Sentinel Node Biopsy in Post-neoadjuvant Chemotherapy Breast Cancer Patients Using Pre-chemotherapy Breast Tattooing

Abstract

The role of sentinel lymph node biopsy (SLNB) in locally advanced breast cancer (LABC) post-neoadjuvant chemotherapy (NACT) is debatable. We conducted a novel pilot study in which pre-NACT tattooing of breast lumps in LABC patients resulted in black tattoos being deposited in the axillary node. We hypothesized that this black node was the sentinel node. The identification rate (IR) of the black node in our pilot study was 100%, and the false-negative rate (FNR) was 0%. This study aims to evaluate our hypothesis that the black node is the sentinel node in post-NACT LABC patients after pre-NACT breast tattooing. This is a cross-sectional study of prospectively collected data of women with LABC undergoing surgery after NACT. Patients underwent tattooing of breast primarily using black tattoo ink prior to NACT. Women who progressed on NACT were excluded. All patients underwent axillary dissection. Intraoperatively identified black nodes were sent separately for pathological evaluation. The accuracy of the black nodes was assessed using IR and FNR. Of the 214 patients, a complete clinical response was present in 36%. Black node IR was 88.8% and FNR was 17.4%. In pre-NACT cN0 and cN1 patients, IR was 100% and 96.6%, and FNR was 0% and 4.63%, respectively. SLNB using pre-NACT tattooing in LABC patients has a high IR and FNR. In the subset with low pre-NACT axillary burden (cN0 or cN1), SLNB by pre-NACT breast tattooing has a high IR and low FNR.

Introduction

The advent of sentinel lymph node biopsy (SLNB) has revolutionalized the surgical management of the axilla in node-negative early breast cancer in the primary setting [1, 2]. Patients with limited axillary disease may avoid further axillary surgery based on the results of the ACOSOG Z0011 and AMAROS trials [3, 4]. However, the de-escalation of the axilla in the post-neoadjuvant chemotherapy (NACT) setting is still debatable. Various studies using SLNB alone such as the ACOSOG Z1071, SENTINA, and SN FNAC have a high false-negative rate (FNR) and low identification rate [5–7]. Targeted axillary dissection has been shown to have a low FNR [8]. There are multiple markers which can be used to target the axillary node including clips, radioiodine seeds, and magnetic seeds each with their own advantages and disadvantages [8–10]. The cost of these agents and the need for skilled manpower and training are major hurdles in low-resource settings. In most of the low-resource centres in our country including our centre, axillary lymph node dissection (ALND) continues to be offered to all patients post-NACT even with clinical complete nodal response.

Our patients with locally advanced breast cancer (LABC) routinely undergo tattooing around the breast primarily prior to NACT using Indian ink intradermally. This helps in response assessment, planning of breast conservation surgery post-NACT, and orientation of the pathological specimens. We have published our pilot study in which the black tattoo ink deposited in the axillary node was visualized as a black node during axillary dissection [11]. We postulated that the black tattoo ink would travel through the lymphatics and get deposited in the first echelon node. This black dye would permanently stain the node black and would not be affected by chemotherapy. We hypothesized that this black node should serve as the sentinel node. The black dye in the node was visualized in the capsular sinusoids of the black node in our pilot study and the images have been published [11]. The identification rate was 100% and the FNR was 0% in our pilot study. This is a very cost-effective, simple, and scientific method without any additional burden on the patient or the institute. No additional cost or procedures are required for this technique. This technique would be very useful in low-resource settings. This study aims to further evaluate the role of the black node as the sentinel node in LABC patients post-NACT.

Material and Methods

This is a cross-sectional study of prospectively collected data conducted at a tertiary reference centre in Kerala, India, from the 1st of July 2020 to the 31st of December 2021. This study was conducted after clearance from the Institutional Review Board (IRB No 06/2020/11) and Human Ethics Committee clearance was exempted as the study did not involve any deviation from the routine practice of the breast clinic. This study included women with biopsy-proven LABC (cT3 or 4, N0–3 or any T, N 2–3) who received NACT and were planned to undergo surgery. Women with inflammatory breast cancer and those who progressed on NACT were excluded. All patients with breast cancer planned for NACT regularly undergo tattooing of a minimum 4 points around the margins of the breast primary intradermally using black tattoo ink (Suluape black, Intenze, Mumbai, Maharashtra, India) at our centre (Fig. 1). The patients underwent breast conservation surgery (BCS) or modified radical mastectomy (MRM) as per the treatment plan. During the axillary dissection, the black node if identified was sent separately for histopathology. The ALND was completed including all three levels. Data collected included the patient’s age, menopausal status, T and N stage at presentation, quadrant of the tumour, grade, biology, response to treatment, post-chemotherapy clinical T and N stage, type of surgery (BCS or MRM), and final histopathology report. The accuracy of the black nodes was assessed using identification rate and FNR. The association between two categorical variables was assessed using a chi-square test. A p-value < 0.05 was considered significant.

Fig. 1.

Breast tattoo using black tattoo ink

Results

Overall, 584 patients underwent surgery post-NACT during the study period. Of the 223 patients with LABC who underwent surgery, 9 patients were excluded due to tumour progression on chemotherapy. Thus, 214 patients were included in the analysis (Fig. 2).

Fig. 2.

Consort diagram of the study population

The clinic-pathological characteristics of the patients and the surgical aspects are detailed in Table 1.

Table 1.

Patient, tumour, and surgery characteristics

	Number of patients	Percentage (%)
Age in years
≤ 40	16	7.5
41–60	172	80.4
> 60	26	12.1
Menopausal status
Premenopausal	68	31.8
Postmenopausal	146	68.2
Pre-chemotherapy T stage
T1	13	6.1
T2	123	57.5
T3	43	20.1
T4	35	16.4
Pre-chemotherapy N stage
N0	7	3.3
N1	119	55.6
N2	69	32.2
N3	19	8.9
Laterality
Left	133	62.1
Right	81	37.9
Quadrant
Upper outer	108	50.5
Upper inner	44	20.6
Lower inner	29	13.6
Lower outer	13	6.1
Central	20	9.3
Grade
2	34	15.9
3	180	84.1
Biology
Hormone receptor–positive HER2-negative	94	43.9
Hormone receptor–positive HER2-positive	44	20.6
Hormone receptor–negative HER2-positive	27	12.6
Triple-negative	49	22.9
Post-chemotherapy T stage
T0	104	48.6
T1	69	32.2
T2	22	10.3
T3	2	0.9
T4	17	7.9
Post-chemotherapy N stage
N0	146	68.2
N1	65	30.4
N2	3	1.4
Clinical response
Complete clinical response	77	36
Partial clinical response	129	60.7
Stable disease	7	3.3
Surgery
Breast conservation surgery	131	61.2
Modified radical mastectomy	83	38.8

The commonest age group was 41 to 60 years (80.4%). Postmenopausal women constituted 67.8% of the women. The commonest side was left (62.1%) and the commonest quadrant was the upper outer quadrant (50.5%). The commonest pre-chemotherapy T and N stages were cT3 (57.5%) and cN1 (55.6%), respectively. There were no cases of isolated internal mammary node positivity without the presence of axillary nodes. All the patients with cN2 disease had matted axillary nodes. Of the 19 patients with cN3 disease, 4 patients had supraclavicular node positivity and 15 patients had both axillary and internal mammary artery nodes present. Hormone receptor–positive HER2-negative was the commonest biology (43.9%). Complete clinical response was present in 36% of the patients. BCS was performed in 61.2% of the women while the rest underwent MRM. The commonest post-chemotherapy T and N stages were ycT0 (48.6%) and ycN0 (68.2%), respectively. The pCR rate was 34.6%. Nodal pCR was achieved in 67.3% of the patients.

Black node was identified during the axillary dissection (Fig. 3) in 190/214 patients (88.8%). It was sent separately for histopathological evaluation (Fig. 4). The FNR was 17.4%. The median number of black nodes identified was 1. The median number of nodes identified on axillary dissection was 12 (range 4–29). In the pre-NACT cN0 (n = 7), the patients’ identification rate was 100% and the FNR was 0%. In the pre-NACT cN1 (n = 119) subset of patients, the identification rate was 96.6% (115/119 patients) and FNR was 4.63%. In the pre-NACT cN2 (n = 69) subset of patients, the identification rate was 82.6% (57/69 patients) and FNR was 35.8%. In the pre-NACT cN3 (n = 19) subset of patients, the identification rate was 94.7% (18/19 patients) and FNR was 27.8%. In the ycN0 subset (n = 146), the identification rate was 89% (130/146 patients) and the FNR was 15.4% (Fig. 5).

Fig. 3.

Black node identified during axillary dissection

Fig. 4.

Black node

Fig. 5.

Identification rate and false negative rate of black node

There was a significant correlation between the cN and ycN status and the FNR of the black node (p < 0.05). There was no significant association between the age, menopausal status, T stage at presentation, quadrant of the tumour, biology, response to treatment, post-chemotherapy clinical T stage, type of surgery, and the FNR of black nodes.

Discussion

The management of the node-negative axilla in early breast cancer has been revolutionalized by the development of sentinel lymph node techniques. As per the meta-analysis by DiSipio et al., the arm lymphedema rate following ALND was four times higher than that following SLNB [12]. Multiple studies have been conducted to investigate the role of SLNB in the post-NACT setting [5–7]. In the ACOSOG Z1071 study, in patients with cN1 nodal status and 2 or more nodes identified at SLNB, the identification rate was 92.9% and the FNR was 12.6% [5]. In the arm C of the SENTINA trial, the identification rate was 80.1% and the FNR was 14.2% [6], while in the SN FNAC trial, the identification rate was 87.6% and the FNR was 8.4% [7]. These trials concluded that the use of dual tracers, identification of more than 2 nodes, and use of immunohistochemistry for the detection of metastasis may optimize the identification rate and FNR of SLNB after NACT [13]. Targeted axillary dissection is a relatively new procedure in which the positive nodes are marked prior to the neoadjuvant therapy and excised along with the sentinel nodes during the axillary dissection [8]. It was initially described by Caudle et al., by using clips to mark the positive nodes pre-chemotherapy followed by SLNB and selective removal of the clipped node using iodine-125 seed localization at surgery [8]. Multiple other techniques of localization of the node like radioiodine seeds, magnetic seeds, carbon tattooing of the node, RFID tagging, wire localization, and ultrasound visible clip have been described each with their own advantages and disadvantages [8–10, 14–18]. A recent systematic review and meta-analysis of the accuracy of different surgical axillary staging procedures compared with ALND in clinically node-positive axilla following NACT of 20 studies found that in the targeted procedure, the identification rate was 100% and FNR ranged from 2 to 4% [19]. The economic constraints and the need for manpower are major issues in low-resource settings. In most of the low-resource centres in our country including our centre, ALND continues to be offered to all patients post-NACT even with clinical complete nodal response.

Our patients undergo tattooing with tattoo ink around the primary tumour prior to NACT. We have previously reported our technique in which the black tattoo ink deposited in the axillary node was visualized as a black node during axillary dissection [11]. This black dye acts as a permanent marker on the node and was hypothesized to be the sentinel node. As the black pigmentation node is not affected by NACT, this black node could be visualized post-NACT in the axilla. As per the principles of SLNB, this dye would travel through the lymphatics, deposit itself in the sentinel node capsular sinusoids, and stain the node black. The size of the carbon particles would prevent further transport into the non-sentinel nodes. There are multiple case reports of tattoo pigment mimicking malignant melanoma in patients with tattoos who underwent surgery [19–24]. The principle of the black node in our study is the same as that of the case reports.

The current study was conducted based on the results of the pilot study with an identification rate of 100% and FNR of 0% [11]. This study consisted of 214 patients, the identification rate was 88.8%, and the FNR was 17.4%. In pre-NACT cN0 patients, the identification rate was 100% and the FNR was 0%, and in the pre-NACT cN1 patients, the identification rate was 96.6% and FNR was 4.63%. The results of our study are on par with the results of the ACOSOGZ1071 study where the FNR was lower in cN1 patients [5]. The role of SLNB in cN0 patients post-NACT setting has already been established with the use of dual tracers [25, 26]. In the GANEA 2 study, the identification rate of SLNB using dual tracers was 97.6%, and the axillary recurrence was seen in only 1/419 in cN0 patients [25]. Similarly, in our study, the identification rate was 100% and the FNR was 0%, but the sample size of this subset was very small and further studies in this subset are planned. Our study included patients with pre-NACT N1, N2, and N3 nodal status. Previous studies have already proven that SLNB has a lower FNR and higher identification rate in cN1 patients [5–7]. Our subset of cN1 patients had a high identification rate and low FNR on par with that of the targeted axillary dissection [8]. Their procedure requires clipping of the node prior to NACT and then again tagging of the clipped node with radionucleotide markers using image guidance prior to surgery to intraoperatively localize the node. This adds to the cost, manpower, and additional invasive procedures for the patient. In our technique, the skin is tattooed during the core cut biopsy itself with a low-cost dye which does not need any additional manpower or procedures. Pre-NACT radioactive agent labelling requires the expertise and availability of the radioactive agent and the safe handling of the marker. There are doubts regarding radiation safety as the radiolabeled marker remains in the body throughout the NACT period [9, 17]. Other agents like the Magseed and RFID-tagged seeds are not cost-effective in our setting [10, 18]. The hydrogel-based marker requires intraoperative ultrasound to be detected which is time-consuming and requires additional training and equipment [16]. The wire localization of the node is cost-effective but requires an additional procedure and causes discomfort to the patient [15]. Tattooing of the node using carbon tattoo is cost-effective but may lead to misidentification of the node due to the perinodal tissue tattooing [14]. Compared to these techniques, our technique is simple and cost-effective; does not require additional training, visits, or manpower; and is based on scientific principles.

There are some limitations of our study. The tattooing of the breast leads to permanent pigmentation of the skin of the breast which affects the cosmesis. Although this is routine practice in our institute, some patients may not be comfortable with the same. During BCS, we excise the tattooed tissue along with the primary so that the cosmesis is not affected. This is not an issue in patients undergoing MRM as the overlying skin with the tattoo is excised in the specimen. Secondly, this technique cannot be combined with SLNB using blue dye alone as the blue dye masks the black node. However other techniques of SLNB using radionucleotide dyes or indocyanine green should not interfere with the identification of the black nodes [27, 28]. Further studies with the combined technique are warranted to study if the FNR and identification rate can be improved further. Our technique cannot be used in patients with tattoos on the body areas that drain into the axilla as the nodes may already be stained black due to the tattoo ink. The black dye fails to stain beyond the very first echelon nodes. The use of other agents like indocyanine green or radiotracers as a dual tracer in addition to the black node can alleviate this limitation. In spite of these limitations, our study has identified the subsets in whom this technique is effective, and a prospective study of black node identification in cN0 and cN1 patients is planned.

Conclusions

SLNB by pre-NACT tattooing of the breast primary in LABC patients has a high identification rate and FNR. In the subset with the low pre-chemotherapy axillary burden (cN0 or cN1), SLNB using pre-NACT breast tattooing has a high identification rate and low FNR. Further studies in this subset are indicated to predict if the black nodes can be considered sentinel nodes in this subset.

Data Availability

Data is available on request to the corresponding author.

Declarations

Ethics Approval

The study was cleared by the Institutional Review Board of our cancer centre (IRB No 06/2020/11).

Conflict of Interest

The authors declare no competing interests.