Targeted Imaging of Endometriosis and Image-Guided Resection of Lesions Using Gonadotropin-Releasing Hormone Analogue-Modified Indocyanine Green

Abstract

Objective. In this study, we utilized gonadotropin-releasing hormone analogue-modified indocyanine green (GnRHa-ICG) to improve the accuracy of intraoperative recognition and resection of endometriotic lesions. Methods. Gonadotropin-releasing hormone receptor (GnRHR) expression was detected in endometriosis tissues and cell lines via immunohistochemistry and western blotting. The in vitro binding capacities of GnRHa, GnRHa-ICG, and ICG were determined using fluorescence microscopy and flow cytometry. In vivo imaging was performed in mouse models of endometriosis using a near-infrared fluorescence (NIRF) imaging system and fluorescence navigation system. The ex vivo binding capacity was determined using confocal fluorescence microscopy. Results. GnRHa-ICG exhibited a significantly stronger binding capacity to endometriotic cells and tissues than ICG. In mice with endometriosis, GnRHa-ICG specifically imaged endometriotic tissues (EMTs) after intraperitoneal administration, whereas ICG exhibited signals in the intestine. GnRHa-ICG showed the highest fluorescence signals in the EMTs at 2 h and a good signal-to-noise ratio at 48 h postadministration. Compared with traditional surgery under white light, targeted NIRF imaging-guided surgery completely resected endometriotic lesions with a sensitivity of 97.3% and specificity of 77.8%. No obvious toxicity was observed in routine blood tests, serum biochemicals, or histopathology in mice. Conclusions. GnRHa-ICG specifically recognized and localized endometriotic lesions and guided complete resection of lesions with high accuracy.

1. Introduction

Endometriosis is a common benign gynecological disease that seriously affects the physical and mental health of women of childbearing age. Laparoscopic surgery is currently the dominant approach for relieving symptoms and promoting fertility in patients with endometriosis [1]. However, the recurrence rate is as high as 40-50% within 5 years of surgery [2]. Recurrent endometriotic lesions can arise from incompletely removed or unrecognized lesions [2–4]. To effectively reduce the recurrence rate, accurate intraoperative identification and complete resection of endometriotic lesions are critical [5–7].

Traditional imaging approaches such as ultrasound [8–10] and MRI [11, 12] lack sufficient sensitivity and specificity for endometriosis, and it is difficult to recognize subtle lesions and provide real-time intraoperative imaging [8]. As an emerging imaging technology, NIRF imaging has been widely employed in clinical practice in recent years and is more suitable for intraoperative real-time imaging because of its deeper tissue penetration and weaker autofluorescence background signals [13].

Indocyanine green (ICG), the only clinically approved near-infrared (NIR) fluorophore, has been used in intraoperative imaging for endometriosis, but it cannot improve the detection of endometriotic lesions compared to traditional white light imaging owing to its nonspecificity [14]. Specific imaging agents are required to detect endometriosis.

GnRHR is mainly expressed in the pituitary and pelvic-abdominal reproductive systems [15], and a high expression of GnRHR has been reported in endometriotic cells and tissues [16–19]. Hence, GnRHR is a promising imaging target in endometriosis. As the binding ligands of GnRHR, GnRH analogues (both agonists and antagonists) have been proven to have a high affinity for tumor cells [20]. Conjugates of GnRH analogues with cytotoxic drugs have shown potential for targeted cancer therapy [21]. The ¹¹¹In-labeled GnRH peptides effectively detected human prostate cancer [22] and breast cancer [23]. Our previous study indicated that GnRHa-modified ICG can specifically recognize and image peritoneal metastases in ovarian cancer [24].

In this study, the clinical value of GnRHa-ICG was first investigated in endometriosis. The expression of GnRHR was detected in endometriosis tissues and cell lines to evaluate the feasibility of GnRHR as an imaging target for endometriosis, and subsequently, the binding capacities of GnRHa, GnRHa-ICG, and ICG were determined in vitro and in vivo. The performance of intraoperative imaging and image-guided resection was assessed in mouse models of endometriosis. The biodistribution and safety of GnRHa-ICG were investigated in vivo.

2. Materials and Methods

2.1. Cell Culture

The immortalized human endometriotic epithelial cell line 12Z (cat. No. T0764) and human immortalized endometrial stromal cells (ESC) (cat. No T0533) purchased from ABM (Richmond, BC, Canada) were used for the experiments. The 12Z cells were cultured in Prigrow III medium (Cat. No TM003, ABM), and HESC were cultured in DMEM/F-12 medium, both of which were supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Cells were grown at 37°C, 5% CO2 in a humidified environment.

2.2. Immunohistochemistry

Human tissue samples comprising 45 ovarian endometriomas, 17 deep infiltrating endometrioses, and 6 normal endometria samples were obtained from the human tissue bank of the Obstetrics and Gynecology Hospital of Fudan University with the Ethics Committee’s approval. Formalin-fixed and paraffin-embedded sections were stained with hematoxylin-eosin (HE). For immunohistochemistry (IHC), samples were incubated with a 1 : 100 dilution of GnRHR antibody (ab183079, Abcam) and then incubated with a 1 : 500 dilution of secondary antibody (anti-rabbit HRP IgG, 111-035-003, Jackson ImmunoResearch). DAB substrate solution (8059, CST) was used as the chromogen.

2.3. Western Blot

Western blot analysis was performed as described [24]. Proteins were visualized by chemiluminescence using the ImageQuant LAS4000 system (GE Healthcare), and each immunoblot was performed at least three times.

2.4. Fluorescence Microscopy

The cells were cultured to 60–70% confluence in an 8-well chamber slide (Ibidi). GnRHa peptide was synthesized as the sequence of GnRH antagonist cetrorelix, and the N-terminus was labeled with fluorescein isothiocyanate (FITC) to synthesize GnRHa-FITC (GL Biochem Ltd., Shanghai, China). 12Z cells were incubated with 10 μmol/L GnRHa-FITC for 60 min at 37°C, and 12Z and HESC cells were incubated with 10 μmol/L GnRHa-ICG or ICG (Sigma-Aldrich) for 30 min at 37°C. After incubation, the cells were washed 3 times with PBS and fixed in 4% paraformaldehyde for 10 min, mounted with DAPI. TCS SP5 confocal microscope (Leica) was used to analyze the samples (excitation wavelength 633 nm and emission 780 nm long-pass filter, ×100 oil objective, ×1000 magnification), and ImageJ software (version 1.50 g) was used to calculate the mean fluorescence intensity for quantitative analysis.

2.5. Flow Cytometry

12Z and HESC cells were cultured to 70-80% confluency in 6-well plates. 12Z cells were first incubated with 10 μmol/L GnRHa-FITC for 60 min at 37°C, and 12Z and HESC cells were incubated with GnRHa-ICG/ICG (10 μmol/L) for 30 min at 37°C. For the blocking experiments, 12Z cells were pretreated with 100 μmol/L GnRHa peptide for 10 min and then incubated with 10 μmol/L GnRHa-ICG for 30 min at 37°C. The samples were measured on a CytoFLEX flow cytometer (Beckman Coulter). All the samples were examined in triplicate. The data were analyzed using FlowJo software (version X 10.0.7).

2.6. Cell Viability Assay

12Z cells were cultured to 40–50% confluence in 96-well plates. The cells incubated with GnRHa-ICG were divided into four groups at different concentrations (0, 1, 10, and 100 μmol/L), and each group contained four subwells. Cell viability was assessed after exposure for 48 h using the Cell Counting Kit-8 assay (Dojindo). Absorbance at 450 nm (reference wavelength: 630 nm) was measured using a microplate reader.

2.7. Animal Model

Six- to eight-week-old female green fluorescent protein- (GFP-) transgenic C57BL/6 mice (Slack Experimental Animals Co. LT; Shanghai, China) were used in this study. All mice were maintained under controlled conditions with a light/dark (12/12 h) cycle and had ad libitum access to chow and water. The animal experiments were approved by the Institutional Animal Care and Use Committee of Fudan University. GFP transgenic C57BL/6 mice were selected as donors of uterine tissue fragments and were initially treated i.m. with 100 mg/kg estradiol benzoate (Animal Medicine Factory, Hangzhou, China) twice a week after one week of acclimation. Wild-type female C57BL/6 mice were designated as recipients. One week after treatment with estrogen, the uteri from the sacrificed donor mice were harvested in a petri dish, washed twice with sterile saline, and then split longitudinally. Finally, the uterine tissues were minced into fragments with diameters smaller than 1 mm, and then, the uterine fragments were suspended in sterile saline and injected into the abdominal cavities of recipient mice. The uterus of each donor mouse was injected in equal amounts into two recipient mice. The successful establishment of the model was indicated when nodules were palpable from the abdominal wall after 3–4 weeks of induction.

2.8. Near-Infrared Fluorescence Imaging

The synthesis and characterization of GnRHa-ICG are described in the study by Liu et al. [24]. The endometriotic model mice were sacrificed at the indicated times after intraperitoneal injection of GnRHa-ICG (1 mg/kg) or ICG (0.37 mg/kg), and the GFP and NIRF signals were measured using the IVIS Lumina K imaging system (PerkinElmer, Waltham, US). Each group had 3 mice, and the probe was administered after 24 h of fasting. NIRF images were obtained using a 780 nm excitation filter and an 845 nm emission filter. For ex vivo imaging, allografts and organs were dissected and analyzed immediately after sacrifice. Fluorescence signals were quantified as the average radiant efficiency ( $[p / s / c m^{2} / sr] / [μ W / c m^{2}]$ ) using Living Image software. Fluorescence intensity was measured by drawing a region of interest (ROI) around the area. The signal-to-noise ratio (SNR) was calculated as the average fluorescence intensity of the tumor divided by that of skeletal muscle or intestine.

2.9. Target Image-Guided Resection of Endometriotic Foci

The targeted probe GnRHa-ICG (1 mg/kg) was injected intraperitoneally into model mice at the indicated times, and the abdominal cavities were exposed after sacrifice. NIRF image- and white light-guided surgeries were performed using a clinically used fluorescence navigation system (FloNavi, Optomedic Technique Inc., Guangdong, China). Endometriotic foci were resected under white light and NIRF guidance ( $n = 8$ per group). The number and maximum diameter of the foci confirmed to be endometriotic by histopathology were recorded. To evaluate the accuracy of the probe, ten endometriotic model mice were intraperitoneally injected with GnRHa-ICG, followed by GFP and NIRF imaging. The GFP-positive tissues were resected under the guidance of NIRF imaging, and the sensitivity and specificity of the probe were calculated with the histopathology examination as a reference.

2.10. In Vivo Toxicity Tests

GnRHa-ICG toxicity was determined in BALB/c mice ( $n = 3$ ). Two groups received intraperitoneal injections of 1.0 mg/kg GnRHa-ICG and were followed for 2 and 120 h. The control group received the vehicle alone. Blood was drawn to assess alanine transaminase (ALT), aspartate transaminase (AST), blood urea nitrogen (BUN), creatinine (CREA), white blood cells (WBC), and red blood cells (RBC). Heart, lung, liver, spleen, and kidney tissues were harvested for HE staining.

2.11. Statistical Analysis

Student’s $t$ -test was used to compare the intensities. Differences were considered significant at $P < 0.05$ and are reported as the $mean \pm SD$ . All statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, Inc.).

3. Results

3.1. GnRHR Expression in Human Endometriotic Tissues and Cell Lines

GnRHR expression was observed in 100% (6/6) of endometria (EM), 88.9% (40/45) of ovarian endometrioma (OE), and 88.4% (15/17) of deep endometriosis (DIE) samples by IHC analysis (Figure 1(a)). GnRHR was highly expressed in the human immortalized endometriotic epithelial cell line 12Z and negatively expressed in human endometrial stromal cell ESC, as determined by western blotting (Figure 1(b)).

Figure 1

Expression of GnRHR in the endometriotic tissues and cell lines. (a) Representative IHC staining for GnRHR in human normal endometrium (EM), ovary endometriosis (OE), and deep infiltrating endometriosis (DIE). (b) Western blot showing GnRHR expression in immortalized human endometriotic epithelial cell line 12Z and endometrial stromal cells (ESC). (c) Representative fluorescence microscopy images and (d) flow cytometry analysis of 12Z cells after incubation with GnRHa-FITC. Scale bar, 50 μm.

(b)

(c)

(d)

3.2. The Binding Capacity of GnRHa Peptide to Endometriotic Cells

To evaluate whether GnRHa peptide binds to endometriotic cell lines, we labeled GnRHa peptide with FITC and incubated GnRHa-FITC with immortalized human endometriotic epithelial 12Z cells. Strong fluorescence signals of GnRHa-FITC were detected in 12Z cells via fluorescence microscopy and flow cytometry (Figures 1(c) and 1(d)).

3.3. Enhanced Cell Binding Capacity of GnRHa-ICG to Endometriotic Cells

To evaluate the binding capacity of GnRHa-ICG to endometriotic cells, we incubated GnRHa-ICG with the 12Z and ESC cell lines. A significant difference in fluorescence intensity was detected between GnRHR-positive 12Z cells and GnRHR-negative ESC (Figure 2(a)). The mean fluorescence intensity of GnRHa-ICG was 2.36-fold higher than that of ICG in the endometriotic 12Z cells (Figure 2(b)). After GnRHR blockade with unlabeled GnRHa peptides, GnRHa-ICG binding was significantly suppressed in GnRHR-positive 12Z cells (Figure 2(c)). Altogether, these results demonstrated the GnRHR-specific binding capacity of GnRHa-ICG.

Figure 2

Cell binding capacity of GnRHa-ICG. (a) Comparison between GnRHa-ICG and ICG in 12Z and ESC cells. Scale bar, 50 μm; magnification, ×1000. (b) Flow cytometry analysis of 12Z cells after incubation with GnRHa-ICG and ICG. (c) Flow cytometry analysis of 12Z cells after the GnRHa block ( $^{*} P < 0.01$ ; $^{* *} P < 0.001$ ; $^{* * *} P < 0.0001$ ).

(b)

(c)

3.4. Targeted Imaging of GnRHa-ICG in Mice Bearing Endometriosis

The fluorescence intensities of EMT and background tissues (muscle and intestine) per dose group are shown in Figure S1, indicating a low background in normal tissues. A dose of 1.0 mg/kg was used for the following experiments because of the higher SNR compared with 0.5 mg/kg and similar SNR with 1.5 mg/kg. Two hours postinjection with GnRHa-ICG or ICG and GFP and NIR fluorescence images of mouse abdomens were obtained and analyzed using the IVIS Lumina K imaging system (Figure 3(a)). More EMT foci were imaged under NIRF in the GnRHa-ICG group than in the white light group. The NIRF signals of the EMT foci were consistent with the GFP fluorescence signals from donor mice. Moreover, GnRHa-ICG specifically localized to EMT foci, whereas accumulation of ICG was mainly observed in the intestine.

Figure 3

GnRHa-ICG targets endometriotic lesions. (a) Representative GFP and NIRF images of the pelvic and peritoneal cavity 2 h after injection of GnRHa-ICG, ICG, and NaCl in GFP mouse models for endometriosis. Yellow dotted lines indicate the EMT locations. (b) Ex vivo near-infrared fluorescence images of the EMTs and mouse organs 2 h after intraperitoneal injection of GnRHa-ICG, ICG, and NaCl. (c) Histopathological analysis of an EMT slice showing the colocalization of endometriotic cells, GnRHR expression, and GnRHa-ICG fluorescence. Scale bar, 250 μm. (d) Comparison of fluorescence intensities between EMT, muscle, and intestine.

(b)

(c)

(d)

Ex vivo NIRF images of the EMT foci and other organs demonstrated that the fluorescence signals were mainly concentrated in the EMT foci and liver after intraperitoneal injection of GnRHa-ICG, whereas they were predominantly concentrated in the intestine, liver, uterus, and kidney in the ICG group (Figure 3(b)). The fluorescence signal intensities of the ex vivo tissues and organs are shown in Figure S2. The EMT foci imaged by GnRHa-ICG were also confirmed by pathology and showed GnRHR expression (Figure 3(c)). Furthermore, GnRHa-ICG exhibited the highest fluorescence signals and SNR in the EMT foci at 2 h postinjection and a stable SNR between 0.5 and 48 h. However, the fluorescence signals of ICG were mainly concentrated in the intestine and decreased exponentially (Figure 3(d)). The poor SNR and rapid decay of ICG signals influenced the detection of the EMT foci.

3.5. Targeted NIRF Imaging-Guided Resection of Endometriotic Lesions

To further evaluate the clinical value of GnRHa-ICG, EMT resections were performed under the guidance of white light and NIRF 2 h after intraperitoneal injection of GnRHa-ICG (Figure 4(a)). Minute residual endometriotic lesions in the abdominal cavity after white light-guided surgery were found on NIRF imaging via GnRHa-ICG. These residual foci were resected under NIRF imaging and were pathologically confirmed to be endometriotic tissues (Figure 4(b)). In vivo and ex vivo imagings using a clinically applied intraoperative fluorescence navigation imaging system were the same as those of the IVIS Lumina K imaging system (Figure S3). The median diameter of the EMT lesions resected in the NIRF-guided group ( $3.8 \pm 2.4$ mm, $n = 37$ ) was significantly smaller than that in the white light-guided group ( $7.0 \pm 2.9$ mm, $n = 28$ ) (Figure 4(c)). Under the guidance of NIRF using GnRHa-ICG, the minimum detectable lesions were 1 mm in diameter (Figure 4(d)). GnRHa-ICG provided excellent sensitivity (97.3%) and acceptable specificity (77.8%), with pathological examination as a reference.

Figure 4

Targeted recognition capacity of GnRHa-ICG. (a) Comparison of recognition capacity to the EMTs between visible-guided surgery and NIRF-guided surgery. After visible surgical resection, no visible lesions were observed, but there were remaining hidden lesions removed under the guidance of NIRF. Yellow arrows indicate the EMT locations. (b) The tissues resected with NIRF-guided surgery were confirmed to be endometriotic tissues by histopathology. (c) Comparing the maximum diameters of the EMTs resected by imaging-guided surgery (red circles) and visible-guided surgery (blue circles) (37 vs 28 biopsies, $n = 8$ mice), and the median size is $3.8 \pm 2.4$ and $7.0 \pm 2.9$ mm, respectively ( $^{*} P < 0.01$ ). (d) The recognition accuracy of GnRHa-ICG indicates that the probe can identify the lesions as small as 1 mm in diameter.

(b)

(c)

(d)

3.6. Biodistribution and Dynamics of GnRHa-ICG in Mice Bearing Endometriosis

To evaluate the biodistribution of the GnRHa-ICG probes, ex vivo EMTs and organs were collected 2 h postadministration. Confocal imaging confirmed the high accumulation of GnRHa-ICG in the EMT and liver, while ICG showed strong fluorescence signals in the intestine and liver and a weak signal in the EMT (Figure 5). These results are consistent with the in vivo fluorescence imaging.

Figure 5

Comparison of biodistribution between GnRHa-ICG and ICG. Confocal microscopy of frozen sections of the EMTs and other organs in mice injected with (a) GnRHa-ICG and (b) ICG. Scale bar, 75 μm.

(b)

The dynamics of GnRHa-ICG and ICG in mice with endometriosis were evaluated. Mice were intraperitoneally injected with GnRHa-ICG or ICG and monitored for 48–120 h. As shown in Figures 6(a) and 6(b), the fluorescence signals of GnRHa-ICG reached the highest peak in EMTs and the liver at 2 h postadministration and then decreased gradually with a good SNR within 48 h. Liver clearance was the main metabolic pathway and was almost completely metabolized at 120 h postadministration. Although normal uterus and ovary tissues also showed a slight increase in fluorescence signals, indicating GnRHR-specific binding of the probe, the EMTs were high enough to differentiate them from those of normal reproductive tissues. Few fluorescent signals were detected in the intestines, kidneys, and other normal tissues. In contrast, the fluorescence signals of ICG were mainly concentrated in the intestine and liver and almost completely metabolized at 48 h postadministration (Figures 6(c) and 6(d)). In addition to the EMTs, the uterus and kidney also showed fluorescence signals owing to the nonspecificity of ICG.

Figure 6

Dynamics and biodistribution of GnRHa-ICG and ICG. Representative fluorescence images of ex vivo endometriotic tissues and organs after injection of GnRHa-ICG (a) or ICG (c). Biodistribution of GnRHa-ICG (b) or ICG (d).

(b)

(c)

(d)

3.7. Toxicity of GnRHa-ICG In Vitro and In Vivo

To address concerns of possible side effects caused by GnRHa-ICG, we assessed the viability of GnRHR-positive 12Z cells after dose-gradient incubation with GnRHa-ICG. Exposure to a high concentration of GnRHa-ICG did not decrease the cell viability (Figure S4). In mice intraperitoneally injected with GnRHa-ICG or vehicle control, no significant differences were observed in WBC, RBC, ALT, AST, CREA, or BUN at 2 h and 120 h postinjection (Figure 7(a)). Histological examination of the main organs did not reveal any obvious damage (Figure 7(b)).

Figure 7

Toxicity of GnRHa-ICG in BALB/c mice. (a) Assessment of the liver panel, kidney panel, and blood cells. (b) HE staining of the heart, lung, liver, spleen, and kidney. Scale bar, 200 μm.

(b)

4. Discussion

In our study, we first reported intraoperative imaging and imaging-guided surgery using the targeted near-infrared fluorescent probe GnRHa-ICG in an endometriotic mouse model, which suggested a strong potential in clinical intraoperative imaging for endometriosis.

Over the past few decades, imaging approaches have been reported for the intraoperative detection of endometriosis; however, their diagnostic performance has been limited. Methylene blue, a widely used thiazide dye in surgical imaging, exhibits blue staining of peritoneal endometriotic lesions, but the colored peritoneal areas were not histopathologically confirmed [25, 26]. 5-Aminolevulinic acid (5-ALA), an endogenous nonproteinogenic amino acid, has been used to visualize bladder cancer [27], early-stage lung cancer [28], and malignant gliomas [29] using photodynamic detection. It has also been reported to help in the detection of red and white peritoneal endometriotic lesions, but not pigmented lesions [30], and its high rate of topical phototoxicity cannot be ignored [31].

Autofluorescence, the natural emission of light by biological structures, such as mitochondria and lysosomes [32], has been proven to be advantageous in the diagnosis of precancerous lesions in the bronchial tract [33] and bladder [34]. Buchweitz et al. reported the diagnostic value of autofluorescence imaging in the detection of red and occult peritoneal endometriotic lesions compared with white light, but no help in endometriomas and DIE [35]. Therefore, a specific imaging probe needs to be developed to improve diagnostic performance in endometriosis.

Targeted imaging of endometriosis can be performed by conjugating imaging agents with endometriotic-specific ligands (such as small molecules, peptides, proteins, and antibodies) [36, 37]. In 2006, 3-aminoethyl estradiol (EDL) was conjugated to glutamate peptide (GAP) to yield GAP-EDL (a functional estrogen ligand) and then labeled with ^99mTc to develop a radioactive imaging probe ^99mTc-GAP-EDL, which was first reported to target endometriotic lesions in a rabbit model of endometriosis [38]. Considering the superexpression of vascular endothelial growth factor (VEGF) in endometriosis [39–42], ^99mTc-labeled monoclonal anti-VEGF antibody (bevacizumab-^99mTc) was used as a radiopharmaceutical for endometriosis imaging in an endometriotic rat model in 2015 [43]. Moses et al. constructed silicon naphthalocyanine- (SiNc-) loaded polymeric nanoparticles (SiNc-NP) that efficiently delineated endometriotic lesions with NIR fluorescence signals in a mouse model of endometriosis in 2020 [44]. However, the reported targeted imaging probes have the universal shortcomings of radioactivity, insufficient targeting, or toxicity side effects, which restrict their clinical application.

Gonadotropin-releasing hormone receptor, which was mentioned in our study as GnRHR-I, is mainly expressed in the anterior pituitary and reproductive tissues, such as the breast, ovary, and prostate [45–47]. Higher GnRHR expression was observed in reproductive tissues than in other tissues from the GTEx portal database (data not shown). Moreover, high expression of GnRHR has been reported in endometriotic cells and tissues [18, 19], which is consistent with our results showing a nearly 90% expression rate in both ovarian endometrioma and deep infiltrating endometriosis. This finding suggests its potential as a novel imaging target in endometriosis.

NIRF imaging is an emerging real-time imaging technology that has been widely used in fluorescence-guided surgery because of its high penetration depth and low autofluorescence background [48]. ICG, the FDA-approved near-infrared fluorophore, has been widely used in clinical angiography and lymphography [49, 50]. ICG can visualize highly neovascularized endometriotic lesions under near-infrared fluorescence imaging, but its performance in the detection of endometriosis during surgery is inconsistent [51]. In our study, ICG showed high fluorescence in the liver and intestine and weak signals in the endometriotic lesions, indicating nonspecific imaging.

The use of NIRF imaging for molecularly guided surgical resection of cancers could reduce residual tumor burden and surgical morbidity associated with excising sufficient tissues to avoid positive surgical margins [52]. Based on our previous research [24], we conjugated GnRH peptides with a high affinity for binding to GnRHR with ICG to synthesize a targeted imaging probe, GnRHa-ICG. We demonstrated that GnRHa-ICG could specifically recognize endometriotic lesions from pelvic and peritoneal normal tissues and easily distinguish EMTs in mouse models of endometriosis. After intraperitoneal administration, GnRHa-ICG was mainly concentrated in the EMTs and liver, whereas ICG was mainly concentrated in the intestine and liver. We also observed that ICG exhibited a weak NIRF signal in the EMTs, but its fluorescence intensity decreased quickly within 24 h.

To further evaluate the target recognition capability of GnRHa-ICG for endometriosis, white light and NIRF-guided surgeries were performed. GnRHa-ICG could recognize tiny endometriotic lesions that could not be identified under white light, and the minimum detectable lesions were up to 1 mm in diameter. GnRHa-ICG has the potential to help localize tiny residual lesions that the naked eye cannot identify during surgery. Moreover, the signals of ICG decreased exponentially in mice after intraperitoneal administration, whereas GnRHa-ICG reached the highest concentration in the EMTs at 2 h and still had a good SNR within 48 h of administration. This period provides flexible administration time for clinical intraoperative imaging.

The toxicity of GnRHa-ICG was assessed, and no hematotoxicity and hepatotoxicity were observed in the study. The short-term administration of GnRHa-ICG is safe in vivo.

5. Conclusion

In summary, GnRHa-ICG specifically recognized and localized endometriotic lesions and showed high fluorescence signals from 2 to 48 h postadministration. Targeted NIRF imaging-guided surgery completely resected the endometriotic lesions with high accuracy. Future clinical applications might help surgeons accurately resect lesions to reduce the postoperative recurrence of endometriosis.

Footnotes

Data Availability

All datasets generated for this study are included in the article/supplementary material.

Conflicts of Interest

QL, CX, and XZ have a patent for GnRHa-ICG licensed to the Obstetrics and Gynecology Hospital of Fudan University. The remaining authors have no conflicts of interest.

Authors’ Contributions

Jing Peng conducted the data curation, writing of the original draft preparation, visualization, and investigation; Qiyu Liu conducted the investigation, software, and validation; Tao Pu, Mingxing Zhang, Meng Zhang, and MingDu conducted the investigation; Guiling Li carried out resources; Xiaoyan Zhang conducted the conceptualization and methodology and wrote, reviewed, and edited the paper; Congjian Xu conducted the conceptualization and methodology. Jing Peng and Qiyu Liu have contributed equally to this work.

Acknowledgments

We express our thanks to Guangdong OptoMedic Technologies Inc. for their assistance with the fluorescence navigation system (FloNavi). We are grateful to Ms. Xiao Guo of the Joint Live Small Animal Imaging Laboratory of Fudan University Shanghai Medical College-PerkinElmer Company for technical support with the use of the IVIS Lumina K imaging system. This work was supported by the National Natural Sciences Foundation of China (grant no. 82172747), the National Key R&D Program of China (grant no. 2016YFC1303100), and the Shanghai Medical Center of Key Programs for Female Reproductive Diseases (grant no. 2017ZZ01016).

Supplementary Materials

References

Marcoux

Maheux

Bérubé

Laparoscopic surgery in infertile women with minimal or mild endometriosis

New England Journal of Medicine 1997 337

6674054

4 217 222

10.1056/NEJM199707243370401

2-s2.0-0030811896

Guo

S.-W.

Recurrence of endometriosis and its control

Human Reproduction Update 2009 15

6674054

4 441 461

10.1093/humupd/dmp007

2-s2.0-66249098615

Roman

Hennetier

Darwish

Badescu

Csanyi

Aziz

Tuech

J.-J.

Abo

Bowel occult microscopic endometriosis in resection margins in deep colorectal endometriosis specimens has no impact on short-term postoperative outcomes

Fertility and Sterility 2016 105

6674054

2 423 429.e7

10.1016/j.fertnstert.2015.09.030

2-s2.0-84948845609

26474734

Guo

S.-W.

Martin

D. C.

The perioperative period: a critical yet neglected time window for reducing the recurrence risk of endometriosis?

Human Reproduction 2019 34

6674054

10 1858 1865

10.1093/humrep/dez187

2-s2.0-85074118062

31585460

Vercellini

Frontino

De Giorgi

Aimi

Zaina

Crosignani

P. G.

Comparison of a levonorgestrel-releasing intrauterine device versus expectant management after conservative surgery for symptomatic endometriosis: a pilot study

Fertility and Sterility 2003 80

6674054

2 305 309

10.1016/S0015-0282(03)00608-3

2-s2.0-0042065171

12909492

Alborzi

Momtahan

Parsanezhad

M. E.

Dehbashi

Zolghadri

Alborzi

A prospective, randomized study comparing laparoscopic ovarian cystectomy versus fenestration and coagulation in patients with endometriomas

Fertility and Sterility 2004 82

6674054

6 1633 1637

10.1016/j.fertnstert.2004.04.067

2-s2.0-10044274463

15589870

Vignali

Bianchi

Candiani

Spadaccini

Oggioni

Busacca

Surgical treatment of deep endometriosis and risk of recurrence

Journal of Minimally Invasive Gynecology 2005 12

6674054

6 508 513

10.1016/j.jmig.2005.06.016

2-s2.0-28744441609

Nisenblat

Bossuyt

P. M. M.

Farquhar

Johnson

Hull

M. L.

Cochrane Gynaecology and Fertility Group

Imaging modalities for the non-invasive diagnosis of endometriosis

Cochrane Database of Systematic Reviews 2016 2016

6674054

2, article CD009591

10.1002/14651858.CD009591.pub2

2-s2.0-84961891999

Guerriero

Ajossa

Minguez

J. A.

Jurado

Mais

Melis

G. B.

Alcazar

J. L.

Accuracy of transvaginal ultrasound for diagnosis of deep endometriosis in uterosacral ligaments, rectovaginal septum, vagina and bladder: systematic review and meta-analysis

Ultrasound in Obstetrics & Gynecology 2015 46

6674054

5 534 545

10.1002/uog.15667

2-s2.0-84946088921

26250349

10.

Noventa

Saccardi

Litta

Vitagliano

D'Antona

Abdulrahim

Duncan

Alexander-Sefre

Aldrich

C. J.

Quaranta

Gizzo

Ultrasound techniques in the diagnosis of deep pelvic endometriosis: algorithm based on a systematic review and meta-analysis

Fertility and Sterility 2015 104

6674054

2 366 383.e2

10.1016/j.fertnstert.2015.05.002

2-s2.0-84938293745

26051099

11.

Bazot

Darai

Hourani

Thomassin

Cortez

Uzan

Buy

J.-N.

Deep pelvic endometriosis: MR imaging for diagnosis and prediction of extension of disease

Radiology 2004 232

6674054

2 379 389

10.1148/radiol.2322030762

2-s2.0-3242776854

15205479

12.

Medeiros

L. R.

Rosa

M. I.

Silva

B. R.

Reis

M. E.

Simon

C. S.

Dondossola

E. R.

da Cunha Filho

J. S.

Accuracy of magnetic resonance in deeply infiltrating endometriosis: a systematic review and meta-analysis

Archives of Gynecology and Obstetrics 2015 291

6674054

3 611 621

10.1007/s00404-014-3470-7

2-s2.0-84925465593

25288268

13.

Weissleder

Ntziachristos

Shedding light onto live molecular targets

Nature Medicine 2003 9

6674054

1 123 128

10.1038/nm0103-123

2-s2.0-0037236238

12514725

14.

Lier

M. C. I.

Vlek

S. L.

Ankersmit

van de Ven

P. M.

Dekker

J. J. M. L.

Bleeker

M. C. G.

Mijatovic

Tuynman

J. B.

Comparison of enhanced laparoscopic imaging techniques in endometriosis surgery: a diagnostic accuracy study

Surgical Endoscopy 2020 34

6674054

1 96 104

10.1007/s00464-019-06736-8

2-s2.0-85065010740

31028547

15.

Harrison

Wierman

Nett

Glode

Gonadotropin-releasing hormone and its receptor in normal and malignant cells

Endocrine-Related Cancer 2004 11

6674054

4 725 748

10.1677/erc.1.00777

2-s2.0-11244260381

16.

Imai

Ohno

Iida

Fuseya

Furui

Tamaya

Presence of gonadotropin-releasing hormone receptor and its messenger ribonucleic acid in endometrial carcinoma and endometrium

Gynecologic Oncology 1994 55

6674054

1 144 148

10.1006/gyno.1994.1264

2-s2.0-0027944974

7959256

17.

Ikeda

Taga

Kurogi

Minaguchi

Gene expression of gonadotrophin-releasing hormone, but not its receptor, in human endometrium and decidua

Molecular and Cellular Endocrinology 1997 135

6674054

2 165 168

10.1016/S0303-7207(97)00202-5

2-s2.0-0031565820

18.

Borroni

Di Blasio

A. M.

Gaffuri

Santorsola

Busacca

Viganò

Vignali

Expression of GnRH receptor gene in human ectopic endometrial cells and inhibition of their proliferation by leuprolide acetate

Molecular and Cellular Endocrinology 2000 159

6674054

1-2 37 43

10.1016/S0303-7207(99)00199-9

2-s2.0-0033958986

10687850

19.

Khan

K. N.

Kitajima

Hiraki

Fujishita

Nakashima

Ishimaru

Masuzaki

Cell proliferation effect of GnRH agonist on pathological lesions of women with endometriosis, adenomyosis and uterine myoma

Human Reproduction 2010 25

6674054

11 2878 2890

10.1093/humrep/deq240

2-s2.0-77958133481

20829343

20.

Maleksabet

Jaliani

H. Z.

Asgari

Ramezani

Erfani

Specific targeting of recombinant human pancreatic ribonuclease 1 using gonadotropin-releasing hormone targeting peptide toward gonadotropin-releasing hormone receptor-positive cancer cells

Iranian Journal of Medical Sciences 2021 46

6674054

4 281 290

10.30476/ijms.2020.83234.1219

34305240

21.

Zhou

Liu

Targeting drug delivery system based on gonadotropin-releasing hormone analogs: research advances

Journal of International Pharmaceutical Research 2014 41

6674054

2 140 148

22.

Fengb

Miao

Evaluation of novel ¹¹¹In-labeled gonadotropin-releasing hormone peptides for human prostate cancer imaging

Bioorganic & Medicinal Chemistry Letters 2017 27

6674054

20 4647 4651

10.1016/j.bmcl.2017.09.004

2-s2.0-85029232336

28917649

23.

Guo

H. X.

Hathaway

Royce

M. E.

Prossnitz

E. R.

Miao

Y. B.

Synthesis and evaluation of novel gonadotropin-releasing hormone receptor-targeting peptides

Bioconjugate Chemistry 2011 22

6674054

8 1682 1689

10.1021/bc200252j

2-s2.0-80051744241

21749045

24.

Liu

Zhou

Feng

Kang

Zhang

Gonadotropin-releasing hormone receptor-targeted near-infrared fluorescence probe for specific recognition and localization of peritoneal metastases of ovarian cancer

Frontiers in Oncology 2020 10

6674054

266

10.3389/fonc.2020.00266

32185134

25.

Manhes

Shulman

Haag

Canis

Demontmarin

J. L.

Infertility due to diseased pelvic peritoneum: laparoscopic treatment

Gynecologic and Obstetric Investigation 1994 37

6674054

3 191 195

10.1159/000292557

2-s2.0-0028216361

26.

Lessey

B. A.

Higdon

H. L.

III Miller

S. E.

Price

T. A.

Intraoperative detection of subtle endometriosis: a novel paradigm for detection and treatment of pelvic pain associated with the loss of peritoneal integrity

Journal of Visualized Experiments 2012 70

10.3791/4313

23288271

27.

Kriegmair

Baumgartner

Knüchel

Stepp

Hofstädter

Hofstetter

Detection of early bladder cancer by 5-aminolevulinic acid induced porphyrin fluorescence

The Journal of Urology 1996 155

6674054

1 105 110

10.1016/S0022-5347(01)66559-5

2-s2.0-0029656086

28.

Baumgartner

Huber

R. M.

Schulz

Stepp

Rick

Gamarra

Leberig

Roth

Inhalation of 5-aminolevulinic acid: a new technique for fluorescence detection of early stage lung cancer

Journal of Photochemistry and Photobiology. B 1996 36

6674054

2 169 174

10.1016/S1011-1344(96)07365-4

2-s2.0-0030294025

9002254

29.

Stummer

Stocker

Wagner

Stepp

Fritsch

Goetz

A. E.

Kiefmann

Reulen

H. J.

Intraoperative detection of malignant gliomas by 5-aminolevulinic acid-induced porphyrin fluorescence

Neurosurgery 1997 42

6674054

3 518 526

30.

Al-Taher

Hsien

Schols

R. M.

Hanegem

N. V.

Bouvy

N. D.

Dunselman

G. A. J.

Stassen

L. P. S.

Intraoperative enhanced imaging for detection of endometriosis: a systematic review of the literature

European Journal of Obstetrics & Gynecology and Reproductive Biology 2018 224

6674054

108 116

10.1016/j.ejogrb.2018.03.020

2-s2.0-85044128902

29573627

31.

Malik

Meyhöfer-Malik

Trutenau

Diddens

Küpker

Diedrich

Pilotstudie zur photodynamischen Diagnostik der Endometriose mittels 5-Aminolävulinsäure

Geburtshilfe und Frauenheilkunde 1998 58

6674054

8 420 425

10.1055/s-2007-1022737

32.

Monici

Cell and tissue autofluorescence research and diagnostic applications

Biotechnology Annual Review 2005 11

6674054

227 256

10.1016/S1387-2656(05)11007-2

2-s2.0-27144515959

33.

Häussinger

Becker

Stanzel

Kreuzer

Schmidt

Strausz

Cavaliere

Herth

Kohlhäufl

Müller

Huber

R. M.

Pichlmeier

Bolliger ChT

Autofluorescence bronchoscopy with white light bronchoscopy compared with white light bronchoscopy alone for the detection of precancerous lesions: a European randomised controlled multicentre trial

Thorax 2005 60

6674054

6 496 503

10.1136/thx.2005.041475

2-s2.0-20444487818

15923251

34.

Zheng

Lau

Cheng

Soo

K. C.

Olivo

Optimal excitation-emission wavelengths for autofluorescence diagnosis of bladder tumors

International Journal of Cancer 2003 104

6674054

4 477 481

10.1002/ijc.10959

2-s2.0-0037457476

12584746

35.

Buchweitz

Staebler

Tio

Kiesel

Detection of peritoneal endometriotic lesions by autofluorescence laparoscopy

American Journal of Obstetrics and Gynecology 2006 195

6674054

4 949 954

10.1016/j.ajog.2006.02.044

2-s2.0-33748796406

16677595

36.

Luo

Zhang

Cheng

Shi

A review of NIR dyes in cancer targeting and imaging

Biomaterials 2011 32

6674054

29 7127 7138

10.1016/j.biomaterials.2011.06.024

2-s2.0-79960835810

37.

Haque

Faizi

M. S. H.

Rather

J. A.

Khan

M. S.

Next generation NIR fluorophores for tumor imaging and fluorescence-guided surgery: a review

Bioorganic & Medicinal Chemistry 2017 25

6674054

7 2017 2034

10.1016/j.bmc.2017.02.061

2-s2.0-85014653750

28284863

38.

Takahashi

Yang

D. J.

Kohanim

C.-S.

D.-F.

Azhdarinia

Kurihara

Zhang

Chang

J. Y.

Kim

E. E.

Targeted functional imaging of estrogen receptors with 99mTc-GAP-EDL

European Journal of Nuclear Medicine and Molecular Imaging 2007 34

6674054

3 354 362

10.1007/s00259-006-0191-6

2-s2.0-33847138138

17021817

39.

Taylor

R. N.

Torres

P. B.

Schickedanz

A. C.

Park

J. K.

Mueller

M. D.

Sidell

Mechanistic and therapeutic implications of angiogenesis in endometriosis

Reproductive Sciences 2009 16

6674054

2 140 146

10.1177/1933719108324893

2-s2.0-59849120875

19001553

40.

Machado

D. E.

Abrao

M. S.

Berardo

P. T.

Takiya

C. M.

Nasciutti

L. E.

Vascular density and distribution of vascular endothelial growth factor (VEGF) and its receptor VEGFR-2 (Flk-1) are significantly higher in patients with deeply infiltrating endometriosis affecting the rectum

Fertility and Sterility 2008 90

6674054

1 148 155

10.1016/j.fertnstert.2007.05.076

2-s2.0-44949171444

17765237

41.

Machado

D. E.

Berardo

P. T.

Palmero

C. Y.

Nasciutti

L. E.

Higher expression of vascular endothelial growth factor (VEGF) and its receptor VEGFR-2 (Flk-1) and metalloproteinase-9 (MMP-9) in a rat model of peritoneal endometriosis is similar to cancer diseases

Journal of Experimental & Clinical Cancer Research 2010 29

6674054

1 1 9

10.1186/1756-9966-29-4

2-s2.0-77449141720

42.

Machado

D. E.

Júnior

A. P.

Santos

J. M.

Mattos

R. M.

dos Santos

T. A.

Seabra

S. H.

Boldrini

L. D.

Machado

J. A.

Nasciutti

L. E.

A GFP endometriosis model reveals important morphological characteristics of the angiogenic process that govern benign and malignant diseases

Histology and Histopathology 2014 29

6674054

43.

Machado

D. E.

Perini

J. A.

Orlando

M. M. C.

Santos-Oliveira

Developing a noninvasive procedure using labeled monoclonal antibody anti-VEGF (bevacizumab) for detection of endometriosis

BioMed Research International 2015 2015

6674054

751460

10.1155/2015/751460

2-s2.0-84937710086

26240826

44.

Moses

A. S.

Taratula

O. R.

Lee

Luo

Grenz

Korzun

Lorenz

A. S.

Sabei

F. Y.

Bracha

Alani

A. W. G.

Slayden

O. D.

Taratula

Nanoparticle-based platform for activatable fluorescence imaging and photothermal ablation of endometriosis

Small 2020 16

6674054

18 e1906936

10.1002/smll.201906936

32250034

45.

Aguilar-Rojas

Huerta-Reyes

Human gonadotropin-releasing hormone receptor-activated cellular functions and signaling pathways in extra-pituitary tissues and cancer cells (review)

Oncology Reports 2009 22

6674054

5 981 990

10.3892/or_00000525

2-s2.0-70449448175

19787210

46.

Gründker

Emons

The role of gonadotropin-releasing hormone in cancer cell proliferation and metastasis

Frontiers in Endocrinology 2017 8

6674054

187

10.3389/fendo.2017.00187

2-s2.0-85027103512

28824547

47.

Limonta

Marelli

M. M.

Mai

Motta

Martini

Moretti

R. M.

GnRH receptors in cancer: from cell biology to novel targeted therapeutic strategies

Endocrine Reviews 2012 33

6674054

5 784 811

10.1210/er.2012-1014

2-s2.0-84870839178

22778172

48.

Hong

Antaris

A. L.

Dai

Near-infrared fluorophores for biomedical imaging

Nature Biomedical Engineering 2017 1

6674054

10.1038/s41551-016-0010

2-s2.0-85029866713

49.

Danhier

To exploit the tumor microenvironment: since the EPR effect fails in the clinic, what is the future of nanomedicine?

Journal of Controlled Release 2016 244

6674054

Part A 108 121

10.1016/j.jconrel.2016.11.015

2-s2.0-84997047916

50.

Maeda

Sawa

Matsumura

Hori

Tumor vascular permeability and the EPR effect in macromolecular therapeutics: a review

Journal of Controlled Release 2000 65

6674054

1-2 271 284

10.1016/S0168-3659(99)00248-5

2-s2.0-0034000453

10699287

51.

Ianieri

M. M.

Della Corte

Campolo

Cosentino

Catena

Bifulco

Scambia

Indocyanine green in the surgical management of endometriosis: a systematic review

Acta Obstetricia et Gynecologica Scandinavica 2021 100

6674054

2 189 199

10.1111/aogs.13971

52.

Zhu

Sevick-Muraca

E. M.

A review of performance of near-infrared fluorescence imaging devices used in clinical studies

The British Journal of Radiology 2015 88

6674054

1045 20140547

10.1259/bjr.20140547

2-s2.0-84919741015

25410320