIBRS that will potentially delay the introduction of systemic or local therapy 2) complications associatedto radiotherapy after IBRS.

� 2014 The Authors. Published by Elsevier Ltd. All rights reserved.

Introduction

Pre-operative extensive pure ductal carcinoma in situ of thebreast (DCIS) diagnosed by the combination of mammography and

r the terms of the CreativeWorks License, which per-

ion in any medium, provided

2000@yahoo.fr (E. Guillot).

Published by Elsevier Ltd. All righ

preoperative biopsies are associated with an increased risk ofmicro-invasive (MIC) or invasive ductal carcinoma (IDC) identifiedon the post-operative specimen but also with an increased proba-bility of axillary lymph node (ALN) involvement. ALN metastaseswere found in 0%e14% of patients with a final diagnosis of pureDCIS [1e5]. Between 5 and 14% of patients pre-operatively diag-nosed with pure DCISs present aMIC component on the final report[6e8] and the prevalence of lymph node metastases in these pa-tients varies from 0 to 20% [9e22]. Many studies reported that theproportion of IDC on specimen resection from patients with a pre-operative diagnosis of DCIS, ranged from 6% to 41% [1,2,4,5,8e

ts reserved.

E. Guillot et al. / The Breast 23 (2014) 97e10398

10,23e26] (Table 5). As MIC and IDC are frequently identified inpatients with extensive DCIS who undergo mastectomy, recom-mendation have been published to simultaneously perform asentinel lymph node (SLN) procedure as lymphatic mapping isunreliable aftermastectomy [27] and therefore exposewomenwithinvasive carcinoma on their final report to the need for an axillarylymph node dissection with an increased risk of morbidity. Thepresence of IDC and the involvement of axillary lymph nodes arekey determinants of breast cancer adjuvant therapy. As most pa-tients with an indication of radical surgery for extended pure DCIScan undergo mastectomy with immediate reconstruction [28], it isof great value to be able to identify on pre-operative features pa-tients at risk for IDC and ALN metastasis. Several clinical settingshave been identified: 1) complications related to immediatereconstructive surgery could potentially delay the introduction ofsystemic therapy, 2) patients with triple negative, HER2 positive,high grade, lymph node positive tumors should not suffer a delay oftheir adjuvant systemic treatment, 3) prothesis-based breast re-constructions followed by radiotherapy are prone to post-operativecomplications and poor esthetic results. Identification of preoper-ative risk factors for IDC and ALN metastases would allow to adaptthe treatment scheme and to consider the likelihood of adjuvantsystemic treatment and/or radiotherapy before starting anyreconstructive surgery. At least the surgeon will be able to warnpatients with an increased risk of IDC or ALN involvement beforethe procedure. The aim of this study was to establish risk factors forinfiltrating carcinoma and lymph node involvement in patientswith an initial diagnosis of extensive pure DCIS, in order to proposethe most adequate multidisciplinary treatment.

Materials and methods

Study population

Between January 2000 and December 2009, 241 patients with apre-operative diagnosis of pure extensive DCIS had underwentmastectomy at Institut Curie, Paris, France. Patients with micro-invasive or invasive ductal carcinoma at initial diagnosis wereexcluded from our study. ALN exploration with SLN and/or axillarydissectionwere performed in 221 patients (92%). The SLN procedurewasperformedwithbluepatentonly. Axillary lymphnodedissectionwas simultaneously performed when the SLN was positive byimprint cytology or frozen section. We considered, as lymph nodepositivity, macrometastases (>2 mm), micrometastases (<2 mm)and positive tumors cells by immunohistochemistry. A secondoperation was indicated when either hematoxylin-eosin staining orimmunohistochemistry post-operatively revealed tumor cells in theSLN, including isolated tumor cells [29]. Clinical and imaging featuresincluding bodymass index (BMI),menopausal status, familial cancerhistory, clinical palpation of the tumor, opacity, microcalcifications,multifocality at mammography were recorded and type of preoper-ative biopsies (core biopsy, vacuum assisted biopsy or small lump-ectomy). Pathologic features analyzed on pre-operative biopsies(DCIS nuclear grade, presence/absence of necrosis) and on surgicalspecimens (presence/absence of invasion, tumor size, histologicaltype of invasive tumor, invasion, histological grade, multifocality,multicentricity, lympho-vascular invasion, estrogen receptor status(ER), progesterone receptor status (PR) and HER2 status were rec-ommended. Estrogen receptor status (ER), progesterone receptorstatus (PR) and HER2 status were only obtained on invasive compo-nent. The number of positive lymph nodes with macro-metastasis,micro-metastasis and isolated cells were recorded. Tumors wereconsidered to be positive for either ER or PR if at least 10% of theinvasive tumor cells in a section showed nuclear staining [30].

Statistical analyses

Number (percentage) described categorical variables; median(range) described continuousvariables. Comparisons betweengroupswere performed using a Chi-square or a Fisher Exact tests for cate-gorical variables and a Student’s t-test for continuous variables. Uni-variate analysis was performed to investigate association betweeninvasive carcinoma, axillary metastases lymph node and selectedvariables.P-valueswere considered significantwhen<0.05. Statisticalanalyses were performed using R software (www.cran.r-project).

Results

Patients and pre-operative characteristics

The patients’ characteristics are presented in Table 1. Median ageat diagnosis was 51 years (range 28e82 years). BMI was less than orequal to 25 in 186 (77.5%) women and 111 (46%) were menopausal.At diagnosis, the majority of patients had no palpable tumor (199,83%). The disease was discovered by mammography, with micro-calcifications in 233 cases (97%), opacities on mammography in34 patients (15%). Lesions were unifocal lesions in 153 patients(64%). The preoperative diagnosis was obtained mainly by macro-biopsy (65%). Most tumors were high grade DCIS (126, 53%) withnecrosis (154, 74%) on pre-operative biopsy.

Surgery and post-operative characteristics

All patients had a mastectomy. Immediate breast reconstruction(IBRS) was performed in 167 patients (69%) mainly by implant(81%) (Table 1).

ALN staging was performed in 221 patients (92%) with SLN bi-opsy in 109 patients (45%), ALND in 93 patients (39%) or a combi-nation of both methods in the remaining 19 patients (8%). Twentypatients (8%) had no ALN staging as ALN staging was not systematicwhen the patient had a lumpectomy following by a mastectomyafter for positive margins (Table 2).

Histological examination of the surgical specimen showed 156DCISs (65%), 35 DCISs with micro-invasion (14%) and 50 DCISs withinvasive carcinoma (21%).

DCISs component was of high grade in 128 (54%) tumors, withnecrosis in 151 (84%) tumors. Invasive carcinomas were mainlyestrogen receptor positive (34, 71%), grade 2 tumors (27, 54%) withlympho-vascular invasion (13, 30%). Eleven (22%) tumors were ofGrade 1 and 7 (14%) tumors of Grade 3. Nine (33%) out of the 27(54% of DCISs with invasive carcinoma) patients tested showed anover-expression of HER2 (Table 1).

Twentypatients (9%)hadametastatic axillary lymphnode.Outof128 SLN procedure, 11 showed metastases (1 macro-metastasis, 2micro-metastases and 8 presented positive tumors cells by immu-nohistochemistry). Out of 50 patients with an invasive carcinomacomponent, 17 (33%) showed ALN involvement (Tables 1 and 2).

Adjuvant treatment

Thirty-seven (15%) patients received an adjuvant treatment:

- 20 (54%) a single therapy (8 patients received radiotherapy, 11patients hormonal therapy and 1 patient chemotherapy)

- 17 (46%) a combination of radiotherapy and systemic treatment(chemotherapy and/or hormonal therapy).

- Chemotherapy was delivered in 13 (26%) of 50 patients withfinal diagnosis of invasive carcinoma. Chemotherapy was asso-ciated most of the time to radiotherapy and/or hormonotherapy

Table 1Pre and post-operative characteristics.

Preoperative features N (%)

Clinical featuresAge (years)Median (range) 51 (28e82)<40 years 27 (11.2%)40e50 years 87 (36.1%)>50 years 127 (52.7%)

Menopause 111 (46)Body index mass (n ¼ 240)Median (range) 23 (16e39)�25 186 (77.5%)>25 54 (22.5%)

Palpable tumor (n ¼ 240) 41 (17%)Mammogram opacity (n ¼ 227) 34 (15.0%)Microcalcifications (nn ¼ 241) 233 (97%)Multifocal tumors (n ¼ 238) 85 (36%)Multicentric tumors (n ¼ 238) 43 (18%)

Preoperative biopsyVacuum assisted biopsy and/or core biopsy 185 (77%)Vacuum assisted biopsy only 158 (65%)Small lumpectomy 56 (23%)

Pathological characteristicsGradeLow 17 (7%)Intermediate 94 (40%)High 126 (53%)

NecrosisYes 154 (74%)Moderate 48 (31%)Extended 106 (69%)

No 54 (16%)Unknown 33

Surgery N (%)

BreastSimple mastectomy 74 (31%)Skin sparing mammectomy þ IBRS 167 (69%)Implant 136 (81%)Implant þ latissimus dorsal flap 10 (6%)Latissimus dorsal flat 11 (7%)TRAM 9 (5%)Implant þ TRAM 1 (1%)

AxillaSentinel lymph node biopsy 109 (45%)Axillary lymph node dissection 93 (39%)ALN þ SLNB 19 (8%)None 20 (8%)

Post operative features N (%)

Tumor typeDCIS 156 (65%)DCIS þ microinvasion 35 (14%)DCIS þ invasion 50 (21%)

DCISSize median mm (range) 40 (0e95)

Grade (n ¼ 237)Low grade 26 (11%)Intermediate grade 83 (35%)High grade 128 (54%)

NecrosisYes 151 (84%)Moderate 30 (20%)Extended 121 (80%)

No 29 (16%)MultifocalityYes 84 (34.9%)No 157 (65.1%)

DCIS þ Invasion (n ¼ 50)Invasive size: median mm, (range) 6 (2e50)

Invasive hormonal statusHR positive 34 (71%)HR negative 14 (29%)

Invasive Grade Elston EllisI 11 (22%)II 27 (54%)

Table 1 (continued )

Post operative features N (%)

III 7 (14%)Unknown 5 (10%)

Lympho vascular invasion 13 (30%)Invasive HER2 positive (n ¼ 27) 9 (33%)Axillary metastasis (n ¼ 221) 20 (9%)Sentinel lymph nodes (n ¼ 128)Negative 117 (91%)Metastatic 11 (9%)Macrometastatic 1 (1%)Micrometastatic 2 (2%)

Positive tumors cells by immune-histochemistry 8 (6%)

IBRS ¼ immediate breast reconstruction surgery, TRAM ¼ Transverse RectusAbdominis Myocutaneous, ALN: Axillary Lymph Node Dissection. SLNB: SentinelLymph Node biopsy.

E. Guillot et al. / The Breast 23 (2014) 97e103 99

- 3 patients (2%) with final diagnosis of DCIS and 2 patients withfinal diagnosis of DCIS þ MI (5%) received radiotherapy becauseof close margins.

- 18 patients with IBRS (10%) a combination of radiotherapy andsystemic treatment (chemotherapy and/or hormonal therapy)and 2 (1%), only radiotherapy.

Prediction of IDC and ALN involvement

Univariate analysis showed that palpable tumor (p ¼ 0.002),opacity on mammography (p ¼ 0.019), pre-operative high gradeDCIS diagnosis (p ¼ 0.002) were significant predictors of invasivecarcinoma associated with DCIS at final diagnosis (Table 3).

Moreover, the univariate analysis showed that BMI > 25(p ¼ 0.007), palpable tumor (p ¼ 0.012), opacity on mammography(p ¼ 0.044) were statistically significant predictors of ALN metas-tases (Table 4).

Risk factors of recurrence

Univariate analysis showed that palpable tumor (p ¼ 0.06),opacity on mammography (p ¼ 0.01) and microcalcifications onmammography (p ¼ 0.02) were significant risk factors of breastrecurrences.

Univariate analysis showed that palpable tumor (p ¼ 0.08) andBMI>25 (p ¼ 0.03) were significant risk factors of breast and axil-lary recurrences.

Outcomes

Median follow-up was 30 months [26e36]. Five years survivalwas 98%.

Table 2Axillary lymph node status distribution by histological tumor type.

ALN only SLNB � ALN

Histologicaltype

N PositiveN (%)

N Positive

N (%) IHC/Micro/macroM

DCIS 53 0 91 1 (1.1%) IHC (no ALN)DCIS þ MIC 15 0 21 3 (14.3%) 3 IHC (no ALN in one case)DCIS þ IC 37 9 (27%) 16 7 (43.8%) 1 Macrometa

2 Micrometa 4IHCTotal 105 9 128 11

ALN: Axillary Lymph Node Dissection. SLNB: Sentinel Lymph Node biopsy. DCIS:ductal carcinoma in situ. MIC: microinvasive carcinoma. IC: invasive carcinoma.IHC: positive tumors cells by immune-histochemistry. Macrometa: macro-metastatic.Micrometa: micrometastatic.

Table 3Risk factors for invasive carcinoma.

Risk factors for invasive carcinoma identified in the surgical specimen

Surgical specimen DCIS DCIS þ MIC DCIS þ IC Total p

N 156 (65%) 35 (14%) 50 (21%)Clinical factorsPalpable tumor 18 (12%) 13 (37%) 10 (18%) 41 (17%) 0.002BMI > 25 36 (23%) 7 (21%) 11 (22%) 54 (22,5%) 0.957Age <40 years 13 (8%) 8 (23%) 6 (12%) 27 (11%) 0.151

40e50 60 (39%) 9 (26%) 18 (36%) 87 (36%)>50 years 83 (53%) 18 (51%) 26 (52%) 127 (53%)

Radiological factorsOpacity 16 (11%) 10 (29%) 8 (18%) 34 (15%) 0.019Multifocality 56 (36%) 14 (40%) 15 (31%) 85 (36%) 0.3526Pre operative pathological factorsGrade Low grade 14 (9%) 0 3 (6%) 17 (7%) 0.002

Intermediate grade 72 (46%) 6 (18%) 16 (33%) 94 (40%)High grade 70 (45%) 27 (82%) 29 (61%) 126 (53%)

Necrosis No necrosis 40 (30%) 5 (17%) 9 (20%) 54 (26%) 0.3765Moderate necrosis 31 (23%) 6 (20%) 11 (24%) 48 (23%)Extended necrosis 62 (47%) 19 (63%) 25 (56%) 106 (51%)

BMI: body mass index, DCIS: ductal carcinoma in situ, MIC: microinvasive carcinoma, IC: invasive carcinoma.Bold values signify the body mass index.

E. Guillot et al. / The Breast 23 (2014) 97e103100

Seven patients had a local recurrence (3%) (Table 6): five breastrecurrences and two axillary recurrences. Five had an immediatebreast reconstruction (p ¼ 0.551).

Three patients (1%) developed metastasis. One had macro-metastatic axillary invasion at the first surgery. Two patients aredead because of the breast disease.

Patients with immediate breast reconstruction surgery (IBRS)compared with patient without reconstruction

Histological examination of the surgical specimen showed 21%DCISs with invasive carcinoma after classic mastectomy and 20%after mastectomy with IBRS. Axillary lymph node invasion wasmore often observed after standard mastectomy (9 patients, 15%)than after IBRS (11 patients, 7%), (p ¼ 0.059).

Univariate analysis showed that patients with IBRS wereyounger (p < 0.00001), thinner (p ¼ 0.0052) and mostly premen-opausal women (p < 0.0001).

Univariate analysis showed that no palpable tumor (p ¼ 0.01),no opacity on mammography (p ¼ 0.07) and a lower preoperative

Table 4Predictors of axillary lymph node metastasis.

Predictive factors of axillary lymph node metastasis identified in the surgicalspecimen (n ¼ 221)

Surgical specimen Lymph nodenegative

Lymph nodepositive

Total p

N 201 20 221Clinical factorsBMI > 25 41 (20.5%) 10 (50%) 51 (23%) 0.007Palpable tumor 30 (15%) 8 (40%) 38 (17%) 0.012Age <40 years 22 (11%) 4 (20%) 26 (12%) 0.385

40e50 years 73 (36%) 8 (40%) 81 (37%)>50 years 106 (53%) 8 (40%) 114 (52%)

Radiological factorsOpacity 25 (13%) 6 (32%) 31 (15%) 0.044Multifocality 76 (38%) 5 (26%) 81 (19%) 0.570Pathological factors (surgical specimens)Grade Low Grade 14 (7%) 1 (5%) 0.928

Intermediate Grade 74 (37%) 8 (40%)High Grade 110 (56%) 11 (55%)

Necrosis No necrosis 44 (26%) 5 (28%) 0.825Few necrosis 39 (23%) 5 (28%)Extended necrosis 89 (51%) 8 (44%)

DCIS grade (p ¼ 0.03) were significant preoperative characteristicsin women with IBRS.

There was less axillary lymph node invasion (p ¼ 0.0598) inwomen with IBRS.

There was no significant difference in terms of local recurrenceafter 5 years after IBRS (4.89%) and after standard mastectomy(1.97%) (p ¼ 0.561).

Discussion

In this study, 21% of patients with an initial diagnosis of pure DCIShad invasive carcinoma and 15% a micro-invasive carcinoma. Theseresults are consistent with the literature [5,25,31]. Our univariateanalysis revealed that presence of clinically palpablemass (p¼ 0.002),opacityonmammography (p¼ 0.019) andhighgradeDCIS (p¼0.002)were significant predictors of concurrent invasive carcinoma.

Radiological features and infiltrating disease

Opacity on mammography is a feature strongly associated withinvasive disease. Veronesi et al. studied 1258 patients with clini-cally occult invasive carcinoma and concluded that opacity,distortion or thickening were present in 24.6% of the cases [32].Subclinical breast cancers revealed by spiculated opacity corre-sponded predominantly to an infiltrating process, in contrast tobreast cancers revealed by microcalcifications, which were mainlyassociated with DCIS [33]. The mammographic tumor size isfrequently found to be a risk factor for invasive disease [5,27].Indeed, Yen et al., in a study of 398 patients, observed that, given aninitial diagnosis of DCIS, the presence of a tumor of mammographicsize larger than 4 cm (p ¼ 0.001) was an independent predictivefactor for invasive cancer on final pathology [25]. In another studyof 624 patients with a preoperative diagnosis of DCIS or MIC, a sizelarger than 5 cm (by clinical examination or imaging, p ¼ 0.04) alsocame up as an independent factor in a multivariate analysis [27].Similar results were reported by other authors, such as Silversteinor Lagios [34,35]. In a study of 53 patients, Lagios et al. concludedthat extensive DCIS was often associated with the presence ofoccult invasion and multi-centricity [35]. Because of missing in-formation about tumor focality on mammography, this parametercould not be integrated into our analysis. We used “clinical mass”,which was found to be a significant risk factor for invasive disease

Table 5Overview of studies on micro-invasion, invasion and axillary lymph node metastasis in patients with preoperative diagnosis of pure DCIS.

Studies on micro-invasion, invasion and axillary lymph node metastasis in patients with preoperative diagnosis of pure DCIS

First author Year Patients with purepre-operative DCIS

Surgical specimen final pathology report

DCIS DCIS plus MIC DCIS plus IDC ALNM

Yen et al. [25] 2005 398 318 (80%) 14 (3.4%) 66 (16.6%) 14/141 (10%)Maffuz et al. [9] 2006 24 14 (58%) 7 (29%) 3 (12.5%) 3/24 (12.5%)Dillon et al. [8] 2006 93 49 (52.7%) 13 (14%) 31 (33%) NAGoyal et al. [2] 2006 587 NA NA 220 (38%) 35/269 (13%)Rutstein et al. [23] 2007 254 226 (89%) 6 (2%) 15 (6%) NAMeijnen et al. [10] 2007 172 127 (73.8%) 10 (11.6%) 35 (41.4%) 11/91 (11.3%)Moran et al. [1] 2007 62 42 (67.7%) NA 20 (32%) 3/35 (8.6%)Dominguez et al. [26] 2008 179 159 (88.8%) 8 (4.5%) 12 (6.7%) 20/177 (11.2%)Schneider et al. [4] 2010 110 95 (86.4%) NA 15 (13.6%) 3/34 (9%)Han et al. [5] 2011 199 126 (63.3%) 11 (5.5%) 41 (20.6%) 18/131 (13.7%)Our study 2011 241 156 (65%) 35 (15%) 50 (21%) 20/221 (9%)Total 2319 1312/1732 (75.7%) 104/1560 (6.6%) 508/2319 (22%) 127/1123 (11.3%)

DCIS: ductal carcinoma in situ, MIC: micro-invasive carcinoma, IDC: invasive ductal carcinoma, ALNM: axillary lymph node metastasis.

E. Guillot et al. / The Breast 23 (2014) 97e103 101

at final pathology review. Similar results were noted in the recentstudy of Han [5]. It is noteworthy that a biopsy of a largemammographic lesion is likely to underestimate the risk of invasivedisease in contrast to a focal lesion. Furthermore, size correlateswith the type of surgery carried out. Patients with smallermammographic DCIS sizes are more prone to undergo lumpectomyand show no evidence of invasive disease on final pathology re-view; conversely, patients with larger tumors on mammographyare more likely to require total mastectomy to obtain free marginsand to have invasive cancer on their final pathology report [25]. Dueto the first treatment by lumpectomy in 56 patients (23%), DCIS inmastectomy specimen was 0e95 mm.

Pathological features and infiltrating disease

High grade DCIS at preoperative biopsy was also found to be arisk factor for invasive or micro-invasive disease in a number ofstudies [20,25,36,37].

ALN involvement prediction

Occult invasive carcinoma could result in metastatic ALNinvolvement. Therefore, it was important to identify patients with

Table 6Characteristics of the seven recurrences.

Mastectomyþ IBRS

Finalhistology

Lymph nodestatus

Recurrencehistology

Treatment

1 NO(SBRS)

DCIS Unknown IC SurgeryRHþ RadiotherapyHER2� Hormonotherapy

2 YES(TRAM)

DCIS þ IC Negative IC SurgeryRHþ RadiotherapyHER2� Hormonotherapy

3 YES(implant)

DCIS þ MIC Negative MIC Surgery

4 YES(implant)

DCIS þ IC Negative IC SurgeryRHþ RadiotherapyHER2� Hormonotherapy

5 NO DCIS þ IC Macrometastatic

IC ChemotherapyRHþHER2�

6 YES(implant)

DCIS þ MIC Negative IC SurgeryRHþ RadiotherapyHER2� Hormonotherapy

7 YES(Latissimusdorsal Flat)

DCIS Negative IC SurgeryRHþ RadiotherapyHER2� Hormonotherapy

IBRS: immediate breast reconstruction, SBRS: secondary breast reconstruction.

risk factors associated with ALN metastases in order to adapt thetreatment scheme. In our study, 20 (9%) patients had ALN metas-tases out of 221 (92%) patients who underwent ALN staging. 6% ofpositive SLN had been detected by immunohistochemistry (iþ). Outof 20 patients with ALN metastases, 16 (80%) and 3 (15%) hadinvasive cancer and micro-invasive cancer respectively, on finalpathology review. Only 1 patient presented a DCIS only disease.These results are similar to the existing literature [5,24,26,38].Combination of both methods (ALND and SLND) were done in 19patients because SLND was not done alone (learning curve). Uni-variate analysis showed that a BMI above 25 (p ¼ 0.007), a palpabletumor (p ¼ 0.012), an opacity on mammography (p ¼ 0.044) wereassociated with a risk of ALN involvement. In Yi’s 2008 study, only aDCIS size larger than 5 cm was found to be an independent(p ¼ 0.002) risk factor in a multivariate analysis [27]. Ko and al.reported that out of 293 patients with breast MIC, 7.5% showed anALN involvement. Lymphatic invasion (p < 0.001) and positive ERstatus (p ¼ 0.03) were independent risk factors for axillary me-tastases [22].

In Wilkie’s 2005 study, out of 613 patients with an initial diag-nosis of DCIS, 55 (9%) tumors were reclassified as invasive carci-nomas. Of the 559 patients with definitive diagnosis of DCIS, 27(5%) had positive SLN, and of them 70% could only be detected byimmunohistochemistry staining. Out of 66 patients who had adefinitive diagnosis of invasive carcinoma, 15 (23%) had SLNmetastasis. Microinvasion, high-grade DCIS at the time of biopsyand mass on mammography were risk factors for invasive carci-noma. The authors hence recommended SLN exploration for pa-tients undergoing mastectomy for DCIS or having DCIS associatedwith these risk factors [20]. Yi detected SLN metastases in 9 (2%) of475 patients with a final diagnosis of DCIS or MIC. The multivariateanalysis showed that a DCIS size larger than 5 cm (p¼ 0.02) was theonly independent risk factor for positive SLN [27]. In another study,Yen observed that of 398 patients operated on, 141 (35%) had un-dergone concomitant SLN biopsy. Out of 14 (10%) patients with apositive SLN, only 11 (79%) had invasive carcinoma on final pa-thology review. A multivariate analysis revealed that the only in-dependent risk factor for positive SLNB was the presence of apalpable tumor (p ¼ 0.042) [25].

In a recent study, Han and al. noted that in 131 patients who hadundergone a SLN biopsy, 18 (13.7%) had metastases. Presence andsize of invasive carcinoma and extent of DCIS on final pathologyspecimen were more frequently associated with positive SLNcompared to negative SLN [5].

The risk factors for SLN metastases such as palpable mass andopacity on mammography, observed in our study, were alreadyreported in the existing literature, apart from obesity. The

E. Guillot et al. / The Breast 23 (2014) 97e103102

expression of ER and that of insulin-like growth factor-I receptorare positively correlated in breast cancer specimens. Abdominalobesity in women is associated with higher concentrations both offree estradiol and free insulin-like growth factor-I. Activation oftheir respective receptors may induce synergistic stimulation ofmammary carcinogenesis, notably in women approaching meno-pause [39]. Indeed, we reported in our study, a median age of 51years and the majority of invasive tumors were hormono-sensitive.80% of positive ALN had invasive carcinomas. We could thereforediscuss in obese patients a secondary mammary reconstruction,with the added benefit of avoiding surgical complications generallyassociated with obesity [40,41].

Currently, SLN biopsy procedure, an alternative to conventionalALND, is routinely used in mastectomy for extensive DCIS[4,24,26,42,43]. During conservative treatment, indications of SLNare still controversial. Yet, it is important to estimate the risk factorsfor ALN metastasis in order to adapt management. The currenttrend is to propose mastectomy with immediate reconstructiontreatment for pure DCIS. The postsurgical discovery of unexpectedinvasive carcinoma may prompt additional axillary surgery andpossible adjuvant treatment such as breast irradiation. So, the im-mediate reconstruction may be inadequate. For that reason, tosingle out patients with risk factors for ALNmetastasis is necessary.

Impact of the IBRS

Into our study, patients with IBRS had less palpable tumor(p ¼ 0.01), less opacity on mammography (p ¼ 0.07) and a lowerpreoperative DCIS grade (p ¼ 0.03).

There was less axillary lymph node invasion (p ¼ 0.0598) inwomen with IBRS.

There was no significant difference in terms of local recurrenceafter 5 years between the two groups: 1.97% after classic mastec-tomy and 4.89% after IBRS (p ¼ 0.561). Our results are similar withthe literature that describes 1e3% local recurrence after 5 years[44]. In Carlson’s 2007 study [44], 223 patients had skin sparingmastectomy and IBRS for DCIS. After 82 months median follow-up,there is 3.3% of local recurrence and mostly invasive carcinoma(85%). Into our study, we described 6 invasive carcinoma re-currences and 1 microinvasive carcinoma recurrence, which thediagnosis wasmostly done by clinical examination.We recommendclosed clinical examination and mammography even after IBRS.

Into our study, patients with IBRS had better initial prognosticfactors except the youthfulness but the rate of final invasive carci-noma is the same. However, we can recommend IBRS whenextended DCIS is diagnosed because this surgical technic is safewithout excess oncologic risk as the literature confirmed [45,46].

Risk factors of recurrence

After 30 months median follow-up, 7 patients (3%) had a localrecurrence. We founded same results in literature [44,47]. Univar-iate analysis showed that palpable tumor (p ¼ 0.06), opacity onmammography (p ¼ 0.01) and microcalcifications on mammog-raphy (p ¼ 0.02) were significant risk factors of breast recurrences.Univariate analysis showed that palpable tumor (p ¼ 0.08) andBMI>25 (p ¼ 0.03) were significant risk factors of breast and axil-lary recurrences. Five years survival was 98% as the literature’s data[48]. Into Carlson’s 2007 study [44], concerning patients with IBRS,multivariate analysis didn’t found significant risk factors of local orloco-regional recurrence (age< 50 y, tumor size> 40mm, necrosis,macrobiopsy, skin sparing mastectomy).

In conclusion, we have confirmed that the incidence of ALNmetastases in “pre-surgical pure extensive DCIS” are correlatedwith the following factors: BMI > 25 kg/m2, palpable tumor and

opacity on mammography. So we recommend that, patients withsuch risk factors and who have an indication of mastectomy forextensive DCIS, would undergo an axillary exploration with SLNdissection. Furthermore, physicians should discuss SLN biopsyprocedure with patients with palpable tumor, opacity onmammography and high grade DCIS at preoperative biopsy even ifa conservative treatment is recommended due to the increasingrisk of occult invasive disease. Identification of these risk factors ofIDC and ALN metastases, would allow one to adapt the treatmentscheme and, in some cases, the possibility to consider adjuvantsystemic treatment and/or radiotherapy before starting anyreconstructive surgery. Skin-sparing mastectomy and IBRS hasbecome increasingly popular, especially for patients with extendedDCIS of the breast. Patients should be informed of the rate of 1)complications associated to IBRS that will potentially delay theintroduction of systemic or local therapy 2) complications asso-ciated to radiotherapy after IBRS. However, loco-regional recur-rence after standard mastectomy or skin sparing mastectomy plusIBRS for extended pure DCIS remained very low after 10 years offollow-up.

Conflict of interest statement

None declared.

References

[1] Moran CJ, Kell MR, Flanagan FL, Kennedy M, Gorey TF, Kerin MJ. Role ofsentinel lymph node biopsy in high-risk ductal carcinoma in situ patients. AmJ Surg 2007;194(2):172e5.

[2] Goyal A, Douglas-Jones A, Monypenny I, Sweetland H, Stevens G, Mansel RE. Isthere a role of sentinel lymph node biopsy in ductal carcinoma in situ?:analysis of 587 cases. Breast Cancer Res Treat 2006;98(3):311e4.

[3] Sakr R, Barranger E, Antoine M, Prugnolle H, Darai E, Uzan S. Ductal carcinomain situ: value of sentinel lymph node biopsy. J Surg Oncol 2006;94(5):426e30.

[4] Schneider C, Trocha S, McKinley B, Shaw J, Bielby S, Blackhurst D, et al. The useof sentinel lymph node biopsy in ductal carcinoma in situ. Am Surg2010;76(9):943e6.

[5] Han JS, Molberg KH, Sarode V. Predictors of invasion and axillary lymph nodemetastasis in patients with a core biopsy diagnosis of ductal carcinoma in situ:an analysis of 255 cases. Breast J 2011;17(3):223e9.

[6] Smigal C, Jemal A, Ward E, Cokkinides V, Smith R, Howe HL, et al. Trends inbreast cancer by race and ethnicity: update 2006. CA Cancer J Clin 2006;56(3):168e83.

[7] Adamovich TL, Simmons RM. Ductal carcinoma in situ with microinvasion. AmJ Surg 2003;186(2):112e6.

[8] Dillon MF, McDermott EW, Quinn CM, O’Doherty A, O’Higgins N, Hill AD.Predictors of invasive disease in breast cancer when core biopsy demonstratesDCIS only. J Surg Oncol 2006;93(7):559e63.

[9] Maffuz A, Barroso-Bravo S, Najera I, Zarco G, Alvarado-Cabrero I, Rodriguez-Cuevas SA. Tumor size as predictor of microinvasion, invasion, and axillarymetastasis in ductal carcinoma in situ. J Exp Clin Cancer Res 2006;25(2):223e7.

[10] Meijnen P, Oldenburg HS, Loo CE, Nieweg OE, Peterse JL, Rutgers EJ. Risk ofinvasion and axillary lymph node metastasis in ductal carcinoma in situdiagnosed by core-needle biopsy. Br J Surg 2007;94(8):952e6.

[11] Cox CE, Nguyen K, Gray RJ, Salud C, Ku NN, Dupont E, et al. Importance oflymphatic mapping in ductal carcinoma in situ (DCIS): why map DCIS? AmSurg 2001;67(6):513e9. discussion 519e21.

[12] De Mascarel I, MacGrogan G, Mathoulin-Pelissier S, Soubeyran I, Picot V,Coindre JM. Breast ductal carcinoma in situ with microinvasion: a definitionsupported by a long-term study of 1248 serially sectioned ductal carcinomas.Cancer 2002;94(8):2134e42.

[13] Intra M, Zurrida S, Maffini F, Sonzogni A, Trifirò G, Gennari R, et al. Sentinellymph node metastasis in microinvasive breast cancer. Ann Surg Oncol2003;10(10):1160e5.

[14] Yang M, Moriya T, Oguma M, De La Cruz C, Endoh M, Ishida T, et al. Micro-invasive ductal carcinoma (T1mic) of the breast. The clinicopathologicalprofile and immunohistochemical features of 28 cases. Pathol Int 2003;53(7):422e8.

[15] Gray RJ, Mulheron B, Pockaj BA, Degnim A, Smith SL. The optimal manage-ment of the axillae of patients with microinvasive breast cancer in thesentinel lymph node era. Am J Surg 2007;194(6):845e8. discussion 848e9.

[16] Fortunato L, Santoni M, Drago S, Gucciardo G, Farina M, Cesarini C, et al.Sentinel lymph node biopsy in women with pT1a or “microinvasive” breastcancer. Breast 2008;17(4):395e400.

E. Guillot et al. / The Breast 23 (2014) 97e103 103

[17] Sakr R, Bezu C, Raoust I, Antoine M, Ettore F, Darcourt J, et al. The sentinellymph node procedure for patients with preoperative diagnosis of ductalcarcinoma in situ: risk factors for unsuspected invasive disease and for met-astatic sentinel lymph nodes. Int J Clin Pract 2008;62(11):1730e5.

[18] Guth AA, Mercado C, Roses DF, Darvishian F, Singh B, Cangiarella JF. Micro-invasive breast cancer and the role of sentinel node biopsy: an institutionalexperience and review of the literature. Breast J 2008;14(4):335e9.

[19] Zavagno G, Belardinelli V, Marconato R, Carcoforo P, Franchini Z, Scalco G,et al. Sentinel lymph node metastasis from mammary ductal carcinoma in situwith microinvasion. Breast 2007;16(2):146e51.

[20] Wilkie C, White L, Dupont E, Cantor A, Cox CE. An update of sentinel lymphnode mapping in patients with ductal carcinoma in situ. Am J Surg2005;190(4):563e6.

[21] Katz A, Gage I, Evans S, Shaffer M, Fleury T, Smith FP, et al. Sentinel lymphnode positivity of patients with ductal carcinoma in situ or microinvasivebreast cancer. Am J Surg 2006;191(6):761e6.

[22] Ko BS, Lim WS, Kim HJ, Yu JH, Lee JW, Kwan SB, et al. Risk factor for axillarylymph node metastases in microinvasive breast cancer. Ann Surg Oncol2012;19(1):212e6.

[23] Rutstein LA, Johnson RR, Poller WR, Dabbs D, Groblewski J, Rakitt T, et al.Predictors of residual invasive disease after core needle biopsy diagnosis ofductal carcinoma in situ. Breast J 2007;13(3):251e7.

[24] Van Deurzen CH, Hobbelink MG, van Hillegersberg R, van Diest PJ. Is there anindication for sentinel node biopsy in patients with ductal carcinoma in situ ofthe breast? A review. Eur J Cancer 2007;43(6):993e1001.

[25] Yen TW, Hunt KK, Ross MI, Mirza NQ, Babiera GV, Meric-Bernstam F, et al.Predictors of invasive breast cancer in patients with an initial diagnosis ofductal carcinoma in situ: a guide to selective use of sentinel lymph node bi-opsy in management of ductal carcinoma in situ. J Am Coll Surg 2005;200(4):516e26.

[26] Dominguez FJ, Golshan M, Black DM, Hughes KS, Gadd MA, Christian R,et al. Sentinel node biopsy is important in mastectomy for ductal carci-noma in situ. Ann Surg Oncol 2008;15(1):268e73.

[27] Yi M, Krishnamurthy S, Kuerer HM, Meric-Bernstam F, Bedrosian I, Ross MI,et al. Role of primary tumor characteristics in predicting positive sentinellymph nodes in patients with ductal carcinoma in situ or microinvasive breastcancer. Am J Surg 2008;196(1):81e7.

[28] Greenberg CC, Lipsitz SR, Hughes ME, Edge SB, Theriault R, Wilson JL, et al.Institutional variation in the surgical treatment of breast cancer: a study of theNCCN. Ann Surg 2011;254(2):339e45.

[29] Fréneaux P, Nos C, Vincent-Salomon A, Genin P, Sigal-Zafrani B, Al Ghuzlan A,et al. Histological detection of minimal metastatic involvement in axillarysentinel nodes: a rational basis for a sensitive methodology usable in dailypractice. Mod Pathol 2002;15(6):641e6.

[30] Balaton AL, Coindre JM, Collin F, Ettore F, Fiche M, Jacquemier J, et al. Rec-ommendations for the immunohistochemical evaluation of hormone re-ceptors on paraffin sections of breast cancer. Study Group on HormoneReceptors using Immunohistochemistry FNCLCC/AFAQAP. National Federationof Centres to Combat Cancer/French Association for Quality Assurance inPathology. Ann Pathol 1996;16:144e8.

[31] Verkooijen HM, Peeters PH, Buskens E, Koot VC, Borel Rinkes IH, Mali WP,et al. Diagnostic accuracy of large-core needle biopsy for nonpalpable breastdisease: a meta-analysis. Br J Cancer 2000;82(5):1017e21.

[32] Veronesi U, Luini A, Botteri E, Zurrida S, Monti S, Galimberti V, et al. Non-palpable breast carcinomas: long-term evaluation of 1,258 cases. Oncologist2010;15(12):1248e52.

[33] Le Bouedec G, Feillel V, de Latour M, Pomel C, Dauplat J. Breast cancer detectedby a stellar opacity. Histologic characteristics: 155 cases. Gynecol Obstet Fertil2000;28(11):798e805.

[34] Silverstein MJ, Waisman JR, Gamagami P, Gierson ED, Colburn WJ, Rosser RJ,et al. Intraductal carcinoma of the breast (208 cases). Clinical factors influ-encing treatment choice. Cancer 1990;66(1):102e8.

[35] Lagios MD, Margolin FR, Westdahl PR, Rose MR. Mammographically detectedduct carcinoma in situ. Frequency of local recurrence following tylectomy andprognostic effect of nuclear grade on local recurrence. Cancer 1989;63(4):618e24.

[36] Hoorntje LE, Schipper ME, Peeters PH, Bellot F, Storm RK, Borel Rinkes IH. Thefinding of invasive cancer after a preoperative diagnosis of ductal carcinoma-in-situ: causes of ductal carcinoma-in-situ underestimates with stereotactic14-gauge needle biopsy. Ann Surg Oncol 2003;10(7):748e53.

[37] Renshaw AA. Predicting invasion in the excision specimen from breast coreneedle biopsy specimens with only ductal carcinoma in situ. Arch Pathol LabMed 2002;126(1):39e41.

[38] Mabry H, Giuliano AE, Silverstein MJ. What is the value of axillary dissectionor sentinel node biopsy in patients with ductal carcinoma in situ? Am J Surg2006;192(4):455e7.

[39] Stoll BA. Oestrogen/insulin-like growth factor-I receptor interaction in earlybreast cancer: clinical implications. Ann Oncol 2002;13(2):191e6.

[40] McCarthy CM, Mehrara BJ, Riedel E, Davidge K, Hinson A, Disa JJ, et al. Pre-dicting complications following expander/implant breast reconstruction: anoutcomes analysis based on preoperative clinical risk. Plast Reconstr Surg2008;121(6):1886e92.

[41] Spear SL, Ducic I, Cuoco F, Taylor N. Effect of obesity on flap and donor-sitecomplications in pedicled TRAM flap breast reconstruction. Plast ReconstrSurg 2007;119(3):788e95.

[42] Tang Y, Xu F, Tao K, Qian N, Toi M. Clinical applications of sentinel lymph nodebiopsy in ductal carcinoma in situ of the breast: a dilemma. Tohoku J Exp Med2011;224(1):1e5.

[43] Collado MV, Ruiz-Tovar J, García-Villanueva A, Rojo R, Latorre L, Rioja ME,et al. Sentinel lymph node biopsy in selected cases of ductal carcinoma in situ.Clin Transl Oncol 2010;12(7):499e502.

[44] Carlson GW, Page A, Johnson E, Nicholson K, Styblo TM, Wood WC, et al. Localrecurrence of ductal carcinoma in situ after skin-sparing mastectomy. J AmColl Surg 2007;204(5):1074e8.

[45] Patani N, Mokbel K. Oncologic and aesthetic considerations of skin-sparingmastectomy. Breast Cancer Res Treat 2008 Oct;111(3):391e403.

[46] Slavin SA, Schnitt SJ, Duda RB, Houlihan MJ, Koufman CN, Morris DJ, et al.Skin-sparing mastectomy and immediate reconstruction: oncologic risks andaesthetic results in patients with early-stage breast cancer. Plast ReconstrSurg 1998;102(1):49e62.

[47] Boyages J, Delaney G, Taylor R. Predictors of local recurrence after treatmentof ductal carcinoma in situ: a meta-analysis. Cancer 1999;85(3):616e28.

[48] Ernster VL, Barclay J, Kerlikowske K, Wilkie H, Ballard-Barbash R. Mortalityamong women with ductal carcinoma in situ of the breast in the population-based surveillance, epidemiology and end results program. Arch Intern Med2000;160(7):953e8.