Sentinel lymph node biopsy in microinvasive ductal carcinoma in situ

Abstract

Background

Microinvasive breast cancer is an uncommon pathological entity. Owing to the rarity of this condition, its surgical axillary management and overall prognosis remain controversial.

Methods

A database was analysed to identify patients with microinvasive ductal carcinoma in situ (DCIS) who had surgery for invasive breast cancer at the European Institute of Oncology, Milan, between 1998 and 2010. Women who had undergone axillary staging by sentinel lymph node biopsy were included in the study.

Results

Of 257 women with microinvasive breast cancer who underwent sentinel lymph node biopsy (SLNB), 226 (87·9 per cent) had negative sentinel lymph nodes (SLNs) and 31 had metastatic SLNs. Twelve patients had isolated tumour cells (ITCs), 14 had micrometastases and five had macrometastases in sentinel nodes. Axillary lymph node dissection was performed in 16 of the 31 patients with positive SLNs. After a median follow-up of 11 years, only one regional first event was observed in the 15 patients with positive SLNs who did not undergo axillary lymph node dissection. There were no regional first events in the 16 patients with positive SLNs who had axillary dissection.

Conclusion

Good disease-free and overall survival were found in women with positive SLNs and microinvasive DCIS. This study is in line with studies showing that SLNB in microinvasive DCIS may not be useful, and supports the evidence that less surgery can provide the same level of overall survival with better quality of life.

No added value

Introduction

Microinvasive breast cancer is an uncommon pathological entity, accounting for approximately 1 per cent of all breast cancers^1,2. The definition of microinvasive breast cancer has varied over time^3,4.

Recently, the definition of microinvasion, as given by the seventh edition of the AJCC staging manual⁵, of extension of cancer cells beyond the basement membrane into the adjacent tissue with no focus more than 1 mm in greatest dimension, has gained common acceptance. As a result, the term ‘T1mic’ has now been added to the TNM staging system^5,6.

Owing to the rarity of this condition, questions remain regarding the surgical management of the axilla and the overall prognosis of this entity. In the literature, a large incidence spectrum of axillary metastasis is found. This can be attributable to differing definitions of microinvasive ductal carcinoma in situ (DCIS) over the years and to the varying techniques used to analyse the sentinel node. These differences are probably responsible for the different recommendations on how to manage the axilla in microinvasive DCIS⁷.

DCIS is a disease devoid of invasive behaviour and thus without potential for spread to the axillary lymph nodes. Current practice is to perform sentinel lymph node biopsy (SLNB) only in selected patients with DCIS when there is substantial risk of upgrade of the lesion at final pathology, such as a mass lesion highly suggestive of invasive cancer at imaging and physical examination, patients with a large area of DCIS at imaging (5 cm or greater), or when mastectomy is indicated⁸. However, evidence for this recommendation is inadequate because of the sparsity of data analysed in the literature, also characterized by a lack of long-term follow-up studies and still subject to controversial scientific analysis⁹.

If staging the axilla in DCIS is accepted globally in the above conditions, what remains controversial is the real value of staging the axilla with SLNB in microinvasive DCIS^{4,6–7,10–31}, as reviewed in Table 1^{4,6–7,10–24,27–31}. The incidence of axillary metastasis in sentinel nodes varies in studies from approximately 2 to 20 per cent. This is probably due to the different pathological methods used to examine the sentinel node, as well as differences in the methodology used to section the breast tissue. Factors correlated with axillary nodal positivity in women with DCIS and microinvasive DCIS are younger age, size of DCIS lesion, histological grade, receptor status, human epidermal growth factor receptor (HER) 2 overexpression and lymphovascular invasion^6,23,25.

Table 1.

Literature review of selected studies of microinvasive ductal carcinoma in situ in patients who had sentinel lymph node biopsy

Reference	Year	Total no. of patients with axillary staging	Patients submitted to SLNB	Type of metastasis (AJCC criteria)			SLNB positivity (%)
				ITCs	Micro	Macro
Without defined SLNB status
Cox et al.11	2001	15	15	n.s.	n.s.	n.s.	3 (20)
Camp et al.10	2005	13	13	n.s.	n.s.	n.s.	2 (15)
Wilkie et al.29	2005	51	51	5	n.s.	n.s.	7 (14)
Tunon-de-Lara et al.28	2008	45	45	0	n.s.	n.s.	2 (4)
Fortunato et al.12	2008	77	77	n.s.	n.s.	n.s.	6 (8)
Vieira et al.6	2010	17	14	n.s.	n.s.	n.s.	1 (6)
Parikh et al24	2012	46	4	n.s.	n.s.	n.s.	1 (2)
With defined SLNB status
Zavotsky et al.31	1999	14	14	1	0	1	2 (14)
Klauber-DeMore et al.19	2000	31	31	0	2	1	3 (10)
Intra et al.16	2003	41	41	0	2	2	4 (10)
Katz et al.18	2006	21	21	0	1	1	2(10)
Leidenius et al.21	2006	11	11	1	0	0	1 (9)
Zavagno et al.30	2007	43	43	0	1	3	4 (9)
Gray et al.13	2007	79	77	2	2	2	6 (8)
Guth et al.14	2008	44	20	2	0	3	5 (11)
Sakr et al.27	2008	20	20	0	2	0	2 (10)
Lyons et al.7	2012	112	112	6	5	3	14 (12·5)
Ko et al.20	2012	293	180	6	12	4	22 (7·5)
Kapoor et al.17	2013	45	31	4	4	1	9 (20)
Margalit et al.22	2013	68	53	4	3	0	7 (10)
Matsen et al.23	2014	414	414	0	26	6	32 (7·7)
Hanna et al.15	2014	81	64	2	0	0	2 (2)
Orzalesi et al.4	2016	126	126	10	3	5	18 (14·3)

Values in parentheses are percentages. SLNB, sentinel lymph node biopsy; ITC, isolated tumour cell; micro, micrometastases; macro, macrometastases; n.s., not stated.

To contribute to a better understanding of this surgical aspect, as well as to the prognostic implications of microinvasion, this retrospective observational study examined patients with microinvasive breast cancer who underwent axillary staging via SLNB.

Methods

After institutional review board approval, a database of patients who underwent surgery for invasive breast cancer at the European Institute of Oncology, Milan, Italy, between 1998 and 2010 was analysed, and patients with microinvasive DCIS were identified. Patients who did not undergo axillary surgery were excluded, and the remaining patients with microinvasive breast cancer who had undergone axillary staging by SLNB were included in the analysis.

Sentinel lymph node (SLN) identification was usually performed using a radiocolloid technique (^99mTc-labelled colloidal particles of human albumin). Intraoperative lymph node analysis was conducted using haematoxylin and eosin-stained sections, aided by immunohistochemical staining when necessary, as has been reported previously³².

Based on AJCC classification criteria⁵, axillary lymph node metastases were defined as follows: macrometastases (larger than 2·0 mm), micrometastases (0·2–2·0 mm) or isolated tumour cells (ITCs) (smaller than 0·2 mm). Systemic adjuvant therapy was recommended according to the contemporary St Gallen treatment guidelines^33–37.

The following parameters were used in the analysis: clinical (year of surgery, age, menopausal status), pathology (tumour histology, tumour grade, tumour subtype) and type of treatment (local or systemic). Long-term outcomes were studied via follow-up data recording the first recurrence events, classified as local (ipsilateral breast and chest), regional (ipsilateral axillary or supraclavicular lymph nodes), distant metastasis, contralateral breast cancer, other primary tumour and death as the first-reported event.

Statistical analysis

Demographic and clinical characteristics of the study sample were analysed using descriptive statistics. The association between SLN status, and demographic and clinical variables was evaluated using the χ² test. Cumulative incidences of the first observed relapse (categorized as local recurrence, regional recurrence or distant metastasis) were assessed from the date of surgery to the date of event. In case of no event, the observation was censored at the last follow-up visit. Cumulative incidence functions were estimated according to the method described by Kalbfleisch and Prentice³⁸, taking into account the competing causes of relapse. Gray's test³⁹ was used to assess cumulative incidence differences between groups.

Overall survival (OS) was defined as the time from date of surgery to date of death from any cause; disease-free survival (DFS) was defined, according to standardized definitions for efficacy end points (STEEP) criteria⁴⁰, as the time from surgery to events such as relapse (including ipsilateral breast tumour recurrence), appearance of a second primary cancer (including contralateral breast cancer) or death, whichever occurred first. OS and DFS curves were estimated using the Kaplan–Meier method, and the log rank test was used to assess differences between groups.

Median follow-up was calculated using the reverse Kaplan–Meier method⁴¹. All analyses were performed using SAS® software version 9.4 (SAS Institute, Cary, North Carolina, USA). All statistical tests were two-sided.

Results

Of 22 120 patients in the database, 310 with microinvasive DCIS were identified. Fifty-three were excluded as they did not undergo axillary surgery, and the remaining 257 patients (82·9 per cent) with microinvasive breast cancer who had axillary staging by SLNB were included in the analysis. Of these 257 women, 161 (62·6 per cent) had only one SLN, 57 (22·2 per cent) had two SLNs, 26 (10·1 per cent) had three SLNs and 13 patients (5·1 per cent) had more than three SLNs removed.

Sentinel node metastasis and tumour characteristics

Negative SLNs were found in 226 of the 257 women (87·9 per cent). In one of these 226 patients, axillary dissection was performed owing to micrometastasis in an additional level 1 lymph node removed at the time when this still was an institutional criterion for axillary dissection. A total of 31 women presented with metastatic SLNs: 12 with ITCs, 14 with micrometastases and five with macrometastases. Thus, the overall rate of metastasis in the SLN was 12·1 per cent (31 of 257), with macrometastasis in 1·9 per cent, micrometastasis in 5·4 per cent and ITCs in 4·7 per cent (Table S1, supporting information). All patients with metastatic SLNs had ductal histology of the breast cancer. A higher proportion with positive SLNs were found in luminal B (31 per cent) and triple-negative (21 per cent) subtypes compared with other subtypes (Table 2).

Table 2.

Patient characteristics according to sentinel lymph node status

	SLN status*			All patients (n = 257)†
	Negative (n = 226)	Positive (n = 31)	P
Year of surgery			0·915
Before 2003	42 (89)	5 (11)		47 (18·3)
2003–2006	111 (88·1)	15 (11·9)		126 (49·0)
2007–2010	73 (87)	11 (13)		84 (32·7)
Age (years)			0·079
< 50	91 (82·7)	19 (17·3)		110 (42·8)
50–59	69 (91)	7 (9)		76 (29·6)
≥ 60	66 (93)	5 (7)		71 (27·6)
Menopausal status			0·030
Premenopausal	99 (83·2)	20 (16·8)		119 (46·3)
Postmenopausal	127 (92·0)	11 (8·0)		138 (53·7)
Histology			0·177
Ductal	203 (86·8)	31 (13·2)		234 (91·1)
Lobular	8 (100)	0 (0)		8 (3·1)
Other	15 (100)	0 (0)		15 (5·8)
Grade			0·511
G1	36 (88)	5 (12)		41 (16·0)
G2	87 (90)	10 (10)		97 (37·7)
G3	83 (85)	15 (15)		98 (38·1)
Unknown	20 (95)	1 (5)		21 (8·2)
Subtype			0·387
Unknown	23 (88)	3 (12)		26 (10·1)
Luminal A	70 (91)	7 (9)		77 (30·0)
Luminal B (Ki67 ≥ 20%)	22 (81)	5 (19)		27 (10·5)
Luminal B (HER2-positive)	22 (88)	3 (12)		25 (9·7)
HER2-positive	66 (90)	7 (10)		73 (28·4)
Triple negative	23 (79)	6 (21)		29 (11·3)
Local treatment			0·206
Mastectomy without radiotherapy	49 (82)	11 (18)		60 (23·3)
Mastectomy with radiotherapy	27 (87)	4 (13)		31 (12·1)
Quadrantectomy with radiotherapy	150 (90·4)	16 (9·6)		166 (64·6)
Systemic treatment			< 0·001
None	109 (93·2)	8 (6·8)		117 (45·5)
Endocrine therapy	110 (89·4)	13 (10·6)		123 (47·9)
Chemotherapy	5 (36)	9 (64)		14 (5·4)
Chemotherapy + endocrine therapy	2 (67)	1 (33)		3 (1·2)

Values in parentheses are percentages of

^*row and.

^†column. SLN, sentinel lymph node; HER2, human epidermal growth factor receptor 2.

Axillary surgery

Axillary dissection was performed in 16 of the 31 women with positive SLNs: one patient with ITCs, ten with micrometastasis and five with macrometastasis. The five patients with macrometastasis had no more than three positive lymph nodes at final histological examination (pN1a). The remaining 15 women (11 with ITCs and 4 with micrometastasis of the SLN) were diagnosed in the later period (from 2004 onwards) and were thus not subjected to axillary dissection. Table 2 shows the clinical and pathological characteristics of the women in the study, according to lymph node status.

Breast surgery

Breast-conserving surgery (BCS) was performed in 166 of the 257 women (64·6 per cent); of these, 150 (90·4 per cent) had negative and 16 (9·6 per cent) had positive SLNs. A total of 91 women (35·4 per cent) had a mastectomy, with conservation of the nipple–areola complex and immediate reconstruction in most cases. Of these 91 women, 76 (84 per cent) had negative and 15 (16 per cent) had positive SLNs. Of the 31 women who had nipple-sparing mastectomy and received intraoperative radiotherapy of the nipple–areola complex, 27 had negative and four had positive SLNs.

Adjuvant treatment

Adjuvant endocrine treatment alone was given to 123 of the 257 women (47·9 per cent) who underwent SLNB; 14 (5·4 per cent) received chemotherapy alone and three women (1·2 per cent) had both chemotherapy and endocrine therapy (Table 2). The distribution of treatment by SLN status is shown in Table S1 (supporting information).

Of the 53 women who did not undergo SLNB, 21 received endocrine therapy alone; eight of these were classified as having the luminal A subtype. Four patients received chemotherapy alone: one had the luminal B subtype with Ki67 of 20 per cent or above, one patient had HER2+ cancer, one had the triple-negative subtype, and in one patient information to determine tumour subtype was missing. Five patients received endocrine therapy plus chemotherapy: two had the luminal A subtype, one had the luminal B subtype (HER2+), one had the triple-negative subtype, and one patient had insufficient information to ascertain the tumour subtype. The remaining 23 patients did not receive adjuvant treatment in accordance with pathological tumour stage.

Recurrence and survival

The median duration of follow-up was 11 years, with 2765 cumulative person-years. At median follow-up, 14 deaths and 69 first events were observed. Seventeen local recurrences, six regional recurrences and six distant metastases were observed among the 226 SLN-negative patients. In the SLN-positive group without further axillary dissection, two local events, one regional event and one case of distant metastasis were observed, whereas in the SLN-positive group that had subsequent axillary dissection, there were two local events and one case of distant metastasis. Fig. 1a shows the cumulative incidence of events over 15 years of follow-up. The estimated 10-year cumulative incidence of local, regional and distant recurrence was 7·0, 2·9 and 3·2 per cent respectively. DFS and OS are shown in Fig. 1b and 1c respectively. The estimated 10-year DFS rate was 77·5 per cent, and the estimated 10-year OS rate was 94·8 per cent. The cumulative incidence of regional and local recurrence in relation to SLN status and its associated surgical axillary treatment (SLN-negative or SLN-positive followed or not by axillary dissection) is shown in Fig. 2a,b.

Fig. 1.

Cumulative incidence of local, regional and distant events, disease-free and overall survival. a Local, regional and distant events, b disease-free survival and c overall survival

Fig. 2.

Cumulative incidence of regional and local events according to sentinel lymph node status and axillary dissection. a Regional events, b local events. SLN, sentinel lymph node. aP = 0·495, bP = 0·628 (Gray's test)

Discussion

In the present study, the incidence of SLNB metastasis was 12·1 per cent in patients with microinvasive breast cancer, which falls within the range described in the literature. The rate of macrometastasis was low (1·9 per cent). Moreover the long-term outcomes were favourable (median follow-up 11 years) with a very low rate of regional recurrence in patients with positive SLNs. There was only one regional recurrence among patients with positive SLNs who did not undergo axillary dissection, which was not significantly different from recurrence in the group of patients who had axillary dissection after a positive SLNB finding. No correlation was found between the incidence of SLN metastasis and type of breast surgery, conservation of the breast with or without radiotherapy, or mastectomy without radiotherapy. Most interesting is the discovery of a higher rate of regional recurrence in patients with microinvasive DCIS with negative SLNs, but with a specific molecular pattern.

The findings of this study indicate that SLNB may not be useful in microinvasive DCIS owing to the low risk of lymph node metastasis and good prognosis. The good prognosis may be explained by the theory¹⁵ that the major rate of positivity could correspond to an iatrogenic transit of tumour/epithelial cells to lymph nodes, without the significance of real metastasis. Level 1 evidence shows that, in patients with SLN-positive breast cancer, axillary dissection may be avoided when there is a low axillary metastatic burden (Z0011)⁴² and in patients undergoing BCS with radiotherapy; this also supports the conclusion that SLNB in microinvasive DCIS may not be useful. In particular, in the women in the present study who underwent BCS and axillary dissection for positive sentinel nodes, the total number of positive nodes, including sentinel nodes, was less than three, including those women who met the American College of Surgeons Oncology Group Z0011 criteria. An important consideration in staging the axilla in these patients is the possible implication for systemic therapy. In this study, however, adjuvant treatment was largely decided based on cancer biology.

Microinvasive breast cancer is a rare form of breast cancer defined by the presence of 1 mm of invasive cancer in a background of DCIS, and comprises 0·6–3·4 per cent of all breast cancer^1,39,41. In the AJCC staging system, it is considered a subset of T1 disease (T1mi)³⁹. A precise and more complete definition is the WHO classification of clearly separate microscopic foci of infiltration of tumour cells into the mammary stroma, each 1 mm or less in size. No further extension beyond the specialized intralobular stroma is required, the number of invasive foci and their proportion among all the carcinoma cells are irrelevant, and sizes of different foci are not to be added together⁴³. Invasive cells are generally found in the context of DCIS in the background with microinvasive cancer found in 10–20 per cent of patients with DCIS². This consideration could justify the fact that it is often defined as DCIS with microinvasion^7,44. The sole presence of an invasive breast carcinoma of 1 mm or less, with no in situ background, is rare and should be regarded as an invasive carcinoma of that specific diameter¹.

A number of relevant studies^{6,22,24,45–47} have investigated the histopathological characteristics and clinical outcomes of microinvasive DCIS; how patient survival and the biological behaviour of this rare form of breast carcinoma differ from DCIS remain controversial. Microinvasive DCIS is frequently found in a high nuclear grade comedo DCIS setting, and less frequently with other types of DCIS or lobular carcinoma in situ⁴⁸.

The present findings, of low positive SLN rates in women with good DFS and OS, and the lack of influence on selection of adjuvant treatment, are in line with other studies^4,7,15,20,23 showing that SLNB in microinvasive DCIS may not be useful. This study supports the evidence that less surgery, combined with adequate presurgical clinical/histological information allowing the planning of a correct, personalized, clinical pathway for each patient, may provide the same level of OS with better patient quality of life.

Supplementary Material

bjs11079-sup-0001-TableS1

Table S1 Distribution according to sentinel lymph node status