The efficacy of intra-operative injection of [99mTc] labelled colloidal nanoparticles for sentinel lymph nodes identification in breast cancer.

1
Introduction
Sentinel lymph node (SLN) biopsy plays a crucial role in the management of breast cancer allowing for appropriate staging and tailoring of the treatment plan [[1], [2], [3]]. Compared to axillary lymph node dissection (ALND), SLN biopsy is a less invasive procedure which results in faster recovery and reduces the risk of complications such as lymphedema, nerve damage, and shoulder mobility issues [4], contributing to a better long-term quality of life for patients undergoing breast cancer treatment.
The identification rates for SLN with dual tracer (radiocolloid and blue dye) are over 90 %, with low false negative rates and favorable recurrence rates [3,5]. Unlike blue dye, which is injected intraoperatively under anesthesia, Tc-99m labelled colloidal nanoparticles is typically administered preoperatively in the nuclear medicine suite, often causing significant pain and unnecessary stress [6,7].
Approval to inject radio-tracers within the operating room was obtained in our institution at 2019 allowing the surgeons to inject the radioactive tracer to the breast while the patient is under anesthesia, inattentive to the injection pain, similar to the standard practice with dye injection. This method has been described in a relatively few studies from individual centers [[8], [9], [10], [11], [12], [13], [14], [15]]. Although these studies report good results in hundreds of patients, most medical centers still administer the radio-tracer prior to surgery at the nuclear medicine suite. Rambam Health Care Campus is the only hospital in Israel where Tc-99m labelled colloidal nanoparticles is injected in the operating room after anesthesia. This study aimed to evaluate the performance of intraoperative radio-tracer injection in terms of SLN mapping accuracy, while additionally providing information on its financial implications, procedural efficiency, safety, and potential advantages relevant to its consideration as an alternative to standard preoperative protocols.

2
Materials and Methods
2.1
Study population and setting
This was a single-arm retrospective study, which included data of patients aged ≥18 years who underwent breast cancer surgery at Rambam Health Care Center, Haifa, Israel, between 2020 and 2023. The study's protocol was authorized by the local research ethics committee (0176-24-RMB-D, 2024). The need for informed consent was waived due to the study's retrospective nature.
In 2019, regulatory approval was obtained to administer radio-tracers intraoperatively, replacing preoperative injection done previously in our nuclear medicine institute. Three surgeons were trained and approved by the hospital radiation safety officer for handling and injecting radioactive materials. A protocol was established dictating the safe transfer of radioactive material between the nuclear medicine department and the operating theater. Syringes with unfiltered Tc-99m sulphur colloid were stored within a dedicated lead-shielded case (Fig. 1), which was placed in a predetermined closet file within the operating theater.
All patients requiring SLN biopsy were eligible. We excluded from further analysis patients with ipsilateral breast cancer recurrence who had previously undergone ipsilateral SLN biopsy and were in need of preoperative lymphoscintigraphy localization.

2.2
Procedure
SLN biopsy was performed in clinically node-negative (cN0) patients, while SLN biopsy and targeted axillary dissection (TAD) were used for initially node-positive (cN1) patients who underwent neoadjuvant therapy (NAT). Immediately following anesthesia, 1 mCi (0.2 ml) of unfiltered Tc-99m sulphur colloid were injected intradermally into the nipple areolar complex. Diluted patent blue was injected subcutaneously. SLN retrieval was performed in the usual manner, detecting the SLNB by blue dye and a hand-held gamma detector. Lymphoscintigrams were not performed.

2.3
Data collection
Demographic, clinical, surgical, and pathological data were retrospectively collected and securely stored on a password-protected database.

2.4
Data analysis
Data were recorded in Excel (Microsoft Excel ©2010 Microsoft Corporation, Redmond, WA, USA) and analyzed using GraphPad Prism version 6.00 for Windows (GraphPad Software, La Jolla, CA, USA) and IBM SPSS Statistics for Windows, Version 28.0 (released 2021; IBM Corp., Armonk, NY, USA). Descriptive statistics assessed SLN identification rates, stratified by intraoperative Tc-99m success and dye use. Associations with SLN detection were analyzed using Fisher's exact test, chi-square, and Mann-Whitney tests. Significant variables underwent logistic regression, with odds ratios (OR) and 95 % confidence intervals (CI) reported. Candidate variables that could reasonably affect SLN identification were included in the association analyses, including patient factors (age, BMI, ethnicity, prior surgery on the same breast), tumor factors (T stage, N stage, receptor status), and treatment factors (neoadjuvant therapy and breast surgical procedure). Statistical significance was set at p < 0.05.

3
Results
A total of 350 patients who underwent 352 intraoperative unfiltered Tc-99m sulphur colloid injections were included in this retrospective analysis. The demographic and clinical characteristics of the cohort are detailed in Table 1.
The median age of the cohort was 59.8 years, with the majority of patients being female (98.3 %) and postmenopausal (62.2 %). Invasive ductal carcinoma was the most common histological subtype (80.8 %). The majority of patients presented with early-stage disease and clinically negative lymph nodes. One-hundred-thirty-five (38.6 %) patients were treated with NAT, mostly chemotherapy (85.9 %). Breast-conserving surgery was performed in 66.2 % of cases. One patient, referred for risk-reduction mastectomies, underwent sentinel lymph node biopsy following a preoperative diagnosis of atypical ductal hyperplasia on core biopsy in one of the breasts. Four patients with long-standing stable oligometastatic disease underwent SLN biopsy following multidisciplinary team decision.
Table 2 describes the detection rates for SLNs’ with the radio-tracer and patent blue. For the entire cohort (upfront surgery and post NAT), radioactive lymph nodes were identified following intraoperative unfiltered Tc-99m sulphur colloid injections in 89.5 % (315/352) procedures. In 13 additional procedures, the use of patent blue dye successfully identified the SLNs increasing the identification rate with dual tracers to 93.2 % (328/352).
In twenty-four patients (24 procedures), in which SLNs were not identified following dual tracing with the radio-tracer and patent blue, axillary dissection was performed in twenty. No further dissection was performed in three patients following successful identification and surgical removal of previously biopsied positive nodes, confirmed by intraoperative specimen imaging demonstrating the clip (i.e., targeted axillary dissection). In one patient, no further exploration of the axilla was performed. Metastatic lymph nodes were diagnosed in seven patients within this subgroup.
Significant differences were observed in the successful identification of SLNs between patients undergoing upfront surgery and those undergoing surgery following NAT. SLNs identification rate following radio-tracer injection in the upfront surgery setting was 92.2 % vs. 85.1 % for those who underwent prior NAT (p = 0.048). Adding blue dye increased the identification rate to 95.9 % and 88.8 % respectively (p = 0.015).
Subanalysis of 136 procedures in 135 patients treated with NAT reveals differences in the successful identification of SLNs between patients whose preoperative clinical staging was N0 and N1-3. Successful identification with radiotracer was 94.3 % and 78.3 % for N0 and N1-3, respectively (p = 0.014). Additional blue dye led to successful identification in 96.2 % and 83.1 %, respectively (p = 0.027).
Possible risk factors for failure of intraoperative radio-tracer injection are summarized in Table 3. Failure of intraoperative unfiltered Tc-99m sulphur colloid injection to identify the SLNs was associated with Arab or Druze ethnicity, advanced T- and N-stages, and NAT. When these factors were evaluated together in a multivariate analysis, only T-stage and N-stage remained significantly associated with failure to identify the SLNs.
SLNs identification rate improved within the years whether in patients undergoing upfront surgery or following NAT (Fig. 2). By 2023, identification rates of SLNs with dual tracing was 98.2 %, 94.6 %, and 96.7 % in upfront surgeries, surgeries following neoadjuvant treatment, and all surgeries, respectively.
Throughout the study period and up to the time of writing this manuscript, no radiation safety incidents, procedural errors, or malpractice events were reported. This reflects the effectiveness and reliability of the safety protocols implemented in our institution, as well as the strict adherence to radiation handling guidelines by all members of the clinical team. When considering the additional costs associated with imaging and physician interpretation of the images, the 352 procedures included in this study correspond to an estimated cost avoidance of approximately 387,200 Israeli Shekels (about USD 112,640).

4
Discussion
The status of axillary lymph nodes remains a critical prognostic factor guiding therapeutic decisions in breast cancer management. SLN biopsy has replaced traditional ALND as the gold standard for axillary staging in most cases. Our study shows that using intraoperative Tc-99m radiocolloid injection yields SLN detection rates of 89.5 % and when combined with blue dye injection 93.2 % was achieved. The results for the subgroup of patients undergoing upfront surgery were comparable with the detection rate shown in early-stage breast cancer [3]. In patients who underwent NAT using dual tracing, our detection rate was 88.8 %, largely due to lower detection rates in those with metastatic lymph node disease before NAT.
Intraoperative injection of Tc-99m radiocolloid for SLN detection has emerged as a viable alternative to preoperative administration and several studies have demonstrated its efficacy (Table 4). The largest study by Berrocal J et al. is a single center study that over two decades included 2338 patients who received intraoperative injection of Tc-99m labelled colloidal nanoparticles with an overall success rate of 99.8 % in identifying at least one sentinel lymph node [14]. Similar success rates were identified by others who reported smaller series [[8], [9], [10], [11], [12], [13],15]. Our detection rate was slightly lower than that of the above-mentioned intraoperative radio-tracer injection studies. This could partially be attributed to the learning curve we experienced as depicted in Fig. 2, and the fact that we included patients who underwent neoadjuvant therapy.
Our findings suggest that advanced tumor stage (T3-T4) and positive lymph node involvement (N+) are the most significant predictors of failure to identify the SLN using intraoperative Tc-99m labelled colloidal nanoparticles injection. Differences in SLN identification between patients undergoing surgery first or SLN biopsy following NAT were also reported in the SENTINA trial [16] in which patients were allocated to several treatment arms, two of which are relevant to our discussion. All of the procedures in this trial were performed following radiocolloid injection. Additional blue dye injection was optional. The first relevant arm included patients with clinically node-negative disease who underwent surgery first. The overall surgical detection rate with radiocolloid was 98.8 %, which is slightly superior compared to the 95.9 % surgical detection rate following radiocolloid injection in our study. The second relevant arm included patients with clinically node-positive disease who underwent SLN biopsy following NAT. The overall surgical detection rate in these patients was 77.4 %, with radioactive colloid alone. This identification rate is similar when compared to 78.3 % observed in our study. One must take into account that, unlike the SENTINA trial, not all of our patients allocated to SLN biopsy following NAT had clinically node-positive disease. Of note, the median number of lymph nodes removed in the SENITNA trial was 2, comparable to our results.
Avoiding unwarranted discomfort for the patients by the preoperative injection of the radiotracer to the breast led us to abandon the radiotracer's injection before surgery in favor of an injection after anesthesia in the majority of our patients. Intraoperative Indocyanine Green (ICG) Fluorescence is an alternative that may be injected intraoperatively with or without concomitant injection of the radiotracer [17,18]. In these two studies, SLN identification rate with ICG alone was 97.9 % and 91.5 % in all patients and in patients following neoadjuvant chemotherapy, respectively.
Current guidelines recommend omitting routine SLN biopsies in selected patients who are 50 years or older, postmenopausal, with grade 1–2, hormone-receptor-positive, HER2-negative breast cancer, and undergoing breast conserving therapy [19]. These recommendations are based on the SOUND and INSEMA trials, in which SLN biopsy was omitted in 697 patients with T1–2 node-negative tumors and in 962 patients with T1 node-negative tumors, respectively [20,21]. When outcomes were assessed against those of patients who underwent SLN biopsies, no differences in disease survival were observed after a median follow-up of 5.7 and 6.1 years. Relative to the original trial inclusion criteria, the ASCO guidelines limited eligibility based on the observation that most patients enrolled in these trials were 50 years or older, and had small, hormone-responsive tumors [19]. In this patient population with favorable tumor biology, the probability of a significant axillary disease burden is low, and axillary staging has a questionable impact on adjuvant therapy decisions [20,22]. Although current guidelines support omitting SLN biopsy in many of the patients included in our study, this does not diminish the relevance of our primary objective, which was to evaluate the efficacy of intraoperative radiotracer injection in identifying SLNs. In all other patients who do not meet the eligibility criteria for omission of SLN biopsy, avoiding axillary dissection when SLNs are not identified should be considered sparingly, particularly when the morbidity of the operation outweighs omitting axillary staging.
Radio-tracer injection following anesthesia does not allow isotope imaging. Preoperative imaging may be important in certain circumstances, mainly recurrent breast cancer, where aberrant drainage is more common [23]. Otherwise, and unlike other malignancies such as melanoma, the objective of this procedure in most of the patients undergoing breast cancer surgery is to identify the sentinel lymph nodes located in the axilla. Injecting the radio-tracer in the operating room and a search using the probe (without imaging) does not impair the ability to find the sentinel node.
Intraoperative injection offers financial as well as logistical advantages that should not be overlooked; advantage in scheduling and without travelling at the day of surgery. According to Thompson et al., preoperative injection in the US at 2008 already incurred an additional charge of $1325 attributed to imaging, injection, and interpretation of images by a physician [11]. In our medical center, the savings associated with intraoperative Tc-99m labelled colloidal nanoparticles injection amount to approximately 1100 Israeli Shekels (approximately USD 320) per patient, derived from multiple factors that streamline the process.

5
Conclusions
Our study demonstrates that intraoperative injection of Tc-99m labelled colloidal nanoparticles is an acceptable alternative to preoperative injection in terms of detecting SLNs. These findings suggest that the established practice of preoperative injection can be safely implemented, offering means to reduce patient pain and anxiety as well as associated costs.

CRediT authorship contribution statement
Or Barkai: Writing – original draft, Visualization, Investigation, Data curation. Itamar Ashkenazi: Writing – review & editing, Visualization, Validation, Investigation, Formal analysis. Aviad Hoffman: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Conceptualization.

Limitation
The study has several limitations, including its retrospective design and reliance on a single-center database.

Declarations
AI and AI-assisted technologies in the writing process.
As non-native English speakers, we utilized ChatGPT and Grammarly to verify the accuracy of our grammar and spelling. All the data is original. We checked the content to ensure that our original messages were not distorted and the citations correctly represent the source. We take full responsibility for the content of the published article.

Studies in humans statement
All procedures were performed in compliance with relevant laws and institutional guidelines and have been approved by institutional research ethics committee (0176-24-RMB-D, 2024). The privacy rights of human subjects were observed. Identifiers were removed and data was collected in a secured computer platform. The need for informed consent was waived by the research ethics committee.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.