A new fluorescenttargeting tracer contrasts dual tracers in sentinel lymph node biopsy of breast cancer

Abstract

Purpose: To explore the clinical application value of indocyanine green (ICG)–rituximab in sentinel lymph node biopsy. Methods: This study included 156 patients with primary breast cancer: 50 patients were enrolled in dose-climbing test, and 106 patients were enrolled in verification test. This was to compare the consistency of ICG–rituximab and combined method in the detected lymph nodes. Results: According to the verification test, the imaging rate of ICG–rituximab was 97.3%. Compared with the combined method, the concordance rate of fluorescence method was 0.991 (28 + 78/107; p < 0.001). Conclusion: For ICG–rituximab as a fluorescent targeting tracer, the optimal imaging dose of ICG 93.75 μg/rituximab 375 μg can significantly reduce the imaging of secondary lymph nodes. Compared with the combined method, it has a higher concordance rate.

Keywords: :

• breast cancer
• dose-climbing test
• fluorescent targeting tracer
• ICG
• ICG–rituximab
• rituximab
• sentinel lymph node
• SLNB
• tracer
• verification test

In patients with clinically negative lymph nodes, sentinel lymph node biopsy (SLNB) instead of axillary lymph node dissection (ALND) has no difference in regional control, disease-free and overall survival, and can improve quality of life and upper limb function [1–4]. The combination of patent blue and radioisotope can provide the success rate of sentinel lymph node (SLN) detection [3], but these approaches present specific advantages and drawbacks.

The fluorescent tracer indocyanine green (ICG) has advantages such as economy, safety and visibility [5,6], but ICG is a small molecule tracer, which has the shortcomings of short trace time and high rate of nonsentinel lymph node (n-SLN) imaging. ICG was combined with rituximab directly to produce a new receptor-targeted tracer (ICG–rituximab), which specifically binds to CD20 molecules on the surface of B lymphocytes in the SLN and has SLNB targeting [7]. The aim of this study was to establish the optimal dose and feasibility of ICG–rituximab as a tracer of SLNB in breast cancer, which were tested by combined (methylene blue and a radioisotope) method.

Materials & methods

Patients

The sample size was calculated by the sample size formula (n = [μα/δ]²[1-p]p). Setting: the false negative rate of SLNB was less than 10% and the positive rate of SLNs was 30%. μα: 1.960; δ: 0.1; p: 0.9; n: 105.

This study included 156 patients with primary breast cancer in the Breast Disease Center of Shandong Cancer Hospital from January 2016 to September 2019: 50 patients were enrolled in the ICG–rituximab dose-climbing test and 106 patients were enrolled in the ICG–rituximab verification test. The protocol and consent forms were approved by the Institutional Ethics Committee (No. SDTHEC20110324).

The enrollment criteria included clinical physical examination and ultrasonography of primary breast cancer patients with negative axillary lymph nodes, and no distant metastasis was found in imaging examination. The exclusion criteria included breast cancer patients who had received neoadjuvant therapy, had previous axillary surgery history and had inflammatory breast cancer.

Indocyanine green–rituximab dose-climbing test

The coupling mass ratio of ICG to rituximab is 1:4 [8], and the 0.05 ml of rituximab solution (10 mg/ml) and ICG solution (2.5 mg/ml) were extracted and mixed with a 1 ml syringe, and diluted with sterilized water for injection to 1.0 ml (125 μg/500 μg) as mother solution. A total of 100, 75, 50, 20 and 10% of the mother liquor were extracted and diluted to 1.0 ml with sterilized injection water as five groups of reagents. The mass ratio of each group of reagents was as follows: group A 625 μg (125 μg/500 μg); group B 468.75 μg (93.75 μg/375 μg); group C 312.5 μg (62.5 μg/250 μg); group D 125 μg (25 μg/100 μg) and group E 62.5 μg (12.5 μg/50 μg).

Indocyanine green–rituximab verification test

Rituximab solution (10 mg/ml) 0.1 ml and ICG solution (2.5 mg/ml) 0.1 ml were extracted with a 1 ml syringe, then mixed and diluted to 0.8 ml with sterilized water for injection. A total of 0.3 ml was taken as the injection dose, and the mass was 468.75 μg (93.75 μg/375 μg).

Instruments & reagents

The test instruments and reagents are shown in Table 1.

Table 1. Instruments and reagents.

	Specifications/models	Provider
Radioisotope	^99mTc-labeled sulfur colloid	Beijing New Kostar company, prepared by nuclear medicine department of Shandong Cancer Hospital (Beijing, China)
Dye	Methylene blue 20 mg 2 ml	Jichuan Pharmaceutical Group Co., Ltd (Jiangsu, China)
Indocyanine green	25 mg	Dandong Yichuang Pharma, Ltd (Dandong, China)
Rituximab	100 mg	Shanghai Roche Pharmaceuticals (Shanghai, China)
Near-infrared fluorescence imaging system	MI-1 fluorescence imaging system	Mingde Biotech, Ltd (Hebei, China)
Gamma probe detection	Neoprobe, Neo2000 gamma detection system	Johnson and Johnson (Shanghai, China)

Axillary sentinel lymph node biopsy

Indocyanine green–rituximab dose-climbing test

From 3 to 18 h before operation, different doses of ICG–rituximab were injected into the glandular layer of the affected breast (group A 625 μg; group B: 468.75 μg; group C: 312.5 μg; group D: 125 μg; group E: 62.5 μg). The volume of the injection reagent was 1 ml.

Indocyanine green–rituximab verification test

In each patient, ^99mTc-SC (1.0–1.2 ml/22.2–55.5 MBq) was injected into the mammary gland at 6 and 12 o’clock of the areola surrounding area with the guidance of ultrasound and ICG–rituximab 0.3 ml (ICG 93.75 μg/rituximab 375 μg) was injected intradermally near the areola in the upper outer quadrant 3–18 h before surgery. After general anesthesia, near-infrared fluorescence imaging system was used to label the luminous lymphatic vessels and lymph nodes percutaneously, and gamma detection probe was used to label the ’hot spots’ of lymph nodes. SLNB was performed 10–15 min after subcutaneous injection of 1% methylene blue 4 ml on the surface of primary tumor. Gamma detection probe identified all nodes with count rates greater than 10% of the nodes with the highest count rate, that is, hot nodes. Blue dye nodes and hot nodes were defined as SLNs. Palpation of the axilla was performed, and palpated enlarged hard lymph nodes were removed and recorded as palpation SLNs. The SLNs were detected by near-infrared fluorescence imaging system, and the gray values of SLNs were recorded by ICG–rituximab imaging.

Mode of operation

SLNB has become the standard procedure of breast cancer surgery, so it is difficult to verify the diagnostic efficiency of ICG–rituximab by ALND. We cannot compare ALND to obtain an accurate false negative rate. Patients who agree to replace ALND with SLNB are not allowed to undergo ALND if there is no metastasis in SLN diagnosed by frozen pathology and print cytology. Patients without metastasis of SLNs only underwent SLNB. Patients with intraoperative or postoperative diagnosis of SLN metastases undergo ALND.

Pathologic evaluation

All SLNs were examined by intraoperative frozen section and printed cytology and postoperative routine paraffin pathological examination. According to the 8th edition of American Joint Committee on Cancer, macrometastases, micrometastases and isolated tumor cells are defined as SLN positive.

Drug safety

The allergic reaction symptoms such as skin swelling, papule, palpitation and respiratory allergy were observed after ICG–rituximab was injected into the breast. The eosinophils were measured on the first day after operation.

Statistical analysis

SPSS 22.0 was used for statistical analysis to compare the agreement of ICG–rituximab and combined method in the detection of lymph nodes in SLNB. t-test, chi-square test or Fisher exact probability method were used to analyze the difference of mean between continuous variables and agreement rate. The difference was statistically significant when α = 0.05 (p < 0.05).

Results

General information & security results

The general data of 156 patients in this study – classification of tumor size, location of tumor, pathological type, mode of operation, BMI, etc. – are shown in Tables 2 & 3. There was no allergic reaction after injection of ICG–rituximab, and the eosinophil count did not increase after operation.

Table 2. Characteristics of indocyanine green–rituximab dose-climbing test cases (n = 50).

Characteristic	A	B	C	D	E	p-value
n	10	10	10	10	10
Age						0.757
<40 years	2	1	2	3	1
From 40 to approximately 59 years	7	7	6	6	5
>59 years	1	2	2	1	4
Tumor size (cm)	2.283	2.018	2.267	2.152	2.087	0.341
Tumor site						0.872
Upper outer quadrant	5	6	5	5	4
Lower outer quadrant	1	2	1	1	1
Upper inner quadrant	2	1	2	1	0
Lower inner quadrant	1	1	1	1	4
Central	1	1	1	2	1
Histological grading						0.315
Carcinoma in situ	4	1	1	3	1
I	0	0	0	0	1
II	3	6	7	7	7
III	3	3	2	0	1

Table 3. Characteristics of indocyanine green–rituximab verification test cases (n = 110).

Characteristic	n (%)
Pathology:   Noninvasive carcinoma   Ductal   Lobular   Others	16 (14.5) 86 (78.2) 1 (0.9) 7 (6.3)
Tumor site:   Upper outer quadrant   Lower outer quadrant   Upper inner quadrant   Lower inner quadrant   Central	69 (62.7) 13 (11.8) 19 (17.3) 5 (4.5) 4 (3.6)
BMI:   BMI<18.5   18.5≤BMI<25   25≤BMI<30   30≤BMI	3 (2.7) 70 (63.6) 29 (26.4) 8 (7.3)
Mode of operation:   Mastectomy   Breast conserving surgery   Prosthesis implantation	65 (59.1) 41 (37.3) 4 (3.6)
Tumor staging:   Ductal carcinoma in situ   T1   T2   T3	16 (14.5) 63 (57.3) 29 (26.4) 2 (1.8)

Imaging results of indocyanine green–rituximab dose-climbing test

In the ICG–rituximab dose-climbing test, the imaging rate of all the five groups was 100%, and the imaging rate of SLN tracer was compared as shown in Figure 1. The fluorescence imaging rate of n-SLNs in five groups was 54.1, 6.7, 3, 0 and 0%.

Figure 1. The development of five groups at different doses.

ICG-Rit: Indocyanine

Imaging results of indocyanine green–rituximab verification test

SLNB was performed in 110 cases of breast cancer by fluorescence, nuclide and dye methods, including 107 cases of ICG–rituximab imaging, 108 cases of nuclide imaging, 101 cases of dye imaging and 110 cases of combined imaging. SLN metastasis was found in two of three patients whose ICG–rituximab imaging failed. A total of 270 lymph nodes were detected by ICG–rituximab, with the highest gray value of 254, most of which were concentrated in the range from 254 to 220. Those less than 220 were not easily found during the operation, and the gray values were not continuous. The lymph nodes with different gray values are shown in Figure 2.

Figure 2. Different gray values of lymph nodes in ICG-Rit imaging.

The gray values were 254 (A), 226 (B) and 209 (C).

ICG-Rit: Indocyanine green–rituximab.

Agreement evaluation of comparative combined method

Lymph node metastases were detected in 31 SLNs detected by SLNB, of which 30 were detected by the nuclide method, 27 were detected by the dye method, 28 were detected by ICG–rituximab and two failed imaging. According to Table 4, compared with the combined method, the agreement rate of fluorescence method was 0.991 (28 + 78/107; p < 0.001).

Table 4. Detection rate of sentinel lymph node by indocyanine green–rituximab and combined method.

Number	Combined +	Combined -	Total ICG	p-value
ICG+	28	0	28	≤ 0.001
ICG-	1	78	79
Total combined	29	78	107

Combined tracer: methylene blue and ^99mTc-SC.

ICG: Indocyanine green.

Discussion

In recent years, scholars at home and abroad have tried to study new tracers for SLNB, including ultrasound contrast agents, superparamagnetic iron oxide and fluorescent tracer. ICG, the most commonly used reagent in clinic, can observe the subcutaneous lymphatic drainage pathway and SLN position through the fluorescence imager [5,6], which is economical and visual, and has a high detection rate (93.1–100.0%) [9]. However, ICG imaging time is short (5–15 min), there are more secondary lymph nodes appear in imaging, and the number of lymph nodes detected is more than that of radionuclide or blue dye method [10], which may increase the risk of postoperative upper limb complications and lack of standardized procedures such as optimal concentration, dose and injection time, which hinders its clinical promotion.

Rituximab is a specific humanized monoclonal antibody, which can specifically bind to CD20 molecules on the membrane of B lymphocytes in lymph nodes, and is not easy to dissociate. The chemical structure contains a domain that binds to small molecules and can couple with ICG [11,12]. When ICG and rituximab were coupled at a mass ratio of 1:4, the labeling rate reached 100%, while the integrity and immune activity of the antibody molecules were maintained, and there was no sterility, no pyrogen and no acute toxicity [7,8]. ICG–rituximab has the advantages of uniform molecular weight, close binding to lymph nodes and controllable secondary lymph node imaging.

ICG and rituximab are physically coupled and maintain their original drug characteristics. ICG has been used as a fluorescent agent in clinic for 50 years. It has good safety [13] and has been approved by the US FDA. The main adverse reactions of rituximab are neutropenia and rash (≥10%). The therapeutic dose of rituximab for lymphoma is 630 mg (60 kg, 170 cm). The dose of ICG for choroidal angiography is 25 mg; however, the maximum dosage in this study is ICG 125 μg and rituximab 500 μg, ICG is 0.5% of the above dose, and rituximab is 0.79% of the therapeutic dose. In the trial, ICG–rituximab is used in 156 patients, no allergic reaction is found, and the eosinophil count do not increase after operation. Its security is verified.

In the ICG–rituximab dose-climbing test group, according to the dose decreasing order, the SLN imaging rates of the five groups of ICG–rituximab were 91.7, 76.7, 67.7, 37.1 and 7.4% (p < 0.05). Compared with the combined method (nuclides + dyes), group A is slightly higher than the combined method (p < 0.05); group B is similar to the combined method (p = 0.485); and the three groups C, D and E are lower than the combined method (p < 0.05). ALND was performed on patients with SLN metastasis in five groups, and fluorescence imaging of n-SLN was detected. The fluorescence imaging rate of n-SLNs in five groups was 54.1, 6.7, 3, 0 and 0%. In group A, the imaging rate of SLN was 91.7%, and the imaging rate of n-SLN was 54.1%, indicating that it had labeled too many secondary lymph nodes. In group B, the imaging rate of SLN was 76.7%, and that of n-SLN was 6.7%. The above results show that there is no statistical difference in imaging rate between group B and the combined method, and the imaging rate for n-SLN is low (6.7%). This group of doses (93.75 μg/375 μg) can be used as the optimal imaging dose.

In the ICG–rituximab verification test, the detection rates of ICG–rituximab, nuclide, dye and combined methods were 97.3, 98.2, 91.8 and 100%, respectively. The mean SLN of ICG–rituximab imaging was 2.44 and the median was 2, which was significantly lower than that of SLN detected by radionuclide method (2.83 and 3; p < 0.001). A total of 1736 patients were included in 12 studies to compare the SLNB diagnostic efficacy of ICG and nuclide method. The SLN detection mean of ICG was 1.5–3.4, and that of nuclide method was 1.35–2.3. The number of lymph nodes detected by ICG was more than that of nuclide method, but there was no statistical difference between them [14]. From the above results, it is inferred that using ICG–rituximab as a SLNB tracer, the number of lymph node visible in imaging is lower than that of ICG imaging, which reduces the imaging of secondary lymph nodes, which may reduce the risk of postoperative axillary complications.

SLN metastasis was detected in 31 patients by frozen pathology, printed cytology and paraffin pathology, of which 30 cases were detected by radionuclide method, 27 cases by dye method, 31 cases by combined method and 28 cases by ICG–rituximab imaging. Compared with the combined method, the agreement rate of the fluorescence method was 0.991 (28 + 78/107; p < 0.001). It was inferred from the agreement rate of the fluorescence method that this was in excellent agreement with the combined method.

Previous ICG studies showed that the factors of ICG failure included obesity (BMI ≥30) and SLN macrometastasis. A recent study comparing nuclide tracers showed that the ICG imaging rate was 81.9% and the false negative rate was 34.7%. The patients with overweight (BMI >25) or SLN macrometastases (>2 mm) were related to the low detection rate of ICG (p = 0.02) [15]. In this study, ICG–rituximab imaging failed in three cases, and the failed cases were overweight patients. In one of the three patients, three SLNs were detected by nuclide method, and the SLN with the lowest radionuclide count was pathologically macrometastatic lymph nodes, which was similar to previous studies. However, among the patients in this study, the effect of BMI on the number of SLNs detected by fluorescence method was not statistically significant (p = 0.995). In the study of Mazouni et al., 46% of the cases were overweight (BMI >25) [15]; according to Table 3, 29 cases (26.4%) were overweight and eight cases (7.3%) were obese in our study. If the number of overweight cases is expanded, the imaging rate may decrease and the failure rate may increase.

Limitations

Our research is single-center, and the sample size is not very sufficient. The ICG–rituximab is a new tracer developed by our team. At present, it is only used in our hospital. Although the number of patients participating in the trial reaches the estimated sample size, there is still data bias, and more patients are needed to reduce the bias and obtain more accurate results. In our center, SLNB has completely replaced ALND for patients with intraoperative and postoperative pathological N0. It is impossible to verify the diagnostic efficacy of fluorescent tracers through ALND of each patient. Taking the combined tracers recommended in the current guidelines as the standard for noninferiority comparison to indirectly test the diagnostic efficacy of fluorescent tracers makes the final results biased. In the follow-up trial, the fluorescence tracer tracing after ALND should be recorded to further verify its diagnostic energy efficiency, but the process of enrolling patients would then become slow. The technical bottleneck of fluorescent tracer is its weak penetration, and the photoacoustic imaging technology with high contrast resolutions and deep imaging depth may help the fluorescent tracer to break through this bottleneck and expand its penetration range [16]. With this technique, ICG–rituximab will not need to look for SLNs along the lymphatic vessels of subcutaneous imaging, but can directly explore the lymph nodes of axillary imaging and detect SLNB. The fluorescent tracer lacks the threshold value as radionuclides (the nuclide count is more than 10% of the maximum nuclide count) to judge whether the imaging lymph node is a SLN. In this study, we tried to use the gray value of lymph node imaging to evaluate the degree of ICG–rituximab labeling lymph nodes, but the gray value of detected lymph nodes was not continuous, concentrated in 254–220, and the gray threshold could not be established to distinguish SLN from n-SLN.

Conclusion

In this study, the optimal imaging dose of ICG–rituximab was detected. Compared with ICG, it can reduce the imaging of secondary lymph nodes. Compared with the combined method, it has higher accuracy, specificity and coincidence rate, and lower false negative rate. As a SLNB tracer, ICG–rituximab has good safety and clinical application value.

Summary points

• Neoadjuvant therapy has become the standard treatment for patients with inoperable as well as some invasive and high-risk breast cancers.

• Indocyanine green (ICG)–rituximab was used to set up five groups of dose-increasing test to compare the success rate of development and to select the optimal dose.

• Sentinel lymph node biopsy was performed by using the optimal dose of ICG–rituximab and the combined method to test the agreement between them.

• ICG–rituximab has good safety.

• There may be a decrease in the success rate of ICG–rituximab in obese women, which needs to be confirmed by further trial results.

• Axillary lymph node dissection cannot be performed in the test, and there is a lack of data such as the false negative rate of ICG–rituximab.

• The fluorescence penetration of ICG is weak. It can be combined with photoacoustic imaging technology to optimize fluorescence imaging and improve the detection rate and success rate.

• Photoacoustic imaging technology will be introduced in the next test to detect the imaging results of ICG rituximab in breast sentinel lymph node biopsy again.

Author contributions

QS Zhang, PP Li and S Wu were responsible for study conception and design. X Sun, BB Cong and YS Wang were responsible for acquisition of data and analysis. S Wu, PP Li and YS Wang were responsible for drafting of the manuscript. All authors were responsible for interpretation of data/results and revision of the manuscript.

Financial disclosure

This work was supported by Hebei Medical Science Research Project (20220190). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Open access

This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/