Original Research  n  Molecular

Imaging

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

Fluorescent Iodized Emulsion for Pre- and Intraoperative Sentinel Lymph Node Imaging: Validation in a Preclinical Model1 Honsoul Kim, MD, PhD Sang Kil Lee, MD, PhD Yoo Min Kim, MD Eun-Hye Lee, BS Soo-Jeong Lim, PhD Se Hoon Kim, MD, PhD Jaemoon Yang, PhD Joon Seok Lim, MD, PhD Woo Jin Hyung, MD, PhD

j1  From the Department of Radiology (H.K., J.Y., J.S.L.), Research Institute of Radiological Science (H.K., J.S.L.), Gastric Cancer Clinic (H.K., S.K.L., J.S.L., W.J.H.), and Department of Internal Medicine (S.K.L.), Department of Pathology (S.H.K.), YUHS-KRIBB Medical Convergence Research Institute (J.Y.), Department of Surgery (W.J.H.), and Robot and MIS Center (W.J.H.), Severance Hospital, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea; Department of Surgery, Bundang CHA Hospital, CHA University College of Medicine, Seongnam, Republic of Korea (Y.M.K.); Department of Bioscience and Bioengineering, Sejong University, Seoul, Republic of Korea (E.H.L., S.J.L.). From the 2013 RSNA Annual Meeting. Received May 30, 2014; revision requested July 22; final revision received August 4; accepted August 28; final version accepted September 10. Supported by the National Research Foundations of Korea (grants 2010-0021549 and 2012R1A2A2A01046171), funded by the Ministry of Education, Science and Technology, Republic of Korea. Address correspondence to J.S.L. (e-mail: [email protected]).

Purpose:

To validate the usefulness of a newly developed tracer for preoperative gastric sentinel lymph node (LN) (SLN) mapping and intraoperative navigation after a single preoperative submucosal injection in rat and beagle models.

Materials and Methods:

This study was approved by the Experimental Animal Ethical Committee of Yonsei University College of Medicine according to the eighth edition of the Guide for the Care and Use of Laboratory Animals published in 2011. An emulsion was developed that contained indocyanine green in iodized oil, which can be visualized with both computed tomography (CT) and near-infrared (NIR) optical imaging and has the property of delayed washout. This emulsion was injected into the footpad of rats (n = 6) and the gastric submucosa of beagles (n = 8). CT lymphography was performed. The degree of enhancement of popliteal LNs was measured in rats, and the enhancing LNs were identified and the degree of enhancement of the enhancing LNs was measured in beagles. Next, NIR imaging was performed in beagles during open, laparoscopic, and robotic surgery to identify LNs containing the fluorescent signals of indocyanine green. The enhanced LNs detected with CT lymphography and NIR imaging were matched to see if they corresponded.

Results:

Preoperative CT lymphography facilitated SLN mapping, and 26 SLNs were identified in eight beagles. NIR imaging enabled high-spatial-resolution visualization of both SLNs and the intervening lymphatic vessels and was useful for intraoperative SLN navigation.

Conclusion:

SLN mapping with fluorescent iodized oil emulsion is effective and feasible for both CT and NIR imaging.  RSNA, 2014

q

Online supplemental material is available for this article.

 RSNA, 2014

q

196

radiology.rsna.org  n  Radiology: Volume 275: Number 1—April 2015

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

S

ystemic lymph node (LN) dissection with gastric resection is the standard surgery for treating gastric cancer (1). However, most patients with early gastric cancer are free of nodal metastasis. Therefore, uniformly performing radical gastrectomy raises concerns of potential overtreatment (2,3). To minimize LN dissection for patients with early gastric cancer while maintaining the same oncologic outcomes, we need an indicator that can confirm the absence of nodal metastasis with high accuracy (4). Sentinel LN (SLN) mapping may provide a solution (5–7). In spite of the complexity of gastric lymphatic drainage, accumulating evidence indicates that gastric SLN biopsy can be performed safely and effectively when it is performed in selected patients: The dual tracer method, which combines vital dyes and radioactive colloids, is the method of choice for detecting SLNs in early gastric cancer because of its high sensitivity and specificity (8–10). Such combinative strategies are

Advances in Knowledge nn We developed a fluorescent iodized oil emulsion that contains indocyanine green and used it for hybrid multimodality imaging (CT and near infra-red [NIR] optical imaging) of gastric sentinel lymph nodes (LNs) (SLNs) in beagles to achieve both SLN preoperative mapping and intraoperative navigation. nn Preoperative CT lymphography facilitated anatomy-oriented preoperative SLN mapping in rat and beagle models. nn Intraoperative NIR optical imaging enabled the detection of SLNs and visualization of the intervening lymphatic vessels of the draining lymphatic system. nn Our emulsion enabled delayed phase imaging, as it did not rapidly wash away after entering the draining LNs and therefore allowed sufficient time for SLN evaluation.

recommended to compensate for the intrinsic limitations of individual tracers: Vital dyes transit and wash away rapidly and are difficult to detect in areas covered by dense fat tissue (9,11), whereas radioactive tracers demonstrate poor spatial resolution and poor sensitivity for detection of LNs near the injection site because of high background scattering (shine-through effect) (12–14). Moreover, these methods carry risks associated with radiation exposure in the operating room and require separate endoscopic sessions during surgery, which considerably increases the surgical operation time. Recently, several groups, including ours, have reported promising results for preoperative computed tomographic (CT) imaging of SLNs using iopamidol (Isovue; Bracco Diagnostics, Princeton, NJ) (15–17), iodized oil emulsion (18), and ethiodized oil (14) in patients with gastric cancer or animal models. Preoperative CT lymphography for SLN mapping has the advantages of providing anatomic information about SLNs without intraoperative procedures and of reducing the surgical operation time. However, during the surgical operation, additional procedures for identifying SLNs are still necessary (14,15,17,18). We hypothesized that a tracer that can be detected with both CT and optical imaging and can serve as a vital dye would enable both preoperative SLN mapping and intraoperative SLN navigation. Such tracers could overcome the limitations of dual tracer methods using vital dyes and radioactive colloids. Therefore, we developed a tracer fluorescent iodized oil emulsion containing indocyanine green (ICG) that can be simultaneously imaged with CT and near-infrared (NIR) optical imaging systems. The purpose

Implication for Patient Care nn Our lymphography method can be potentially integrated into current diagnostic and/or therapeutic workflow of early gastric cancer and therefore expand the application to gastric SLN assessment.

Radiology: Volume 275: Number 1—April 2015  n  radiology.rsna.org

Kim et al

of this study was to validate the usefulness of a new tracer for preoperative gastric SLN mapping and intraoperative navigation after a single preoperative endoscopic submucosal injection in rat and beagle models.

Materials and Methods This study was approved by the Experimental Animal Ethical Committee of Yonsei University College of Medicine (Seoul, Republic of Korea), according to the eighth edition of the Guide for the Care and Use of Laboratory Animals published by the National Academies Press (Washington, DC) in 2011.

Preparation and Characterization of Fluorescent Iodized Oil Emulsion The emulsion was prepared by using a homogenization method, as previously reported (19). Briefly, surfactants (127 mg polyoxyethylene sorbitan monooleate [Tween 80; Sigma-Aldrich, St Louis, Mo], 14 mg sorbitan trioleate [Span 85; Sigma-Aldrich]), 5 mg ICG (Dongindang, Siheung, Korea), and 300 mL of iodized oil (Lipiodol; Guerbet, Villepinte, France) were dissolved in ethanol. It is generally known that the combination of surfactants with a high hydrophiliclipophilic balance value (Tween 80) and a low hydrophilic-lipophilic balance value

Published online before print 10.1148/radiol.14141159  Content codes: Radiology 2015; 275:196–204 Abbreviations: ICG = indocyanine green LN = lymph node NIR = near-infrared SLN = sentinel LN Author contributions: Guarantors of integrity of entire study, J.Y., J.S.L.; study concepts/study design or data acquisition or data analysis/ interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, H.K., S.J.L., J.S.L., W.J.H.; clinical studies, S.K.L., Y.M.K.; experimental studies, H.K., S.K.L., E.H.L., S.J.L., S.H.K., J.Y., J.S.L., W.J.H.; and manuscript editing, H.K., J.S.L., W.J.H. Conflicts of interest are listed at the end of this article.

197

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

(Span 85) produces a stable and fine emulsion (20). The organic solvent was removed with reduced pressure until a pre-emulsion concentrate was obtained, and 5% d-glucose solution (0.7 mL) was added drop by drop while vortexing took place. The mixture was sonicated at 37°C for 30 minutes and placed in an ice bath, homogenized for 10 minutes at 11 000 rpm with a homogenizer (IKA Werke, Staufen, Germany), and next held for 5 minutes at room temperature. This process was repeated three times to reduce the emulsion size. Unloaded ICG was removed from the emulsion dispersion by using dialysis (Fig E1a [online]). The emulsion was dark green and produced a strong CT attenuation effect (. 3000 HU) (Fig E1b [online]). The process we went through to confirm the technical adequacy during development of the emulsion and the methods we used to measure the basic physicochemical characteristics of the emulsion are provided in Appendix E1 (online) (Figs E1, E2 [online]).

Emulsion as a Tracer for SLNs in a Rat Model Male 9-week-old Crj/Bgi SpragueDawley rats obtained from a supplier (Charles River Laboratories Japan, Yokohama, Japan) weighing 430–550 g (n = 6) were used. The rats were anesthetized with an intraperitoneal injection of tiletamine and zolazepam (Zoletil; Vibrac Laboratories, Carros, France), 50 mg/kg, and xylazine (Rompun; Bayer, Seoul, Korea), 10 mg/kg. Popliteal LN lymphography was performed by subcutaneously injecting 0.3 mL of emulsion into the right footpad. Whole-body CT images were obtained before the injection and at 1, 2, 4, 8, 24, and 72 hours after the injection (Fig 1a, 1b) by using a commercial 64-channel multidetector CT scanner (Discovery CT750 HD; GE Healthcare, [Milwaukee, Wis]) with the following parameters: 120 kV, 250 mA, 0.625mm beam collimation, and 1.25-mm axial reconstruction thickness. Field of view was adjusted to include the entire abdomen and lower extremities. The lower extremity was dissected, and the popliteal LNs were 198

Kim et al

exposed for gross inspection and NIR imaging (eXplore Optix; Advanced Research Technologies GE Healthcare, Montreal, Quebec, Canada) after 24 hours (n = 3) or 72 hours (n = 3) (Fig 1c). Afterward, the rats were euthanized with an overdose of zolazepam.

Preclinical Application of Emulsion in a Canine Model Adult male beagles (18 months old, n = 8) obtained from a supplier (Orient, Seoul, Korea) weighing 10.5–13.5 kg were used. A mixture of zolazepam (Zoletil 50; Vibrac), 5 mg/kg, and xylazine, 0.2 mg/kg, was intravenously injected to induce and maintain adequate levels of anesthesia. After CT imaging, the animals were intubated, and general anesthesia was maintained with enflurane (JW Pharmaceutical, Seoul, Korea) in oxygen (2.0% end-tidal concentration). For endoscopic injection, the emulsion was endoscopically injected into the gastric submucosa by an experienced gastroenterologist (S.K.L., with 12 years of experience), as previously described (18). In brief, a flexible endoscope (GIF-Q260; Olympus, Tokyo, Japan) was inserted to inject emulsion into the gastric submucosa (three 0.3mL injections, with 0.9 mL in total) within a narrow area surrounding a specific target. This simulates the clinical setting for early gastric cancer, in which multiple injections are delivered at the margins of each lesion of limited size. Different target locations were set up at the proximal half (n = 4) and distal half (n = 4) of the stomach. For CT lymphography, multiple CT scans (with the same parameters used in rats) were obtained before emulsion injection (precontrast imaging) and immediately after injection (at approximately 3–5 minutes) and at 30, 60, 90, and 120 minutes after injection (Fig E3a, E3b [online]). Image Analysis All CT images were analyzed at a picture archiving and communicating system workstation (Centricity; GE Healthcare, Milwaukee, Wis) by a gastrointestinal radiologist (H.K.,

with 3 years of experience). To quantify LN enhancement, a region of interest was placed on each target LN, avoiding the inclusion of perinodal fat tissue, and the mean Hounsfield unit was determined. For rats, enhancement of right popliteal LNs was quantitatively measured. On CT lymphography images in beagles, we defined SLN as an LN that contained nodular or curvilinear highattenuation foci, with attenuation greater than 30 HU on any contrast material–enhanced (postcontrast) image compared with the corresponding precontrast image (15,21). The location of each SLN was recorded on the basis of the classification of LN stations of the Japanese Classification of Gastric Carcinoma (1).

Surgery SLN navigation surgery was performed by two experienced surgeons (W.J.H. and Y.M.K., with 13 years and 2 years of experience in gastric cancer surgery, respectively) who were informed of the CT lymphography results before the surgical operations. We attempted to simulate various types of gastric surgery by performing open (n = 3), laparoscopic (n = 3), and robotic surgery (n = 2). The emulsion served as a vital dye, and the surgeons searched for LNs stained green to identify SLNs. Standard laparoscopic gastrectomy was performed by using a high-definition laparoscopy system without NIR imaging (1288 HD; Stryker Medical, Portage, Mich). Robotic surgery was conducted by using a surgical system (da Vinci Si Surgical System; Intuitive Surgical, Sunnyvale, Calif ) equipped with a NIR imaging system (22). For open and laparoscopic surgery, the surgical operation began 3–4 hours after emulsion injection and lasted approximately 1 hour. Robotic surgery began approximately 4 hours after emulsion injection and lasted approximately 2 hours, on average. After the experiment, all beagles were euthanized with saturated potassium chloride and an overdose of zolazepam.

radiology.rsna.org  n Radiology: Volume 275: Number 1—April 2015

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

Kim et al

Figure 1

Figure 1:  Emulsion lymphography of rat popliteal LNs (n = 6). (a) Representative CT images of a popliteal LN (arrow, LN is outlined by adjacent fat tissue) obtained at different times ( 72 hours) after unilateral footpad injection of the emulsion. Some of the injected emulsion remained in the subcutaneous compartment. (b) Graph shows degree of enhancement of popliteal LNs ( Hounsfield unit: subtraction of the measured Hounsfield unit at a specific phase from that of the precontrast phase, expressed as mean 6 standard deviation [error bars]). One-half of the rats were sacrificed at 24 hours; therefore, the graph shows data for six rats until 24 hours and data for three rats at 72 hours. (c) Gross image (left) and NIR optical image (right) of an exposed popliteal LN (arrow) 24 hours (top) and 72 hours (bottom) after emulsion injection in the footpad. Insets show NIR optical image of the same LN after dissection.

Postresection Specimen Assessment The LNs were retrieved from the specimen, and LN stations were labeled by two experienced surgeons (W.J.H. and Y.M.K. with 13 years and 2 years of experience in gastric cancer surgery, respectively). NIR optical images of the retrieved LNs were obtained with an optical imaging system (eXplore Optix; Advanced Research Technologies

GE Healthcare) to identify the LNs that were labeled with the fluorescent ICG (Fig E3c [online]). The enhanced LNs, each detected by using CT lymphography and NIR imaging, were matched to see whether the enhancement in each LN corresponded. Histologic assessment (Fig E3d [online]) was performed after hematoxylin-eosin staining (S.H.K.) to confirm that LNs were correctly sampled.

Radiology: Volume 275: Number 1—April 2015  n  radiology.rsna.org

Results Physicochemical Characteristics of the Fluorescent Iodized Oil Emulsion Measurement results for technical validation during the development of the emulsion and the basic physicochemical characteristics of it are provided in Appendix E1 (online) and Figures E1 and E2 (online). 199

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

Kim et al

Figure 2

Figure 2:  SLN enhancement patterns at CT lymphography in eight beagles. Degree of enhancement ( Hounsfield unit: subtraction of the measured value of Hounsfield unit at a specific phase from that of the precontrast phase) for (a) all SLNs together or (b) perigastric SLNs (stations 1–6, red line) and extraperigastric SLNs (stations 7–14, blue line), shown separately, is expressed as mean 6 standard deviation (error bars). Dotted line = mean value of all SLNs.

Validation of Emulsion as a Tracer in a Rat Model We observed a pattern of gradually increasing enhancement of the ipsilateral popliteal LNs, which peaked at 8 hours. Partial washout was observed at 24 hours, but a considerable degree of enhancement remained. Further washout of the emulsion was observed at 72 hours; however, the degree of enhancement still exceeded 30 HU, compared with that of the precontrast image (Fig 1a, 1b). At both times, the popliteal LNs were green, indicating the presence of ICG. At NIR imaging, the LNs showed obvious ICG signal at 24 hours, which was decreased at 72 hours (Fig 1c). Preclinical Validation in a Canine Model Preoperative CT lymphography.—After endoscopic gastric submucosal injection of the emulsion, we obtained multiphasic dynamic CT images (Fig E3a, E3b [online]). At preoperative CT lymphography, 26 LNs were identified with enhancement greater than 30 HU in eight beagles (Figs E3a, E3b; 200

E4 [online]), which we defined as SLNs (15,21). The median number of SLNs per beagle was three, with a range of two to six. Overall, the SLNs showed a pattern of gradually increasing enhancement over time (Fig 2a). The enhancement of extraperigastric SLNs (stations 7–12) seemed to show a tendency of being relatively weaker compared with that of perigastric SLNs (stations 1–6) (Fig 2b). In three of the beagles (38%), emulsion was directly visualized in a lymphatic channel draining into a nearby LN immediately after injection (Fig E3b [online]). SLN navigation in different types of surgery.—In all animals, the emulsion injection site was visible from the peritoneal surface of the stomach. However, even when preoperative CT revealed enhanced LNs, it was often difficult to find green LNs at the corresponding locations during surgery without NIR imaging. During laparoscopic gastrectomy (n = 3), regions showing enhanced LNs at preoperative CT (Fig 3a) were carefully examined. The magnification and

illumination of the laparoscope facilitated detection of the stained LNs. We were also able to trace the stained intervening lymphatic vessels (Fig 3b) that connected the primary injection site and SLNs (Fig 3c). Ex vivo NIR optical imaging of the retrieved LNs also showed ICG signals in the LNs (Fig 3c, inset). However, detection of the green stain was occasionally difficult, especially when the draining lymphatic structures were located in deep fat tissue. Robotic surgery (n = 2) by using the NIR imaging–equipped system was performed. Intraoperative navigation based on real-time NIR imaging of ICG improved signal detection and tissue penetration. In both beagles, we found intraoperative real-time NIR imaging to be a feasible and confirmative method to assess emulsion-containing lymphatic structures. Preoperative CT lymphography revealed anatomic locations of SLNs for the surgeon to examine during the surgical operation (Fig 4a). At intraoperative NIR imaging, we identified ICG fluorescence signals (Fig 4b–4d) in all six enhanced LNs revealed by preoperative

radiology.rsna.org  n Radiology: Volume 275: Number 1—April 2015

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

Kim et al

Figure 3

Figure 3:  Laparoscopic surgery after emulsion injection into the anteroinferior wall of the gastric antrum in beagle 4. (a) CT image at 120 minutes after injection revealed two SLNs at station 6 (arrows). (b, c) Magnified images of the operation field through the laparoscope camera approximately 4 hours after emulsion injection. The draining lymphatic vessels (arrowheads), SLN directly connected to the injection site (yellow arrow), and two station 6 LNs (white arrows) are stained green. Ex vivo NIR optical imaging of the retrieved tissue (inset) shows the two corresponding fluorescence signals (arrows).

CT lymphography in both animals (Fig E4c [online]). Moreover, intraoperative NIR imaging revealed two additional ICG-retaining LNs (stations 11 and 8 in beagle 7) that did not meet the criteria of an SLN at CT lymphography. Of note, nearly the entire length of the intervening lymphatic vessels could be directly visualized and traced, enabling the assessment of lymphatic connections between the emulsion injection site and the draining LNs (Fig 4b, 4c; Movie). The enhanced lymphatic visualization capability was beneficial not only for SLN detection, but also for the differentiation of true SLNs (having a direct connection with the injection site) from second-tier LNs and beyond (Fig 4d). Correlation of preoperative CT and postoperative NIR imaging.—At ex vivo examination of the retrieved LNs, ICG fluorescence was observed in 24 of 26 (92%) enhanced LNs at preoperative CT lymphography (Fig E4 [online]). The two mismatches appeared to be the result of sampling error, as these extraperigastric LNs had not been retrieved from the specimen (station 11 in beagle 5 and station 14 in beagle 4).

Discussion In this study, we developed a tracer containing iodized oil and fluorescent dye, ICG, together; both components already

have been approved and are actively used in regular practice. This emulsion was designed to exhibit physicochemical characteristics adjusted to prevent rapid washout from SLNs, as well as features such as small mean particle size, low polydispersity index, high emulsion stability, and increased chemical stability of ICG. Our tracer enables lymphography for SLN assessment in both preoperative CT and intraoperative NIR fluorescent imaging even at delayed phases, indicating that it can overcome the limitations of current SLN tracers. One reason for developing this tracer was to prolong tracer retention in the lymphatic system and, therefore, to allow sufficient time for examination. The flow kinetics of a tracer appears to be determined largely by particle size and stability (23). If the particle size is small, a considerable portion could drain directly into capillaries or easily pass through the SLNs. If the particle size is too large, the tracer is not readily absorbed into the lymphatic system (23,24). Because radiocolloids with particle size between 100 and 200 nm appeared to be effective for SLN identification (25), we tested a number of formulas to produce emulsions of various particle sizes, ending up with the current version (particle size, approximately 200 nm). Unlike water-soluble vital dyes or iopamidol (9,11,15,26,27),

Radiology: Volume 275: Number 1—April 2015  n  radiology.rsna.org

our emulsion showed prolonged accumulation in the draining lymphatic system. A tracer showing long retention enhances procedure feasibility, because it increases the window of time for SLN evaluation (28). We were able to accomplish preoperative CT and intraoperative NIR SLN imaging throughout the entire surgical process. One endoscopic injection before surgery, which we assume can be performed during routine preoperative endoscopy, was sufficient to deliver the emulsion. We believe that the process of endoscopic emulsion injection and CT lymphography could be integrated into routine preoperative endoscopy and CT procedures. CT lymphography is promising in that it can preoperatively depict SLNs (or the SLN basin) and provide detailed anatomic information (14,15,26,27). In our previous study, we performed CT lymphography with a water-soluble nonparticulate iodinated contrast agent (iopamidol) but experienced rapid tracer washout and consequently low SLN detection rate (15). To overcome this issue, we used iodized oil emulsion as a CT lymphography tracer, which provided sufficient and sustained contrast enhancement of the draining LNs or reticuloendothelial organs in rat and canine models (18,20,29). Similarly, researchers in recent studies described iodized oil–based CT lymphography as 201

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

Kim et al

Figure 4

Figure 4:  Robotic surgery with a unit equipped with an integrated NIR-capable imaging system in beagle 7. Emulsion was injected into the posterosuperior wall of the gastric antrum. Preoperative CT lymphography images of the same beagle are shown in Figure E3 (online). (a) Magnified CT images at 120 minutes after injection (images selected from Figure E3a [online]). Numbers in parentheses represent the measured Hounsfield units at CT. (b) Intraoperative representative images of corresponding fields acquired by using robot camera with white light (left) and NIR optical device (right). (c) Optimal tissue exposure and magnification achieved with real-time NIR optical device guidance. Station 3 LN (yellow arrow), station 6 LN (green arrow), and station 8 LN (orange arrow) are demonstrated. The intervening lymphatic vessels (arrowheads) and LNs containing emulsion (arrows) were clearly visualized and traceable by using real-time NIR optical imaging. Arrows and arrowheads of the same color indicate corresponding structures throughout a–d. Brightness of NIR images has been adjusted. (d) Schematic map depicts the draining lymphatic system based on intraoperative findings (Movie). We identified station 3 (yellow arrow) and 6 (green arrow) LNs as true SLNs (red dotted circles) because they were directly connected to the injection site. Station 8 and 11 LNs appeared to be second-tier (stations 8, 11) and third-tier (station 8) LNs, as their connection with the injection site was established through the station 3 LN. Numbers in each circle indicate the station of the LN.

a feasible and effective SLN detection method that avoids the problem of rapid tracer washout (14,30). However, CT lymphography itself cannot replace intraoperative SLN 202

detection procedures; therefore, conventional dye methods that are based on intraoperative endoscopy are still required (14,15,18,30). Moreover, some discrepancies between SLN detection

by using iodized oil–based CT lymphography and the dual tracer method were noticed (30). We speculated that multiple tracers delivered in a single particle could overcome this limitation. The ICG

radiology.rsna.org  n Radiology: Volume 275: Number 1—April 2015

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

component not only served as a vital dye, but also enabled intraoperative SLN navigation by NIR imaging during robotic surgery. This improves the sensitivity and resolution of the ICG signal and enhances the detectability of SLNs (5,31,32), and we were able to visualize intervening lymphatic vessels over nearly their entire length and assess lymphatic connections between draining LNs. We acknowledge several limitations of this study. First, our preclinical study did not evaluate the safety of the emulsion. Thus, biologic safety evaluation is necessary although we did not observe any adverse events in the animals that underwent lymphography. Second, we were not able to identify an ideal time for preoperative CT lymphography. Thus, further study is needed to identify optimal scan time for simple and efficient CT lymphography. Third, our study was performed on SLN evaluation in a model not bearing tumor. We plan to evaluate the diagnostic performance in tumor models before attempting clinical validation in patients with early gastric cancer. In conclusion, we developed a fluorescent iodized oil emulsion that can be used as a hybrid multimodality imaging lymphography tracer to achieve gastric SLN assessment by using preoperative mapping with CT and intraoperative navigation with NIR optical imaging. The emulsion shows prolonged retention in SLNs, enabling delayed phase imaging, and no additional injections were required. Consequently, this approach improves the diagnostic performance and overcomes the limitations of tracers currently used for SLN mapping. SLN assessment on the basis of this emulsion is effective and feasible, suggesting that it can expand the application of SLN navigation into gastric cancer treatment. Disclosures of Conflicts of Interest: H.K. disclosed no relevant relationships. S.K.L. disclosed no relevant relationships. Y.M.K. disclosed no relevant relationships. E.H.L. disclosed no relevant relationships. S.J.L. disclosed no relevant relationships. S.H.K. disclosed no relevant relationships. J.Y. disclosed no relevant relationships. J.S.L. Activities related to the present article: institution received

a grant from the National Research Foundations of Korea. Activities not related to the present article: disclosed no relevant relationships. Other relationships: disclosed no relevant relationships. W.J.H. disclosed no relevant relationships.

References 1. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 3rd English edition. Gastric Cancer 2011;14(2):101–112. 2. Kitagawa Y, Kitajima M. Lymphatic mapping for upper gastrointestinal malignancies. Semin Oncol 2004;31(3):409–414. 3. Kitagawa Y, Ohgami M, Fujii H, et al. Laparoscopic detection of sentinel lymph nodes in gastrointestinal cancer: a novel and minimally invasive approach. Ann Surg Oncol 2001;8(suppl):86S–89S. 4. Lee JH, Lee HJ, Kong SH, et al. Analysis of the lymphatic stream to predict sentinel nodes in gastric cancer patients. Ann Surg Oncol 2014;21(4):1090–1098. 5. Takeuchi H, Kitagawa Y. New sentinel node mapping technologies for early gastric cancer. Ann Surg Oncol 2013;20(2):522–532. 6. Bembenek A, Gretschel S, Schlag PM. Sentinel lymph node biopsy for gastrointestinal cancers. J Surg Oncol 2007;96(4):342–352. 7. Can MF, Yagci G, Cetiner S. Systematic review of studies investigating sentinel node navigation surgery and lymphatic mapping for gastric cancer. J Laparoendosc Adv Surg Tech A 2013;23(8):651–662. 8. Park J, Kim HH, Park YS, et al. Simultaneous indocyanine green and (99m)Tc-antimony sulfur colloid-guided laparoscopic sentinel basin dissection for gastric cancer. Ann Surg Oncol 2011;18(1):160–165. 9. Kitagawa Y, Fujii H, Kumai K, et al. Recent advances in sentinel node navigation for gastric cancer: a paradigm shift of surgical management. J Surg Oncol 2005;90(3):147– 151; discussion 151–152. 10. Kitagawa Y, Takeuchi H, Takagi Y, et al. Sentinel node mapping for gastric cancer: a prospective multicenter trial in Japan. J Clin Oncol 2013;31(29):3704–3710. 11. Kitagawa Y, Saikawa Y, Takeuchi H, et al. Sentinel node navigation in early stage gastric cancer: updated data and current status. Scand J Surg 2006;95(4):256–259. 12. Kitagawa Y, Fujii H, Mukai M, Kubota T, Otani Y, Kitajima M. Radio-guided sentinel node detection for gastric cancer. Br J Surg 2002;89(5):604–608. 13. Mariani G, Moresco L, Viale G, et al. Radioguided sentinel lymph node biopsy in

Radiology: Volume 275: Number 1—April 2015  n  radiology.rsna.org

Kim et al

breast cancer surgery. J Nucl Med 2001; 42(8):1198–1215. 14. Kim YH, Lee YJ, Park JH, et al. Early gastric cancer: feasibility of CT lymphography with ethiodized oil for sentinel node mapping. Radiology 2013;267(2):414–421. 15. Hyung WJ, Kim YS, Lim JS, Kim MJ, Noh SH, Kim KW. Preoperative imaging of sentinel lymph nodes in gastric cancer using CT lymphography. Yonsei Med J 2010;51(3):407–413. 16. Suga K, Shimizu K, Kawakami Y, et al. Lymphatic drainage from esophagogastric tract: feasibility of endoscopic CT lymphography for direct visualization of pathways. Radiology 2005;237(3):952–960. 17. Tsujimoto H, Yaguchi Y, Sakamoto N, et al. Computed tomography lymphography for the detection of sentinel nodes in patients with gastric carcinoma. Cancer Sci 2010;101(12):2586–2590. 18. Lim JS, Choi J, Song J, et al. Nanoscale iodized oil emulsion: a useful tracer for pretreatment sentinel node detection using CT lymphography in a normal canine gastric model. Surg Endosc 2012;26(8):2267–2274. 19. Lee EH, Hong SS, Kim SH, Lee MK, Lim JS, Lim SJ. Computed tomography-guided screening of surfactant effect on blood circulation time of emulsions: application to the design of an emulsion formulation for paclitaxel. Pharm Res 2014;31(8):2022–2034. 20. Chung YE, Hyung WJ, Kweon S, et al. Feasibility of interstitial CT lymphography using optimized iodized oil emulsion in rats. Invest Radiol 2010;45(3):142–148. 21. Suga K, Yuan Y, Ueda K, et al. Computed tomography lymphography with intrapulmonary injection of iopamidol for sentinel lymph node localization. Invest Radiol 2004;39(6):313–324. 22. Rossi EC, Ivanova A, Boggess JF. Robotically assisted fluorescence-guided lymph node mapping with ICG for gynecologic malignancies: a feasibility study. Gynecol Oncol 2012;124(1):78–82. 23. Wilhelm AJ, Mijnhout GS, Franssen EJ. Radiopharmaceuticals in sentinel lymphnode detection: an overview. Eur J Nucl Med 1999;26(suppl):S36–S42. 24. Leidenius MH, Leppänen EA, Krogerus LA, Smitten KA. The impact of radiopharmaceutical particle size on the visualization and identification of sentinel nodes in breast cancer. Nucl Med Commun 2004;25(3):233–238. 25. Mariani G, Erba P, Villa G, et al. Lym phoscintigraphic and intraoperative detection of the sentinel lymph node in breast

203

MOLECULAR IMAGING: Gastric Sentinel Lymph Node Imaging with Fluorescent Emulsion

cancer patients: the nuclear medicine perspective. J Surg Oncol 2004;85(3):112– 122. 26. Suga K, Ogasawara N, Yuan Y, Okada M, Matsunaga N, Tangoku A. Visualization of breast lymphatic pathways with an indirect computed tomography lymphography using a nonionic monometric contrast medium iopamidol: preliminary results. Invest Radiol 2003;38(2):73–84. 27. Hayashi H, Tangoku A, Suga K, et al. CT lymphography-navigated sentinel lymph node biopsy in patients with superficial esophageal cancer. Surgery 2006;139(2):224–235.

204

Kim et al

28. Emerson DK, Limmer KK, Hall DJ, et al. A receptor-targeted fluorescent radiopharmaceutical for multireporter sentinel lymph node imaging. Radiology 2012;265(1):186–193. 29. Kweon S, Lee HJ, Hyung WJ, Suh J, Lim JS, Lim SJ. Liposomes coloaded with iopamidol/lipiodol as a RES-targeted contrast agent for computed tomography imaging. Pharm Res 2010;27(7):1408–1415. 30. Lee JH, Park J, Kim YH, Shin CM, Lee HS, Kim HH. Clinical implementations of preoperative computed tomography lymphography in gastric cancer: a comparison with dual tracer methods in sentinel node navi-

gation surgery. Ann Surg Oncol 2013;20(7): 2296–2303. 31. Nimura H, Narimiya N, Mitsumori N, Yamazaki Y, Yanaga K, Urashima M. Infrared ray electronic endoscopy combined with indocyanine green injection for detection of sentinel nodes of patients with gastric cancer. Br J Surg 2004;91(5):575–579. 32. Ishikawa K, Yasuda K, Shiromizu A, Etoh T, Shiraishi N, Kitano S. Laparoscopic sentinel node navigation achieved by infrared ray electronic endoscopy system in patients with gastric cancer. Surg Endosc 2007;21(7):1131– 1134.

radiology.rsna.org  n Radiology: Volume 275: Number 1—April 2015

Fluorescent iodized emulsion for pre- and intraoperative sentinel lymph node imaging: validation in a preclinical model.

To validate the usefulness of a newly developed tracer for preoperative gastric sentinel lymph node (LN) (SLN) mapping and intraoperative navigation a...
1MB Sizes 0 Downloads 7 Views