Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 May 15.
Published in final edited form as: Clin Cancer Res. 2019 Sep 3;25(22):6653–6661. doi: 10.1158/1078-0432.CCR-19-1686

Fluorescence lifetime-based tumor contrast enhancement using an EGFR antibody labeled near infrared fluorophore

Rahul Pal 1, Homan Kang 2, Hak Soo Choi 2, Anand TN Kumar 1,*
PMCID: PMC6858968  NIHMSID: NIHMS1538765  PMID: 31481509

Abstract

Purpose

Imaging techniques for highly specific detection of cancer cells in-vivo can have applications ranging from preclinical drug discovery studies to clinical cancer diagnosis and surgical therapy. While fluorescence imaging using cancer targeted antibodies has shown promise, non-specific probe accumulation in tissue results in significant background fluorescence, reducing detection sensitivity using traditional intensity-based continuous wave (CW) fluorescence imaging. Here we demonstrate that fluorescence lifetime (FLT) imaging can provide significant tumor contrast enhancement over CW intensity in preclinical models of human breast cancer.

Experimental Design

Mice bearing MDA-MB-231 tumors were injected with anti-epidermal growth factor receptor (EGFR) antibody conjugated to the fluorescent dye IRDye800CW (anti-EGFR-800). Time domain fluorescence imaging was performed in vivo and in situ up to 48 hours post dye injection.

Results

Mice injected with anti-EGFR-800 showed a significantly longer FLT (0.7±0.03 ns) compared to the FLT of non-specific probe uptake in liver (0.63±0.05 ns), providing a dramatic improvement in sensitivity and specificity compared to CW intensity. IgG antibody conjugated IRDye800CW did not show an increased FLT compared to normal tissue, suggesting that the FLT increase of anti-EGFR-800 in tumors was due to receptor binding. Using serial surgery, we show that FLT allows the detection of smaller residual tumors in the surgical bed than possible using CW intensity.

Conclusions

Our data suggest that FLT can significantly enhance tumor contrast using fluorescently labeled antibodies, thereby accelerating the efficient clinical application of these probes for margin assessment in image guided surgery and for highly specific detection of tumor receptors in-vivo.

Keywords: Time-domain imaging, Fluorescence lifetime, Cancer detection, Intra-operative imaging, Margin assessment, In-vivo imaging, EGFR antibody

Introduction

Fluorescence imaging of solid tumors has gained significant momentum in recent years, primarily due to advances in optical imaging technologies and development of cancer targeted fluorescent probes. Cancer cell surface marker proteins are attractive targets for cancer detection, effective drug delivery, and therapeutic interventions (1). The epidermal growth factor receptor (EGFR), a member of the ErbB family of trans-membrane tyrosine kinase receptors, is a well-established key regulator of growth, invasion and metastasis of many solid tumors including colorectal cancers (2), non-small cell lung cancer (NSCLC) (3), triple negative breast cancers (TNBC) (4), and head & neck cancers (5). Naturally, EGFR is a suitable target for tumor detection using fluorescence imaging. EGFR targeted fluorescence imaging can be either based on small molecule tyrosine kinase inhibitors (TKI) (e.g. Erlotinib, Gefitinib etc.) (6, 7) or monoclonal antibodies (mAbs) of EGFR (e.g. Cetuximab, Panitumumab etc.) (8, 9) tagged with fluorophores. In phase II/III clinical trials in combination with chemotherapy and radiotherapy, mAbs showed successful EGFR inhibition (10). Additionally, mAbs induce immune response to cancer cells (4), including antibody-dependent cell-mediated cytotoxicity and T-cell-medicated immune response.

Several studies have shown the promise of fluorescence imaging of anti EGFR antibodies conjugated to fluorescent molecules such as Alexa Fluor 488 (11), Cy5.5 (12, 13) and IRDye800CW (14, 15). Preclinical studies (16, 17) and clinical trials (14, 18) have employed antibody-based fluorescence detection of EGFR expression level (19), examination of anti EGFR therapeutic response (16) and tumor margin assessment during surgery (14, 17). Cetuximab and Panitumumab have also shown tolerable safety profiles in humans (9) after conjugation with fluorescent molecules, making them attractive candidates for targeted imaging of cancer in vivo.

Despite their significant promise, a major drawback with the use of antibodies for imaging is a slow clearance from the body, potentially due to their large molecular weight (20, 21). Anti EGFR mAbs clear through the hepatobiliary system, which is usually a slow process (20). The non-specific antibody accumulation, particularly from clearance organs such as the liver (21, 22) can result in significant background fluorescence. Previous studies primarily employed continuous wave (CW) fluorescence imaging (23, 24), which detects the total emitted fluorescence intensity and cannot distinguish non-specific accumulation of contrast agents (such as in liver) (25), from tumor specific uptake, on an absolute scale. CW intensity is also strongly dependent on imaging conditions, such as laser power, detection efficiency and probe uptake. The strong CW intensity from non-specific accumulation may interfere with tumor specific signal in a clinical setting (26–28), thereby lowering sensitivity, increasing false positives, and limiting the size of tumors that can be resected.

An alternative approach to CW imaging is time domain (TD) fluorescence imaging, which allows the detection of fluorescence lifetime (FLT). FLT is a photophysical quantity that refers to the average time spent by a molecule (ñanoseconds) in its excited state, following laser excitation (29). Unlike CW intensity, FLT is largely unaffected by experimental conditions such as excitation power, probe concentration (30) and tissue uptake and is often uniquely indicative of the local tissue environment (31). Our previous work has demonstrated a dramatic improvement in tumor/background contrast using FLT over CW imaging (32) of tumors labeled with indocyanine green (ICG), a non-targeted tumor contrast agent. Although ICG is FDA approved and has been applied for tumor imaging (33–35), it is not designed for tumor targeting and is therefore non-specific to tumor-specific molecular expression. Here we demonstrate, for the first time, FLT-based contrast enhancement of EGFR overexpressing tumors in presence of high background fluorescence from tissue and non-specific liver uptake, using an anti-EGFR antibody labeled fluorophore. For the fluorescent reporter, we used IRDye800CW (LI-COR Biosciences), a near infra-red (NIR) fluorescent molecule that is easily conjugated to peptides and antibodies via standard NHS ester chemistry and has been previously used in clinical trials with panitumumab and cetuximab (9, 36). Although not explored in this paper, the NIR emission of IRDye800CW can enable deep-tissue imaging of tumors based on the FLT contrast reported in this work. We show tumor specific FLT enhancement of IRDye800CW tagged monoclonal anti EGFR antibody (anti-EGFR-800) and delineation of tumor from surrounding normal tissue with high sensitivity/specificity, in a TNBC mouse tumor model. Additionally, the tumor FLT of anti-EGFR-800 was unique across multiple animals within measurement error, and provided strong tumor/background contrast in presence of nearby organs with strong non-specific uptake of anti-EGFR-800. Using a mock surgery, we show the advantage of FLT over CW for detecting small residual tumors in the surgical bed. We validate these in vivo results using ex-vivo microscopy and pathology.

Materials and Methods

Cell culture

MDA-MB-231 cells were purchased from ATCC (Cat# HTB-26) and maintained in DMEM culture medium supplemented with 10% fetal bovine serum and, 1% penicillin-streptomycin (Life Technologies, CA). Cells were harvested at 70–80% confluency for tumor induction.

Antibody conjugation

Monoclonal anti-EGFR antibody (Clone 225, Cat# BE0278) and isotype control IgG1 antibody (Cat# BE0083) were purchased from Bio X Cell (West Lebanon, NH). The anti-EGFR antibody used in this study is a research grade non-therapeutic biosimilar of the monoclonal antibody drug Cetuximab. A high molecular weight protein labeling kit with IRDye 800CW (cat# 928–38040) was purchased from Li-COR (Lincoln, NE). The antibodies anti-EGFR and IgG1 were conjugated with IRDye 800CW according to the manufacturers protocol. Briefly, antibodies were diluted to 1mg/ml concentration with phosphate buffered saline (PBS). The pH of the resulting protein solution was then raised to 8.5 by adding 1 M potassium phosphate buffer. Antibodies were mixed with IRdye800 (60 μg dissolved in 15μl ultrapure water) and incubated for 2 hours at room temperature in the dark. The reaction product was purified using a Pierce Zeba desalting spin column (Cat# 89891). After purification of IRdye800 conjugated anti EGFR antibody (anti-EGFR-800) and IgG1 (IgG-800), the purification product typically contains less than 5% free dye, and the dye to protein ratio was measured using UV-Vis absorbance spectroscopy. Dye to protein ratio of 2.4:1 was found consistently for both anti-EGFR-800 and IgG-800.

Animal model

All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) in accordance with the animal welfare guidelines at the Massachusetts General Hospital. Eight female Nu/Nu mice (anti-EGFR-800: n=5; IgG-800: n=3) were purchased from Charles River Laboratories Inc (Wilmington, WA) and were housed at the animal facility in MGH. Animals were quarantined for one week and kept in a normal diet with 12-hour light and dark cycle. After one week animals were anesthetized with 3% isoflurane and subcutaneously injected with 2×106 MDA-MB-231 cells in 1:1 PBS:Matrigel mixture. Tumor growth was monitored until 5–10 mm diameter was reached at least in one dimension.

TD Imaging system

A previously published custom-built TD imaging system was used for in-vivo and ex-vivo animal studies (37). The small animal imaging system consisted of a titanium sapphire laser (Mai Tai, Spectra Physics; 80 MHz repetition rate; 100-fs pulses) providing 770-nm excitation, a multimode fiber (Thorlabs) delivering light to the sample, and a gated intensified CCD (LaVision, Picostar, 500 V gain, 0.1 to 1 second integration time, 150 ps steps, 256×344 pixels after 4×4 hardware binning). A diffusing filter (Thorlabs, ED1-S50-MD) was used to expand the output of the optical fiber and delivered to the surface of the animal. The average total power across the illumination area (~5 cm diameter) was 10–20 mW. Fluorescence was collected in reflectance mode using an 835-nm −70 nm band pass filter. TD fluorescence imaging was performed with a gate width of 500 ps and 150 ps steps for a total duration of 6 ns per laser duty cycle of 12.5 ns. In-vivo imaging was performed at multiple timepoints for 48 hours after probe injection. Animals were sacrificed after 48 hours and then imaged in-situ by removing the skin from the top of the tumor and surrounding normal tissue (muscle and liver).

Image processing and data analysis

Algorithms for fluorescence image processing were implemented in MATLAB (Mathworks; Natick). Continuous wave (CW) data was generated from each TD data set by adding the images over all the temporal gates. TD data from individual pixels were plotted as time gate vs. log(intensity) and the fluorescence lifetime (FLT) was obtained by fitting the decay portion of TD fluorescence profiles to a single exponential function, e−t/τ(r), where r denotes pixel location and τ(r) constitutes a lifetime map (e. g., Fig. 2). The white light image of whole animals was used to manually identify and create a region of interest (ROI) for the tumor and surrounding normal tissue (muscle/liver). The CW intensities and FLTs from pixels enclosed by the ROIs were used to calculate probability distributions for pixels as normal or tumor. Receiver Operator Characteristic (ROC) curves were obtained by varying the threshold for CW intensity and FLT, and computing sensitivity and specificity. Sensitivity was calculated as the number of pixels within the tumor ROI above the threshold (CW intensity or FLT), divided by the total number of pixels within the tumor ROI. Specificity was calculated as the number of pixels within the normal ROI below the threshold divided by the total number of pixels within the normal ROI.

Figure 2:

Figure 2:

Open in a new tab

In-vivo TD imaging, 48 hours after i.v. administration of anti-EGFR-800 (a, c, e) or IgG-800 (b, d, f) in MDA-MB-231 tumor bearing mice. (a, b) White light images with tumor boundaries indicated by red dotted outline; (c, d) CW intensity from tumors, surrounding liver and muscle; (e, f) FLT maps obtained from single exponential fits to the decay of TD data; (g, h) Distributions of CW intensities and (i, j) FLT distributions (i, j) in the anti-EGFR-800 (g, i) and IgG-800 (h, j) injected animals, within tumor (red) and normal tissue (green); (k) Receiver operating characteristic (ROC) curves calculated based on CW intensity (blue) and FLT (red) thresholding, for the anti-EGFR-800 (k) and IgG-800 (l) injected animals. The arrow in (i) represents the optimal threshold FLT for tumors used for surgical guidance study shown in Fig. 5.

NIR Fluorescence Microscopy

NIR fluorescence microscopy was performed on a Nikon TE2000 with a QImaging color camera, Hamamatsu Orca R2 camera for fluorescence imaging and IVison software for data collection. Tumors, liver and muscle from animals injected with anti-EGFR-800 or IgG-800 were harvested after sacrifice (48 hours post injection), fixed in 10% formalin and embedded in paraffin. 10 μm thick sections were obtained and images of slides without further staining were collected to identify localization of antibody-dye conjugates. Imaging was performed using a Xenon lamp passed through a 750/50 nm BP excitation filter and a 785 nm LP dichroic mirror, and fluorescence was collected through an 810/40 nm BP emission filter (Chroma Technologies). Exposure times were varied to obtain a similar maximum fluorescence value for each fluorescence image, and all NIR fluorescence images were normalized identically for all conditions of an experiment.

Histopathology and immunohistochemistry

Tumors, muscle and liver tissues were fixed in 10% formalin, embedded in paraffin, sectioned (10 μm thickness), and stained with hematoxylin and eosin (H&E) or processed for immunohistochemistry (IHC) with anti EGFR antibody. For IHC, 10 μm thick paraffin embedded tissue sections were dewaxed in xylene and rehydrated in decreasing concentration of alcohol. Antigen retrieval was performed with EDTA (pH 9.0) at sub boiling temperature for 15 minutes. Tissue sections were incubated in 1:50 dilution of anti EGFR antibody (Cat# 4267, Cell Signaling Tech., MA) overnight at 4 °C. Secondary antibody was applied for 30 minutes at 37 °C and slides were developed with DAKO HRP-compatible DAB (Cat# SF-4100, Vector Laboratories, CA) and counterstained with Harris Hematoxylin. Images of H&E and IHC stained tissue sections was performed using an upright Nikon Eclipse 80i microscope. A 20x, 0.75 NA air objective (Nikon) and a color camera MBF CX9000 was used to capture images.

Statistical Analysis

Statistical analysis was carried out using a one-way ANOVA followed by Bonferroni and Holm multiple comparison. P values less than 0.05 were considered significant: *P <0.05, and **P <0.01. The experiments were not randomized, and the investigators were not blinded to allocation during experiments and outcome assessment. Results were presented as mean ± s.d.

Results and Discussion

Fluorescence lifetime contrast between anti-EGFR-800 in tumors and normal tissue

Figure 1 shows representative TD fluorescence data for tissue autofluorescence, anti-EGFR-800 in-vitro and in-vivo in the tumor, muscle and liver of a nude mouse bearing an MDA-MB-231 tumor, 48 hours after injection of anti-EGFR-800. The mean FLT of anti-EGFR-800 in-vitro (0.5 ns) in the tumor (0.7 ns), muscle (0.62 ns), and liver (0.63 ns) were obtained as single exponential fits to TD data as described in Methods. The autofluorescence FLT, measured before anti-EGFR-800 administration was 0.41 ns. The FLTs in muscle and liver were therefore longer than FLT of autofluorescence, indicating that anti-EGFR-800 was retained in normal tissue 48 hours post injection. However, the FLT of anti-EGFR-800 in tumors was longer than the FLT in muscle and liver, indicating a preferential FLT enhancement of anti-EGFR-800 in tumors.

Figure 1:

Figure 1:

Open in a new tab

TD fluorescence decay signals of tissue autofluorescence (solid line), anti-EGFR-800 (abbreviated as aEGFR-800) in-vitro (solid line - triangles), anti-EGFR-800 in-vivo in tumor (solid line - diamonds), muscle (dashed line), and liver (dotted line) 48 hours after injection. The decay curves are represented as the average of 3×3 pixels in TD reflectance fluorescence images. FLTs obtained from single exponential fits to the decay portion of the curves are shown in the inset.

In-vivo detection of tumors in the presence of non-specific liver uptake

In Figure 2 we demonstrate the advantage of FLT for enhanced tumor contrast in the presence of strong non-specific liver uptake in a live mouse. Shown are representative data from anti-EGFR-800 and control IgG-800 injected mice with subcutaneous tumors close to the liver. White light images (Fig. 2a, b) show the location of subcutaneous tumors outlined by red dotted lines. In-vivo imaging of mice 48-hours post anti-EGFR-800 injection showed that the CW intensity (Fig. 2c) in the tumor (dotted outline) and the adjacent liver (solid arrow) were comparable, making the tumor nearly indistinguishable from the liver. Animals that received IgG-800 (Fig.2d), on the other hand, showed lower tumor fluorescence compared to the anti-EGFR-800 mice. While this is expected given IgG is a non-specific antibody and does not bind to EGFR, it is interesting to note that the tumor was still considerably brighter than the surrounding normal muscle tissue, indicating a higher retention of IgG-800 in tumor microenvironment than in muscle. However, the relative CW intensities between the tumor and liver in the IgG-800 group were comparable, similar to the case of the anti-EGFR-800 group.

Representative FLT maps for mice from the anti-EGFR-800 and IgG-800 groups are shown in Fig. 2e and Fig. 2f, respectively. The FLT of the anti-EGFR-800 labeled tumor was 0.7±0.007 ns (mean of the pixels within the red dotted outline in Fig. 2c), which was longer than the liver FLT (0.64±0.006 ns) or the surrounding muscle (0.63±0.01 ns). In contrast, the tumor FLT (0.64±0.005 ns) in the IgG-800 group was similar to the liver (0.63±0.005 ns) and muscle (0.64±0.005 ns) FLT. These observations were quantified by the histograms and ROC analysis shown in Fig. 2g–l. The histograms show the distribution of CW intensities and FLTs within (red) and outside (green) the tumor boundary as identified in the white light images. It is clear that the CW intensity distributions show considerable overlap between tumor and adjacent liver for both anti-EGFR-800 (Fig. 2g) and IgG-800 (Fig. 2h) injected animals. Tumor CW intensity for IgG-800 was appreciably lower than anti-EGFR-800. However, the distribution of FLTs for anti-EGFR-800 (Fig. 2i) in the normal and tumor regions showed minimal overlap, while IgG-800 (Fig. 2j) showed significant overlap in tumor and normal tissue FLT distributions. It is also noteworthy that the FLTs within the tumor for the anti-EGFR-800 case exhibited a narrow range with a standard error of 1.5%, suggesting that the FLT distribution within the tumor was relatively homogenous. This should be compared with the highly heterogeneous distribution of CW fluorescence intensity within and outside the tumor (Fig. 2g,h). ROC curves generated from the histograms (see Methods) show the sensitivity and specificity of tumor discrimination using CW intensity (blue) and FLT thresholding (red) of the data from the anti-EGFR-800 (Fig. 2k) and IgG-800 (Fig. 2l) groups. In the anti-EGFR-800 group, FLT (AUC=0.99) showed a clear improvement over CW (AUC=0.87). Notably, while tumor discrimination in the IgG-800 group was not expected using either CW or FLT contrast, CW measurements showed strong tumor discrimination (AUC=0.95) while FLT performed poorly (AUC=0.74). Despite the better performance of IgG using CW imaging, it should be noted that the tumor CW contrast of the IgG group is not cellularly specific, as we confirm below using microscopy, and could possibly be due to enhanced permeability and retention in the tumor environment. Furthermore, the CW contrast is dependent on probe uptake, and hence tumor size, thereby limiting the detection of small tumors as we will discuss below in the context of surgical resection.

To evaluate the statistical significance of CW and FLT contrast, we sacrificed multiple animals (n = 4 for anti-EGFR-800 and n = 3 for IgG-800) 48 hours after probe injection and quantified the in-situ FLT and CW intensity for tumor, liver and muscle. Fig. 3a shows that the FLT of anti-EGFR-800 in tumors (0.7±0.03 ns) was significantly longer (p < 0.05) than the FLT in liver (0.63 ± 0.02 ns) and muscle (0.63 ± 0.03 ns). Animals that received IgG-800 (Fig. 3b) did not show a tumor specific FLT increase, with the average FLT in tumor (0.64 ± 0.01 ns), liver (0.64 ± 0.01 ns) and muscle (0.65 ± 0.02 ns) being nearly identical across animals, to within measurement error. Additionally, the average liver FLT and muscle FLT were comparable between the IgG-800 and anti-EGFR-800 groups. This indicated that the FLTs observed from liver and muscle were independent of EGFR expression and were possibly due to non-specific retention.

Figure 3:

Figure 3:

Open in a new tab

In-situ quantification of FLT and CW intensity of anti-EGFR-800 or IgG-800 in tumors, liver and muscle. FLTs for a) anti-EGFR-800, and b) IgG-800 in tumor, liver and muscle; CW intensities for anti-EGFR-800 (c) and IgG-800 (d) in tumor, liver and muscle. * p < 0.05

The CW intensity of anti-EGFR-800 (Fig. 3c) in tumor (1.16 × 104 ± 4114.18 a.u.) and liver (1.32 × 104 ± 6611.63 a.u.) were comparable as was also observed in Fig. 2c, while muscle showed low intensity (0.47 × 104 ± 1614.92 a.u.) compared to the tumor. While the tumor/muscle CW intensity contrast was therefore high for individual animals, a large variation was observed (Fig. 3c) in the tumor and muscle CW intensities across multiple animals, possibly due to inter-animal variations in tumor size and probe uptake. Therefore, statistical comparisons based on the Bonferroni and Holm method did not show a significant difference in average CW intensities between tumor and muscle across multiple animals. For the IgG-800 group (Fig. 3f), the CW intensity was 0.67 × 104 ± 2554.38 a.u. in tumor, 0.59×104 ± 2594.1 a.u. in liver and 0.27 × 104 ± 1440.77 a.u. in muscle. Similar to the anti-EGFR-800 group, the CW intensity in tumor, liver and muscle in the IgG-800 group were not significantly different when averaged over all animals.

Ex-vivo validation of EGFR expression and cellular specificity of anti-EGFR-800 fluorescence from tumors

We performed fluorescence microscopy of excised tissue to confirm tumor localization of anti-EGFR-800 in tumors, liver and muscle. Fluorescence microscopy of unstained tissue sections indicated intracellular localization of anti-EGFR-800 (Fig. 4a), which were observed as discrete speckles in the cytoplasm. Fluorescence intensity in the liver (Fig. 4b) was primarily confined in the sinusoids and central vein, indicating non-specific accumulation of anti-EGFR-800. These microscopy data confirm that the CW intensity observed from the liver in Fig. 2c originates from non-specific anti-EGFR-800 accumulation, while the CW intensity from the tumor is due to cellular uptake. Muscle tissue (Fig 4c) showed low fluorescence in the myocytes, which is consistent with the low muscle fluorescence observed in Fig. 2c. Fluorescence microscopy of tumors that received the control contrast agent IgG-800 (Fig. 4d) did not show uptake of IgG-800 in tumor cells, as expected. H&E staining of the tumor confirmed the presence of cancerous cells (Fig. 4e). Similarly, H&E staining of the liver (Fig. 4f) and the muscle (Fig. 4g) indicated histologically normal liver and muscle tissue. The EGFR expression status in tumor (Fig. 4h), liver (Fig. 4i) and muscle (Fig. 4j) were confirmed with IHC. Strong EGFR expression was only observed in tumors, corroborating the intracellular localization of anti-EGFR-800 in tumors shown in Fig. 4a. The lack of EGFR expression in liver and muscle confirms the non-specific nature of fluorescence observed in these tissues.

Figure 4:

Figure 4:

Open in a new tab

Fluorescence microscopy of ex-vivo tissue sections of a tumor (a), liver (b), and muscle (c) labeled with anti-EGFR-800; (d) fluorescence microscopy of tumor from IgG-800 injected mouse; H&E staining of tumor (e), liver (f) and muscle (g); Immunohistochemical staining of tumor (h), liver (i) and muscle (j). Magnification: 20x. Inset in (h) shows a magnified region in the tumor with predominantly membrane staining for EGFR. All ex-vivo measurements were performed after sacrifice 48 hours post injection.

The in vivo and ex vivo data presented in Figs. 2–4 demonstrate that anti-EFGR-800 is both localized in cells and shows a FLT increase, while IgG-800 is not cellularly localized and does not show a FLT change compared to normal tissue. These results suggest that the FLT increase of anti-EGFR-800 in tumors could be attributed to probe sequestration and degradation by the cells following EGFR binding (38). Tissue environmental factors, such as pH and microviscosity, or binding to calcium and sodium ions could ultimately play a role in altering the FLT of the the sequestered probe in the tumor cells (31, 39–41). The same FLT increase could therefore also occur due to other mechanisms that lead to internalization, including binding of NIR dye conjugated antibodies to other cancer-specific cell surface receptors.

Surgical resection of anti-EGFR-800 labeled tumors based on FLT and CW intensity contrast

As discussed above, the CW fluorescence intensity is strongly dependent on tumor size, since it depends on probe accumulation in the tumor, which increases with tumor volume. The tumors considered so far were sufficiently large to produce reasonable intensity contrast against normal muscle in single animals (although there was large variation between animals). Smaller tumors will show significantly reduced tumor CW intensity due to reduced probe uptake, while the tumor FLT should be largely independent of tumor size, thereby enabling detection of smaller tumors in the surgical bed than possible with CW imaging. To validate this hypothesis, we performed surgical resection of an anti-EGFR-800 labeled subcutaneous MDA-MB-231 tumor. Following sacrifice, the subcutaneous tumor was exposed by removing the skin from the top of the tumor and surrounding muscle and surgery was performed in two stages (stage 1 and stage 2). A residual tumor of size ~1 mm was intentionally left on the surgical bed after stage 1. Then in stage 2, the residual tumor was fully removed, exposing the underlying normal muscle. TD imaging of the whole tumor in-situ was performed before and after the two stages of surgery. For the large initial tumor, both the CW intensity and FLT map of the whole tumor (Fig. 5a) showed strong contrast, which can be observed in the histogram distributions of intensity and FLT within the tumor (red) and surrounding normal muscle (green). CW intensity measurements showed an average tumor intensity of 12315.7 ± 1718.5 a.u. and average muscle intensity of 6607 ± 758.5 a.u. Tumor FLT was 0.7 ± 0.01 ns and muscle FLT was 0.64 ±0.01 ns in the same ROIs used to measure CW intensity. Stage 1 surgery (Fig. 5b) showed a significantly reduced CW intensity from the residual tumor (6753.1±662.1 a.u.), making it difficult to distinguish it from background fluorescence, while FLT in the residual tumor stayed constant within measurement uncertainty (0.69 ± 0.004 ns). Correspondingly, significant overlap of tumor and surrounding muscle is seen in histograms of CW intensity values, while FLT histograms did not show any significant alteration caused by the surgery and reduced tumor mass. Fig. 5c shows the CW intensity and FLT map of the surgical bed following Stage 2 surgery. Histograms indicate the CW intensity range did not alter between Stage 1 and Stage 2. However, the FLT map of the surgical bed showed no residual long FLT component corresponding to the tumor bound anti-EGFR-800, indicating complete removal of tumor. Confirmation of successful surgery was performed by IHC staining of EGFR (Fig. 5d) in tissue sections obtained from the residual tumor (brown membranous staining pattern confirming EGFR expressing tumor cells) and the surgical bed (EGFR staining absent).

Figure 5:

Figure 5:

Open in a new tab

In-situ CW intensity and FLT maps of an MDA-MB-231 tumor before and after surgical removal of tumor mass. The animal was sacrificed 48-hours after i.v. injection of anti-EGFR-800, and TD imaging was performed after the skin was removed from the top of the tumor and surrounding muscle. CW intensity image, histogram of CW intensity, FLT map and histogram of FLT map are shown correspondingly from left to right for the whole tumor (a), residual tumor (b) and surgical bed (c); (d) Immunohistochemistry for EGFR in the residual tumor obtained from (C) (left) and surgical bed (right) are shown. Scale bar: 5 mm unless otherwise mentioned.

Summary

We have shown that FLTs of tumors systemically labeled with an EGFR-antibody labeled fluorescent dye, anti-EGFR-800, are significantly longer than the FLTs of non-specific probe in muscle tissue and liver. We showed that this FLT difference can provide significant tumor contrast and sensitivity/specificity enhancement over traditional intensity-based fluorescence imaging. While studies are under way to further understand the mechanisms behind the FLT increase in tumors compared to normal tissue, our data suggests that the FLT change of anti-EGFR-800 could be attributed to an EGFR specific cellular uptake. Irrespective of the mechanisms underlying the observed FLT changes, our data shows that the enhanced sensitivity and specificity for tumor detection using FLT can be exploited for detecting small or residual tumors that would be difficult to delineate based on fluorescence intensity alone.

Several anti-EGFR antibody-tagged NIR fluorophores are currently in FDA approved clinical trials for intraoperative imaging of various cancers (42–44). In particular, IRDye800 conjugated to Panitumumab, an FDA approved EGFR antibody, has been found to be safe for human use under multiple IND directed clinical studies (9). Our results therefore suggest that TD imaging can serve as a new tool for cancer imaging that can provide enhanced tumor sensitivity and specificity using Panitumumab-IRDye800 and other antibody labelled near infrared fluorophores. Currently, FDA approved TD imaging systems are not available for clinical use. This could be attributed to the higher cost of TD systems compared to CW, but more importantly, to the paucity of studies demonstrating the use of FLT contrast for clinical applications. We hope that our work as well as other recent preclinical studies demonstrating FLT contrast for tumor imaging using exogenous agents (41, 45, 46) will motivate the development and clinical translation of TD imaging systems in the near future. Nevertheless, an example where TD imaging of EGFR-targeted fluorescence is more immediately feasibile is the comprehensive margin assessment of resection specimen from surgery. From a regulatory standpoint (47), specimen imaging studies are less cumbersome than direct intraoperative imaging given that the imaging system will not be used directly on patients. Margin assessment of resection specimen is an important unmet need, given that pathology is slow and does not provide comprehensive assessment.

Following validation of margin assessment in resection specimen, TD imaging can be extended to intraoperative imaging of the surgical bed using exogenous agents. Existing NIR fluorescence guided surgery systems (48) can be readily adapted to incorporate TD detection. Depending on the type of application, these systems could employ wide-field, endoscopic or handheld imaging. With successful validation of FLT imaging for clinical resection specimen and intraoperative imaging, we hope that TD image guided surgery systems will become an integral part of cancer surgery. The use of NIR light can allow penetration of up to 0.5 – 1 cm in tissue using planar reflectance imaging. This should not be a limitation for surgery applications since tumors in the surgical bed are usually partially exposed or at most covered by thin layers of tissue.

Penetration depth can be enhanced up to several cm using NIR transmission imaging, as has been demonstrated using diffuse optical tomography of intrinsic tumor contrast in the human breast (49). FLT contrast using cancer targeted agents can therefore be exploited with TD fluorescence tomography (37) for whole body diagnostic imaging of deep-seated tumors in the breast. Continued efforts in developing molecular targeted probes with distinct fluorescence lifetimes (41) may eventually allow multiplexed imaging for quantifying multiple tumor markers simultaneously in vivo (50).

Statement of Translational Relevance.

Near infrared fluorescence imaging is currently being evaluated for enhancing tumor contrast using cancer targeted probes, to enable accurate surgical resection of tumors while minimizing loss of healthy tissue. Despite enhanced tumor fluorescence intensity, fluorescence from non-specific accumulation of targeted probes in healthy tissue reduces overall contrast, sensitivity and specificity. In this report, we show that fluorescence lifetime detection can dramatically improve tumor sensitivity and specificity in the presence of large background fluorescence, in a mouse tumor model of human breast cancer. Both the tumor model and probe used in this study have direct clinical relevance. At least two anti-EGFR-antibodies, namely, panitumumab and cetuximab, have been shown to be safe for human use in IND directed clinical trials for fluorescence guided surgery. The work presented in this paper therefore has immediate translational relevance for both intraoperative image guided surgery and for comprehensive margin assessment in excised resection specimen post surgery.

Acknowledgements

This work was supported by the National Institutes of health grant R01-CA211084

Financial support for all authors: This work was supported by the National Institutes of health grant R01-CA211084

Footnotes

The authors have no potential conflicts of interest to disclose.

References

  • 1.Divgi CR, Welt S, Kris M, Real FX, Yeh SD, Gralla R, et al. Phase I and imaging trial of indium 111-labeled anti-epidermal growth factor receptor monoclonal antibody 225 in patients with squamous cell lung carcinoma. J Natl Cancer Inst. 1991;83(2):97–104. [DOI] [PubMed] [Google Scholar]
  • 2.Yarom N, Marginean C, Moyana T, Gorn-Hondermann I, Birnboim HC, Marginean H, et al. EGFR expression variance in paired colorectal cancer primary and metastatic tumors. Cancer Biol Ther. 2010;10(5):416–21. [DOI] [PubMed] [Google Scholar]
  • 3.Metro G, Finocchiaro G, Toschi L, Bartolini S, Magrini E, Cancellieri A, et al. Epidermal growth factor receptor (EGFR) targeted therapies in non-small cell lung cancer (NSCLC). Rev Recent Clin Trials. 2006;1(1):1–13. [DOI] [PubMed] [Google Scholar]
  • 4.Nakai K, Hung MC, Yamaguchi H. A perspective on anti-EGFR therapies targeting triple-negative breast cancer. Am J Cancer Res. 2016;6(8):1609–23. [PMC free article] [PubMed] [Google Scholar]
  • 5.Bossi P, Resteghini C, Paielli N, Licitra L, Pilotti S, Perrone F. Prognostic and predictive value of EGFR in head and neck squamous cell carcinoma. Oncotarget. 2016;7(45):74362–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Yang Z, Hackshaw A, Feng Q, Fu X, Zhang Y, Mao C, et al. Comparison of gefitinib, erlotinib and afatinib in non-small cell lung cancer: A meta-analysis. Int J Cancer. 2017;140(12):2805–19. [DOI] [PubMed] [Google Scholar]
  • 7.Lin YT, Chen JS, Liao WY, Ho CC, Hsu CL, Yang CY, et al. Clinical outcomes and secondary epidermal growth factor receptor (EGFR) T790M mutation among first-line gefitinib, erlotinib and afatinib-treated non-small cell lung cancer patients with activating EGFR mutations. Int J Cancer. 2019;144(11):2887–96. [DOI] [PubMed] [Google Scholar]
  • 8.Gao M, Su H, Lin G, Li S, Yu X, Qin A, et al. Targeted imaging of EGFR overexpressed cancer cells by brightly fluorescent nanoparticles conjugated with cetuximab. Nanoscale. 2016;8(32):15027–32. [DOI] [PubMed] [Google Scholar]
  • 9.Gao RW, Teraphongphom N, de Boer E, van den Berg NS, Divi V, Kaplan MJ, et al. Safety of panitumumab-IRDye800CW and cetuximab-IRDye800CW for fluorescence-guided surgical navigation in head and neck cancers. Theranostics. 2018;8(9):2488–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Yazdi MH, Faramarzi MA, Nikfar S, Abdollahi M. A Comprehensive Review of Clinical Trials on EGFR Inhibitors Such as Cetuximab and Panitumumab as Monotherapy and in Combination for Treatment of Metastatic Colorectal Cancer. Avicenna J Med Biotechnol. 2015;7(4):134–44. [PMC free article] [PubMed] [Google Scholar]
  • 11.Mangia A, Saponaro C, Malfettone A, Bisceglie D, Bellizzi A, Asselti M, et al. Involvement of nuclear NHERF1 in colorectal cancer progression. Oncol Rep. 2012;28(3):889–94. [DOI] [PubMed] [Google Scholar]
  • 12.Chen Y, Liu G, Guo L, Wang H, Fu Y, Luo Y. Enhancement of tumor uptake and therapeutic efficacy of EGFR-targeted antibody cetuximab and antibody-drug conjugates by cholesterol sequestration. Int J Cancer. 2015;136(1):182–94. [DOI] [PubMed] [Google Scholar]
  • 13.Gleysteen JP, Newman JR, Chhieng D, Frost A, Zinn KR, Rosenthal EL. Fluorescent labeled anti-EGFR antibody for identification of regional and distant metastasis in a preclinical xenograft model. Head Neck. 2008;30(6):782–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Gao RW, Teraphongphom NT, van den Berg NS, Martin BA, Oberhelman NJ, Divi V, et al. Determination of Tumor Margins with Surgical Specimen Mapping Using Near-Infrared Fluorescence. Cancer Res. 2018;78(17):5144–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.van Keulen S, van den Berg NS, Nishio N, Birkeland A, Zhou Q, Lu G, et al. Rapid, non-invasive fluorescence margin assessment: Optical specimen mapping in oral squamous cell carcinoma. Oral Oncol. 2019;88:58–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Day KE, Beck LN, Heath CH, Huang CC, Zinn KR, Rosenthal EL. Identification of the optimal therapeutic antibody for fluorescent imaging of cutaneous squamous cell carcinoma. Cancer Biol Ther. 2013;14(3):271–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Saccomano M, Dullin C, Alves F, Napp J. Preclinical evaluation of near-infrared (NIR) fluorescently labeled cetuximab as a potential tool for fluorescence-guided surgery. Int J Cancer. 2016;139(10):2277–89. [DOI] [PubMed] [Google Scholar]
  • 18.Teraphongphom N, Kong CS, Warram JM, Rosenthal EL. Specimen mapping in head and neck cancer using fluorescence imaging. Laryngoscope Investig Otolaryngol. 2017;2(6):447–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bernhard W, El-Sayed A, Barreto K, Gonzalez C, Hill W, Parada AC, et al. Near infrared fluorescence imaging of EGFR expression in vivo using IRDye800CW-nimotuzumab. Oncotarget. 2018;9(5):6213–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Mishani E, Abourbeh G, Eiblmaier M, Anderson CJ. Imaging of EGFR and EGFR tyrosine kinase overexpression in tumors by nuclear medicine modalities. Curr Pharm Des. 2008;14(28):2983–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.van Dijk LK, Hoeben BA, Kaanders JH, Franssen GM, Boerman OC, Bussink J. Imaging of epidermal growth factor receptor expression in head and neck cancer with SPECT/CT and 111In-labeled cetuximab-F(ab’)2. J Nucl Med. 2013;54(12):2118–24. [DOI] [PubMed] [Google Scholar]
  • 22.Sihver W, Pietzsch J, Krause M, Baumann M, Steinbach J, Pietzsch HJ. Radiolabeled Cetuximab Conjugates for EGFR Targeted Cancer Diagnostics and Therapy. Pharmaceuticals (Basel). 2014;7(3):311–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Day KE, Sweeny L, Kulbersh B, Zinn KR, Rosenthal EL. Preclinical comparison of near-infrared-labeled cetuximab and panitumumab for optical imaging of head and neck squamous cell carcinoma. Mol Imaging Biol. 2013;15(6):722–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Cohen R, Stammes MA, de Roos IH, Stigter-van Walsum M, Visser GW, van Dongen GA. Inert coupling of IRDye800CW to monoclonal antibodies for clinical optical imaging of tumor targets. EJNMMI Res. 2011;1(1):31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cohen R, Vugts DJ, Stigter-van Walsum M, Visser GW, van Dongen GA. Inert coupling of IRDye800CW and zirconium-89 to monoclonal antibodies for single- or dual-mode fluorescence and PET imaging. Nat Protoc. 2013;8(5):1010–8. [DOI] [PubMed] [Google Scholar]
  • 26.Chen W, Shen B, Sun X. Analysis of Progress and Challenges of EGFR-Targeted Molecular Imaging in Cancer With a Focus on Affibody Molecules. Mol Imaging. 2019;18:1536012118823473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Gong H, Kovar J, Little G, Chen H, Olive DM. In vivo imaging of xenograft tumors using an epidermal growth factor receptor-specific affibody molecule labeled with a near-infrared fluorophore. Neoplasia. 2010;12(2):139–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Keller DS, Ishizawa T, Cohen R, Chand M. Indocyanine green fluorescence imaging in colorectal surgery: overview, applications, and future directions. Lancet Gastroenterol Hepatol. 2017;2(10):757–66. [DOI] [PubMed] [Google Scholar]
  • 29.Chen LC, Lloyd WR 3rd, Chang CW, Sud D, Mycek MA. Fluorescence lifetime imaging microscopy for quantitative biological imaging. Methods Cell Biol. 2013;114:457–88. [DOI] [PubMed] [Google Scholar]
  • 30.Becker W Fluorescence lifetime imaging--techniques and applications. J Microsc. 2012;247(2):119–36. [DOI] [PubMed] [Google Scholar]
  • 31.Berezin MY, Achilefu S. Fluorescence lifetime measurements and biological imaging. Chem Rev. 2010;110(5):2641–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Kumar AT, Carp SA, Yang J, Ross A, Medarova Z, Ran C. Fluorescence lifetime-based contrast enhancement of indocyanine green-labeled tumors. J Biomed Opt. 2017;22(4):40501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Ishizawa T, Bandai Y, Ijichi M, Kaneko J, Hasegawa K, Kokudo N. Fluorescent cholangiography illuminating the biliary tree during laparoscopic cholecystectomy. Br J Surg. 2010;97(9):1369–77. [DOI] [PubMed] [Google Scholar]
  • 34.Yokoyama J, Fujimaki M, Ohba S, Anzai T, Yoshii R, Ito S, et al. A feasibility study of NIR fluorescent image-guided surgery in head and neck cancer based on the assessment of optimum surgical time as revealed through dynamic imaging. Onco Targets Ther. 2013;6:325–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Corlu A, Choe R, Durduran T, Rosen MA, Schweiger M, Arridge SR, et al. Three-dimensional in vivo fluorescence diffuse optical tomography of breast cancer in humans. Opt Express. 2007;15(11):6696–716. [DOI] [PubMed] [Google Scholar]
  • 36.Heath CH, Deep NL, Sweeny L, Zinn KR, Rosenthal EL. Use of panitumumab-IRDye800 to image microscopic head and neck cancer in an orthotopic surgical model. Ann Surg Oncol. 2012;19(12):3879–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rice WL, Shcherbakova DM, Verkhusha VV, Kumar AT. In vivo tomographic imaging of deep-seated cancer using fluorescence lifetime contrast. Cancer Res. 2015;75(7):1236–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Ceresa BP, Vanlandingham PA. Molecular Mechanisms that Regulate Epidermal Growth Factor Receptor Inactivation. Clin Med Oncol. 2008;2:47–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Nakabayashi T, Ohta N. Sensing of intracellular environments by fluorescence lifetime imaging of exogenous fluorophores. Anal Sci. 2015;31(4):275–85. [DOI] [PubMed] [Google Scholar]
  • 40.Lakowicz JR. Principles of fluorescence spectroscopy: Springer Science & Business Media; 2013. [Google Scholar]
  • 41.Sarder P, Maji D, Achilefu S. Molecular probes for fluorescence lifetime imaging. Bioconjug Chem. 2015;26(6):963–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Rosenthal EL, Warram JM, de Boer E, Basilion JP, Biel MA, Bogyo M, et al. Successful Translation of Fluorescence Navigation During Oncologic Surgery: A Consensus Report. J Nucl Med. 2016;57(1):144–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Rosenthal EL, Warram JM, de Boer E, Chung TK, Korb ML, Brandwein-Gensler M, et al. Safety and Tumor Specificity of Cetuximab-IRDye800 for Surgical Navigation in Head and Neck Cancer. Clin Cancer Res. 2015;21(16):3658–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Miller SE, Tummers WS, Teraphongphom N, van den Berg NS, Hasan A, Ertsey RD, et al. First-in-human intraoperative near-infrared fluorescence imaging of glioblastoma using cetuximab-IRDye800. J Neurooncol. 2018;139(1):135–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Abe K, Zhao L, Periasamy A, Intes X, Barroso M. Non-invasive in vivo imaging of near infrared-labeled transferrin in breast cancer cells and tumors using fluorescence lifetime FRET. PLoS One. 2013;8(11):e80269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Ardeshirpour Y, Chernomordik V, Hassan M, Zielinski R, Capala J, Gandjbakhche A. In vivo fluorescence lifetime imaging for monitoring the efficacy of the cancer treatment. Clin Cancer Res. 2014;20(13):3531–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Tummers WS, Warram JM, Tipirneni KE, Fengler J, Jacobs P, Shankar L, et al. Regulatory Aspects of Optical Methods and Exogenous Targets for Cancer Detection. Cancer Res. 2017;77(9):2197–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.AV DS, Lin H, Henderson ER, Samkoe KS, Pogue BW. Review of fluorescence guided surgery systems: identification of key performance capabilities beyond indocyanine green imaging. J Biomed Opt. 2016;21(8):80901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Fang Q, Selb J, Carp SA, Boverman G, Miller EL, Brooks DH, et al. Combined optical and X-ray tomosynthesis breast imaging. Radiology. 2011;258(1):89–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Raymond SB, Boas DA, Bacskai BJ, Kumar AT. Lifetime-based tomographic multiplexing. J Biomed Opt. 2010;15(4):046011. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES