Ann Surg Oncol (2015) 22:11181127

DOI 10.1245/s10434-014-4122-7

ORIGINAL ARTICLE BREAST ONCOLOGY

Impact of Multifocal or Multicentric Disease on Surgery

and Locoregional, Distant and Overall Survival of 6,134 Breast
Cancer Patients Treated With Neoadjuvant Chemotherapy
Beyhan Ataseven, MD1, Bianca Lederer, MD2, Jens U. Blohmer, MD, PhD3, Carsten Denkert, MD, PhD4,
Bernd Gerber, MD, PhD5, Jorg Heil, MD6, Thorsten Kuhn, MD, PhD7, Sherko Kummel, MD8, Mahdi Rezai, MD9,
Sibylle Loibl, MD, PhD2, and Gunter von Minckwitz, MD, PhD2
Department of Gynecology and Gynecologic Oncology, Kliniken Essen-Mitte, Essen, Germany; 2Headquarter, German
Breast Group, Neu-Isenburg, Germany; 3Department of Gynecology and Obstetrics, St. Gertrauden Hospital, Berlin,
Germany; 4Institute for Pathology, Charite, Berlin, Germany; 5Department of Gynecology and Obstetrics, University
Hospital, Rostock, Germany; 6Department for Gynecology and Obstetrics, Interdisciplinary Breast Center, Esslingen,
Germany; 7Department of Gynecology and Obstetrics, University Womens Clinic, Heidelberg, Germany; 8Breast Center,
Luisenkrankenhaus, Dusseldorf, Germany; 9Department of Senology, Kliniken Essen-Mitte, Essen, Germany
1

ABSTRACT
Background. The impact of tumor focality on type of
surgery, local recurrence rate, and survival after neoadjuvant chemotherapy (NACT) for breast cancer is not fully
understood. This study aimed to compare local recurrencefree survival (LRFS), disease-free survival (DFS), and
overall survival (OS) according to focality stratified by
type of surgery and pathologic complete response (pCR),
with a focus on breast conservation.
Methods. Participants (n = 6,134) in the GeparTrio, GeparQuattro, and GeparQuinto trials with operable or locally
advanced tumors receiving NACT were classified as having unifocal (1 lesion), multifocal (C2 lesions in 1
quadrant), or multicentric (C1 lesion in C2 quadrants)
disease. The study investigated LRFS, DFS, and OS
according to focality stratified by type of surgery and
pathologic complete response.
Results. The patients were classified as having unifocal
(n = 4,733, 77.1 %), multifocal (n = 820, 13.4 %), or
multicentric (n = 581, 9.5 %) tumors. The respective pCR
Electronic supplementary material The online version of this
article (doi:10.1245/s10434-014-4122-7) contains supplementary
material, which is available to authorized users.
Society of Surgical Oncology 2014
First Received: 11 June 2014;
Published Online: 9 October 2014
B. Ataseven, MD
e-mail: ataseven@gmx.net

rates were 19.4, 16.5, and 14.4 %. Breast conservation was

performed for 71.6, 58.5, and 30 % of these patients,
respectively (P \ 0.001). The LRFS rate was 92.9 % for
the unifocal, 95.1 % for the multifocal, and 90.4 % for the
multicentric tumors (P = 0.002). The patients with multicentric tumors but not the patients with multifocal tumors
had worse DFS (P \ 0.001) and OS (P = 0.009) than the
patients with unifocal tumors. However, LRFS, DFS, and
OS were not inferior for the patients with multicentric or
multifocal tumors if pCR was achieved or breast conservation was performed after NACT.
Conclusion. Breast conservation is feasible for clinically
multifocal or multicentric breast cancer patients who
undergo NACT without worsening LRFS if tumor-free
margins can be attained or if patients achieve a pCR.

A prevalence of 4 to 65 % is reported for multifocal/

multicentric breast cancer.17 Currently, mastectomy is
considered for almost all patients with multicentric tumors,
whereas breast-conserving surgery is considered for some
multifocal tumors. The recommendation for mastectomy
derived from reports of local recurrence rates after breast
conservation in multifocal/multicentric disease is as high as
40 %.810 However, these results were obtained from
studies published more than two decades ago that included
predominantly small case series from single institutions. In
contrast, recent retrospective studies have reported low
local recurrence rates after breast conservation for patients
with multifocal/multicentric disease. This emphasizes the

Neoadjuvant Therapy in Multifocal Breast Cancer

1119

b
1.0

0.8

0.6

0.4

log-rank p = 0.002

0.2

Proportion local recurrence-free

Proportion local recurrence-free

1.0

0.8

0.6

0.4

log-rank p = 0.030

0.2

Single
Multifocal
Multicentric
0

10

Local recurrence-free survival (years)

No. at risk
Single
4733
Multifocal
820
Multicentric 581

3250
580
389

1807
274
199

601
89
58

Single
Multifocal
Multicentric

57
9
8

0
0
0

Single
4276
Multifocal
822
Multicentric 381

841
219
253

450
109
117

127
38
31

9
6
4

0
0
0

0.8

0.6

0.4

log-rank p = 0.003

0.2

Proportion local recurrence-free

1.0

0.8

0.6

0.4

log-rank p = 0.314

0.2

Single
Multifocal
Multicentric
0

Single
3759
668
Multifocal
Multicentric 487

2576
472
321

1426
227
164

474
76
51

48
7
7

0
0
0

1.0

0.8

0.6

0.4

log-rank p = 0.724
Single
Multifocal
Multicentric
0

No. at risk
Single
907
132
Multifocal
Multicentric 82

10

Local recurrence-free survival (years)

654
100
65

378
47
34

126
13
7

9
2
1

0
No. at risk

0.2

Single
Multifocal
Multicentric

10

Local recurrence-free survival (years)

No. at risk

Proportion local recurrence-free

10

1.0

Proportion local recurrence-free

Local recurrence-free survival (years)

No. at risk

0
0
0

Single
3217
454
Multifocal
Multicentric 163

10

Local recurrence-free survival (years)

2326
341
123

1329
155
77

463
48
24

47
2
3

0
0

1120
b FIG. 1 Local relapse-free-survival according to unifocal, multifo-

cal, and multicentric breast cancer among a all patients, b patients

with mastectomy, c patients without a pathologic complete response,
d patients with breast conservation, and e patients with a pathologic
complete response

importance of tumor biology for predicting local relapse

rates.11,12
Modern neoadjuvant chemotherapy offers a high probability of complete tumor eradication for patients with
large or even locally advanced breast cancer of certain
subtypes, thereby increasing the chance for breast conservation.13 However, for patients with multicentric disease
and most patients with multifocal disease, breast conservation currently is recommended neither as a primary
treatment option nor after neoadjuvant chemotherapy.14
This recommendation does not take into account that local
relapse risk might differ according to the response after
neoadjuvant chemotherapy and that surgical treatment
might be adapted to the extent of residual disease. This
study aimed to evaluate the impact of focality, type of
surgery, and pathologic complete response on the outcome
for patients after neoadjuvant chemotherapy.
PATIENTS AND METHODS
Patients
The
GeparTrio
(NCT00544765),
GeparQuattro
(NCT00288002), and GeparQuinto (NCT00567554) 1522
trials were prospective, multicenter, and randomized phase
3 trials analyzing neoadjuvant chemotherapy administered
to 6,134 women with primary breast cancer between
August 2002 and July 2010 (Supplemental Table).
The diagnosis of invasive breast adenocarcinoma had to
be confirmed histologically for at least one lesion. Biopsy
of additional suspicious foci was left to the investigators
discretion. Tumor classification as unifocal, multifocal, or
multicentric was determined by physical examination,
mammography, and breast ultrasound. Magnetic resonance
imaging (MRI) of the breast was not mandatory. Focality
was determined by site investigators and routinely classified as unifocal (1 lesion), multifocal (C2 lesions in 1
quadrant), or multicentric (C1 lesion in C2 quadrants)
disease. Histologic evaluation at diagnosis and at surgery
was reported by local pathologists. Central review of
reports was performed only to confirm pathologic complete
response.
The patients were grouped as having luminal A-like
disease, luminal B-like disease, luminal B-like/HER2?
disease, nonluminal-like/HER2? disease, or triple-negative breast cancer (TNBC) using baseline hormonereceptor status, HER2 status, and grading.20

B. Ataseven et al.

Treatment
The participants in the GeparTrio study received two
cycles of docetaxel/doxorubicin/cyclophosphamide (TAC).
The patients with an early response were randomized to
four or six further TAC cycles. The patients without an
early response were randomized to four further cycles of
TAC or four cycles of vinorelbine/capecitabine. AntiHER2-therapy was not applied. The participants in the
GeparQuattro study initially received four cycles of epirubicin/cyclophosphamide and then were randomized to one
of the three arms for four cycles of docetaxel, four cycles
of docetaxel/capecitabine, or four cycles of docetaxel
respectively followed by four cycles of capecitabine. For
cases of HER2? disease, trastuzumab was administered for
a total of 1 year, starting with the first neoadjuvant treatment cycle. A similar chemotherapy was administered to
the GeparQuinto participants. However, the patients with
HER2? tumors randomly received either trastuzumab or
lapatinib, and the patients with HER tumors randomly
received bevacizumab or no bevacizumab concomitant to
all chemotherapy cycles. The patients without response to
the first four chemotherapy cycles with or without bevacizumab were randomized to either weekly paclitaxel or
weekly paclitaxel with everolimus. Treatment was discontinued, and further treatment was administered at the
discretion of the investigator in case of local tumor progression during chemotherapy.
All the patients were allocated to surgery after the end of
neoadjuvant chemotherapy. The type of surgery (breast
conservation or mastectomy) was left to the investigators
discretion. In case of breast conservation, a tumor-free
margin was warranted. The recommended postoperative
treatment consisted of radiotherapy to the ipsilateral
remaining breast for patients treated by breast conservation
and radiotherapy to the chest wall with or without regional
nodes in case of mastectomy for patients with an initial
tumor larger than 5 cm or clinically positive suspect axillary nodes before treatment or at least one histologically
involved node after neoadjuvant treatment. Endocrine
treatment for 5 years after surgery was recommended for
patients with hormone-receptor-positive tumors. Detailed
information about actual accomplished adjuvant treatment
and duration was not available. Therefore, adjuvant treatment was not included as a variable in the multivariate
analysis. The patients were followed up according to local
guidelines. Detailed methods and results of the studies have
been published recently.1519 All the patients consented to
participate in the study, and all the necessary authority
approvals were obtained for performance of the clinical
trials and for further use of anonymized data in subsequent
research projects.

Neoadjuvant Therapy in Multifocal Breast Cancer

1121

TABLE 1 Baseline characteristics of patients by tumor focality

Total
n

Unifocal
%

Multifocal

Multicentric

P Value
(uni- vs. multifocal)

P Value
(uni- vs. multicentric)

Age (years)
\35

384

6.3

296

6.3

53

6.5

35

6.0

3539
4049

603
2,165

9.8
35.3

450
1,654

9.5
34.9

92
301

11.2
36.7

61
210

10.5
36.1

5059

1,779

29.0

1,386

29.3

232

28.3

161

27.7

[60

1,203

19.6

947

20.0

142

17.3

114

19.6

Total

6,134

4,733

820

0.233

0.868

\0.001

\0.001

0.015

\0.001

0.276

0.077

0.088

0.140

0.003

0.048

0.014

0.001

0.004

\0.001

581

Clinical tumor stage

cT1

497

8.1

341

cT2
cT3

7.2

105

12.9

51

8.8

3,767

61.5

3,009

63.7

492

60.2

266

46.1

992

16.2

756

16.0

123

15.1

113

19.6

cT4ac

449

7.3

333

7.0

55

6.7

61

10.6

cT4d

416

6.8

288

6.1

42

5.1

86

14.9

Total

6,121

4,727

817

577

Clinical nodal stage

cN0

2,831

46.6

2,285

48.8

352

43.0

194

33.7

cN1

2,928

48.2

2,186

46.7

420

51.3

322

56.0

cN2
cN3

242
72

4.0
1.2

167
42

3.6
0.9

35
11

4.3
1.3

40
19

7.0
3.3

Total

6,073

4,680

818

575

Histologic type
Ductal-invasive
Lobular-invasive
Other invasive
Total

5,011

81.8

3,865

81.7

681

83.3

465

80.0

673

11.0

506

10.7

88

10.8

79

13.6

444

7.2

358

7.6

49

6.0

37

6.4

6,128

4,729

818

581

Grading
G1

209

3.5

161

G2

3,284

55.2

2,496

3,5
54.3

G3

2,457

41.3

1,943

42.2

Total

5,950

4,600

30

3.8

18

460

58.2

328

58.7

301

38.1

213

38.1

791

3,2

559

Hormone receptor status

HR

2,144

35.4

1,706

36.5

254

31.1

184

32.3

HR?

3,916

64.6

2,968

63.5

562

68.9

386

67.7

Total
HER2 status

6,060

4,674

816

Negative

4,165

72.9

3,258

74.1

542

69.9

365

67.2

Positive

1,547

27.1

1,136

25.9

299

30.1

178

32.8

Total

5,712

4,394

570

775

543

Intrinsic subtype
Luminal A

2,011

35.7

1,522

35.1

287

37.5

202

38.1

Luminal B

707

12.6

555

12.8

92

12.0

60

11.3

Luminal HER2?

895

15.9

660

15.2

142

18.6

93

17.5

Nonluminal HER2?

637

11.3

466

10.7

89

11.6

82

15.5

TNBC

1,382

24.5

1,134

26.1

155

20.5

93

17.5

Total

5,632

4,337

765

530

1122

B. Ataseven et al.

TABLE 1 continued
Total
n

Unifocal
%

Multifocal

Multicentric

P Value
(uni- vs. multifocal)

P Value
(uni- vs. multicentric)

\0.001

\0.001

0.050

0.004

Type of surgery
Breast conservation

3,834

66.0

3,217

71.6

454

58.5

163

30.0

Mastectomy

1,979

34.0

1,276

28.4

322

41.5

381

70.0

Total
5,813
Pathologic complete response

4,493

776

544

No

4,914

81.4

3,759

80.6

668

83.5

487

85.6

Yes

1,121

18.6

907

19.4

132

16.5

82

14.4

6,035

4,666

800

569

HR Hormone receptor, TNBC triple-negative breast cancer

Statistical Analysis
Statistical analysis was performed using SPSS software
version 20.0 (SPSS Inc. Chicago, IL, USA). Correlation
between focality and the clinicopathologic parameters was
assessed by v2 testing. Uni- and multivariate Cox regression analyses were conducted to identify factors predicting
survival. The analyses were conducted according to intentto-treat principles and adjusted for age, cT-stage, cN-stage,
histologic type, grading, hormone receptor and HER2-status, type of surgery, and pathologic complete response.
Survival was plotted by KaplanMeier curves, and differences between groups were analyzed using the log-rank
test. All the tests were two-sided, with significance levels
set at 0.05 without correction for multiple testing.

unifocal vs. multifocal disease and P \ 0.001 for unifocal

vs. multicentric disease).
The patients with multifocal disease significantly more
often had cT1 stages (12.9 vs. 7.2 %), cN ? stages (56.9
vs. 51.2 %), hormone-receptor-positive disease (68.9 vs.
63.5 %) and HER2? disease (30.1 vs. 25.9 %) than the
patients with unifocal tumors (Table 1). The patients with
multicentric tumors showed significantly higher cT and cN
stages, and their tumors were HR? and HER2? more
frequently than those of the patients with unifocal tumors.
Survival Outcome

Clinicopathologic Characteristics

The univariate analysis and log-rank test results concerning the survival end points for all the focality groups
are provided overall and by type of surgery or achievement
of pathologic complete response in Table 2. We chose
unifocality as the reference group to identify the impact of
focality on survival.

This prospectiveretrospective study enrolled 6,134

patients (Supplemental Fig. 1). The median follow-up
period was 36 months (range 0.0110.3 months). At
diagnosis, 4,733 patients (77.1 %) presented with unifocal,
820 patients (13.4 %) with multifocal and 581 patients
(9.5 %) with multicentric disease. A pathologic complete
response (defined as no invasive and no in situ tumor
residual in breast and nodes; ypT0/ypN0) was diagnosed in
907 patients (19.4 %) with unifocal, 132 patients (16.5 %)
with multifocal, and 82 patients (14.4 %) with multicentric
disease (unifocal vs. multifocal P = 0.050 and unifocal vs.
multicentric P = 0.004 disease, respectively). Breast conservation after neoadjuvant chemotherapy was performed
for 3,217 patients (71.6 %) with unifocal disease compared
with 454 patients (58.5 %) with multifocal disease and 163
patients (30 %) with multicentric disease (P \ 0.001 for

Local Relapse-Free Survival The 3-year local relapsefree survival rate was 92.9 % for the patients with unifocal
tumor, 95.1 % for the patients with multifocal tumor, and
90.4 % for the patients with multicentric tumor
(P = 0.002) (Fig. 1a). Interestingly, the patients with
multifocal tumors showed fewer local recurrences than
the patients with unifocal tumors (P = 0.031). This effect
was more pronounced among the patients treated with
mastectomy (P = 0.009; Fig. 1b) and those who failed to
achieve a pathologic complete response (P = 0.034;
Fig. 1c). No significant difference between uni- and
multifocality was detected for the patients treated with
breast conservation or the patients with a pathologic
complete response. In general, the patients with
multicentric tumors showed the highest local relapse
rates. However, local recurrence-free survival was not

RESULTS

Neoadjuvant Therapy in Multifocal Breast Cancer

1123

TABLE 2 Locoregional, overall relapse, and death events for all focality groups by type of surgery or achievement of a pathologic complete
response (ypT0 ypN0)
Local relapse-free survival
Events
Total n

P Value
%

Disease-free survival
a

Events
n

Overall survival
P Value

P Valuea

Events
n

All patients
Unifocal

4,733

336

7.1

886

18.7

Multifocal

820

40

4.9

0.031

133

16.2

0.178

89

10.5

0.735

Multicentric

581

56

9.6

0.015

142

24.4

\0.001

89

15.3

0.009

6,134

432

7.0

0.002

1,161

18.9

\0.001

736

12.0

0.025

Total

558

11.8

Type of surgery
Breast conservation
Unifocal

3,217

182

5.7

463

14.4

274

8.5

Multifocal

454

19

4.2

0.260

60

13.2

0.717

37

8.2

0.916

Multicentric

163

13

8.0

0.347

30

18.4

0.316

18

11.0

0.528

3,834

214

5.6

0.314

553

14.4

0.546

328

8.6

0.820

1,276

140

11.0

375

29.3

246

19.3

322
381

19
41

5.9
10.8

0.009
0.994

64
100

20.0
26.5

0.002
0.444

45
60

14.1
15.9

0.070
0.253

1,979

200

10.1

0.030

539

27.2

0.009

350

17.8

0.137

Total
Mastectomy
Unifocal
Multifocal
Multicentric
Total

Pathologic complete response

Unifocal

907

25

2.8

75

8.3

38

4.2

Multifocal

132

1.5

0.430

5.3

0.290

3.8

0.876

82

2.4

0.844

11

13.4

0.104

8.5

0.081

1,121

29

2.6

0.724

93

8.3

0.121

50

4.5

0.202

514

13.7

Multicentric
Total

No pathologic complete response

Unifocal

3,759

309

8.2

803

21.4

Multifocal

668

38

5.7

0.034

125

18.7

0.192

83

12.4

0.604

Multicentric

487

54

11.1

0.020

129

26.5

0.004

81

16.6

0.055

4,914

401

8.2

0.003

1,057

21.5

0.004

678

13.8

0.114

28

3.7

Total

Pathologic complete response/no pathologic complete response in patients treated by type of surgery
Breast conservation and pathologic complete response
Unifocal

759

20

2.6

Multifocal

86

1.2

Multicentric

30

3.3

875

22

2.5

Total

58

7.6

0.432

5.8

0.609

4.7

0.640

0.965

6.7

0.674

3.3

0.798

0.731

65

7.4

0.811

33

3.8

0.858

403

1.7

246

10.0

Breast conservation and no pathologic complete response

Unifocal

2,436

161

6.6

Multifocal

358

18

5.0

0.295

55

1.5

0.690

33

9.2

0.780

Multicentric

131

12

9.2

0.366

28

2.1

0.263

17

13.0

0.526

2,925

191

6.5

0.354

486

16.6

0.471

296

10.1

0.776

16

13.3

10

8.3

Total

Mastectomy and pathologic complete response

Unifocal
Multifocal
Multicentric
Total

120

3.3

37

0.262

2.7

0.093

2.7

0.288

47
204

1
5

2.1
2.5

0.710
0.527

9
26

19.1
12.7

0.209
0.066

6
17

12.8
8.3

0.319
0.254

1124

B. Ataseven et al.

TABLE 2 continued
Local relapse-free survival
Events
Total n

Disease-free survival

P Value

Events

Overall survival
P Value

P Valuea

Events

Mastectomy and no pathologic complete response

Unifocal

1,150

136

11.8

358

31.1

235

20.4

281

19

6.8

0.018

63

22.4

0.010

44

15.7

0.142

330
1,761

40
195

12.1
11.1

0.826
0.051

89
510

27.0
29.0

0.268
0.029

53
332

16.1
18.9

0.160
0.173

Multifocal
Multicentric
Total
a

Unifocal tumors are the reference group and are compared with multifocal and multicentric tumors, respectively. The P values in the total
line refer to log-rank tests over all three groups

TABLE 3 Locoregional events for all focality groups by intrinsic subtype

Luminal A

Luminal B

Event
Total

1,522

60

287

Multicentric
Total

202
2,011

10
77

P Value

0.306

Unifocal
Multifocal

Luminal HER2?

Event
%

Event

Total

3.9

555

31

2.4

92

5.0
3.8

60
707

6
41

10.0
5.8

0.337

Nonluminal HER2?

Event

Total

5.6

660

46

4.3

142

93
895

6
57

6.5
6.4

0.476

TNBC

Total

7.0

466

3.5

89
82
637
0.096

Event
%

Total

42

9.0

1,134

129

11.4

5.6

155

17

11.0

12
59

14.6
9.3

93
1,382

14
160

15.1
11.6

0.386

TNBC, triple-negative breast cancer

lower for the patients with multicentric tumors than for the
patients with uni- and multifocal tumors if pathologic
complete response was achieved or breast conservation was
performed (Fig. 1d, e). The local relapse rates for the three
focality groups did not differ between the breast cancer
subtypes (Table 3).
Disease-Free Survival A total of 1,161 patients (18.9 %)
experienced relapse, including 886 patients (18.7 %) in the
unifocal group, 133 patients (16.2 %) in the multifocal
group, and 142 patients (24.4 %) in the multicentric group.
The difference between the patients with unifocal tumors
and those with multifocal tumors was not significant
(P = 0.178). A statistically significant difference was
detected between the patients with unifocal tumors and
those with multicentric tumors (3-year DFS: 81.3 vs.
75.6 %; P \ 0.001; Fig. 2a). The patients with breast
conservation after neoadjuvant chemotherapy (Fig. 2b,
P = 0.316) and those who achieved a pathologic
complete response (Fig. 2c, P = 0.104) showed no
difference in disease-free survival when the diagnosis
was either unifocal or multicentric tumor. In the cohort of
patients who underwent mastectomy, disease-free survival
was significantly better for the patients with multifocal
tumors than for those with unifocal tumors, whereas no

difference was detected between the patients with unifocal

tumors and those with multicentric tumors.
Overall Survival A total of 736 patients (12 %) died. The
3-year OS was 88.2 % for the patients with unifocal
tumors, 89.5 % for those with multifocal tumors, and
84.7 % for those with multicentric tumors (P = 0.025).
There was no difference between uni- and multifocal
tumors (P = 0.735), but a significant difference was
detected between unifocal and multicentric tumors
(P = 0.009). However, overall survival did not differ
among the patients with unifocal, multifocal, and
multicentric tumors if a pathologic complete response
was achieved or breast conservation was successfully
performed.
Focality as an Independent Predictor of Outcome
In the multivariate, adjusted Cox regression analysis,
local relapse-free survival did not differ significantly
between the patients with unifocal tumors and those with
multicentric tumors (HR 0.94; (95 % CI 0.681.30;
P = 0.709) nor between the patients with multifocal
tumors and those with multicentric tumors (HR 1.51; 95 %
CI 0.952.40; P = 0.082). However, the local relapse-free

Neoadjuvant Therapy in Multifocal Breast Cancer

1125

b
1.0

Proportion disease-free

Proportion disease-free

1.0

0.8

0.6

0.4

log-rank p < 0.001

0.2

0.8

0.6

0.4

log-rank p = 0.546

0.2

Single
Multifocal
Multicentric
0

Single
4733
Multifocal 820
Multicentric 581

3157
560
370

1736
259
183

578
85
55

55
9
8

10

Disease-free survival (years)

No. at risk

Single
Multifocal
Multicentric

No. at risk
0
0
0

Single
3217
Multifocal
454
Multicentric 163

10

Disease-free survival (years)

2277
330
120

1286
150
72

451
47
22

46
2
3

0
0
0

Proportion disease-free

1.0

0.8

0.6

0.4

log-rank p = 0.121

0.2

Single
Multifocal
Multicentric
0
No. at risk
Single
907
Multifocal 132
Multicentric 82

10

Disease-free survival (years)

647
96
60

377
47
31

126
13
7

9
2
1

0
0
0

FIG. 2 Disease-free-survival according to unifocal, multifocal, and multicentric breast cancer among a all patients, b breast conservation
patients, and c patients achieving a pathologic complete response

survival rate differed significantly between the patients

with unifocal tumors and those with multifocal tumors (HR
0.62; 95 % CI, 0.430.88; P = 0.008). Moreover, these
results did not change when a pathologic complete
response was removed as a variable in the model to test the
robustness of the model.
DISCUSSION
The presence of multifocal and multicentric breast
cancer is frequently considered to be an indication for

extended operations or mastectomies. Mastectomy is recommended in general for patients with multicentric disease
irrespective of previous neoadjuvant chemotherapy.14 The
data are insufficient to support the use of breast conservation after neoadjuvant chemotherapy for patients with
multifocal or multicentric disease, even if a complete
pathologic response was reached. We therefore analyzed
the survival of patients treated with neoadjuvant chemotherapy and breast conservation according to tumor focality
at diagnosis. Local relapse-free survival for multifocal and
multicentric disease did not differ from that for unifocal

1126

disease. However, among patients treated with mastectomy, those with multifocal but not multicentric disease
had a local relapse-free and disease-free survival advantage
over those with unifocal disease. This might be explained
by more frequent omission of postmastectomy radiotherapy
for patients with unifocal disease than for patients with
multifocal/multicentric disease. Unfortunately we lacked
information on the use of adjuvant radiotherapy in the
examined cohort, so we cannot support this hypothesis with
evidence. However, neither multifocality nor multicentricity were independent negative prognosticators for local
relapse-free survival when a pathologic complete response
was removed as a variable in Cox regression analysis.
In an earlier study by Oh et al. 706 patients receiving
anthracycline-based neoadjuvant chemotherapy, including
97 patients classified as clinically multifocal or multicentric, were examined.23 These authors found no statistical
difference in the locoregional relapse rate regardless of
locoregional treatment between unifocal and multifocal/
multicentric breast cancer patients. Likewise, no difference
between the two groups was found for 5-year disease-free
or overall survival. The authors concluded that clinically
detected multifocal or multicentric disease did not negatively influence the outcome for patients. Our results
support this finding only for patients treated with breast
conservation and those who achieve a pathologic complete
response. This might be explained by some important
differences between the Oh et al.23 study and ours. First,
our study analyzed a much larger cohort of patients within
each focality group, which provided more statistical power
to evaluate the impact of clinical focality after neoadjuvant
chemotherapy on survival. Second, in our study, 58 % of
the multifocal patients and 29 % of the multicentric
patients underwent breast conservation, whereas in the MD
Anderson Hospital cohort,28 breast conservation was performed only for patients with multifocal disease and not for
those with multicentric disease. Former retrospective
studies analyzing the outcome for patients with multifocal
or multicentric breast cancer in the adjuvant setting gave
controversial results. Weissenbacher et al.24 and Yerushalmi et al.25 reported a significantly better breast
cancer-specific survival for patients with unifocal cancer
than for those with multifocal or multicentric disease. In
contrast, in other studies,11,26,27 multifocal/multicentric
disease was not a predictor for a higher locoregional
relapse rate if adjustment was made for other risk factors.
A similar finding was reported by a recently published
adjuvant study from the MD Anderson Hospital,28 in which
locoregional recurrence-free survival of patients with
multifocal (n = 673) and multicentric (n = 233) disease
was reported. Breast conservation was performed for 62 %
of patients with unifocal tumors and 38 % of patients with
multifocal but not for patients with multicentric tumors. No

B. Ataseven et al.

difference in local recurrence rate between the different

focality groups was detected after a median follow-up
period of 52 months regardless of locoregional treatment
(breast conservation, mastectomy, or mastectomy plus
radiation). The recurrence rate after breast conservation
was very low in both groups (unifocal 1.02 % vs. multifocal 1.95 %; P = 0.68).
Moon et al.29 investigated the impact of focality on
outcome according to the molecular tumor subtype of
patients treated with primary surgery. They found that
patients with multifocal or multicentric cancer had significantly more luminal A disease and less TNBC. However,
they detected a significantly worse outcome for patients
with multifocal/multicentric disease when the tumors were
classified as TNBC. In our study, we found no impact of
different molecular tumor subtypes and focality on local
relapse-free survival. One possible explanation might be
that especially in patients with TNBC, the high rates of
pathologic complete response after neoadjuvant chemotherapy might overrule any negative effects arising from
potentially disadvantageous focality. This was underscored
by our results for local relapse-free survival, which were
very favorable for the patients with a pathologic complete
response independently of surgical treatment or focality
(Fig. 2e).
Our study had some weaknesses due to its retrospective
design and the failure to prove multifocality or multicentricity by histology. This implies that foci for which biopsy
was not performed may have been misjudged as malignant
or, if malignant, may have had tumor characteristics other
than those that had biopsy evaluation. Due to the multicentric design of the trials, central evaluation of all
radiologic findings (e.g., mammography, ultrasound, or
MRI) was not achievable. Additionally, we lacked information on focality and centricity at the time of surgery.
Thus, the extent to which patients with multifocal or
multicentric tumors have had a mixed response to neoadjuvant chemotherapy remains an open question.
Nevertheless, our study had the strength of investigating
prospective trial cohorts whose diagnosis and treatment
was performed according to identical prespecified procedures in concordance with current recommendations while
still taking into account differences in the chemotherapy
and anti-HER2 therapy used in the three studies. Moreover,
our study is, to the best of our knowledge, the largest one to
evaluate the impact of focality on outcome parameters in
patients treated with neoadjuvant chemotherapy.
Our data provide strong evidence that breast conservation
is feasible for patients with clinically multifocal or multicentric breast cancer who undergo neoadjuvant
chemotherapy. Local relapse-free survival was not worsened if tumor-free margins were attained or patients
achieved a pathologic complete response. Nevertheless,

Neoadjuvant Therapy in Multifocal Breast Cancer

patients needing mastectomy after neoadjuvant chemotherapy showed a higher risk for locoregional recurrence and
should therefore be monitored closely. Our findings should
encourage clinicians and patients to consider breast conservation after neoadjuvant chemotherapy if clear margins
and acceptable cosmetic results can be achieved even if the
tumor was initially diagnosed to be multifocal or
multicentric.
DISCLOSURE

None.

REFERENCES
1. Anastassiades O, Iakovou E, Stavridou N, Gogas J, Karameris A.
Multicentricity in breast cancer. A study of 366 cases. Am J Clin
Pathol. 1993;99:23843.
2. Egan RL. Multicentric breast carcinomas: clinical-radiographicpathologic whole-organ studies and 10-year survival. Cancer.
1982;49:112330.
3. Fisher ER, Gregorio R, Redmond C, Vellios F, Sommers SC,
Fisher B. Pathologic findings from the National Surgical Adjuvant Breast Project (protocol no. 4): I. Observations concerning
the multicentricity of mammary cancer. Cancer. 1975;35:24754.
4. McDivitt RW. Breast cancer multicentricity. Monogr Pathol.
1984(25):139148.
5. Sarnelli R, Squartini F. Multicentricity in breast cancer: a submacroscopic study. Pathol Ann. 1986;21(Pt 1):14358.
6. Spinelli C, Berti P, Ricci E, Miccoli P. Multicentric breast
tumour: an anatomical-clinical study of 100 cases. Eur J Surg
Oncol. 1992;18:236.
7. Vaidya JS, Vyas JJ, Chinoy RF, Merchant N, Sharma OP, Mittra
I. Multicentricity of breast cancer: whole-organ analysis and
clinical implications. Br J Cancer. 1996;74:8204.
8. Wilson LD, Beinfield M, McKhann CF, Haffty BG. Conservative
surgery and radiation in the treatment of synchronous ipsilateral
breast cancers. Cancer. 1993;72:13742.
9. Leopold KA, Recht A, Schnitt SJ, et al. Results of conservative
surgery and radiation therapy for multiple synchronous cancers of
one breast. Int J Radiat Oncol Biol Phys. 1989;16:116.
10. Kurtz JM, Jacquemier J, Amalric R, et al. Breast-conserving
therapy for macroscopically multiple cancers. Ann Surg.
1990;212:3844.
11. Gentilini O, Botteri E, Rotmensz N, et al. Conservative surgery in
patients with multifocal/multicentric breast cancer. Breast Cancer Res Treat. 2009;113:57783.
12. Yerushalmi R, Tyldesley S, Woods R, Kennecke HF, Speers C,
Gelmon KA. Is breast-conserving therapy a safe option for
patients with tumor multicentricity and multifocality? Ann Oncol.
2012;23:87681.
13. Gianni L, Baselga J, Eiermann W, et al. Feasibility and tolerability of sequential doxorubicin/paclitaxel followed by
cyclophosphamide, methotrexate, and fluorouracil and its effects
on tumor response as preoperative therapy. Clin Cancer Res.
2005;11(24 Pt 1):871521.

1127
14. Kaufmann M, von Minckwitz G, Mamounas EP, et al. Recommendations from an international consensus conference on the
current status and future of neoadjuvant systemic therapy in
primary breast cancer. Ann Surg Oncol. 2012;19:150816.
15. Huober J, Fasching PA, Hanusch C, et al. Neoadjuvant chemotherapy with paclitaxel and everolimus in breast cancer patients
with nonresponsive tumours to epirubicin/cyclophosphamide
(EC) bevacizumab results of the randomised GeparQuinto
study (GBG 44). Eur J Cancer. 2013;49:228493.
16. Untch M, Loibl S, Bischoff J, et al. Lapatinib versus trastuzumab
in combination with neoadjuvant anthracycline-taxane-based
chemotherapy (GeparQuinto, GBG 44): a randomised phase 3
trial. Lancet Oncol. 2012;13:13544.
17. Untch M, Rezai M, Loibl S, et al. Neoadjuvant treatment with
trastuzumab in HER2-positive breast cancer: results from the
GeparQuattro study. J Clin Oncol. 2010;28:202431.
18. von Minckwitz G, Eidtmann H, Rezai M, et al. Neoadjuvant
chemotherapy and bevacizumab for HER2-negative breast cancer. N Engl J Med. 2012;366:299309.
19. von Minckwitz G, Rezai M, Loibl S, et al. Capecitabine in
addition to anthracycline- and taxane-based neoadjuvant treatment in patients with primary breast cancer: phase III
GeparQuattro study. J Clin Oncol. 2010;28:201523.
20. von Minckwitz G, Blohmer JU, Costa SD, et al. Response-guided
neoadjuvant chemotherapy for breast cancer. J Clin Oncol.
2013;31:362330.
21. von Minckwitz G, Kummel S, Vogel P, et al. Neoadjuvant
vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase III
randomized GeparTrio trial. J Nat Cancer Inst. 2008;100:54251.
22. von Minckwitz G, Kummel S, Vogel P, et al. Intensified neoadjuvant chemotherapy in early-responding breast cancer: phase III
randomized GeparTrio study. J Nat Cancer Inst. 2008;100:552
62.
23. Oh JL, Dryden MJ, Woodward WA, et al. Locoregional control
of clinically diagnosed multifocal or multicentric breast cancer
after neoadjuvant chemotherapy and locoregional therapy. J Clin
Oncol. 2006;24:49715.
24. Weissenbacher T, Zschage M, Janni W, et al. Multicentric and
multifocal versus unifocal breast cancer: is the tumor-nodemetastasis classification justified? Breast Cancer Res Treat.
2010;122:2734.
25. Yerushalmi R, Kennecke H, Woods R, Olivotto I, Speers C,
Gelmon K. Does multicentric/multifocal breast cancer differ from
unifocal breast cancer? An analysis of survival and contralateral
breast cancer incidence. Breast Cancer Res Treat. 2009;117:365
70.
26. Kadioglu H, Yucel S, Yildiz S, et al. Feasibility of breast conserving surgery in multifocal breast cancers. Am J Surg.
2014;208:45764.
27. Lynch SP, Lei X, Chavez-MacGregor M, et al. Multifocality and
multicentricity in breast cancer and survival outcomes. Annl
Oncol. 2012;23:30639.
28. Lynch SP, Lei X, Hsu L, et al. Breast cancer multifocality and
multicentricity and locoregional recurrence. Oncologist. 2013;
18:116773.
29. Moon HG, Han W, Kim JY, et al. Effect of multiple invasive foci
on breast cancer outcomes according to the molecular subtypes: a
report from the Korean Breast Cancer Society. Ann of Oncol.
2013;24:2298304.