Is regional nodes radiotherapy an alternative to surgery?

The Breast 22 (2013) S118eS128

Contents lists available at SciVerse ScienceDirect

The Breast journal homepage: www.elsevier.com/brst

Review

Is regional nodes radiotherapy an alternative to surgery? Birgitte Vrou Offersen a, *, Hanne Melgaard Nielsen a, Marie Overgaard b, Jens Overgaard b a b

Dept of Oncology, Aarhus University Hospital, Denmark Dept of Experimental Clinical Oncology, Aarhus University Hospital, Denmark

a b s t r a c t Keywords: Adjuvant radiotherapy Breast cancer Positive lymph nodes Axillary dissection Randomized clinical trial

Sentinel node biopsy (SN) in breast cancer treatment was introduced in the mid-1990s in order to be able to stage patients before decision of definitive surgery. Since then, both the pathological examinations of the SN and the systemic adjuvant treatment have improved and cause new challenges in the correct decision making regarding whether or not to radically treat the axilla in case of a positive SN. In SN positive patients, current St. Gallen guidelines support no completion ALND (axillary lymph node dissection) in clinically node-negative patients with 1e2 macrometastatic sentinel nodes operated with breast conservation and receiving tangential field adjuvant radiotherapy (RT). ALND is being questioned due to increased morbidity compared with SN biopsy alone, and to limited long term benefit on disease free survival in selected patients. An alternative to ALND is treating the axilla with nodal RT although this treatment is mostly used as adjuvant treatment after ALND in high risk patients. Few studies have investigated the benefit of nodal RT compared to ALND, and no consensus has yet been reached. Clinical decision making regarding treating the axilla should be based on relevant data, and in this review studies aiming at deciding whether or not and how the axilla should be treated in SN positive patients will be discussed. Furthermore treatment choice will be discussed, since besides ALND, both breast irradiation and nodal irradiation might cure residual disease after SN. Also the issue of improved systemic adjuvant treatment will be discussed in relation to eventually no regional axillary treatment. Ó 2013 Elsevier Ltd. All rights reserved.

Introduction The role for a completion axillary lymph node dissection (cALND) in patients operated for a clinically node-negative (cN0) early breast cancer with a positive sentinel node (SN) is now being questioned. The locoregional and systemic treatment of early breast cancer has changed considerably over the last decades. In many countries mammography screening is now routine, resulting in more patients being diagnosed with small cN0 early breast cancer, thus the risk profile of breast cancer has changed in a favourable direction which is reflected in improved outcome [1,2]. Most patients now receive breast conserving surgery (BCS) followed by adjuvant radiotherapy (RT) and also in most cases systemic therapy with either chemotherapy and/or endocrine therapy. Since the introduction of sentinel node biopsy (SNB) as a staging procedure of the axilla, the gold standard has been a

* Corresponding author. Dept of Oncology, Aarhus University Hospital, Noerrebrogade 44, Building 5, DK-8000 Aarhus C, Denmark. Mobile: þ45 28838012; fax: þ45 86197109. E-mail address: [email protected] (B.V. Offersen). 0960-9776/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.breast.2013.07.023

completion levels I and II axillary dissection in cases with metastasis in the SN (except in patients with isolated tumour cells with tumour
0.2 mm (pN0(iþ))) and in patients where the SN could not be localised [3,4]. Based on data from the Surveillance, Epidemiology and End Results (SEER) database in the period 1998e2004 the use of SNB increased from 11 to 59% in the USA, however, in the same period there was also an increase from 20 to 32% in women having metastasis in the SN but no cALND [5]. It was documented that the frequency of having a positive SN but no cALND was 58/50% in cases with T1/T2 tumours with micrometastasis (
2 mm) in SN, and 30/25% in cases with T1/T2 tumours with macrometastasis (>2 mm) in SN. SN positive patients not receiving a cALND were more often elderly women operated for small, low grade tumours with lack of extracapsular extension and with micrometastasis in the axilla. There is only limited data on the optimal locoregional therapy of the clinical node-negative SNþ axilla. We will review the data on outcome after axillary dissection and/or regional RT in these patients. Since there are many reports on limited numbers of patients, we have chosen to limit the studies included in this review to those reporting data from studies of minimum 50 patients. Furthermore, we will discuss if regional RT can replace axillary dissection.

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

Studies Patients with cN0 breast cancer where SNB and ALND generally were not performed and no regional RT was given Non-randomised studies In Table 1 we have listed non-randomised and randomised studies reporting on outcome in cN0 patients who had no regional therapy. The 5 non-randomised studies were all initiated in the 1980s [6e9] and one in 1998 [10], and no axillary dissection was performed except in one of the Milan studies where 26% of the patients had an ALND and metastatic disease was found in the axilla in 33.7% of the patients [6]. Almost all the patients were operated with BCS, and most often this was followed by normofractionated tangential field irradiation (45e50 Gy) plus a 10 Gy tumour bed boost, except in the 2 Milan studies no adjuvant RT was given in 30 and 70% of those patients who had no ALND [6,7]. The risk profile of the patients was in all the studies relatively good, since the far majority of the patients were more than 60 years old and operated for hormone receptor positive pT1 tumours. Also many of the patients received endocrine therapy, with a clear trend towards all patients being treated with tamoxifen in the latest study from Boston. The regional recurrence risk was disturbingly high in the study from the UK, where after a median follow up of 60 months, the actuarial 10-year risk of lymphatic relapse was 22%, prompting the authors to recommend elective treatment of the axilla [9]. However, in the smaller studies from Boston on 92 and 74 patients no regional recurrences were seen [8,10]. In the 2 studies from Milan the regional recurrence risk was significantly associated with tumour size and grade III tumours, but overall the regional recurrence risk was relatively low [6,7]. Randomised trials Two randomised trials have accrued cN0 patients where no SNB was made [11e13]. The characteristics of these trials are listed in Table 1. The trials randomised patients to ALND, and in general the trials were small. The International Breast Cancer Study Group (IBCSG) 10e93 trial was designed to accrue 1020 patients, but accrued only 46% of the planned number of patients, and due to slow accrual the group behind the trial therefore changed the primary endpoint from regional tumour control to assessing quality of life. The patients in both trials were highly selected with median ages 70e74 years, they were operated for small, hormone receptor positive tumours and 85e96% of the patients received tamoxifen for 5 years. Most patients were operated with BCS followed by adjuvant tangential breast RT without a third field to the regional nodes. For both trials the regional recurrence rates were very low, and the regional morbidities and quality of life were significantly in favour of no ALND. Patients with cN0 and SNþ breast cancer, where cALND was generally not performed and regional RT was not given systematically Non-randomised studies Six non-randomised studies have reported data from patients treated in the period 1993e2009, where the SNB was starting as routine in cN0 patients (Table 2) [14e19]. The largest European study was the nationwide Dutch MIRROR study (Micrometastases and Isolated Tumours Cells: Relevant and Robust or Rubbish?), whose purpose was to evaluate the relationship between isolated tumour cells or micrometastasis in the regional nodes and clinical outcome in patients operated with SNB and who did or did not receive systemic adjuvant therapy [15,20]. Half of the 2680 patients had no cALND and 46% of these patients had pN0(iþ) or pN1(mic) disease in

S119

the SN. In 148 patients (5.5%) regional RT was given instead of cALND. The patients were predominantly operated for T1, hormone receptor positive tumours, and 48% of patients with pN1(mic) had systemic therapy. The study categorized patients into 3 groups based on nodal status: Group 1 consisted of patients without nodal metastases and who had no systemic therapy, Group 2 consisted of patients with isolated tumour cells (ITCs) or micrometastases, and who did not receive adjuvant systemic therapy. Group 3 consisted of patients with ITCs or micrometastases and who had systemic therapy. Group 2 had a significantly worse 5-year DFS compared to Group 1 (75.6 vs. 85.7%, P < 0.001), and the same negative impact was seen both in patients with ITCs and micrometastases. The positive influence from systemic therapy was seen when comparing Group 2 with 3, the 5year DFS being 76.5 vs 86.2%, P < 0.001. Moreover, after median 5.1 years follow up there was a significantly higher regional recurrence risk among patients with micrometastatic disease not receiving axillary treatment (cALND or regional RT). The 5 year rate was 2% versus 0.9% (P ¼ NS) for patients with isolated tumour cells in the SN, but for patients with micrometastasis in the SN the corresponding 5 year rates were 5.6% versus 1.0%, with a hazard rate for regional recurrence at 5 years for the pN1(mic) patients of 4.39 (95% CI, 1.46e 13.24). In the 148 patients treated with regional RT no regional recurrences were seen at 5 years. Patients with regional recurrence were more often diagnosed with micrometastatic larger tumours, hormone receptor negative and grade III. From the USA data from nearly 100,000 patients registered in the National Cancer Data Base has recently been published [17]. Patients diagnosed in 1998e2005 with cN0 early breast cancer with SNþ disease were investigated, and 20.8% of the patients had not received a cALND despite metastasis in the SN, which in 18.2% of the cases was micrometastasis and in 81.8% of the cases was macrometastasis. Through the period from 1998 to 2005, a change in practice pattern was documented; a decline was seen in SNþ patients who had the SNB followed by a cALND from 24.2% to 16.7% in cases with macrometastasis in the SN (P < 0.001), and in cases with micrometastasis in SN the frequency of omitting a cALND increased from 24.4% to 45.3% (P < 0.001). In the same period the technique of SNB also changed; for patients diagnosed in 1998e 2000 the median number nodes examined in the SLND only group was 11 (interquartile range 4e16) and for patients having a cALND it was 14 nodes (interquartile range 10e19). In the period 2004e 2005 the median number nodes examined in the SLND group were 3 (interquartile range 2e5) and for the cALND group 13 (interquartile range 9e18). The authors found that the likelihood of not receiving a cALND despite macrometastasis in the SN was associated with older age, being black, having more comorbidity, smaller tumour, low grade, and being operated in a non-National Cancer Institute-designated hospital. After a median follow up of 63 months for those diagnosed 1998e2000, a non-significant trend towards fewer regional recurrences and better overall survival (OS) for patients having a cALND vs SLND alone in the macrometastatic group was identified: axillary recurrence hazard ratio (HR) 0.58 (95% CI, 0.32e1.06) and OS HR 0.89 (95% CI, 0.76e1.04) in favour of cALND. For patients in the micrometastatic group no differences in regional control and OS were seen. There are some limitations of the study, for example there are no details on the RT and the systemic therapy. There are also no details on the type of examination of the SNB, and there may be underreporting of the regional recurrences. It is therefore unfortunately difficult to draw firm conclusions from this large study. In a study from the Memorial SloaneKettering Cancer Center 287 selected patients were cN0 but SNþ and had no cALND [16]. These patients were compared to 1673 consecutive patients treated in the same institution during the same period, only these patients had a cALND. After median 23 months follow up there were

S120 Table 1 Studies reporting on results from minimum 50 clinically node-negative (cN0) patients, where no sentinel node biopsy (SNB) was done, and where axillary lymph node dissection (ALND) and regional radiotherapy (RT) were not provided. Type

N

Year

Patients

HR status

FU median

pT1

Age

Results

Kent study [9]

-R

1984e1994

5 yr

NR

90% >50 yr

-R

1986e1994

94% of pts 50 yr had 2 yr tam 34% had tam

Not reported

Milan study [7]

291 pts All BCS 401 pts cT1-2 96% BCS

65% PgRþ 75% ERþ

5.2 yr

91%

85% >50 yr 59% 60 yr

Milan study [6]

-R

1987e1992

All had 2 yr tam

No ALND: 0.8% ER-/PgRþ 92% ERþ

15 yr

59% (no cALND)

All 70 yr

22% actuarial 10-yr risk of lymphatic relapse 17% 10-yr risk of symptomatic lymphatic relapse. 6.7% had reg recurrence at median 2.6 yr. Closely related to T1a (2.0%). T1b (1.7%), T1c (10.5%), T2 (18.4%) and grade III Crude cumulative 15 yr incidence of axillary rec in no ALND pts 5.8% overall, and 3.7% for pT1 pts No axillary recurrences in þALND pts

Boston study [8]

-R

1988e1993

58% had tam 5 yr

All pts ERþ

4.2 yr

Median 69 yr

0 reg rec

Boston prospective trial [10]

-R

1998e2003

92% had 5 yr tam

All pts ERþ

4.3 yr

83% Median T 12 mm 89% Median T 12 mm

Median 75 yr All 55 yr

0 reg rec

IBCSG Trial 10-93 [11]

þR

1993e2002

96% had tam
5 yr

80% ERþ

6.6 yr

56%

Median 74 yr All 60 yr

Milan trial [12,13]

þR

1996e2000

All tam for 5 yr, 85% completed tam

1% ER-/PgRþ 89% ERþ 91% Gr I þ II

12.5 yr

94%

Median 70 yr All 65 yr

2% had axillary/reg rec as first event (1% vs 3% in ALND vs no ALND group). Sign better arm movement and less pain at first post-operative assessment in favour of no ALND, thereafter no difference between ALND 50% had cALND: 71% pN0, 20% pN1, 8% pN2 0 and 4 axillary rec in ALND and no ALND group. No differences in distant metastasis and breast cancer mortality. 50% had cALND: 77% pN0, 17% pN1, 6% pN2

671 pts 172 þALND 499 pts ALND All BSC 92 pts, clinical stage I/II All BCS 74 pts, clinical stage 1e2 Single arm multicenter trial. All BCS followed by “high tangential” RT, no reg RT 473 pts with indication for tam irrespective of N status Randomisation to ALND Mastectomy 45% BCSþRT 33% BCS-RT 23% 219 pts cT1 All quadrantectomy Randomisation to ALND RT to breast only Target enrolment was all eligible pts within 5 years

Abbreviations: R, randomised; BCS, breast conserving surgery; Tam, tamoxifen; FU, follow up; HR, hormone receptor status; ER, oestrogen receptor; PgR, progesteron receptor.

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

Study

Table 2 Studies reporting on minimum 50 patients with clinically node-negative (cN0) and sentinel node-positive (SNþ) breast cancer, where completion axillary lymph node dissection (cALND) was not performed and regional radiotherapy (RT) not given systematically. Type N

Year

MD Anderson study [14] MIRROR study [15,20]

-R

196 pts BCS 69% 2680 pts, nationwide T <1 cm of any grade or T 1e3 cm grade 1 or 2, pN0, pN0(iþ), pN1(mic), no pN1(macro) BCS 71%, most had WBI No cALND was done in 34% of SNþ pts.

1993e2005 58% RT 82% ERþ 70% had chemotherapy 70% PgRþ 1997e2005 856 pts pN0 and no syst therapy 91% HRþ 856 pts SNþ and no syst therapy 995 SNþ and syst therapy

287 pts cN0, SNþ and no cALND (selection: older age, smaller T, fewer grade III) No reg RT

1997e2004 55% BCS

cN0 and SNþ 97,314 pts, 21% SNB only, Nationwide oncology outcomes registry

1998e2005 No cALND vs cALND: BCS 81% vs 50%, chemotherapy 61% vs 70%, postmastectomy RT 21% vs 33%, all findings significant 1997e2009 55% had chemotherapy 76% had endocrine therapy

-R

MSKCC study [16] -R

National Cancer Data Base in the US [17]

-R

MSKCC study [18] -R

MSKCC study [19] -R

ACOSOG Z0011 trial [21]

þR

IBCSG Trial 23-01 [22]

þR

Patients

326 pts all BCS and 93% had RT. 66 pts had tangential field RT in prone position, 20 pts had supraclavicular fields also 1997e2009 56 vs 68% had chemotherapy 535 pts, 325 BCS and 210 total BCS vs TM, P ¼ 0.005 mastectomy (TM) 77% had endocrine therapy RT given to 94% of BCS vs 5% of TM pts. RT to TM included the chest wall and supraclavicular fossa 856 pts for analysis, cT1eT2, cN0 1999e2004 96e97% had adjuvant systemic therapy and 1e2 SNs with metastasis 89% had WBI Target enrolment 1900 pts. 931 pts cN0, T
5 cm 2001e2010 Endocrine therapy SN biopsy with pN0(iþ) or to 63e67% Chemotherapy 7e9% pN1(mic) 90% BCS 90e92% had RT Target enrolment 1960 pts

HR status

FU median

pT1

Age

Results

2.5 yr

72%

Median 56 yr

5.1 yr

86%

Median 57 yr

73% ER þ 55% PgR þ

1.9 yr

78%

Median 52 yr

Not reported

50% (range Median 56 yr 5.3 yr for pts diagnosed 1.5e3.0 cm) 1998e2000

No axillary rec, 1 recurrence in fossa supraclavicularis pN0: 5 yr reg rec rate 2.3% (no axillary treatment) vs 1.6% (with axillary treatment), P ¼ NS pN0(iþ): 5 yr reg rec rate 2.0% (no axillary treatment) vs 0.9% (with axillary treatment), P ¼ NS pN1(mic): 5 yr reg rec rate 5.6% (no axillary treatment) vs 1.0% (with axillary treatment), HR for reg rec: 4.39 (95% CI, 1.46e13.24) Results compared with 1673 pts with SNþ and cALND from same period and institution. Axillary rec: ALND vs no ALND, 0.4% vs 2%, P ¼ 0.004 Non-sign trend in pN1(macromet) group towards better outcome (axillary rec and OS) after cALND vs SN alone. No benefit from cALND in pN1(micþ) group

80% ERþ

4.6 yr

85%

Median 60 yr

88% (BCS) 83% (TM)

4.8 yr

85% (BCS) 69% (TM), P < 0.001

9 local recurrences (2.8%), 6 regional Median 59 yr recurrences (1.8%), 17 distant failures, (BCS), 55 yr (TM), P ¼ 0.001 49 deaths. No difference in loco-regional control between BCS and TM, but more distant failure among BCS pts

>83% HRþ 6.3 yr

>67%

Median 54e56 yr

5 yr LRR free survival was 96.7% in SLND and 95.7% in cALND (P ¼ 0.28) No difference in DFS or OS

89% ERþ 75% PgRþ

67%

Mean 54 yr

No difference in 5 yr DFS or OS 1% regional recurrence in the undissected axilla Late morbidity in ALND vs no ALND group: Lymphoedema 13% vs 4% Motor neuropathy 8% vs 3%

4.8 yr

3 pts developed regional recurrence

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

Study

Abbreviations: R, randomised; BCS, breast conserving surgery; Tam, tamoxifen; FU, follow up; HR, hormone receptor status; ER, oestrogen receptor; PgR, progesteron receptor; WBI, whole breast irradiation.

S121

S122

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

significantly more regional recurrences among the patients selected for no cALND: 0.4% versus 2%, P ¼ 0.004. The selected patients were significantly older, had smaller tumours, which were more often grade IeII. Interestingly this study points to a higher regional recurrence rate among patients whose SN positivity was based on H&E staining. In this group of patients 5% had a regional recurrence if no cALND was performed. Optimal regional control has been reported from MD Anderson, where 750 consecutive cN0 but SNþ patients were followed median 29.5 months, and 196 patients (26%) had no cALND [14]. Among the patients with no cALND median 3 lymph nodes (range 1e14) were removed, and median 1 (range 1e6) was positive with a median size 1 mm (range 0.1e12.9 mm). Based on the SN evaluation, patients were classified as having either isolated tumour cells, micrometastasis or macrometastasis in 34%, 46% and 20% of the cases, respectively. The diagnosis of the metastasis was in 64.3% and 35.7% of the cases based on H&E staining and immunohistochemistry, respectively. Most patients had BCS, and 58% of the patients had adjuvant RT, however, details on the RT were only available in 12% of the patients. Among these patients, 64% had received RT to regional nodes. Chemotherapy was given to 70%. Only 1 recurrence in the supraclavicular fossa was reported. The two most recent retrospective studies on patients who had SNþ disease but no cALND are from New York [18,19]. In 326 patients operated with BCS and selected for no cALND after a positive SN (58% had ITCs, 35% had micrometastasis, 7% had macrometastasis) only 3 patients had a regional recurrence after median 55 months. Systemic therapy was given to 93% of the patients, 93% of the patients also had adjuvant RT [18]. Among the irradiated patients 66 patients (22%) were treated in prone position, which does not deliver any appreciable dose to the axilla. These patients operated with BCS were in the other study compared to 210 patients operated with total mastectomy (TM) for a SNþ disease but had not received a cALND [19]. Only 5% of the TM patients received RT and this included the chest wall and supraclavicular fossa. After median 4.8 years follow up only 6 regional recurrences in total were found, and this study thus challenges the hypothesis that low axillary irradiation from the tangential breast fields is contributing to the low regional recurrence risks. Randomised trials Two randomised trials have accrued cN0 and SNþ patients where the patients were randomised to cALND and no regional RT, and the characteristics of these trials are listed in Table 2 [21,22]. With a median follow up of 6.3 years data on regional and distant control from the American College of Surgeons Oncology Group (ACOSOG) Z0011 trial was reported in 2011. Inclusion criteria were patients with cT1eT2 tumours, cN0 and 1e2 SN with metastasis on frozen section, touch preparation or H&E staining on permanent section. Macrometastases in the SN were accepted. The trial was active in 115 centres and the target enrolment was 1900 patients, however, due to slow accrual the trial was closed in 2004, where 856 patients (45% of planned) were available for analysis. All patients had BCS, and 89% had tangential field whole breast irradiation, and no third field to the regional nodes. The RT was based on “high tangents”, and the group behind the trial is now in the process of documenting what that meant in detail. It has been suggested that “high tangents” include most of the axillary lymph nodes in levels I and II [23]. Systemic therapy was administered to 96e97% of the patients. The median number positive nodes in both groups were 1, however, the median number of removed nodes were 17 with cALND vs 2 with SLND alone. Both groups had a very low regional recurrence rate with no difference between the groups, also there were no differences in DFS and OS. The authors found significantly more regional morbidity after cALND. Among

the limitations of the study are failure to achieve target enrolment of 1900 patients, and also there was a possible randomization imbalance in favour of the SLND group, e.g. 7% in the SLND group were pN0 versus 1.2% in the ALND group and there were significantly more patients with micrometastasis only in the SLND group compared to the cALND group, 44.8% vs 37.5% (P ¼ 0.05). Furthermore, information on HER2 status is not available. However, based on this study the St. Gallen International Expert Consensus of 2011 recommended no further axillary surgery in patients fulfilling the inclusion criteria of the Z0011 trial [24]. The other randomised trial was initiated in 2001 (the IBCSG 23-01 trial), and it closed in 2010 after accruing 931 (47.5%) of the target enrolment number of 1960 patients. The trial reported data with 5 year median follow up [22]. In contrast to the Z0011 trial only patients with pN0(iþ) and pN1(mic) but not macrometastasis could enter the trial, and they were randomised to  cALND. In 95% of the patients only 1 SN contained metastasis, and 69% of the patients had
1mm disease in the axilla. Mastectomy was allowed, but 91% had BCS, and 97% of these had adjuvant RT. As in the Z0011 trial most patients had systemic therapy; 63e67% had endocrine treatment, 7e9% had chemotherapy, and 22e23% had combination systemic therapy. The results were very promising with no difference between the groups: 2% in both groups had local recurrence and 1 versus 5 patients had a regional recurrence in the cALND versus no cALND group, respectively. There were no differences in distant disease as well, thus axillary dissection can be avoided in selected patients without an effect on survival, whilst the regional morbidity was significantly reduced among patients who had no cALND as was also the case in the Z0011 trial. Patients with cN0 and SNþ breast cancer, where patients either had cALND or regional RT Non-randomised studies Only few studies have directly compared cALND and regional RT (Table 3) [25e28], and two of these were not randomised [25,28]. In the period 1983e2002, patients with T1eT2cN0 tumours (no SN technique) were treated with either ALND (n ¼ 80, median follow up >13 years), no axillary surgery but 2-field tangential breast irradiation (n ¼ 1134, median follow up 4.6 years), or no axillary surgery but 3-field RT (n ¼ 303, median follow up >10 years). All patients had BCS and adjuvant whole breast irradiation to 50 Gy. Chemotherapy (mostly CMF) was given to 80%, and in addition postmenopausal patients with ERþ tumours had tamoxifen. Median tumour size was 2.2e2.4 cm. The 10-year cumulative axillary recurrence rates were 1.3% vs 4.6% (P ¼ 0.21) for ALND compared to the RT-group. The 5-year axillary recurrence rates were 2.5% vs 1.7% (2-field vs 3-field RT) P ¼ 0.18, and the 5-year regional recurrence rates were 4.8% vs 2.4% (2-fields vs 3-fields) P ¼ 0.048. In the group of patients with tumours 3 cm the 5-year axillary and regional recurrence rates were 6.7% vs 1.2% (2 fields vs 3 fields) P ¼ 0.01. In multivariate analysis the risk of regional recurrence was associated with presence of lymphovascular invasion, outer tumour localisation and large tumour size. There was no difference in OS among groups. The other retrospective study included 180 patients 50 years old and operated with BCS for T1/T2 cN0 breast cancer and who had no ALND but regional RT and at least 2 years of tamoxifen [28]. These patients were compared to 341 patients with similar characteristics who had ALND and only in case of pNþ disease nodal RT was given. At 5 years there were very few regional recurrences (Table 3). Randomised trials With a median follow up of 15 years data from 658 patients treated at Institute Curie documented low risks of regional

Table 3 Studies reporting on minimum 50 patients with cN0, not always SNB, where patients were treated with either ALND or regional RT. Type

N

Year

Patients

HR status

FU median

pT1

Age

Results

Kamakura study [25]

-R

1517 pts T1eT2cN0, All BCS, all WBI A) 80 pts ALND B) 1134 pts tangential 2 field RT (no ALND) C) 303 pts 3-field RT (no ALND)

1983e2002

80% had chemotherapy. Postmenopausal pts with ERþ tumour also had tamoxifen.

A) 44/31% ERþ/PgRþ B) 62/62% ERþ/PgRþ C) 51/40% ERþ/PgRþ

A) 13.4 yr B) 4.6 yr C) 10.2 yr

38e47%

Median 43e48 yr

Deventer study [28]

-R

180 pts T1/T2 cN0 no ALND but reg RT þ tamoxifen 341 pts T1/T2 cN0 ALND (if pNþ the pt also had nodal RT)

1991e2000

In ALND pts: pN0:76.8% pN1:20.5% pN2:0.3%

7.2 yr

68% (no ALND) 80% (ALND)

50 yr

Institut Curie trial [26]

þR

658 pts, T <3 cm, cN0, <70 yr All BCS and WBI A) 326 pts. ALND B) 332 pts. No ALND but Ax RT given

1982e1987

HR unknown No ALND: 99% endocrine and 0.6% chemotherapy ALND: 22% endocrine and 10% chemotherapy 77% ER þ A) 19% received CMF, 14% endocrine therapy B) 9% received CMF, 8% endocrine therapy

Group A) 1 pt (1.3%) had axillary recurrence (ax rec) Group B) 35 (3%) had ax rec, 10 yr cumulative ax rec rate: 1.3% vs 4.6% (ALND vs RT) 5 yr ax rec rates 2.5% vs 1.7% (2 field vs 3 field). P ¼ 0.18 5 yr regional rec rates 4.8% vs 2.4% (2 vs 3 fields) P ¼ 0.048 (reg rec ¼ supraclav recþax rec) No difference in OS among groups Regional recurrence at 5 yr: 1.1% vs 1.5% in RT vs no-RT group. OS similar, but DFS better in RT group.

15 yr

67%

Mean age group A/B: 52/50.6 yr

The EORTC 10981 AMAROS trial [27]

þR

647 pts SNþ, cN0, 697 pts planned accrual R: pts SN with T 0.5e3 cm 1) ALND þ WBI 2) WBI þ reg RT Reg RT: 50 Gy/25 fr levels I þ II þ III and medial part of fossa supraclav

2001e2010

74%

Median 57 yr (24e87 yr)

Premeno: 28% Postmeno: 61% Grades I, II, III 28, 43, 25%

Not reported

21% in ALND group were pNþ (57% 1LNþ, 34% 2e3 LNþ, 9%>3 LNþ) At 15 yr isolated reg rec was 1% vs 3% in ALND vs RT group, P ¼ 0.04 Distant metastasis and overall survival was no different Type of met/further nodal disease Macromet 63%/41% Micromet 25%/18% Single cell 12%/18%

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

Study

Abbreviations: R, randomised; BCS, breast conserving surgery; Tam, tamoxifen; FU, follow up; HR, hormone receptor status; ER, oestrogen receptor; PgR, progesteron receptor; WBI, whole breast irradiation

S123

S124

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

recurrences [26]. The cN0 patients were all <70 years old and had BCS and adjuvant RT for tumours <3 cm. The RT was 55 Gy, 6 weeks, tumour bed boost of 10 Gy, and the regional RT was 45e 50 Gy. The randomisation was between cALND versus regional RT. All pts with positive nodes at ALND received RT to supraclavicular and internal mammary nodes as did also patients in both arms with central or medial lesions, thus there was an overlap of regional RT in both arms. In the ALND group 21% were pNþ (57% had 1 metastatic LN, 34% had 2e3 metastatic LN, 9% had >3 metastatic LN). At 15 years there was isolated regional recurrence in 1% vs 3% in ALND vs RT group, P ¼ 0.04, and no difference in distant disease or OS. In the period 2001e2010 the EORTC 10981-22023 AMAROS (After Mapping of the Axilla, Radiotherapy or Surgery?) trial accrued 647 patients being operated with BCS for a breast cancer with a positive SN and T 0.5e3 cm [27]. All patients had adjuvant whole breast irradiation, and they were randomised to either cALND or regional RT defined as RT to levels I þ II þ III and medial part of the supraclavicular fossa. Dose was 50 Gy/25 fractions. In the group of cALND the distribution of macro-/micro- and isolated tumour cells was 63/25 and 12% in the SNs, and after cALND there were further metastatic lymph nodes in the axilla in 41/18 and 18% of the patients, respectively. The SN was evaluated with H&E staining, and in H&E-negative nodes immunohistochemistry (IHC) was also performed. Data on regional control is expected to be presented at ASCO 2013 (personal communication M. Donker, study monitor of the AMAROS trial). Patients with cN0 breast cancer, where patients had no ALND but had regional RT Non-randomised studies In Table 4 we have listed 4 non-randomised studies reporting on the regional lymph node outcome after omitting ALND in cN0 patients, but where regional RT was given [29e32]. In the two oldest studies all patients had BCS followed by whole breast RT (45e55 Gy) and a tumour bed boost, and in most cases the patients also had regional nodes RT. No SNB was made in these studies, and in the study by Halvorsen et al. only 75 patients were treated with regional RT instead of ALND [29,30]. Systemic therapy was only given to 15e 30%, and despite of this the regional recurrence risk in both studies was quite low; in the group of no ALND but regional RT the risk was 2.7% at median 55 months compared to 0.3% in those patients having a median of 14 nodes removed at ALND (P ¼ 0.14) [29]. In the two studies from Boston a SNB was made and found positive. In the first study all patients were treated with no cALND (70%) or max 5 nodes removed (30%) followed by regional nodes RT to all patients [31], whilst in the latest study no patients had cALND and only half of the patients had regional nodes RT (most often with a 3-field technique) [32]. In both studies most of the patients had chemotherapy and/or endocrine therapy and also trastuzumab when indicated. Only very few regional recurrences were seen, in the first study 1.4% developed a regional recurrence at median 8 years [31], and no regional recurrences were seen in the latest study [32]. Randomised trial Only 1 trial has randomised highly selected cN0 patients to regional RT [33]. In the period 1995e1998, 435 patients with minimum age 45 years (age limit chosen in order to better estimate tumour size preoperatively on mammography), maximum tumour size 12 mm, and no palpable axillary nodes (no patient had SNB) were randomised to regional RT. All patients had a quadrantectomy followed by whole breast RT specifically designed to avoid axillary nodes, thus only half of the patients had regional RT 50 Gy/25 fractions. Only 15% of the patients had no systemic therapy. All patients had pT1 tumour (in 91% of the patients the

tumours were max 12 mm) and median age was 57 years. After median follow up of 63 months, only 4 axillary recurrences were identified: 3 cases in the no regional RT arm, and there were no differences in distant tumour control. Discussion The studies listed in Tables 1e4 are all investigating the regional recurrence risk in selected patients who have been treated with a less aggressive strategy towards the axilla than generally recommended, and in general the recurrence risks are very low almost irrespective of the type of regional therapy. However, in the studies with differences in the regional recurrence risk, the risk was always lowest in patients operated with cALND. It is important to notice that in the majority of the studies the patients are highly selected by being elderly women operated for small hormone receptor positive tumours, and systemic therapy has been prescribed to a high extent. Data from large randomised trials with axillary surgery We have data from large randomised trials on early nodepositive breast cancer patients both after mastectomy and lumpectomy demonstrating a significant gain in loco-regional recurrence (LRR) from adjuvant loco-regional RT using 50 Gy in 25 fractions [34e37]. With more than 18 years median follow up data from 3083 stage II and III high risk patients randomised to locoregional RT after mastectomy and axillary clearance in addition to systemic therapy has demonstrated a significant gain from locoregional RT (the target regional nodes included the internal mammary nodes, high axillary and supraclavicular nodes) in locoregional control [36]. The patients were treated in the Danish Breast Cancer Cooperative Group (DBCG) 82b&c protocol in the period 1982e1989 where the axillary surgery was less aggressive than today thus a median of 7 lymph nodes were removed. Premenopausal patients were treated with 9 series of CMF, and postmenopausal patients received 30 mg tamoxifen for 1 year. LRR was identified in 30% of the no-RT patients compared to only 5% in the RT group (P < 0.0001), and in the no-RT group the 18-year risk of isolated axillary recurrence was 10% among patients with 8 nodes removed compared to 18% among patients with fewer nodes removed (P ¼ 0.001). In the no-RT group the 18-year risk of isolated axillary failure was 4, 14 and 20% respectively in patients with 0, 1e 3 and 4þ metastatic lymph nodes (P < 0.0001). Among the 1545 patients in the no-RT group 143 and 34 patients respectively had an axillary and periclavicular recurrence, thus axillary recurrence represented 81% of the regional recurrences, whilst in the 1538 patients in the RT-group 10 and 10 patients, respectively, had axillary and periclavicular recurrence, thus regional RT modifies the pattern of recurrence [36]. Data from the NCIC-CTG MA.20 trial from Canada was presented at ASCO 2011 [37]. Women diagnosed with high-risk node-negative and node-positive early breast cancer treated with BCS and systemic therapy were randomised to whole breast RT versus whole breast RT þ regional nodes RT (45e50 Gy/ 25 fractions). The target regional nodes included the internal mammary nodes, high axillary and supraclavicular nodes. All patients had ALND. High-risk patients were node-negative with tumours 5 cm, or patients with tumours 2 cm and fewer than 10 nodes removed with either ER neg, grade III or lymphovascular invasion. The 1832 randomised patients were included in the period 2000e2007 and had 0 (10%), 1e3 (85%) or 4þ positive nodes, 91% had chemotherapy, and 71% had endocrine therapy. After median 62 months follow up there was a significantly improved locoregional DFS (HR ¼ 0.59, P ¼ 0.02, 5-year risk 96.8 and 94.5%, resp) in favour of regional nodes RT. The distant DFS was better

Table 4 Studies reporting on minimum 50 patients with clinically node-negative (cN0) disease, no axillary lymph node dissection (ALND), where patients were treated with regional radiotherapy (RT). Study

Study type

cN0, no ALND ±RT Mallinckrodt -R Institute study [29]

Year

Patients

HR status

FU median

pT1

Age

Results

511 pts stage IeII: 1) 351 pts: þALND (median 14 LN removed) 2) 74 pts: þALND þ reg RT 3) 75 pts: ALND þ reg RT 4) 21 pts: observation

1958e1988

All BCS and all had WBI 20% had chemotherapy þ endocrine therapy, 10% only endocrine therapy

Not reported

4.6 yr

NR

Median 55 yr

2.1% reg. rec, 1.2% axillary rec, 0.2% infraclavicular rec, 0.6% supraclavicular rec 0.2% IMN rec. The risk was 2.7% in group 3 vs 0.3% in group 1 (P ¼ 0.14), 0 reg rec in group 4 Significantly more arm and breast oedema after breast and reg RT compared to breast RT alone. 8 pts (2.4%) had regional recurrence Arm oedema was measured in 31 pts and there was no difference in measurements (ipsi/contralat) 6 pts (1.4%) developed reg failure In group 1, 3 pts (1%) had reg rec Transient radiation pneumonitis, brachial plexopathy or arm oedema was diagnosed in 1.2%

Yale-New Haven study [30]

-R

327 pts, clinical stage IeII

1962e1987

All BCS followed by RT including RT locoregional

41% ERþ 43% ER unknown 16% ER neg 5% had chemotherapy 10% had tamoxifen

10.2 yr

NR

72% >50 yr

Boston study [31]

-R

418 pts with clinical stage IeII, cN0, SNþ who had either no cALND or limited axillary dissection (LD) with removal of max 5 LN 1) 292 pts no cALND þ reg RT 2) 126 pts LD þ reg RT

1978e1987

In group 2, 84 pts had LD for pN0 disease, and 42 pts had pNþ disease

8 yr

68% (group 1), 50% (group 2, pN0), 45% (group 2, pNþ), P ¼ 0.03

Median 66 yr (group 1), 51 yr (group 2, pN0), 52 yr (group 2, pNþ), P ¼ 0.001

Boston study [32]

-R

130 consecutive pts SNþ and no cALND treated outside protocol 19% pN0(iþ) 53% pN1(mic) 28% pN1 (macro) 88% had adj RT, 51 % had reg RT (3/4 had a 3-field periclavicular field, the rest high tangents)

1999e2007

Number of removed nodes (mean) 2.8e3.5, mean pos nodes 1.1e1.6

4.9 yr

74%

Median 50 yr

No reg rec Selection bias

Milan trial [33]

þR

435 pts Inclusion criteria: >45 yr, cN0, T
12 mm, no SNB, no ALND. Randomisation to regional RT

1995e1998

All BCS  axillary RT (upper margin of clavicle, lat to the anterior axillary fold, medially to vertebral bodies, lower to 0.5 cm cranial to tangential fields)

72% ERþ (group 1) 40% ERþ (group 2, pN0) 66% ERþ (group 2, pNþ) P ¼ 0.0002 Adj Chemo/tamoxifen 2/2% (group 1) 6/1% (group 2, pN0) 81/7% (group 2, pNþ) (P < 0.0001), tam NS 72% Stage I 86% ERþ 16% HER2þ 78% BCS, all had RT and 51% (66 pts) had reg RT also (most with 3 fields, only 14 pts with high tangents) 82% Chemotherapy, 81% endocrine therapy 6% Herceptin 88% adj RT 80% Gr IeII, 81% ER þ 91% endocrine therapy 8% chemotherapy 1% chemo- and endocrine therapy

5.3 yr

100%, 91% were max 12 mm

Median 57 yr, all >45 yr

Axillary rec 3 cases in no RT, 1 case in þRT

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

N

Abbreviations: R, randomised; BCS, breast conserving surgery; Tam, tamoxifen; FU, follow up; HR, hormone receptor status; ER, oestrogen receptor; PgR, progesteron receptor; WBI, whole breast irradiation; IMN, internal mammary nodes.

S125

S126

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

with regional RT (HR ¼ 0.64, P ¼ 0.002, 5-year risk 92.4 and 87%, resp), and there was a trend towards better OS (HR ¼ 0.76, P ¼ 0.07, 5 year risk 92.3 and 90.7%, resp). Late morbidity was also evaluated and regional nodes RT resulted in more grade 2 or worse pneumonitis (1.3 and 0.2%, P ¼ 0.01) and more lymphoedema (7.3 and 4.1%, P ¼ 0.004). These data from randomised trials demonstrate that both after mastectomy and lumpectomy in high risk patients axillary lymph node dissection followed by regional nodes RT can provide a very high degree of regional control. In the 2005 Oxford overview results on 5-year local recurrence risk was provided for postmastectomy lymph node-positive patients randomised to RT, and for pN1 disease the absolute risk reduction by RT was 12% (SE 2%, reduction from 16 to 4%) and for patients with 4 positive nodes it was 14% (SE 2%, reduction from 26 to 12%) [1]. Regarding the node-negative situation, we have data from the EORTC boost versus no boost trial where 5569 randomised patients with 10.8 years median follow up have documented a very low regional recurrence risk. 78% of the patients were pN0. As first event 59 and 56 patients from the no boost and boost arm, respectively, were diagnosed with a regional recurrence, corresponding to 2.1% in total [38]. Benefit of systemic therapy on loco-regional disease The majority of patients included in Tables 1e4 were treated with systemic therapy. In order to get information on the benefit of modern systemic adjuvant treatment on the regional lymph node status it is helpful to review randomized studies comparing prewith postoperative chemotherapy. It is seen that some locoregional control is obtained due to especially combination chemotherapy with anthracycline and taxane containing regiments, and downstaging of the axilla is seen in up to 40% of the patients [39]. The number of pathological lymph nodes in the axilla is related to primary tumour response after preoperative chemotherapy. In the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-18, 751 patients with T1-3, N0-1, M0 were from 1988 to 1993 randomized to pre- or postoperative AC. Preoperative chemotherapy patients had a significantly increased incidence of pathologically negative axillary nodes compared with patients randomly assigned to postoperative chemotherapy (58% vs 42%) [39]. A significant correlation was found between pathological lymph node status and primary tumour response with less pathological lymph nodes found in patients with pCR in the breast [40]. In the EORTC 10902 study 698 patients with T1eT4b, N0-1, M0 breast cancer were randomized from 1991 to 1999 to pre- or postoperative CEF. 23% were downstaged by the preoperative chemotherapy and 63% were pathological lymph node negative compared with 51% of the patients randomized to postoperative chemotherapy [41]. In the NSABP B-27, 2411 patient with T1-3, N0-1, M0 disease were from 1995 to 2000 treated with preoperative AC and randomized to the following: surgery, surgery followed by docetaxel or docetaxel followed by surgery. Also in this study nodal status was highly correlated with the pathologic response of the breast tumour to preoperative chemotherapy. Only 15.5% of patients with a pCR in the breast had positive nodes by pathology; 35.4% of those with cCR had positive nodes by pathology. Among patients who had a cPR, 53.9% had histologically positive nodes; among those with clinically stable disease, 59.2% had histologically positive nodes; and among those with clinically progressive disease, 57.6% had histologically positive nodes [42]. Since postoperative chemotherapy equals preoperative chemotherapy in terms of disease free and overall survival [40e42] it is reasonable to assume that postoperative chemotherapy in a certain number of cases can eliminate nonsentinel positive nodes,

although identification of the best responders of systemic treatment is unknown. Patients should therefore be carefully selected when omitting cALND or regional RT in SNþ patients even with modern systemic adjuvant combination chemotherapy. Morbidity after adjuvant radiotherapy of early breast cancer It is well-documented that the morbidity is lower in patients operated with SNB compared to ALND, and that RT may increase this risk. Range of shoulder movement, oedema, numbness and tingling were investigated in 4000 patients included in the large phase III NSABP B-32 trial and found significantly lower in patients operated with SNB only, and the risk of morbidity was among other factors correlated with radiation to the axilla [43]. In another study more than 700 patients were treated with BCS followed by breast RT or breast and regional nodes RT [44]. In a subgroup of patients who had levels I and II axillary dissection the 10-year risk of lymphoedema was 10.7% for patients treated with breast and regional nodes RT compared to 1.0% in patients treated with breast RT only (P ¼ 0.0003). If the periclavicular lymph nodes are included in the fields the risk of paraesthesia increases, although this risk may be an indirect effect mediated by lymphoedema [45]. Life-threatening side-effects of adjuvant RT have also been described in the Oxford overview analysis [1,2]. Among irradiated compared to non-irradiated women there was a significantly higher risk of contra-lateral breast cancer (rate ratio 1.18, SE 0.06), and also an excess non-breast-cancer-mortality with a rate ratio 1.12, SE 0.04, primarily due to excess heart disease and lung cancer. This was primarily seen in the studies using older methods of RT, and the risk was most pronounced with longer follow up. Residual disease in the axilla SNB can reliably identify node-negative patients, and these patients are thus spared from the morbidities after ALND. The false negative rate of SNB is 10% or less, and the SN can be identified in the far majority of patients [46,47]. The clinical safety of omitting ALND after a negative SNB was documented in the NSABP B-32 trial, where 5611 cN0 patients were randomised to either SNB alone or SNB followed by cALND [48]. The trial documented with 8-year follow up that there was no difference in OS, DFS and locoregional control between the arms. Sentinel node positive patients have previously been offered cALND even if only isolated tumour cells were found in the SN. However, metaanalyses have documented that the risk of residual spread to non-sentinel nodes at cALND was 12% in patients with isolated tumour cells in the SN [49] and 20% in patients with micrometastasis in the SN [50]. After the publication of the Z0011 trial many institutions have changed guidelines, thus fewer SNþ patients are now routinely operated with cALND. Since the Z0011 trial was closed early and only included patients operated with BCS followed by whole breast RT, the results from this trial are not applicable to all breast cancer patients, and caution is thus warranted. In the 2 randomised trials listed in Table 1 the patients randomised to cALND had additional disease in the axilla. In the IBCSG Trial 10-93 the axillary status after surgery was 71% pN0, 20% pN1, 8% pN2 [11], and the corresponding numbers for the Milan trial was 77% pN0, 17% pN1, 6% pN2 [12,13]. In the AMAROS trial the frequence of further nodal disease depended strongly on the type of SN metastasis. In the 63% of the patients with macrometastasis in the SN, 41% had further nodal disease at the axillary operation; in 25% and 18% the SN had micrometastasis and single cells, and in both situations 18% had further nodal disease [27]. Thus it is well-

B.V. Offersen et al. / The Breast 22 (2013) S118eS128

known that residual disease is left in the axilla in quite many patients. There are no estimates on an acceptable risk of leaving residual disease in the axilla, but several groups have proposed models and nomograms to predict the likelihood of residual disease in the axilla [51e53], and lately an international multicenter tool to predict the risk of residual disease in the axilla has been proposed [54]. This model is based on the following important factors: size of SN metastasis (ITC, micrometastasis or macrometastasis), presence of non-SN metastasis, lymphovascular invasion, multifocality, HER-2 status, number of negative and positive SN, T size, and extracapsular extension. The highest weight in the equation is given to the size of the metastasis. Potentially such a nomogram may be used to help the clinician and patient in deciding what is the best strategy for the individual patient.

S127

criteria for patient selection, the authors base their recommendation on 3 different nomograms applied on 5 different patient cases ranging from recommending only breast RT to full nodal RT. Thus absolute criteria for a selection to replace ALND with regional RT are still missing. Once we are able to select the optimal patient for regional RT instead of ALND, we also need to consider what the optimal target is for the regional RT in this situation. Is it necessary to include levels I, II, III and the supraclavicular fossa as in the AMAROS trial, or are levels I and II sufficient? Thus, further studies are highly needed. Conflict of Interest Statement None declared. Acknowledgements

Is regional nodes radiotherapy an alternative to dissection? The regional recurrence risk is fortunately historically low, and this has stimulated an interest into investigating less morbidity associated strategies towards the axilla. The studies listed in Tables 1e4 in this review all report low or very low regional recurrence risks, except in 2 studies which both reported risks higher than 1% per year [7,9]. Over the last decades the improvements in handling and treatment of early breast cancer have been considerable. We have witnessed introduction of mammography screening, a more specialised diagnostic workup of the patients who now almost always have regional nodes ultrasound before surgery. The SNB identifies with high certainty the relevant lymph node thus assuring a correct staging of the patient. The indications for systemic therapy have been modified, and the trend has always been to include more patients in the group to be treated with increasingly effective systemic chemotherapy, endocrine therapy and targeted therapy. The radiotherapy has for the last 10 years in most centres been 3D planned, and target delineation in the planning of RT has now gained focus. Recently the first national consensus for delineation of targets in early breast cancer was published from the DBCG [55], and work is now in progress to reach an ESTRO consensus within this year. All these factors taken together will further improve the prognosis of patients diagnosed with BC. An answer to the question if regional RT may be an alternative to dissection may be given with the results from the AMAROS trial later this year. Judged from the general results listed in Tables 1e4 and based on the fact that 74% of the tumours in the AMAROS trial were pT1, 71% were grades I/II and 61% were postmenopausal we may see that there will be only very few regional recurrences in the trial in total. We need more data on biological characteristics to identify relevant patients to be treated with nodal RT instead of cALND. Perhaps the 21-gene OncotypeDX recurrence score assay or other assays can be used to identify patients with a particularly high risk of regional recurrence [56,57]. Unfortunately, there is a lack of information in all the studies included in Tables 1e4 in this review regarding the regional recurrence risk in relation to biological classification, but hopefully more studies will report on this in the future. In 2010 A. Recht suggested that it could be acceptable to irradiate the breast only in patients operated with BCS but no cALND for a T1 tumour with pN0(iþ) in the SN [58]. Patients with pN1(mic) in 1 SN operated with BCS for a T1 tumour grades 1 or 2, ER pos and without lymphovascular invasion should receive breast and axillary RT, and all other patients should have cALND. A year later the Z0011 trial proved this to be overtreatment, and these patients are now in general treated without axillary RT. Haffty et al. have also proposed criteria for selection of patients to be treated with regional nodes RT instead of cALND [59]. However, rather than proposing absolute

BVO supported by grant from The Danish Cancer Society, CIRRO (The Lundbeck Foundation Centre for Interventional Research in Radiation Research), and The Danish Research Council. References [1] Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans E, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;366:2087e106. [2] Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;365:1687e717. [3] Lyman GH, Giuliano AE, Somerfield MR, Benson AB, Bodurka DC, Burstein HJ, et al. American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 2005;23: 7703e20. [4] Silverstein MJ, Recht A, Lagios MD, Bleiweiss IJ, Blumencranz PW, Gizienski T, et al. Special report: consensus conference III. Image-detected breast cancer: state-of-the-art diagnosis and treatment. J Am Coll Surg 2009;209:504e20. [5] Rescigno J, Zampell JC, Axelrod D. Patterns of axillary surgical care for breast cancer in the era of sentinel lymph node biopsy. Ann Surg Oncol 2009;16: 687e96. [6] Martelli G, Miceli R, Daidone MG, Vetrella G, Cerrotta AM, Piromalli D, et al. Axillary dissection versus no axillary dissection in elderly patients with breast cancer and no palpable axillary nodes: results after 15 years of follow-up. Ann Surg Oncol 2011;18:125e33. [7] Greco M, Agresti R, Cascinelli N, Casalini P, Giovanazzi R, Maucione A, et al. Breast cancer patients treated without axillary surgery: clinical implications and biologic analysis. Ann Surg 2000;232:1e7. [8] Wong JS, Recht A, Beard CJ, Busse PM, Cady B, Chaffey JT, et al. Treatment outcome after tangential radiation therapy without axillary dissection in patients with early-stage breast cancer and clinically negative axillary nodes. Int J Radiat Oncol Biol Phys 1997;39:915e20. [9] McKinna F, Gothard L, Ashley S, Ebbs S, Yarnold J. Selective avoidance of lymphatic radiotherapy in the conservative management of women with early breast cancer. Radiother Oncol 1999;52:219e23. [10] Wong JS, Taghian AG, Bellon JR, Keshaviah A, Smith BL, Winer EP, et al. Tangential radiotherapy without axillary surgery in early-stage breast cancer: results of a prospective trial. Int J Radiat Oncol Biol Phys 2008;72:866e70. [11] Rudenstam CM, Zahrieh D, Forbes JF, Crivellari D, Holmberg SB, Rey P, et al. Randomized trial comparing axillary clearance versus no axillary clearance in older patients with breast cancer: first results of International Breast Cancer Study Group Trial 10-93. J Clin Oncol 2006;24:337e44. [12] Martelli G, Boracchi P, De PM, Pilotti S, Oriana S, Zucali R, et al. A randomized trial comparing axillary dissection to no axillary dissection in older patients with T1N0 breast cancer: results after 5 years of follow-up. Ann Surg 2005;242:1e6. [13] Martelli G, Boracchi P, Ardoino I, Lozza L, Bohm S, Vetrella G, et al. Axillary dissection versus no axillary dissection in older patients with T1N0 breast cancer: 15-year results of a randomized controlled trial. Ann Surg 2012;256: 920e4. [14] Hwang RF, Gonzalez-Angulo AM, Yi M, Buchholz TA, Meric-Bernstam F, Kuerer HM, et al. Low locoregional failure rates in selected breast cancer patients with tumor-positive sentinel lymph nodes who do not undergo completion axillary dissection. Cancer 2007;110:723e30. [15] de Boer M, van Deurzen CH, van Dijck JA, Borm GF, van Diest PJ, Adang EM, et al. Micrometastases or isolated tumor cells and the outcome of breast cancer. N Engl J Med 2009;361:653e63. [16] Park J, Fey JV, Naik AM, Borgen PI, Van Zee KJ, Cody HS. A declining rate of completion axillary dissection in sentinel lymph node-positive breast cancer

S128

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

B.V. Offersen et al. / The Breast 22 (2013) S118eS128 patients is associated with the use of a multivariate nomogram. Ann Surg 2007;245:462e8. Bilimoria KY, Bentrem DJ, Hansen NM, Bethke KP, Rademaker AW, Ko CY, et al. Comparison of sentinel lymph node biopsy alone and completion axillary lymph node dissection for node-positive breast cancer. J Clin Oncol 2009;27: 2946e53. Setton J, Cody H, Tan L, Morrow M, Hudis C, Catalano JS, et al. Radiation field design and regional control in sentinel lymph node-positive breast cancer patients with omission of axillary dissection. Cancer 2012;118:1994e2003. Milgrom S, Cody H, Tan L, Morrow M, Pesce C, Setton J, et al. Characteristics and outcomes of sentinel node-positive breast cancer patients after total mastectomy without axillary-specific treatment. Ann Surg Oncol 2012;19: 3762e70. Pepels MJ, de Boer M, Bult P, van Dijck JA, van Deurzen CH, MenkePluymers MB, et al. Regional recurrence in breast cancer patients with sentinel node micrometastases and isolated tumor cells. Ann Surg 2012;255: 116e21. Giuliano AE, Hunt KK, Ballman KV, Beitsch PD, Whitworth PW, Blumencranz PW, et al. Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: a randomized clinical trial. JAMA 2011;305:569e75. Galimberti V, Bernard FC, Zurrida S, Viale G, Luini A, Veronesi P, et al. Axillary dissection versus no axillary dissection in patients with sentinel-node micrometastases (IBCSG 23-01): a phase 3 randomised controlled trial. Lancet Oncol 2013;14:297e305. Schlembach PJ, Buchholz TA, Ross MI, Kirsner SM, Salas GJ, Strom EA, et al. Relationship of sentinel and axillary level IeII lymph nodes to tangential fields used in breast irradiation. Int J Radiat Oncol Biol Phys 2001;51:671e8. Goldhirsch A, Wood WC, Coates AS, Gelber RD, Thurlimann B, Senn HJ. Strategies for subtypes e dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol 2011;22:1736e47. Fujimoto N, Amemiya A, Kondo M, Takeda A, Shigematsu N. Treatment of breast carcinoma in patients with clinically negative axillary lymph nodes using radiotherapy versus axillary dissection. Cancer 2004;101:2155e63. Louis-Sylvestre C, Clough K, Asselain B, Vilcoq JR, Salmon RJ, Campana F, et al. Axillary treatment in conservative management of operable breast cancer: dissection or radiotherapy? Results of a randomized study with 15 years of follow-up. J Clin Oncol 2004;22:97e101. Straver ME, Meijnen P, Van TG, van der Velde CJ, Mansel RE, Bogaerts J, et al. Sentinel node identification rate and nodal involvement in the EORTC 1098122023 AMAROS trial. Ann Surg Oncol 2010;17:1854e61. Spruit PH, Siesling S, Elferink MA, Vonk EJ, Hoekstra CJ. Regional radiotherapy versus an axillary lymph node dissection after lumpectomy: a safe alternative for an axillary lymph node dissection in a clinically uninvolved axilla in breast cancer. A caseecontrol study with 10 years follow up. Radiat Oncol 2007;2:40. Halverson KJ, Taylor ME, Perez CA, Garcia DM, Myerson R, Philpott G, et al. Regional nodal management and patterns of failure following conservative surgery and radiation therapy for stage I and II breast cancer. Int J Radiat Oncol Biol Phys 1993;26:593e9. Haffty BG, McKhann C, Beinfield M, Fischer D, Fischer JJ. Breast conservation therapy without axillary dissection. A rational treatment strategy in selected patients. Arch Surg 1993;128:1315e9. Galper S, Recht A, Silver B, Bernardo MW, Gelman R, Wong J, et al. Is radiation alone adequate treatment to the axilla for patients with limited axillary surgery? Implications for treatment after a positive sentinel node biopsy. Int J Radiat Oncol Biol Phys 2000;48:125e32. Barkley C, Burstein H, Smith B, Bellon J, Wong J, Gadd M, et al. Can axillary node dissection be omitted in a subset of patients with low local and regional failure rates? Breast J 2012;18:23e7. Veronesi U, Orecchia R, Zurrida S, Galimberti V, Luini A, Veronesi P, et al. Avoiding axillary dissection in breast cancer surgery: a randomized trial to assess the role of axillary radiotherapy. Ann Oncol 2005;16:383e8. Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson M, Bach F, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997;337:949e55. Overgaard M, Jensen MB, Overgaard J, Hansen PS, Rose C, Andersson M, et al. Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial. Lancet 1999;353:1641e8. Nielsen HM, Overgaard M, Grau C, Jensen AR, Overgaard J. Loco-regional recurrence after mastectomy in high-risk breast cancer e risk and prognosis. An analysis of patients from the DBCG 82 b&c randomization trials. Radiother Oncol 2006;79:147e55. Whelan T, Olivotto IA, Ackerman, Chapman JW, Chau B, Nabid A, et al. NCICCTG MA.20: an intergroup trial of regional nodal irradiation in early breast cancer. J Clin Oncol 6.6.2011.

[38] Bartelink H, Horiot JC, Poortmans PM, Struikmans H, van den Bogaert W, Fourquet A, et al. Impact of a higher radiation dose on local control and survival in breast-conserving therapy of early breast cancer: 10-year results of the randomized boost versus no boost EORTC 22881-10882 trial. J Clin Oncol 2007;25:3259e65. [39] Rastogi P, Anderson SJ, Bear HD, Geyer CE, Kahlenberg MS, Robidoux A, et al. Preoperative chemotherapy: updates of National Surgical Adjuvant Breast and Bowel Project Protocols B-18 and B-27. J Clin Oncol 2008;26:778e85. [40] Wolmark N, Wang J, Mamounas E, Bryant J, Fisher B. Preoperative chemotherapy in patients with operable breast cancer: nine-year results from National Surgical Adjuvant Breast and Bowel Project B-18. J Natl Cancer Inst Monogr 2001:96e102. [41] Van Der Hage JA, van de Velde, Julien JP, Tubiana-Hulin M, Vandervelden C, Duchateau L. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001;19:4224e37. [42] Bear HD, Anderson S, Brown A, Smith R, Mamounas EP, Fisher B, et al. The effect on tumor response of adding sequential preoperative docetaxel to preoperative doxorubicin and cyclophosphamide: preliminary results from National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol 2003;21:4165e74. [43] Ashikaga T, Krag DN, Land SR, Kirsner SM, Salas GJ, Strom EA, et al. Morbidity results from the NSABP B-32 trial comparing sentinel lymph node dissection versus axillary dissection. J Surg Oncol 2010;102:111e8. [44] Coen JJ, Taghian AG, Kachnic LA, Assaad SI, Powell SN. Risk of lymphedema after regional nodal irradiation with breast conservation therapy. Int J Radiat Oncol Biol Phys 2003;55:1209e15. [45] Lundstedt D, Gustafsson M, Steineck G, Alsadius D, Sundberg A, Wilderang U, et al. Long-term symptoms after radiotherapy of supraclavicular lymph nodes in breast cancer patients. Radiother Oncol 2012;103:155e60. [46] Clarke D, Newcombe RG, Mansel RE. The learning curve in sentinel node biopsy: the ALMANAC experience. Ann Surg Oncol 2004;11:211Se5S. [47] McMasters KM, Wong SL, Chao C, Woo C, Tuttle TM, Noyes RD, et al. Defining the optimal surgeon experience for breast cancer sentinel lymph node biopsy: a model for implementation of new surgical techniques. Ann Surg 2001;234: 292e9. [48] Krag DN, Anderson SJ, Julian TB, Brown AM, Harlow SP, Constantino JP, et al. Sentinel-lymph-node resection compared with conventional axillary-lymph-node dissection in clinically node-negative patients with breast cancer: overall survival findings from the NSABP B-32 randomised phase 3 trial. Lancet Oncol 2010;11:927e33. [49] van Deurzen CH, de Boer M, Monninkhof EM, Bult P, van der Wall E, TjanHeijnen VC, et al. Non-sentinel lymph node metastases associated with isolated breast cancer cells in the sentinel node. J Natl Cancer Inst 2008;100: 1574e80. [50] Cserni G, Gregori D, Merletti F, Sapino A, Mano MP, Ponti A, et al. Metaanalysis of non-sentinel node metastases associated with micrometastatic sentinel nodes in breast cancer. Br J Surg 2004;91:1245e52. [51] Van Zee KJ, Manasseh DM, Bevilacqua JL, Boolbol SK, Fey JV, Tan LK, et al. A nomogram for predicting the likelihood of additional nodal metastases in breast cancer patients with a positive sentinel node biopsy. Ann Surg Oncol 2003;10:1140e51. [52] Coutant C, Olivier C, Lambaudie E, Fondrinier E, Marchal F, Guillemin F, et al. Comparison of models to predict nonsentinel lymph node status in breast cancer patients with metastatic sentinel lymph nodes: a prospective multicenter study. J Clin Oncol 2009;27:2800e8. [53] Tvedskov TF, Jensen MB, Lisse IM, Ejlertsen B, Balslev E, Kroman N. High risk of non-sentinel node metastases in a group of breast cancer patients with micrometastases in the sentinel node. Int J Cancer 2012;131:2367e75. [54] Meretoja TJ, Leidenius MH, Heikkila PS, Boross PS, Sejben I, Regitnig P, et al. International multicenter tool to predict the risk of nonsentinel node metastases in breast cancer. J Natl Cancer Inst 2012;104:1888e96. [55] Nielsen MH, Berg M, Pedersen AN, Andersen K, Glavicic V, Jakobsen EH, et al. Delineation of target volumes and organs at risk in adjuvant radiotherapy of early breast cancer: national guidelines and contouring atlas by the Danish Breast Cancer Cooperative Group. Acta Oncol 2013;52:703e10. [56] Mamounas EP, Tang G, Fisher B, Paik S, Shak S, Costantino JP, et al. Association between the 21-gene recurrence score assay and risk of locoregional recurrence in node-negative, estrogen receptor-positive breast cancer: results from NSABP B-14 and NSABP B-20. J Clin Oncol 2010;28:1677e83. [57] Oliveira M, Cortes J, Bellet M, Balmana J, De Mattos-Arruda L, Gomez P, et al. Management of the axilla in early breast cancer patients in the genomic era. Ann Oncol 2013;24:1163e70. [58] Recht A. Can radiotherapy replace axillary dissection for patients with positive sentinel nodes? Breast Dis 2010. [59] Haffty BG, Hunt KK, Harris JR, Buchholz TA. Positive sentinel nodes without axillary dissection: implications for the radiation oncologist. J Clin Oncol 2011;29:4479e81.