2.1.

Fluorescent Contrast Agents

A previously developed NIR nerve-specific fluorophore, LGW05-75, was used to provide nerve-specific contrast.²³ LGW05-75 was formulated in DSPE-PEG micelles to increase solubility as previously described.^25,26 LGW05-75 was selected from a previously synthesized library of oxazine-based fluorophores, where it was highlighted as a lead NIR fluorophore [excitation max = 640 nm, emission max = 665 nm, and molecular weight $(\mathrm{MW})=369.42\mathrm{g}/\mathrm{mol}$].²³ A previously developed tumor-targeting Affibody molecule ($\mathrm{MW}\approx 8\mathrm{kDa}$, Affibody Medical, Solna, Sweden) targeted to EGFR and conjugated to the NIR fluorophore IRDye 800CW (LI-COR Biosciences, Lincoln, Nebraska, United States) was used to provide tumor-specific contrast.²⁷ The NIR-labeled anti-EGFR Affibody, termed ABY-029, was selected due to its safety profile in early-phase clinical trials.^27–29

2.2.

Mice

All animal work was approved by the Institutional Animal Care and Use Committee (IACUC) at Oregon Health and Science University. All mice were housed and treated in accordance with an IACUC-approved protocol. Mixed male and female (50/50) CD-1 mice ($n=10$) were used to quantify the PK profile of DSPE-PEG micelle–formulated LGW05-75. Mixed male and female (50/50) athymic nude mice ($n=20$) were used to quantify the LGW05-75 PK profile in the context of tumors, enabling nerve-to-tumor contrast quantification. Mixed male and female (50/50) athymic nude mice ($n=20$) were used to evaluate multi-color nerve and tumor imaging using LGW05-75 and ABY-029, respectively. At the end of the imaging study, mice were humanely euthanized using carbon dioxide inhalation as the primary method and cervical dislocation as the secondary method of euthanasia.

2.3.

Cell Culture and Tumor Implantation

The human epithelial pharyngeal carcinoma cell line, Detroit 562, was purchased from American Type Culture Collection (ATCC; Manassas, Virginia, United States). The Detroit 562 cell line was cultured in Eagle’s minimum essential medium (Corning, Glendale, Arizona, United States) formulated with 10% fetal bovine serum (FBS, Seradigm, Sanborn, New York, United States) and 1% penicillin-streptomycin (Thermo Fisher Scientific, Waltham, Massachusetts, United States). Cells were maintained at 37°C and 5% ${\mathrm{CO}}_2$. The cells were grown to $\sim 80\%$ to 90% confluence for xenograft implantation. Detroit 562 cells were implanted at a concentration of $1\times {10}^6\text{cells}/100\mu \mathrm{L}$. Athymic nude mice were anesthetized using 2% isoflurane. A 25-gauge needle was used to implant bilaterally into the cheek area to grow near the buccal and marginal facial nerves. Tumors were monitored and not allowed to grow larger than $0.75{\mathrm{cm}}^3$ as per the approved IACUC protocol. Tumors were used for imaging studies at sizes ranging from 0.5 to $0.75{\mathrm{cm}}^3$.

2.4.

Imaging and Quantification

All images were collected using a custom-built, previously reported small animal imaging system capable of real-time color and fluorescence imaging.^{23,25,26,30–32} Briefly, the imaging system consisted of a QImaging EXi Blue monochrome camera (Surrey, British Columbia, Canada) for fluorescence detection with a removable Bayer filter permitting the collection of co-registered color and fluorescence images. A PhotoFluor II light source (89 North, Burlington, Vermont, United States) was focused onto the surgical field through a liquid light guide and used unfiltered for white light illumination. For LGW05-75 fluorescence imaging, the PhotoFluor II was filtered with a $620\pm 30\text{-}\mathrm{nm}$ bandpass excitation filter. The resulting fluorescence was collected with a $700\pm 37.5\text{-}\mathrm{nm}$ bandpass emission filter. For ABY-029 fluorescence imaging, the PhotoFluor II was filtered with a $740\pm 20\text{-}\mathrm{nm}$ bandpass excitation filter. The resulting fluorescence was collected with a 780-nm-long pass emission filter. Camera and light source positions were unchanged throughout the imaging studies, which allowed for a quantitative comparison of fluorescence intensities. Camera exposure times ranged from 10 to 1000 ms, where all images were normalized for exposure time. Tissue fluorescence intensities were quantified using custom-written MATLAB code permitting region of interest (ROI) selection for nerves, muscle, and tumor tissues on the white light images but blinded to the fluorescence images. The white light image–selected ROIs were superimposed on the matched fluorescence images enabling unbiased quantification of tissue intensities, which were used to calculate nerve-to-muscle (N/M), nerve-to-tumor (N/T), tumor-to-muscle (T/M), and tumor-to-nerve (T/N) signal-to-background ratios (SBRs).

2.5.

Pharmacokinetic Profile of DSPE-PEG Micelle–Encapsulated LGW05-75

The PK profile of DSPE-PEG micelle–formulated LGW05-75 was quantified in non-tumor-bearing CD-1 mice and athymic nude mice bearing Detroit 562 tumors. CD-1 and nude mice were fed a chlorophyll-free diet (Animal Specialties, Quakertown, Pennsylvania, United States) for 7 days prior to imaging studies. The CD-1 or tumor-bearing nude mice were injected with DSPE-PEG micelle–formulated LGW05-75 at a dose of $7.5\mu \mathrm{mol}/\mathrm{kg}$ intravenously in the lateral tail vein. The PK profile of DSPE-PEG-formulated LGW05-75 was evaluated 0.5, 1, 2, 4, and 6 h after intravenous (IV) administration ($n=10$ CD-1 mice, two mice per time point, 50/50, male/female; $n=20$ nude mice, four mice per time point, 50/50 male/female). The brachial plexus and facial nerves of the CD-1 mice were exposed and imaged to quantify LGW05-75 biodistribution. The marginal and buccal facial nerves as well as the brachial plexus region of the tumor-bearing nude mice were exposed and imaged for LGW05-75 fluorescence intensity. Six tumor-bearing nude mice (50/50, male/female) were also imaged 4 h after vehicle injection (i.e., blank DSPE-PEG micelles) to quantify nerve and tumor autofluorescence. A one-way analysis of variance (ANOVA) was completed to determine the optimal imaging time point for LGW05-75 nerve imaging after IV administration in CD-1 and tumor-bearing nude mice.

2.6.

Co-administration of DSPE-PEG Micelle–Encapsulated LGW05-75 and ABY-029

Following completion of the PK study for LGW05-75, the optimal administration to imaging time point in the buccal region was found to be between 2 and 4 h. Previous studies showed the optimal administration to imaging time point for ABY-029 started at 4 h.²⁸ Using 4 h as the optimal time point for both LGW05-75 and ABY-029, $n=20$ nude mice (50/50, male/female) bearing Detroit 562 tumors were co-administered $7.5\mu \mathrm{mol}/\mathrm{kg}$ of LGW05-75 and 15 nmol of ABY-029 via lateral tail vein injection. Imaging was completed 4 h after systemic administration as a proof of concept that nerves and tumors could be readily visualized using these targeted, spectrally distinct probes at the same time point. An unpaired parametric $t$-test with Welch’s correction was used to determine if there was a statistically significant difference among LGW05-75 administered alone at 4 h versus when it was co-administered with ABY-029 in tumor-bearing mice.

3.1.

DSPE-PEG Micelle–Encapsulated LGW05-75 Nerve PK Profile

The NIR nerve-specific fluorophore, LGW05-75, was formulated with the previously optimized DSPE-PEG micellar dispersion^25,26 to enable quantification of the PK profile. CD-1 mice were administered the DSPE-PEG micelle–encapsulated LGW05-75 intravenously 0.5, 1, 2, 4, or 6 h prior to fluorescence imaging. Nerve-specific contrast was visible in the brachial plexus as well as facial nerves as early as 0.5 h after systemic administration of DSPE-PEG micelle–formulated LGW05-75. Some fluorescence signals in the surrounding muscle in both the brachial plexus and buccal regions were visible at the early imaging time points [i.e., 0.5 and 1 h; Fig. 2(a)]. By 2 to 4 h post-administration, there was high nerve-to-muscle SBR [Figs. 2(b) and 2(c)]. The nerve and muscle tissue fluorescence intensities steadily decreased over the evaluated 6 h time period post-systemic administration. The brachial plexus nerve showed significantly higher N/M SBR versus control tissue autofluorescence throughout the 6-h imaging period [Fig. 2(b)]. Notably, the facial nerves showed significantly higher N/M SBR 1 to 6 h after systemic administration [Fig. 2(c)]. Both the brachial plexus and facial nerve PK profiles showed peak N/M SBR 4 h after systemic administration.

Fig. 2

PK profile of DSPE-PEG micelle–formulated LGW05-75 in CD-1 mice. (a) Brachial plexus (top, white arrow, $\sim 1\mathrm{mm}$ diameter)³² and facial (bottom, white arrowheads, $\sim 0.1\mathrm{mm}$ diameter)³³ nerve color and fluorescence images from representative CD-1 mice at 0.5, 1, 2, 4, and 6 h after intravenous (IV) administration of DSPE-PEG micelle–formulated LGW05-75. A representative control (Ctrl), vehicle-injected mouse with its brachial plexus and facial nerves is also shown. The eye (purple arrow) showed variable fluorescence intensity, which was likely autofluorescence because it was also seen in the control group. Quantification of the nerve and muscle tissue fluorescence (FL) intensities enabled the calculation of the N/M SBR for the (b) brachial plexus and (c) facial nerves. Data were compared with the control autofluorescence to determine the significance of the N/M SBR using a one-way ANOVA not assuming sphericity with the Geisser–Greenhouse correction and an uncorrected Fisher’s least significance difference (LSD) multiple comparison test, where *$P<0.05$ and **$P<0.01$.

3.2.

DSPE-PEG Micelle–Encapsulated LGW05-75 Nerve-to-Tumor PK Profile

The PK profile of DSPE-PEG micelle–encapsulated LGW05-75 was evaluated in tumor-bearing mice, where the tumors were grown in the buccal area to enable ready quantification of tumor, nerve, and surrounding normal tissue fluorescence. Tumor-bearing mice were administered the DSPE-PEG micelle–encapsulated LGW05-75 intravenously 0.5, 1, 2, 4, or 6 h prior to fluorescence imaging. Similar to the PK study of LGW05-75 in non-tumor bearing animals (Fig. 2), some muscle tissue uptake was seen at the early time points post-administration [i.e., 0.5 and 1 h; Fig. 3(a)]. The nerve and muscle tissue fluorescence intensity steadily decreased over the evaluated 6 h time period post-systemic administration [Figs. 3(b) and 3(c)]. The brachial plexus nerve showed significantly higher N/M SBR versus control tissue autofluorescence throughout the 6-h imaging period [Fig. 3(b)]. In the buccal region of interest, all time points showed significant N/M SBR compared with control tissue autofluorescence. The buccal, marginal, and zygomatic facial nerves were readily visible at the 4 h time point, which had the highest N/M SBR [Fig. 3(c)]. Although there was clear nerve specificity at the 2-, 4-, and 6-h time points [Fig. 3(a)], the 4-h time point showed the highest SBR in both the brachial plexus and facial nerve regions [Figs. 3(b) and 3(c)]. The nerve-to-tumor contrast was found to be significant at 2 and 4 h post-administration [Fig. 3(c)]. The optimal imaging time point for DSPE-PEG micelle–encapsulated LGW05-75 was determined to be 4 h after systemic administration based on the quantified N/M SBR.

Fig. 3

PK profile of DSPE-PEG micelle–encapsulated LGW05-75 in tumor-bearing nude mice. (a) Brachial plexus (top, white arrow) and facial (bottom, white arrowheads) nerve color and fluorescence images from representative athymic nude mice bearing a Detroit 562 tumor (purple arrow) at 0.5, 1, 2, 4, and 6 h after IV administration of DSPE-PEG micelle–formulated LGW05-75. A representative Ctrl, vehicle-injected mouse with its brachial plexus and facial nerves is also shown. Quantification of the nerve and muscle tissue fluorescence intensities enabled the calculation of the N/M signal-to-background ratio (SBR) for the (b) brachial plexus and (c) facial nerves. Statistical analysis was completed to determine the optimal time point for LGW05-75 imaging in the presence of a tumor by comparing data from the 0.5-, 1-, 2-, 4-, and 6-h time points to the control, vehicle-injected group. Data were compared with a one-way ANOVA not assuming sphericity with the Geisser–Greenhouse correction followed by an uncorrected Fisher’s LSD multiple comparisons test, where *$P<0.05$, **$P<0.01$, ***$P<0.001$, and ****$P<0.0001$.

3.3.

Nerve and Tumor Highlighting Using Co-administration of DSPE-PEG Micelle–Encapsulated LGW05-75 and ABY-029

The optimal nerve imaging time point following LGW05-75 administration was utilized to visualize murine nerve tissue in animals bearing head and neck tumors. To enable spectrally distinct visualization of the tumor tissue, mice were co-administered EGFR-targeted ABY-029 and nerve-specific LGW05-75. Four hours after systemic co-administration of DSPE-PEG micelle–formulated LGW05-75 and ABY-029, the tumor and facial nerves were exposed for fluorescence imaging. Images were collected with spectrally distinct excitation/emission filters centered at 620/700 nm for LGW05-75 (i.e., 700 nm) and 740/780 nm for ABY-029 (i.e., 800 nm) showing nerve and tumor specificity, respectively [Fig. 4(a)]. Nerve and tumor tissue fluorescence intensities were high in their respective imaging channels, whereas surrounding normal muscle tissue fluorescence intensity was low. Notably, nerve tissue fluorescence intensity was low in the ABY-029 imaging channel, and conversely, tumor tissue fluorescence intensity was low in the LGW05-75 imaging channel [Fig. 4(b)]. The tissue-specific accumulations of the LGW05-75 and ABY-029 probes in the nerve and tumor tissues, respectively, resulted in nerve-to-muscle and nerve-to-tumor tissue $\mathrm{SBR}>1.5$ as well as tumor-to-muscle and tumor-to-nerve $\mathrm{SBRs}>1.5$ [Fig. 4(c)].

Fig. 4

Co-administration of DSPE-PEG micelle–formulated LGW05-75 and ABY-029. (a) Representative images of Detroit 562 tumor–bearing athymic nude mice imaged in color and the optimal FL imaging channels to detect LGW05-75 (700 nm) and ABY-029 (800 nm) 4 h after systemic co-administration of vehicle (Ctrl, first column) or both NIR fluorescent probes (05-75 + ABY-029, second to fourth columns). (b) Quantified tissue fluorescence intensities of the nerve, muscle, and tumor tissues for both vehicle-injected control and fluorophore–administered mice in each fluorescence imaging channel (i.e., 700 nm = LGW05-75 and 800 nm = ABY-029) were used to calculate the (c) nerve and tumor SBR (N/M, N/T, T/M, and T/N) in their respective imaging channels (i.e., 700 nm = LGW05-75 and 800 nm = ABY-029).

The nerve-to-muscle contrast was not statistically different when LGW05-75 was administered alone or in combination with ABY-029 ($P=0.14$) to tumor-bearing mice. In addition, the nerve-to-tumor contrast was not statistically different when LGW05-75 was administered alone or in combination with ABY-029 ($P=0.53$) to tumor-bearing mice. Representative examples of multi-color fluorescence imaging false colored to visualize nerve and tumor tissue showed tissue-specific signal, where nerve tissue was highlighted by LGW05-75 and tumor tissue was highlighted by ABY-029, with a low accumulation of LGW05-75 in the tumor and some non-specific ABY-029 in surrounding normal tissues (Fig. 5).

Fig. 5

Two-color fluorescence nerve and tumor visualization. Representative images of athymic nude mice with Detroit 562 head and neck tumor xenografts 4 h after co-administration of DSPE-PEG-formulated LGW05-75 and ABY-029 co-administration. Images are shown in color (top), green false color for the 700-nm LGW05-75 imaging channel (second row), red false color for the 800-nm ABY-029 imaging channel (third row), and with the 700- and 800-nm false-colored fluorescence images merged (merged FL). The LGW05-75 highlighted the facial nerves, which are shown in the color image by the white arrowheads. The ABY-029 highlighted the Detroit 562 HNSCC tumor, which is shown in the color image by the white arrow.