Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Jun 1.
Published in final edited form as: J Am Coll Surg. 2012 Apr 27;214(6):997–1007.e2. doi: 10.1016/j.jamcollsurg.2012.02.009

An LED Light Source and Novel Fluorophore Combinations Improve Fluorescence Laparoscopic Detection of Metastatic Pancreatic Cancer in Orthotopic Mouse Models

Cristina A Metildi 1, Sharmeela Kaushal 1, Claudia Lee 2, Chanae R Hardamon 1, Cynthia S Snyder 1, George A Luiken 3, Mark A Talamini 1, Robert M Hoffman 1,4, Michael Bouvet 1
PMCID: PMC3360832  NIHMSID: NIHMS358033  PMID: 22542065

Abstract

Background

The aim of this study was to improve fluorescence laparoscopy of pancreatic cancer in an orthotopic mouse model with the use of an LED light source and an optimal fluorophore combination.

Study Design

Human pancreatic cancer models were established with fluorescent FG-RFP, MiaPaca2-GFP, BxPC-3-RFP, and BxPC-3 cancer cells implanted in 6-week-old female athymic mice. Two weeks post-implantation, diagnostic laparoscopy was performed with a Stryker L9000 LED light source or a Stryker X8000 xenon light source 24 hours after tail vein injection of CEA antibodies conjugated with Alexa 488 or Alexa 555. Cancer lesions were detected and localized under each light mode. Intravital images were obtained with the OV-100 Olympus Small Animal Imaging System and Maestro CRI Small Animal Imaging System, serving as a positive control. Tumors were collected for histologic review.

Results

Fluorescence laparoscopy with a 495-nm emission filter and an LED light source enabled real-time visualization of the fluorescence-labeled tumor deposits in the peritoneal cavity. The simultaneous use of different fluorophores (Alexa 488 and Alexa 555) conjugated to antibodies brightened the fluorescence signal, enhancing detection of sub-millimeter lesions without compromising background illumination. Adjustments to the LED light source permitted simultaneous detection of tumor lesions of different fluorescent colors and surrounding structures with minimal autofluorescence.

Conclusions

Using an LED light source with adjustments to the red, blue and green wavelengths, we can simultaneously identify tumor metastases expressing fluorescent proteins of different wavelengths, which greatly enhanced the signal without compromising background illumination. Development of this technology for clinical use can improve staging and treatment of pancreatic cancer.

Keywords: pancreatic cancer, orthotopic mouse models, fluorescence, laparoscopy, LED light source

Introduction

Despite medical and surgical advances over the past decade, pancreatic cancer remains a lethal disease with a 5-year survival rate of 6%.1 The majority of patients are reported, by most series, to have inoperable, advanced disease at the time of presentation. While 40–50% are diagnosed with metastatic disease, an additional 40% are found to have locally advanced disease leaving only 10–20% of patients eligible for curative surgery.2 The advanced stage at presentation combined with the aggressiveness of this disease speaks to the importance of accurately staging patients with pancreatic cancer and therefore properly identifying patients with resectable disease while sparing those with incurable disease an unnecessary operation.3

To date, a high quality pancreatic computed tomography (CT) scan can be predictive at identifying unresectable disease in patients with pancreatic cancer. However, more recently published reviews report a 20–48% unresectability rate in pancreatic cancer patients found to be resectable by preoperative CT.4–10 As a result, most centers have adopted additional imaging techniques, such as magnetic resonance imaging, endoscopic ultrasound, or positron emission tomography scan, to improve the prediction of resectability. Although the diagnostic accuracy of preoperative staging is improved with combinations of these imaging modalities, a significant portion of patients are still found at exploration to have small size metastatic disease (<1cm) or vascular invasion, precluding them from curative resection.6, 7, 10–14

Although laparoscopy was incorporated early in the staging of pancreatic cancer patients15, its current role remains controversial. Studies have repeatedly demonstrated the ability of laparoscopy to identify small peritoneal or liver implants not seen on preoperative staging, documenting a detection rate just under 40%.4–8, 10, 12, 14, 16, 17 However, these same studies discuss the limitations of this technique in the detection of vascular invasion and deep hepatic metastases, resulting in false negative rates as high as 38%.

We have previously demonstrated that fluorescence laparoscopy can improve the ability to detect genetically-engineered green fluorescent tumors.18, 19 Our laboratory further enhanced the utility of fluorescence laparoscopy with the addition of fluorophore-conjugated anti-CEA antibodies for the rapid and accurate detection of primary and metastatic lesions of CEA-expressing pancreatic cancer in an orthotopic mouse model.20 Fluorescence laparoscopy was shown to be a significant improvement over standard laparoscopy for detection of small metastases.20 The aim of the present study was to develop a more translational method of fluorescence laparoscopy by replacing our original light source (a 300-W Xenon lamp) with an LED light source with adjustable blue, green and red LED lights, thereby allowing visualization of tumors of different fluorescent wavelengths, a benefit if more than one fluorophore is used for optimal tumor detection. The CEA antibody was used to make tumors glow as CEA has a long track record for tumor specificity in pancreatic cancer.21 In addition, we set out to identify the optimal fluorophore that would improve visualization of tumors without compromising background illumination.

Methods

Cell culture

Human BxPC-3 and BxPC-3-Red Fluorescent Protein (RFP) pancreatic cancer cells were maintained as previously described.22 Human FG-RFP and MiaPaca2-Green Fluorescent Protein (GFP) pancreatic cancer cells were maintained as previously described.23

Antibody conjugation

Monoclonal antibody specific for carcinoembryonic antigen (CEA) was purchased from RayBiotech, Inc.(Norcross, GA). The antibody was labeled with the AlexaFluor 488 or 555 Protein Labeling Kit (Molecular Probes Inc., Eugene, OR) according to the manufacturer’s instructions and as previously described.21

Animal care

Female athymic nu/nu nude mice were maintained in a barrier facility on high-efficiency particulate air filtered racks and fed with autoclaved laboratory rodent diet (Teckland LM-485; Western ResearchProducts, Orange, CA). All surgical procedures were performed under anesthesia with an intramuscular injection of 100 μL of a mixture of 100 mg/kg ketamine and 10 mg/kg xylazine. All animal studies were approved by the UCSD Institutional Animal Care and Use Committee (IACUC) and conducted in accordance with the principles and procedures outlined in the National Institutes of Health (NIH) Guide for the Care and Use of Animals.

Orthotopic model

Human pancreatic cancer cells (106) were implanted orthotopically in the tail of the pancreas as previously described.18 The following orthotopic models were established: BxPC-3 labeled with Alexa 488-conjugated anti-CEA antibody or labeled with Alexa 555-conjugated anti-CEA antibody; FG-RFP; FG-RFP co-implanted with BxPC-3 labeled with Alexa 488-conjugated anti-CEA antibody, BxPC-3-RFP labeled with Alexa 488-conjugated anti-CEA antibody, MiaPaca2-GFP co-implanted with BxPC-3 labeled with Alexa 555-conjugated anti-CEA antibody. The orthotopic models used in the present study involved injection of pancreatic cancer cells into the tail of the pancreas is an exact orthotopic model allowing spontaneous metastases. A splenic injection would allow metastasis to the liver but is not exactly orthotopic and allows for what are termed “experimental metastasis” to the liver.24

Carcinomatosis model

Human pancreatic cancer cells were resuspended at a concentration of 1×106 cells per 100 μL of serum- free medium and placed on ice before intra-peritoneal injections directly into the peritoneal cavity of 6-week-old female nude mice using a 27-gauge needle as previously described.19

Fluorescence Laparoscopy

A standard laparoscopic tower provided by Stryker (Stryker, San Jose, California) was slightly modified in the following manner to achieve fluorescence laparoscopy (FL): the excitation light source, a Stryker L9000 LED lamp, was filtered through a glass emission filter (Schott GG495) placed between the laparoscope and the Stryker 1288 HD camera. Using the computer software system provided by Stryker (L9Calibration0.03DOT3), adjustments to the red, blue and green components of the Stryker L9000 LED light source were made to allow visualization of the fluorescent tumors. A Stryker X8000 Xenon light source was used for bright field laparoscopy (BL) (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Stryker laparoscopic setup. A standard laparoscopic tower was modified to achieve a fluorescence light mode that would permit detection of fluorescence signals while still allowing visualization of the background. The LED light source (Stryker L9000 LED lamp) was filtered through a glass emission filter (Schott GG495) that was placed between the laparoscope and the 1288 HD camera. With alterations to red, blue and green components of the LED light source, tumors of different fluorescent wavelengths were simultaneously visualized. A Stryker X8000 Xenon light source was used for bright field laparoscopy.

Mouse laparoscopy

Laparoscopy on mice was performed as previously described.18, 19 A standard staging laparoscopy procedure was performed in which all four quadrants of the peritoneal cavity were examined in a systematic way. Each mouse was examined under both light modes, fluorescence laparoscopy (FL) using the Stryker L9000 LED light source and bright light (BL) using the Stryker X8000 Xenon light source. At termination of laparoscopy, the mice were sacrificed and their abdominal cavities exposed for OV-100 (Olympus, Tokyo, Japan) and Maestro (CRI, Hopkinton, MA) imaging. Tumors were then collected for histological evaluation when possible.

Tumor labeling

Four weeks following orthotopic implantation of human pancreatic cancer cells, selected mice harboring BxPC-3 cells received an injection of 75 μg of anti-CEA-Alexa 488 or 555 conjugates via tail vein 24 hours prior to laparoscopy. The dose and timing of the injection were derived from our previous in vivo study of mouse imaging.25

Tissue histology

At necropsy, all identified lesions were collected when possible for histological preparation with hematoxylin and eosin (H&E) staining. Small tumor foci observed during in vivo fluorescence laparoscopic imaging were localized post mortem in fresh ex vivo organ blocks using both brightfield and fluorescence imaging.

Data processing

Images obtained during laparoscopy were not processed in any way. Representative frames are presented. Intravital images were obtained with the Maestro and OV-100 imaging systems, serving as the positive control.

Results

Fluorescence Laparoscopy

Two to four weeks after orthotopic implantation of human pancreatic cancer cells, female athymic mice bearing tumor were anesthetized before undergoing a diagnostic laparoscopy under both bright light and fluorescence modes. Mice with non-fluorescent tumor received, via tail vein, an injection of 75 μg of anti-CEA-Alexa 488 or 555 conjugates 24 hours prior to laparoscopy. Pneumoperitoneum was established to a pressure of 2 mm Hg, which allowed ample distension of the abdominal wall to enable adequate visualization and navigation within the abdomen of the mouse. All quadrants were evaluated in a systematic manner to identify primary tumor and evaluate for metastasis. The L9000 LED lamp provided by Stryker served as the excitation light source. With alterations to the red, blue and green components of the LED light through a computer software system (L9Calibration0.03DOT3), we were able to simultaneously visualize fluorescence-labeled tumor lesions of different wavelengths in the peritoneal cavity without significantly compromising background illumination.

In previous publications, we showed that one key element in fluorescence-guided surgery is to provide adequate background for navigation. Preferably the observed spectra of the background can mimic what is observed with white lighting. When the signal is observed with typical RGB CCD camera, the blue-skewed illumination for exciting fluorescence is mostly corrected through filtering with a 495-nm glass filter (Schott GG495). Natural looking background and enhanced fluorescence can be achieved by carefully balancing the amount of detection in red, green and blue channels. As shown in Figure 2a, white light filtered with a 480 nm short pass filter to enhance the relative intensity in blue color and blue LED can both be perceived as having the same amount of leakage in the green despite not having identical spectra. The major difference is in the red channel where filtered white light has more relative intensity. While the blue LED is capable of exciting green fluorescence and providing background in channels other than red, additional red-skewed illumination, such as from a red LED, is important for providing a color-balanced background.

Figure 2.

Figure 2

Open in a new tab

(A) Spectrum of two possible light sources for fluorescence guided surgery. The solid line illustrates the spectrum of a filtered Xenon lamp previously described by our lab [21]. The dash line illustrates a typical blue LED spectrum. The color blocks mark the spectral range of red, green and blue channels on common RGB CCDs. (B) Overlapping emission and excitation spectra of GFP/Alexa- 488 and RFP/ Alexa -555 fluorocent proteins and fluorophores. Blue and salmon color peaks represent the excitation and emission spectra of GFP and Alexa 488, respectively. Pink and red peaks represent excitation and emission spectra of RFP and Alexa 555, respectively. This graphic demonstrates the utility of the overlapping spectra of these fluorophores in the spectral range of GFP for detection of tumor and metastases, while maintaining adequate visualization of surrounding structures for spatial orientation and surgical navigation. Filtering an LED light source through a 495 glass filter creates a GFP bandwidth through which tumors labeled with spectrally distinct fluorophores are visualized simultaneously.

As seen in Figure 2b, the excitation spectra of GFP, 488, RFP, and 555 fluorescence overlap below 495 nm. With maximal intensity of blue light from the LED light source, which generally distributes around 475 nm with a 40 nm bandwidth, the 495 nm long pass filter allowed sufficient contrast between the excitation and the two emission wavelengths. Peak emission wavelengths for GFP and RFP are 508 nm and 583 nm, respectively. The large separation allows the two fluorescent signals to be distinguished while observed spontaneously through typical RGB cameras. Adjustments to red and green components of the LED light source permitted adequate “white” light to leak through to maintain background illumination for spatial orientation and navigation. Overall, the LED light source, with maximal blue light and adjustments to red and green light, produced a spectrum of light transmission that resulted in proper color balance and adequate background illumination while enhancing the fluorescence signal to background ratio enabling real-time visualization of fluorescence-labeled tumor and simultaneous detection of differently fluorescent tumors.

To achieve the diagnostic purpose of fluorescence laparoscopy in previous experiments, adjustments to exposure time and gain had to be optimized to compensate for the lack of intensity from the Xenon lamp previously used. Increasing the gain and exposure time, however, resulted in significant dynamic delay impairing navigation. Replacing our original light source with the LED lamp virtually eliminated the need for adjustments to exposure time or gain, overall improving rapid detection of fluorescent tumor while also improving visualization of surrounding tissue for surgical navigation.

Upon termination of laparoscopy, the mice were sacrificed and their abdominal cavities were exposed. Intravital images of all mice were obtained with the Maestro and OV-100 imaging systems using GFP and RFP filters. These images served as a control against which to compare the laparoscopic findings. Figure 3a presents intravital images representative of the mouse models harboring different fluorescent tumors used in this experiment. Spectral analysis was performed using Maestro images generating the illustration seen in Figure 3b. This graph illustrates the emission spectra of GFP-expressing tumor (green lines), RFP-expressing tumor (red lines) as well as the autofluorescence of the mouse (grey lines) through 490 nm and 560 nm emission filters. The lower emission filter allows for distinction between the fluorescent signals, permitting simultaneously visualization of differently fluorescent tumors. This confirms our ability to simultaneously view tumors of different fluorescent wavelengths in the abdomen of a mouse with fluorescence laparoscopy using a 495 nm glass filter without compromising background illumination for surgical navigation.

Figure 3.

Figure 3

Open in a new tab

Maestro intravital images of tumor-bearing mice post laparoscopy. (A) Intravital images of representative tumor-bearing mouse models illustrating fluorescent tumors. (B) Spectral unmixing of intravital images obtained by the Maestro. Gray curves represent the mouse background emission or autofluorescence. Green curves represent Alexa 488/GFP emission spectra while the red curves represent Alexa 555/RFP emission through blue 490 nm and 560 nm emission filters. This graphic illustrates the overlap of background emission spectrum with Alexa 488/GFP and Alexa 555/RFP emission spectra in the GFP bandwidth established through a 495 nm emission filter. This demonstrates the ability to visualize red and green fluorescent tumors simultaneously while maintaining adequate background illumination.

Tumor Detection & Visualization of Metastases

An orthotopic model was established by implanting non-fluorescent BxPC-3 human pancreatic cancer cells into female, athymic mice. Twenty-four hours prior to laparoscopy, tumor-bearing mice were injected via tail vein with 75 μg of anti-CEA-Alexa 488. As demonstrated in Figure 4, fluorescence laparoscopy (FL) permitted improved localization of the human pancreatic tumor labeled with a fluorophore-conjugated antibody directed against CEA compared to bright light laparoscopy (BL). In addition, detection of metastatic disease was greatly enhanced by FL, which revealed small metastatic lesions hidden in the mesentery and fat pad of the mouse that went undetected with BL. The use of an LED light source further enhanced the applicability of FL, by not only improving visualization of fluorescent tumors, but also maintaining adequate background illumination for proper spatial orientation and navigation.

Figure 4.

Figure 4

Open in a new tab

Comparative identification of tumor foci under bright field and fluorescence laparoscopy. (a,b) OV-100 bright field and GFP intravital images from a representative mouse with non-fluorescent BxPC-3 labeled with anti-CEA Alexa-488. Corresponding laparoscopic images of a mouse under BL (c–e) and FL (f–h). Images c and f are corresponding laparoscopic images of the left upper quadrant under BL and FL, respectively. The green fluorescence of the primary lesion is distinct under FL, whereas under BL the tumor resembles normal pancreatic tissue and is more difficult to identify. Images d and g are of the right upper quadrant demonstrating the improved detection of small metastatic tumor foci under FL compared to BL. Images e and h are corresponding images of the lower abdomen of a mouse under BL and FL, respectively. Under BL, the tumor foci are indistinguishable from normal tissue in the fat pad of the mouse and thus not identifiable. However, the green fluorescence is easily detected and identified. BL, bright light laparoscopy; C, cecum; FL, fluorescence laparoscopy; L, liver; P, pancreas; S, spleen.

In a second orthotopic model, non-fluorescent BxPC-3 and MiaPaca-2 GFP were co-implanted into athymic mice. Twenty-four hours prior to laparoscopy, tumor-bearing mice were injected via tail vein with 75 μg of anti-CEA-Alexa 555. MiaPaca-2 GFP-expressing tumor yielded a brighter fluorescence, resulting in better detection of tumor microfoci compared to non-fluorescent BxPC-3 labeled with anti-CEA-Alexa 555. The fluorescence signal of anti-CEA-Alexa 488, however, was comparable to the green fluorescence signal of the MiaPaca-2 GFP-expressing tumor. Fluorescence laparoscopy again afforded overall improved identification and localization of submillimeter tumor deposits that went undetected under bright light laparoscopy. Although not as bright, anti-CEA-Alexa 555 still provided an adequate fluorescence signal for clear distinction of miniscule tumor deposits from surrounding normal tissue (Figure 5). The brightness of fluorescence signal was comparative to genetically-engineered FG Ds RFP-expressing tumor mouse model as illustrated in Figure 5. However, to optimize simultaneous visualization of the red fluorescent tumor and Alexa 555 fluorophore conjugate, background illumination had to be slightly compromised, making spatial orientation less than optimal.

Figure 5.

Figure 5

Figure 5

Open in a new tab

Orthotopic mouse models harboring MiaPaca-2-GFP (i, ii), FG-RFP (iii, iv), and non-fluorescent BxPC-3 anti-CEA Alexa 488 (v, vi) and Alexa 555 (vii, viii) human pancreatic cancers. (A) The green and red fluorescent tumor microfoci are readily detectable under fluorescence laparoscopy (images ii and vi, iv and viii, respectively). These same tumor microfoci (arrows) are not readily detectable in the corresponding bright field laparoscopic images (i and v, iii and vii). The red fluorescent tumor provides improved detection of micrometastases under FL compared to BL, however, at the expense of background illumination. Spatial orientation is better observed with the green fluorescent tumor. The fluorescence signal obtained by labeling nonfluorescent tumors with anti-CEA Alexa 488 (vi) and 555 (viii) is comparable to that of GFP-expressing (ii) and RFP-expressing tumors (iv), respectively. L, liver; S, stomach; C, cecum. (B) Laparoscopic images of the left upper quadrant in representative mouse models of human pancreatic cancer of different fluorescent wavelengths. Fluorescence laparoscopy with the LED light source allows identification and localization of human pancreatic tumors of different fluorescent wavelengths simultaneously with improved accuracy. The combination of RFP-expressing tumor labeled with anti-CEA Alexa 488 afforded the brightest signal of the mouse models.

Optimal Fluorophore

To improve on the tumor-to-background fluorescence ratio, we implanted combinations of differently fluorescent tumors in athymic mice to establish the optimal fluorophore that would increase the fluorescence signal without compromising background illumination. The following combinations of human pancreatic cancer were established: FG-RFP-expressing tumor co-implanted with non-fluorescent BxPC-3 labeled with anti-CEA-Alexa 488; BxPC-3-RFP-expressing tumor additionally labeled with anti-CEA-Alexa 488; and Mia-Paca2 GFP-expressing tumor co-implanted with non-fluorescent BxPC-3 labeled with anti-CEA-Alexa 555. The fluorophore-conjugated antibodies were injected via tail vein 24 hours prior to laparoscopy. Figure 5B shows laparoscopic images of the right upper quadrant in representative mouse models of human pancreatic cancer used in this experiment further illustrating the capabilities of FL with an LED light source to detect tumors of different fluorescent wavelengths simultaneously.

By allowing simultaneous visualization of tumors with different fluorescent wavelengths, this novel adjustable LED light source enabled us to establish the optimal fluorophore that would permit accurate identification and localization of tumor deposits without losing background illumination (see Supplemental video online). The fluorescence tumor signal-to-background ratio was determined in genetically engineered human pancreatic cancer cells that expressed GFP and RFP, as well as non-fluorescent tumors labeled with anti-CEA-Alexa 488 or -555 conjugates.

The combination of red and green fluorescence permitted more efficient identification of metastasis, with a greatly enhanced fluorescent signal. Laparoscopic images obtained from the mouse model of BxPC-3-RFP-expressing tumor labeled with anti-CEA-Alexa 488 (see Figure 5B) demonstrated the increased brightness of the combined fluorescence signal compared to the other mouse models. The mean signal intensity of the BxPC-3-RFP-expressing tumor is 165 while that of non-fluorescent BxPC-3 labeled with anti-CEA-Alexa 488 is 97. However, the combination of RFP-expressing tumor labeled with anti-CEA-Alexa 488 increased the mean signal intensity to 196 (Supplemental Figure 1, online only). This increased intensity of the combined fluorescence signal increased the effectiveness of fluorescence laparoscopy.

An orthotopic model of non-fluorescent BxPC-3 doubly labeled with anti-CEA-Alexa 488 and anti-CEA-Alexa 555 was made (Figure 6). Along with the improved brightness of the combined fluorescence signal, background illumination remained optimal and thus tumor localization was overall improved. The in vitro mean signal intensity of Alexa 555 and Alexa 488 are 85 and 90, respectively. The mean signal intensity is increased when the fluorophore conjugates are combined to 152 (Supplemental Figure 2, online only). Because of overlap of excitation spectra of red and green fluorescence through a 495 nm glass filter, the combination increases the bandwidth of fluorescence light transmission and thus intensifies the signal. In vivo, the mean signal intensities for Alexa 555 and Alexa 488 were 80 and 114, respectively. The dual-fluorophore conjugate mouse model yielded a combined mean signal intensity of 148. The optimal fluorophore was determined to be a combination of green (488) and red (555) conjugates. Overall, the combination brightens the fluorescence signal, enhancing tumor detection, while permitting adequate visualization of surrounding structures for surgical navigation and tumor localization.

Figure 6.

Figure 6

Open in a new tab

Othotopic mouse model harboring non-fluorescent BxPC-3 tumor doubly labeled with anti-CEA Alexa 488 and 555. (A) Laparoscopic images of representative mouse specimen with non-fluorescent BxPC-3 dually labeled with anti-CEA Alexa 488 and 555. The combination of red and green fluorophores creates a significantly brighter fluorescence signal without compromising background illumination. i–iii are laparoscopic images of the left upper quadrant. iv and v are laparoscopic images of metastatic tumor deposits hidden within the mesentery of the mouse. These deposits were virtually undetectable under BL. (B) i–iii are intravital OV-100 images of the same mouse specimen under (i) GFPa (excitation BP 460–490; emission BA 510–550), (ii) RFP (excitation BP 535–555; emission BA 570–623) and (iii) GFP (excitation 460–490; emission BA 510F) filters. The bottom image (iii) resembles the GFP bandwidth through which fluorescence laparoscopy is viewed. iv–vi are the corresponding intravital Maestro images. iv and v are spectral unmixing images of the (vi) compositive image obtained through a (iv) GFP and (v) RFP filter sets, respectively. These images confirm the double labeling of non-fluorescence BxPC-3 tumor with Alexa 488 and 555.

Post-Laparoscopy Analysis

After imaging, representative lesions were collected for histological correlation. Figure 7A demonstrates a superficial fluorescent lesion suspicious for metastatic tumor. Histological review of this focus (Figure 7B) confirms the presence of a thin layer of atypical cells adherent to the liver capsule. The histology correlates well with the fluorescence image; the finding is consistent with early seeding of fluorescent tumor cells to the liver.

Figure 7.

Figure 7

Open in a new tab

(a) Arrow indicates a metastatic deposit in the liver of the FG RFP tumor and the BxPC-3 Alexa 488 tumor superimposed. The red fluorescent tumor, with superimposed green fluorescence, allows for improved detection of microfoci deposits with improved background illumination. (b) Histologic confirmation of tumor lesion on liver surface. Histology performed on the lesion confirms the presence of a thin deposit of atypical cells consistent with tumor seeding to the liver capsule. L, liver; S, stomach, T, tumor

Discussion

Our laboratory has previously demonstrated the ability to illuminate tumors and metastases by the delivery of fluorophore-conjugated antibodies directed against surface antigens unique to pancreatic cancer cells, such as CEA and CA 19-9 21, 25 or by labeling tumors with GFP using a tumor-specific adenovirus.26–28 We further demonstrated the utility of in situ tumor labeling by developing a fluorescence laparoscope, with appropriate excitation and emission filters, to identify and localize fluorescently-labeled primary and metastatic tumors while maintaining adequate visualization of the surrounding tissues in orthotopic mouse models.18–20

Previously, several groups have investigated the utility of ALA-induced PPIX to highlight potentially malignant lesions.29–32 However, ALA is not cancer-specific and may erroneously label benign lesions. Additionally, Winer, et al. investigated the near-infrared (NIR) fluorescence properties of methylene blue to create a tumor contrast in order to improve localization and subsequent resection of small tumors and occult metastases of insulinomas.33 Although the dosing of methylene blue used in this study was clinically translatable, the limited optimal viewing time of one hour for adequate signal-to-background ratio limits this method for current use. Other investigators have experienced the similar limitation of poor specificity of the targeting ligands.34

The use of FL with fluorophore-conjugated antibodies significantly reduced the time to identify primary pancreatic tumor when compared to BL in an orthotopic model of pancreatic cancer.20 Furthermore, FL significantly increased the sensitivity and accuracy of staging laparoscopy to a cumulative rate of 96% from a rate of 40% obtained for BL. These results demonstrated the potential clinical importance of fluorophore-conjugated antibodies in staging of pancreatic adenocarcinoma.

Despite these improvements, fluorescence laparoscopy used in our earlier studies had some limitations. To obtain adequate visualization of the background necessary for spatial orientation and navigation of the laparoscope within the mouse abdomen, exposure time and gain had to be increased to compensate for the lack of light intensity from the Xenon lamp used in prior studies. This resulted in significant dynamic delay and impaired surgical navigation. By replacing the Xenon lamp with an LED light source, this limitation was virtually eliminated. Furthermore, with adjustments to the blue, red and green components of the LED light, real-time visualization of differently fluorescent tumors was enhanced along with background illumination. The ability to improve visualization of spectrally distinct fluorescent tumors enabled us to establish the optimal fluorophore combination to brighten the fluorescence signal, enhance the fluorescence signal-to-background ratio and still maintain optimal illumination of surrounding structures for spatial orientation and surgical navigation.

The LED light source also added clinical translatability to our method of fluorescence laparoscopy. Minimal adjustments were made to the Stryker laparoscopic tower that is commonly used in standard operating rooms today. The excitation light source, the Stryker L9000 LED lamp is typically found in standard laparoscopic towers. Through a 495 nm glass emission filter (Schott GG495) and with alterations to the red, blue and green components of the LED light using a computer software system provided by Stryker (L9Calibration0.03DOT3), fluorescent tumors were visualized. These same adjustments would require minimal effort in current operating rooms.

Making tumors glow offers very important potential advantages for tumor detection such as by laparoscopy and for fluoresence-guided surgery.35 The significant clinical potential of the developments described in the present report, whereby fluorescence laparoscopy is used with multiple light sources, and multiple fluorophores illuminating tumors, allows for ultra-high tumor detection for the goal of making patients NED. The present study used both spectrally distinct multiple light sources and tumors labeled with spectrally-distinct fluorescent probes. This allowed ultra-high resolution tumor detection as well as background normal-tissue detection. This novel approach of multi-spectral laparoscopy is a major step toward translation to the clinic.

Conclusions

We have demonstrated the improved utility of fluorescence laparoscopy with an LED light source to simultaneously identify tumor metastases expressing fluorescent proteins of different wavelengths without compromising background illumination. The new fluorescence laparoscopy method described here is a significant improvement over standard laparoscopy and has also improved our previously-described method of fluorescence laparoscopy. The simultaneous use of different fluorophores (Alexa 488 and Alexa 555) conjugated to antibodies further brightens the fluorescent signal and thus enhances detection without compromising background illumination. More importantly, this novel method is clinically translatable and thus has potential to solidify the role diagnostic laparoscopy in the staging algorithm of patients with pancreatic cancer.

Supplementary Material

01
02

Acknowledgments

This work was supported in part by grants from the National Cancer Institute CA142669 and CA132971 (to Dr Bouvet and AntiCancer, Inc) and T32 training grant CA121938-5 (to Dr Metildi).

Footnotes

Disclosure Information: Dr Lee receives a salary from UVP (Ultra Violet Products), LLC; Dr Luiken is the Chief Scientific Officer and stockholder in OncoFluor, Inc; and Dr Hoffman is the President and stockholder in AntiCancer, Inc. All other authors have nothing to declare.

Abstract presented at the American College of Surgeons 97th Annual Clinical Congress, Surgical Forum, San Francisco, CA, October 2011.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Siegel R, Ward E, Brawley O, Jemal A. Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin. 2011;61:212–236. doi: 10.3322/caac.20121. [DOI] [PubMed] [Google Scholar]
  • 2.Stefanidis D, Grove KD, Schwesinger WH, Thomas CR., Jr The current role of staging laparoscopy for adenocarcinoma of the pancreas: a review. Ann Oncol. 2006;17:189–199. doi: 10.1093/annonc/mdj013. [DOI] [PubMed] [Google Scholar]
  • 3.Katz MH, Wang H, Fleming JB, et al. Long-term survival after multidisciplinary management of resected pancreatic adenocarcinoma. Ann Surg Oncol. 2009;16:836–847. doi: 10.1245/s10434-008-0295-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Doran HE, Bosonnet L, Connor S, et al. Laparoscopy and laparoscopic ultrasound in the evaluation of pancreatic and periampullary tumours. Dig Surg. 2004;21:305–313. doi: 10.1159/000080885. [DOI] [PubMed] [Google Scholar]
  • 5.Ellsmere J, Mortele K, Sahani D, et al. Does multidetector-row CT eliminate the role of diagnostic laparoscopy in assessing the resectability of pancreatic head adenocarcinoma? Surg Endosc. 2005;19:369–373. doi: 10.1007/s00464-004-8712-5. [DOI] [PubMed] [Google Scholar]
  • 6.Maire F, Sauvanet A, Trivin F, et al. Staging of pancreatic head adenocarcinoma with spiral CT and endoscopic ultrasonography: an indirect evaluation of the usefulness of laparoscopy. Pancreatology. 2004;4:436–440. doi: 10.1159/000079617. [DOI] [PubMed] [Google Scholar]
  • 7.Morganti AG, Brizi MG, Macchia G, et al. The prognostic effect of clinical staging in pancreatic adenocarcinoma. Ann Surg Oncol. 2005;12:145–151. doi: 10.1245/ASO.2005.02.021. [DOI] [PubMed] [Google Scholar]
  • 8.Nieveen van Dijkum EJ, Romijn MG, Terwee CB, et al. Laparoscopic staging and subsequent palliation in patients with peripancreatic carcinoma. Ann Surg. 2003;237:66–73. doi: 10.1097/00000658-200301000-00010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Pisters PW, Lee JE, Vauthey JN, et al. Laparoscopy in the staging of pancreatic cancer. Br J Surg. 2001;88:325–337. doi: 10.1046/j.1365-2168.2001.01695.x. [DOI] [PubMed] [Google Scholar]
  • 10.Yoshida T, Matsumoto T, Morii Y, et al. Staging with helical computed tomography and laparoscopy in pancreatic head cancer. Hepatogastroenterology. 2002;49:1428–1431. [PubMed] [Google Scholar]
  • 11.Clarke DL, Thomson SR, Madiba TE, Sanyika C. Preoperative imaging of pancreatic cancer: a management-oriented approach. J Am Coll Surg. 2003;196:119–129. doi: 10.1016/s1072-7515(02)01609-5. [DOI] [PubMed] [Google Scholar]
  • 12.Conlon KC, Dougherty E, Klimstra DS, et al. The value of minimal access surgery in the staging of patients with potentially resectable peripancreatic malignancy. Ann Surg. 1996;223:134–140. doi: 10.1097/00000658-199602000-00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kwon AH, Inui H, Kamiyama Y. Preoperative laparoscopic examination using surgical manipulation and ultrasonography for pancreatic lesions. Endoscopy. 2002;34:464–468. doi: 10.1055/s-2002-32002. [DOI] [PubMed] [Google Scholar]
  • 14.Schachter PP, Avni Y, Shimonov M, et al. The impact of laparoscopy and laparoscopic ultrasonography on the management of pancreatic cancer. Arch Surg. 2000;135:1303–1307. doi: 10.1001/archsurg.135.11.1303. [DOI] [PubMed] [Google Scholar]
  • 15.Cuschieri A, Hall AW, Clark J. Value of laparoscopy in the diagnosis and management of pancreatic carcinoma. Gut. 1978;19:672–677. doi: 10.1136/gut.19.7.672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Jimenez RE, Warshaw AL, Rattner DW, et al. Impact of laparoscopic staging in the treatment of pancreatic cancer. Arch Surg. 2000;135:409–414. doi: 10.1001/archsurg.135.4.409. discussion 414–415. [DOI] [PubMed] [Google Scholar]
  • 17.Liu RC, Traverso LW. Diagnostic laparoscopy improves staging of pancreatic cancer deemed locally unresectable by computed tomography. Surg Endosc. 2005;19:638–642. doi: 10.1007/s00464-004-8165-x. [DOI] [PubMed] [Google Scholar]
  • 18.Tran Cao HS, Kaushal S, Lee C, et al. Fluorescence laparoscopy imaging of pancreatic tumor progression in an orthotopic mouse model. Surg Endosc. 2011;25:48–54. doi: 10.1007/s00464-010-1127-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Tran Cao HS, Kaushal S, Menen RS, et al. Submillimeter-resolution fluorescence laparoscopy of pancreatic cancer in a carcinomatosis mouse model visualizes metastases not seen with standard laparoscopy. J Laparoendosc Adv Surg Tech A. 2011;21:485–489. doi: 10.1089/lap.2011.0181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Tran Cao HS, Kaushal S, Metildi CA, et al. Fluorophore-conjugated anti-CEA antibody improves accuracy of staging laparoscopy in a mouse model of pancreatic cancer. Ann Surg Oncol. 2011;18:S40. [Google Scholar]
  • 21.Kaushal S, McElroy MK, Luiken GA, et al. Fluorophore-conjugated anti-CEA antibody for the intraoperative imaging of pancreatic and colorectal cancer. J Gastrointest Surg. 2008;12:1938–1950. doi: 10.1007/s11605-008-0581-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Bouvet M, Wang J, Nardin SR, et al. Real-time optical imaging of primary tumor growth and multiple metastatic events in a pancreatic cancer orthotopic model. Cancer Res. 2002;62:1534–1540. [PubMed] [Google Scholar]
  • 23.Katz MH, Takimoto S, Spivack D, et al. A novel red fluorescent protein orthotopic pancreatic cancer model for the preclinical evaluation of chemotherapeutics. J Surg Res. 2003;113:151–160. doi: 10.1016/s0022-4804(03)00234-8. [DOI] [PubMed] [Google Scholar]
  • 24.Bouvet M, Tsuji K, Yang M, et al. In vivo color-coded imaging of the interaction of colon cancer cells and splenocytes in the formation of liver metastases. Cancer Res. 2006;66:11293–11297. doi: 10.1158/0008-5472.CAN-06-2662. [DOI] [PubMed] [Google Scholar]
  • 25.McElroy M, Kaushal S, Luiken GA, et al. Imaging of primary and metastatic pancreatic cancer using a fluorophore-conjugated anti-CA19-9 antibody for surgical navigation. World J Surg. 2008;32:1057–1066. doi: 10.1007/s00268-007-9452-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kishimoto H, Aki R, Urata Y, et al. Tumor-selective, adenoviral-mediated GFP genetic labeling of human cancer in the live mouse reports future recurrence after resection. Cell Cycle. 2010;10:2737–2741. doi: 10.4161/cc.10.16.16756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kishimoto H, Urata Y, Tanaka N, et al. Selective metastatic tumor labeling with green fluorescent protein and killing by systemic administration of telomerase-dependent adenoviruses. Mol Cancer Ther. 2009;8:3001–3008. doi: 10.1158/1535-7163.MCT-09-0556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Kishimoto H, Zhao M, Hayashi K, et al. In vivo internal tumor illumination by telomerase-dependent adenoviral GFP for precise surgical navigation. Proc Natl Acad Sci U S A. 2009;106:14514–14517. doi: 10.1073/pnas.0906388106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Gahlen J, Prosst RL, Pietschmann M, et al. Laparoscopic fluorescence diagnosis for intraabdominal fluorescence targeting of peritoneal carcinosis experimental studies. Ann Surg. 2002;235:252–260. doi: 10.1097/00000658-200202000-00014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Hoda MR, Popken G. Surgical outcomes of fluorescence-guided laparoscopic partial nephrectomy using 5-aminolevulinic acid-induced protoporphyrin IX. J Surg Res. 2009;154:220–225. doi: 10.1016/j.jss.2008.12.027. [DOI] [PubMed] [Google Scholar]
  • 31.Loning M, Diddens H, Kupker W, et al. Laparoscopic fluorescence detection of ovarian carcinoma metastases using 5-aminolevulinic acid-induced protoporphyrin IX. Cancer. 2004;100:1650–1656. doi: 10.1002/cncr.20155. [DOI] [PubMed] [Google Scholar]
  • 32.Zopf T, Schneider AR, Weickert U, et al. Improved preoperative tumor staging by 5-aminolevulinic acid induced fluorescence laparoscopy. Gastrointest Endosc. 2005;62:763–767. doi: 10.1016/j.gie.2005.05.020. [DOI] [PubMed] [Google Scholar]
  • 33.Winer JH, Choi HS, Gibbs-Strauss SL, et al. Intraoperative localization of insulinoma and normal pancreas using invisible near-infrared fluorescent light. Ann Surg Oncol. 2010;17:1094–1100. doi: 10.1245/s10434-009-0868-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Stiles BM, Bhargava A, Adusumilli PS, et al. The replication-competent oncolytic herpes simplex mutant virus NV1066 is effective in the treatment of esophageal cancer. Surgery. 2003;134:357–364. doi: 10.1067/msy.2003.244. [DOI] [PubMed] [Google Scholar]
  • 35.Bouvet M, Hoffman RM. Glowing tumors make for better detection and resection. Sci Transl Med. 2011;3:110fs10. doi: 10.1126/scitranslmed.3003375. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS358033-supplement-01.jpg (47.6KB, jpg)
02
NIHMS358033-supplement-02.jpg (57.6KB, jpg)

RESOURCES