Sentinel Lymph Node Mapping of the Colon and ... - Springer Link

Annals of Surgical Oncology, 11(7):674–681

DOI: 10.1245/ASO.2004.06.007

Sentinel Lymph Node Mapping of the Colon and Stomach Using Lymphoseek in a Pig Model Scott J. Ellner, DO, MPH, Jeanette Méndez, MD, David R. Vera, PhD, Carl K. Hoh, MD, † William L. Ashburn, MD, and Anne M. Wallace, MD, FACS

Background: Lymphoseek is a radiopharmaceutical designed for sentinel lymph node (SLN) mapping. The purpose of this study was to compare Lymphoseek colon and gastric pharmacokinetics with filtered [99mTc]sulfur colloid (fTcSC). Methods: Eight anesthetized pigs received an endoscopic injection of Lymphoseek or fTcSC in the stomach and colon. Scintigraphy was obtained of both administration sites at 15-minute intervals up to 3 hours after injection, after which all SLNs were identified by a handheld gamma probe through a laparotomy incision. Isosulfan blue was administered at the injection site 5 minutes before SLN mapping. The percentage of injected dose (%ID) was measured for all harvested nodes, and the clearance half-life (Tc) was calculated for all injection sites. Results: The mean Lymphoseek clearance for colon (Tc, 2.56 ⫾ 1.04 hours) and gastric (Tc, 3.83 ⫾ 1.18 hours) injection sites was statistically faster (P ⫽ .030) compared with fTcSC (colon Tc, 14.98 ⫾ 3.41 hours; stomach Tc, 14.52 ⫾ 4.08 hours). After 3 hours, Lymphoseek exhibited a mean SLN %ID of 1.32% ⫾ 1.71% in the colon and 2.04% ⫾ 2.12% in the stomach; this was not statistically different from fTcSC (colon, .63% ⫾ .39%; stomach, 2.35% ⫾ 2.90%). SLN uptake of Lymphoseek was significantly different from second-echelon node %ID for the colon (P ⫽ .011) and gastric (P ⫽ .029) injection sites. All SLNs exceeded 10 times background, and there was no discordance between isosulfan blue and Lymphoseek or fTcSC. Conclusions: Three hours after colon stomach administration, Lymphoseek demonstrated rapid injection site clearance, detectable SLN uptake, and low second-echelon node uptake. Key Words: Sentinel lymph node—Colon cancer—Gastric cancer—Radiopharmaceutical— [99mTc]DTPA-mannosyl-dextran—Lymphoseek.

gastric and colon malignancies relies on the meticulous detection of lymph node metastases. The application of the sentinel lymph node (SLN) concept4 to gastrointestinal cancer is consistently proposed in various articles.5–10 The SLN approach, as in other cancers, may facilitate the accurate staging of patients with gastric and colon cancer by enabling the pathologist to perform a more thorough analysis of nodes likely to harbor occult metastases by using cytokeratin immunohistochemistry, reverse transcriptase-polymerase chain reaction assays, or both.11–13 The clinical relevance of the SLN status may determine whether extensive lymphadenectomy is necessary for early-stage gastric cancer, thus sparing the patient the morbidity of a radical lymph node dissection.14 Furthermore, focused examination of the SLN with ultrastaging techniques has reportedly upstaged colon cancer patients from American Joint

Colon and gastric cancer combined represent the third most frequent cancer incidence and the second most common cause of cancer deaths in the United States.1 An important prognostic indicator for gastrointestinal cancers is the presence or absence of tumor in the nodal drainage basin.2,3 Accordingly, the precise staging of Received June 3, 2003; accepted April 7, 2004. From the Departments of Surgery (SJE, AMW), Radiology (JM, DRV, CKH, WLA), and Medicine (JM) and the University of California, San Diego Comprehensive Cancer Center (SJE, DRV, CKH, AMW), University of California, San Diego, La Jolla, California. † Deceased. Address correspondence and reprint requests to: Anne M. Wallace, MD, FACS, Theodore Gildred Cancer Facility, 200 West Arbor Drive, San Diego, CA 92103-8421; Fax: 619-543-6644; E-mail: [email protected]. Published by Lippincott Williams & Wilkins © 2004 The Society of Surgical Oncology, Inc.

674

GASTROINTESTINAL SENTINEL NODE MAPPING VIA LYMPHOSEEK Committee on Cancer stage I/II (node-negative) to stage III (node-positive) disease in some series by 18% to 24%15,16; this warrants adjuvant chemotherapy in this group. Moreover, aberrant SLN drainage patterns of the colon beyond the associated vascular pedicle have been identified.17 These reports validate the importance of SLN mapping to colon and gastric cancer staging. Lymphoseek (Neoprobe Corp., Dublin, OH), also known as [99mTc]diethylenetriamine pentaacetic acidmannosyl-dextran, is a receptor-specific binding radiotracer designed for SLN mapping.18 A phase I clinical trial of SLN detection in patients diagnosed with breast cancer demonstrated a high concordance of Lymphoseek with isosulfan blue (Lymphazurin; US Surgical, Norwalk, CT) and a faster injection site clearance, with equivalent SLN uptake compared with filtered [99mTc] sulfur colloid (fTcSC).19 In a previous study, endoscopic administration of Lymphoseek into pigs demonstrated high colon and gastric lymph node accumulation within 10 minutes of injection into the submucosa of the colon and stomach. An ideal SLN agent should demonstrate rapid clearance from the administration site, rapid accumulation with prolonged retention in the SLN, and low uptake in second-echelon lymph nodes. Rapid clearance from the administration site minimizes the confounding effects of injection site scatter, which complicates gamma-guided SLN identification. Rapid SLN accumulation facilitates gamma probe detection of gastrointestinal SLNs without extending operative time. Prolonged SLN retention provides a wide procedural time window during which unexpected drainage patterns can be successfully mapped. Low additional or second-echelon node accumulation permits a focused histopathologic analysis of fewer nodes that are likely to harbor metastases. In this study, we examined the procedural window of colon and gastric SLN mapping with Lymphoseek by using an interval of 3 hours. We hypothesized that over 3 hours from injection to SLN identification, Lymphoseek exhibits rapid injection site clearance from the colon and stomach, high SLN detectability, and low migration to additional nodes. We tested this hypothesis by comparing the injection site clearance data with those of a standard lymphatic mapping agent, fTcSC, and by comparing the SLN accumulation data with standard detectability criteria for SLN mapping. MATERIALS AND METHODS Agent Preparation Synthesis of the radiopharmaceutical Lymphoseek was performed as previously described.20 The mean mo-

675

lecular diameter of the radiopharmaceutical was 7.0 nm, and the molecular weight was 36,000 g/mol. Formulation and quality control were performed as previously described.21 The radiotracer solution consisted of a mixture (1:1 v/v) of marking agent (India Ink; Permark Inc., Edison, NJ) and Lymphoseek (.1 mCi; 1.5 nmol) or fTcSC (.1 mCi). The radiopharmaceutical was administered within 1 hour of radiolabeling. A 10-␮L imaging standard and a 2-␮L counting standard were prepared from the injectate. All Lymphoseek labeling yields were ⬎98%. The labeling yields for fTcSC, measured after filtration by instant thin-layer chromatography, ranged from 90% to 98%.18 Animal Preoperative Preparation Eight inbred adolescent female pigs with average age and weight of 6 months and 15 kg were used. Before surgery, the pigs were fasted for 18 hours. Intramuscular injection of ketamine (20 mg/kg), atropine (.04 mg/kg), and xylazine (2.2 mg/kg) was administered for preanesthesia sedation. Induction of anesthesia occurred by infusion of sodium pentobarbital (30 mg/kg intravenously) through a 22-gauge catheter into a posterior auricular vein. Once anesthetized, the animal was intubated and mechanically ventilated at 8 breaths/min. Heart rate, blood pressure, respiratory rate, temperature, and electrocardiogram tracings were monitored and recorded. Additional boluses of anesthesia (sodium pentobarbital 2 mL intravenous push) were given for any movement, increase in vital signs (heart rate or blood pressure), or response to noxious stimulation (i.e., toe pinch). Hydration was maintained by an intravenous infusion of lactated Ringer’s solution at a continuous rate of 100 mL/ hour. The protocol was approved by the University of California, San Diego Animal Subjects Committee. Agent Administration Endoscopic administration of the radiotracer has been previously described.20 Briefly, a flexible video endoscope (Olympus America Ltd., Melville, NY) was advanced down the esophagus to the lesser curvature of the stomach. A tangential injection of .1 mL of the radiopharmaceutical into the gastric submucosa was performed with a standard endoscopic sclerotherapy needle (23 gauge ⫻ 6 mm; Olympus America Ltd.) and sheath (165 cm). The endoscope and needle were removed from the upper digestive tract once a raised wheal was visualized in the mucosa. Similarly, .1 mL of the same agent was injected into the submucosa of the colon 20 cm from the anal verge. Because the colon wall is thinner than the stomach wall, a shorter sclerotherapy needle (23 gauge Ann Surg Oncol, Vol. 11, No. 7, 2004

676

S. J. ELLNER ET AL.

⫻ 4 mm) was used. Once a raised wheal was seen in the mucosa of the colon, the endoscope was removed. Nuclear Imaging A 20 ⫻ 20-cm field of view gamma camera fitted with a low-energy, high-sensitivity, parallel-hole collimator was used for the imaging protocol (model 2020tc; Digirad Inc., San Diego, CA). An imaging standard with a known dilution of injectate was placed within the field of view. Images of the injection sites in the stomach and colon were acquired .25, .5, .75, 1, 2, 2.5, and 3 hours after injection. All images were stored on an imageprocessing computer as a 256 ⫻ 256 ⫻ 16 matrix. The serial images of the radiotracer injection sites for the stomach and colon acquired at the different time intervals were examined to identify a region of interest around the injection site. The number of counts within each region of interest was calculated by using standard nuclear medicine software (Digirad Inc.). The logarithmic values were then plotted as a function of time to perform a linear regression yielding a slope m (JMP version 5; SAS Institute, Cary, NC). Clearance half-lives (Tc) for Lymphoseek and fTcSC were then calculated as ⫺.693 divided by m. Lymph Node Excision and Radioactivity Assay A generous midline laparotomy incision was made 3 hours after the endoscopic injections. With the aid of a handheld gamma probe (Neoprobe 2000; Neoprobe Corp.), the colon and stomach injection sites were identified. The marking agent was used to facilitate identification of the gastric and colon injection sites. This was confirmed by a radioactive signal measured by the handheld gamma probe over the tattoo in the serosa of the stomach and colon. By using a tuberculin syringe, .1 mL of isosulfan blue was injected (with the exception of study 1) into the subserosa of the previous injection sites to confirm drainage to the SLN. The probe was placed near any blue or nonblue nodes in the drainage basin. For each procedure, a background measurement in counts per second (cps) was taken at three sites in the abdomen with the probe pointed away from the sites of radiotracer injection. All radioactive or blue nodes were then excised within 5 minutes of isosulfan blue administration. An SLN was identified by a signal-background ratio ⬎10 as measured by the gamma probe or by a blue color. We then excised lymph nodes that were neither hot nor blue and that were in the same lymphatic chain distal to the SLN. We refer to these nodes as non-SLN or secondechelon notes because of their location relative to the SLN. The analogy would be a level I axillary node next to an axillary SLN. Ann Surg Oncol, Vol. 11, No. 7, 2004

All excised lymph nodes were assayed for radioactivity by a gamma well counter with a 100- to 200-keV energy window (Gamma 8000; Beckman Instruments, Fullerton, CA). SLN and non-SLN uptake were calculated as the percentage of injected dose (%ID) by dividing a counting standard into the value given from the radioactivity assay multiplied by 100. Statistical Analysis Comparisons between unpaired groups were made by using a nonparametric Wilcoxon’s signed rank test. Injection site clearance, probe count rate, and %ID were compared between Lymphoseek and fTcSC. A P value ⬍.05 was considered statistically significant. All data were analyzed with JMP version 5 statistical software. Discordance between Lymphoseek and isosulfan blue and between fTcSC and isosulfan blue was tested by calculation of ␬, the proportion of chance-expected disagreements, as outlined by Cohen.22 RESULTS The mean background measurement by the gamma probe for the 16 procedures was 18 cps. Excised colon SLNs had a mean signal-background ratio of 712 (Lymphoseek; n ⫽ 6) and 103 (fTcSC; n ⫽ 6). The gastric SLN specimens had a mean signal-background ratio of 787 (Lymphoseek; n ⫽ 4) and 88 (fTcSC; n ⫽ 4). Ex vivo counts of colon and gastric second-echelon lymph nodes were lower than background measurements. The colon injection site clearance and lymph node uptake for blue dye, Lymphoseek, and fTcSC are listed in Table 1. The uptake data for primary and secondary SLNs and second-echelon lymph nodes include the ex vivo count rate, the presence of blue color, and the %ID. On the basis of radioactivity 10 times greater than background, a mean of 1.5 colon SLNs were identified per animal with both Lymphoseek and fTcSC. The colon injection site Tc for Lymphoseek (range, 1.78 – 4.09 hours; mean ⫾ SD, 2.56 ⫾ 1.04 hours) was significantly different (P ⫽ .030) from that for fTcSC (range, 12.21– 19.34 hours; mean ⫾ SD, 14.98 ⫾ 3.41 hours). Injection site clearance data from studies 4 and 7 are plotted in Fig. 1; the Tc values for each study were 4.09 ⫾ .08 hours and 12.33 ⫾ .12 hours. Approximately 70% of Lymphoseek was cleared from the colon injection site over 3 hours (study 4), whereas only ⬍20% of fTcSC (study 7) was removed over that same time. The ex vivo count rate for the colon primary and secondary SLNs after Lymphoseek injection ranged from 600 to 26,400 cps, with a mean of 7,011 cps. The range for the fTcSC SLNs was 200 to 3600 cps, with a mean of 1522 cps. A concor-

GASTROINTESTINAL SENTINEL NODE MAPPING VIA LYMPHOSEEK

677

TABLE 1. Colon injection site clearance and lymph node uptake of Lymphoseek and filtered [99mTc]sulfur colloid at 3 hours Lymph nodes Sentinela Study

Radiopharmaceutical

Clearance half-life (Tc) (h)

1

Lymphoseek

2.06 ⫾ .24

2 3

Lymphoseek Lymphoseek

1.78 ⫾ .09 2.30 ⫾ .14

4 5 6

Lymphoseek fTcSC fTcSC

4.09 ⫾ .08 12.21 ⫾ .28 19.34 ⫾ .31

7 8

fTcSC fTcSC

12.33 ⫾ .12 16.04 ⫾ .21

Lymphoseek fTcSC

2.56 ⫾ 1.04 14.98 ⫾ 3.41 .030

Mean ⫾ SD P value

Probe count rate (cps)

Blue ⫹/⫺

1,400 600 10,800 ⫹ 26,400 ⫹ 1,466 ⫹ 1,400 ⫹ 1,000 ⫹ 1,500 ⫹ 830 ⫹ 200 ⫹ 3,600 ⫹ 2,000 ⫹ 7,011 ⫾ 10,248 1,521 ⫾ 1,187 .57

Distalb % Injected dose

Probe count rate (cps)

.52 .34 .83 4.77 .99 .46 .54 .54 .45 .15 1.30 .80 1.32 ⫾ 1.71 .63 ⫾ .39 .67

2

Blue ⫹/⫺

% Injected dose .003

13 12

⫺ ⫺

.008 .002

12 13 15

⫺ ⫺ ⫺

.005 .001 .015

15 16

⫺ ⫺

.001 .005

9.8 ⫾ 1.3 14.8 ⫾ 1.2 .040

.005 ⫾ .003 .006 ⫾ .007 .77

cps, counts per second; fTcSC, filtered [99mTc]sulfur colloid. a Probe count rate ⬎10 times background. b Distal to the sentinel lymph node in the precolonic chain.

dance rate of 100% was calculated for SLN uptake with blue dye and radioactivity. Of the three Lymphoseek studies that used isosulfan blue 5 minutes before SLN mapping, four lymph nodes were identified as SLNs by Lymphoseek; these same nodes were the only blue nodes identified by isosulfan blue. Lymphoseek uptake as measured by the %ID in colon lymph nodes (n ⫽ 6) ranged from .34% to 4.77% (mean ⫾ SD, 1.32% ⫾ 1.71%). This was not significantly different (P ⫽ .67) from fTcSC uptake by colonic lymph nodes (n ⫽ 6; range, .15%–1.30%; mean ⫾ SD, .63% ⫾ .39%). Second-ech-

FIG. 1. Lymphoseek (diamonds) exhibited significantly faster clearance from both tissues than filtered [99mTc]sulfur colloid (triangles). Colon (closed) injection site clearance from studies 4 and 7 and gastric (open) injection site clearance are shown from studies 1 and 6 over 3 hours.

elon or distal node uptake was significantly less than SLN uptake for Lymphoseek from colon (P ⫽ .011) and gastric (P ⫽ .029) injection sites. Table 2 lists the gastric injection site clearance and lymph node uptake of blue dye, Lymphoseek, and fTcSC. The uptake data for gastric SLNs and distal non–SLNs include the ex vivo count rate, the presence of blue dye, and the %ID. The gastric injection site Tc for Lymphoseek (range, 2.08 – 4.63 hours; mean ⫾ SD, 3.83 ⫾ 1.18 hours) was statistically different (P ⫽ .030) from that for fTcSC (range, 8.47–17.09 hours; mean ⫾ SD, 14.52 ⫾ 4.08 hours). The gastric injection site clearances for studies 1 and 6 are presented in Fig. 1; the Tc values were 4.15 ⫾ .71 hours and 15.71 ⫾ .22 hours, respectively. A lymphoscintigram of the Lymphoseek gastric injection site with an SLN at 15 and 180 minutes after injection is shown in Fig. 2. The typical amount of Lymphoseek remaining in the gastric injection site at 3 hours was 40% ID (Fig. 1; study 1), whereas 90% of the fTcSC dose remained in the injection site at 3 hours (Fig. 1; study 6). The gastric SLN count rate after Lymphoseek injection ranged from 280 to 25,960 cps, with a mean of 10,942 cps; fTcSC ranged from 680 to 2,950 cps (excluding study 8), with a mean of 1,562 cps. Similar to the colon, there was no discordance between radioactivity and blue dye for gastric SLNs. Of the three Lymphoseek studies that used isosulfan blue before SLN mapping, three lymph nodes were identified as SLNs; these same nodes were the only blue gastric nodes identified by isosulfan blue. Lymphoseek uptake as measured by Ann Surg Oncol, Vol. 11, No. 7, 2004

678

S. J. ELLNER ET AL.

TABLE 2. Gastric injection site clearance and lymph node uptake of Lymphoseek and filtered [99mTc] sulfur colloid at 3 hours Lymph nodes a

Distalb

Sentinel Study

Radiopharmaceutical

Clearance half-life (Tc) (h)

1 2 3 4 5

Lymphoseek Lymphoseek Lymphoseek Lymphoseek fTcSC

4.15 ⫾ .71 2.08 ⫾ .12 4.63 ⫾ .86 4.46 ⫾ .07 16.81 ⫾ .18

6 7 8 Mean ⫾ SD

fTcSC fTcSC fTcSCc Lymphoseek fTcSC

15.71 ⫾ .22 8.47 ⫾ .28 17.09 ⫾ .30 3.83 ⫾ 1.18 14.52 ⫾ 4.08 .030

P value

Probe count rate (cps)

Blue ⫹/⫺

1,250 25,960 ⫹ 16,280 ⫹ 280 ⫹ 1,220 ⫹ 680 ⫹ 2,950 ⫹ 1,400 ⫹ NAc ⫺ 10,942 ⫾ 12,405 1,546 ⫾ 974 .67

% Injected dose .46 4.62 2.94 .15 .73 .21 6.58 1.90 NAc 2.04 ⫾ 2.12 2.35 ⫾ 2.90 .88

Probe count rate (cps)

Blue ⫹/⫺

% Injected dose

5 18 15 15 18

⫺ ⫺ ⫺ ⫺

.002 .002 .004 .005 .006

23 ⫺ 23 ⫺ NA ⫺ 13.3 ⫾ 5.7 21.3 ⫾ 2.9 .070

.014 .011 NA .003 ⫾ .002 .010 ⫾ .004 .049

cps, counts per second; fTcSC, filtered [99mTc] sulfur colloid; NA, not applicable. a Probe count rate ⬎10 times background. b Distal to the sentinel lymph node in the perigastric chain. c Tumor at gastroesophageal junction.

%ID by gastric lymph nodes (n ⫽ 4) ranged from .15% to 4.62%; the mean ⫾ SD was 2.04% ⫾ 2.12%, which was not statistically different (P ⫽ .88) from that from fTcSC gastric lymph nodes (n ⫽ 4; range, .73%– 6.58%; mean ⫾ SD, 2.35% ⫾ 2.90%). One animal (study 8) injected with fTcSC had no lymph node uptake. The average number of gastric SLNs identified on the basis of radioactivity criteria with Lymphoseek was 1.0 and with fTcSC was 1.25 per animal. Distal node uptake was significantly less than SLN uptake for fTcSC uptake from colon (P ⫽ .010) and gastric (P ⫽ .052) injection sites. DISCUSSION The results of this study support the hypothesis that 3 hours after endoscopic administration, Lymphoseek exhibits rapid clearance from the colon and gastric injection sites. Moreover, during this time, Lymphoseek exhibits high SLN detectability and low migration to additional lymph nodes. These properties, in combination with the rapid SNL uptake demonstrated by our previous study,20 establish a 5- to 180-minute window during which the SNL activity will maintain a signalbackground ratio in excess of 10 and distal nodes will not exhibit activity above background. We adopted the endoscopic procedure proposed by Méndez et al.20 and achieved comparable results for Lymphoseek 3 hours after administration. The former study demonstrated high SLN accumulation, low secondechelon node uptake, and node concordance with blue Ann Surg Oncol, Vol. 11, No. 7, 2004

dye within 5 minutes of administration. Initially, we attempted to use a mixture of isosulfan blue and either Lymphoseek or fTcSC; however, this was changed to a marking agent after we studied the first pig, because the blue dye was no longer visualized in the lymphatic basin at 3 hours after injection. For the remainder of this study, once a hot colon or gastric SLN was identified with either Lymphoseek or fTcSC at the 3-hour postinjection time point, a subserosal injection of blue dye was then performed at the injection site, as described by Trocha et al.23 There was 100% concordance of blue and hot colon and gastric SLNs for all of the animal subjects. In pig 8, gastric SLNs were not detected by blue dye and fTcSC. Pathologic analysis of a biopsy specimen taken from the gastroesophageal junction of this pig revealed a marked inflammatory response. The combined results from the study by Méndez et al.20 and this investigation indicate that Lymphoseek is a promising radiotracer for SLN mapping of gastrointestinal organs because of the early accumulation in the lymphatic basin at 5 minutes and its persistence for 3 hours. Furthermore, the organ site clearance of Lymphoseek was considerably faster, thus reducing the amount of adjacent scatter at 3 hours compared with fTcSC. The relatively low amount of shine-through with Lymphoseek facilitated SLN detection at 3 hours. The pharmacokinetics of Lymphoseek allow greater flexibility compared with isosulfan blue. The blue dye requires immediate attention after administration. An intraoperative injection of Lymphoseek will require far less mobilization, which can disrupt lymph flow.24 Additionally,

GASTROINTESTINAL SENTINEL NODE MAPPING VIA LYMPHOSEEK

FIG. 2. Lymphoscintigram of gastric injection site and sentinel lymph node (arrow) at 15 minutes after injection (A) and at 180 minutes after injection (B).

the wide time window will permit easier coordination with an endoscopic administration and, therefore, greater potential for laparoscopic procedures. Previous clinical studies identified hot gastrointestinal SLNs on the basis of a signal-background ratio ranging from 2 to 10.23,25–27 By following the standard set by Kitagawa et al.,25 we used the most stringent SLN criterion of 10 for the signal-background ratio. In this study, all lymph nodes mapped at 3 hours after injection by Lymphoseek achieved this criterion. It is important to note that these nodes were also identified as SLNs by isosulfan blue. Equally important was that none of the lymph nodes adjacent to the blue SLNs exceeded the background radioactivity. Although this was not stati-

679

cally significant because of wide variability in uptake by both agents, Lymphoseek exhibited a 7-fold increase in the signal-background ratio compared with fTcSC, indicating greater SLN accumulation and retention of Lymphoseek compared with labeled particles. The rapid injection site clearance and long SLN retention demonstrated in this study are consistent with data obtained from our phase I clinical trial of Lymphoseek in breast cancer.19,28 Rapid clearance and persistent SLN retention are due to the molecular properties of Lymphoseek. Unlike fTcSC, Lymphoseek is not a particle; it belongs to a new class of radiotracers called receptorbinding radiopharmaceuticals. These imaging agents bind to their target tissue on the basis of a specific biochemical affinity for a receptor specific to that tissue. The molecular nature of Lymphoseek provides it with rapid entrance into the lymph channels, and receptor specificity to mannose-binding protein permits high retention by the reticuloendothelial cells of SLNs.18 Although only eight pigs were studied, eight Lymphoseek and seven fTcSC injections were sufficient to prove the statistical significance of Lymphoseek injection site clearance and SLN detectability. The goal of this study was to demonstrate rapid injection site clearance and high SLN detectability after Lymphoseek administration to the colon and gastric submucosa. Each injection generated clearance data with six data points and demonstrated a statistically significant difference compared with fTcSC. Ten SLNs were detected at 3 hours from both gastrointestinal injections of Lymphoseek; the mean %ID was 1.6%, and the average probe response was 8583 cps, compared with an average background of 18 cps. The lowest probe count rate was 280 cps, which is 16 times the background count rate. The Neoprobe Model 2200 uses a counting duration that stops at 6 seconds or when 100 counts are collected, which yields a relative SD of .1. Therefore, the absolute error of the background was 1.8 cps. To achieve a confidence level of 99%, a signal must exceed the background by 3 times the absolute error of the background.29 Our lowest signal (280 cps) exceeded background by 145 times the absolute error. Consequently, all of the 10 lymph nodes mapped by Lymphoseek, which were also stained blue, achieved statistical significance for radioactive detection. We propose that a properly designed radioactive tracer will accelerate the development of SLN mapping in gastrointestinal cancer. Early preliminary studies7,30,31 in colorectal cancer reported wide ranges in identification rates and high false-negative rates. Implementation of immunohistochemistry8 and refinements of the technique as it pertains to the colon32 have reduced the falsenegative rates to levels that suggest clinical releAnn Surg Oncol, Vol. 11, No. 7, 2004

680

S. J. ELLNER ET AL.

vance.33,34 Results for a large series by Saha et al.15 confirmed the consensus that a long learning curve was required to reach acceptable false-negative and detection rates. This situation is similar to that with breast35 SLNs before the introduction of mapping with a radiopharmaceutical,36 which, when combined with blue dye, improved detection and reduced the learning curve.37 Later studies demonstrated a reduction in false-negative rates; a multi-institutional example is the University of Louisville Breast Sentinel Lymph Node Study Group, which demonstrated a reduction from 11.2% to 5.8% with the dual technique.38 Although after mastering the technique some surgeons elect to use only a single agent,39 the major contribution of the dual technique was to propel the implementation of the SLN concept. We propose that the combination of blue dye and Lymphoseek will also accelerate the development of SLN mapping of gastrointestinal cancers. A practical dual-agent technique for gastrointestinal SLN mapping will require an intraoperative co-injection with a wide detection time window that starts within minutes of administration. Therefore, the demonstration in this study that Lymphoseek SLN activity persists for at least 3 hours is critical to the application of dual-agent SLN mapping of gastrointestinal cancers. CONCLUSIONS Lymphoseek demonstrated rapid injection site clearance and, at 3 hours after administration, exhibited high SLN detectability and low second-echelon node uptake in a pig model. A previous study demonstrated high SLN detectability and low distal node uptake within 5 minutes. These demonstrations establish a 5- to 180-minute window for effective SLN mapping with Lymphoseek. This wide window will permit co-injection with isosulfan blue, thus providing visual mapping with blue dye and sustained SLN labeling with radioactivity. In breast cancer and melanoma, such a strategy significantly enhanced the success of SLN mapping. We propose that Lymphoseek will provide similar benefits for SLN mapping of gastrointestinal cancers. ACKNOWLEDGMENTS The acknowledgments are available online in the fulltext version at www.annalssurgicaloncology.org. They are not available in the PDF version. REFERENCES 1. Jemal A, Thomas A, Murray T, Thun M. Cancer statistics 2002. CA Cancer J Clin 2002;52:23– 47.

Ann Surg Oncol, Vol. 11, No. 7, 2004

2. Maruyama K, Gunven P, Okbayashi K, Sasako M, Kinoshita T. Lymph node metastases of gastric cancer. Ann Surg 1989;210: 596 – 602. 3. Gervasoni JEJ, Taneja C, Chung MA, Cady B. Biologic and clinical significance of lymphadenectomy. Surg Clin North Am 2000;80:1631–74. 4. Morton DL, Wen D-R, Wong JH, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992;127:392–9. 5. Saha S, Nora D, Wong JA, Weise D. Sentinel lymph node mapping in colorectal cancer—a review. Surg Clin North Am 2000;80: 1811–20. 6. Kitagawa Y, Kubota K, Nobutoshi A, et al. The role of the sentinel lymph node in gastrointestinal cancer. Surg Clin North Am 2000; 80:1799 –1810. 7. Esser S, Reilly T, Riley LB, Eyvazzadeh C, Arcona S. The role of sentinel lymph node mapping in staging of colon and rectal cancer. Dis Colon Rectum 2001;44:850 – 6. 8. Paramo JC, Summerall J, Wilson C, et al. Intraoperative sentinel lymph node mapping in patients with colon cancer. Am J Surg 2001;182:40 –3. 9. Aikou T, Higashi H, Natsugoe S, Hokita S, Baba M, Tako S. Can sentinel node navigation surgery reduce the extent of lymph node dissection in gastric cancer? Ann Surg Oncol 2001;8:90S–93S. 10. Tsioulias GJ, Wood T, Spirit M, Morton DL, Bilchik AJ. A novel lymphatic mapping technique to improve localization and staging of early colon cancer during laparoscopic colectomy. Am Surg 2002;68:561–5. 11. Wood TF, Tsioulias GJ, Morton DL, et al. Focused examination of sentinel lymph upstages early colorectal carcinoma. Am Surg 2000; 66:998 –1003. 12. Wiese D, Saha S, Badin J, et al. Pathologic evaluation of sentinel lymph nodes in colorectal carcinoma. Arch Pathol Lab Med 2000; 124:1759 – 63. 13. Bilchik AJ, Saha S, Wiese D, et al. Molecular staging of early colon cancer on the basis of sentinel node analysis: a multicenter phase II trial. J Clin Oncol 2001;19:1128 –36. 14. Hiratsuka M, Miyashiro I, Ishikawa O, et al. Application of sentinel node biopsy to gastric cancer surgery. Surgery 2001;129:335– 40. 15. Saha S, Bilchik AJ, Wiese D, et al. Ultrastaging of colorectal cancer by sentinel lymph node mapping technique—a multicenter trial. Ann Surg Oncol 2001;8:94S–98S. 16. Wood TF, Hora DT, Morton DL, et al. One hundred consecutive cases of sentinel lymph node mapping in early colorectal carcinoma: detection of missed micrometastases. J Gastrointest Surg 2002;6:322–30. 17. Bilchik AJ, Saha S, Tsioulias GJ, Wood TF, Morton DL. Aberrant drainage and missed micrometastases: the value of lymphatic mapping and focused analysis of sentinel lymph nodes in gastrointestinal neoplasms. Ann Surg Oncol 2001;8:82S– 85S. 18. Vera DR, Wallace AM, Hoh CK, Mattrey RF. A synthetic macromolecule for sentinel node detection: [99mTc]DTPA-mannosyldextran. J Nucl Med 2001;42:951–9. 19. Wallace AM, Hoh CK, Vera DR, Darrah D, Schulteis G. Lymphoseek: a molecular radiopharmaceutical for sentinel node detection. Ann Surg Oncol 2003;10:531– 8. 20. Méndez J, Wallace AM, Hoh CK, Vera DR. Detection of gastric and colonic sentinel nodes via endoscopic administration of Lymphoseek in pigs. J Nucl Med 2003;44:1677– 81. 21. Hoh CK, Wallace AM, Vera DR. Preclinical studies of [99mTc] DPTA-mannosyl-dextran. Nucl Med Biol 2003;30:457– 64. 22. Cohen J. A coefficient of agreement for nominal scales. Educ Prob Pyschol Meas 1960;20:37– 46. 23. Trocha SD, Nora DT, Saha SS, Morton DL, Wiese D, Bilchik AJ. Combination probe and dye-directed lymphatic mapping detects micrometastases in early colorectal cancer. J Gastrointest Surg 2003;7:340 – 6.

GASTROINTESTINAL SENTINEL NODE MAPPING VIA LYMPHOSEEK 24. Kitagawa Y, Kitajima M. Gastrointestinal cancer and sentinel node navigation surgery. J Surg Oncol 2002;79:188 –93. 25. Kitagawa Y, Watanabe M, Hasegawa H, et al. Sentinel node mapping for colorectal cancer using radioactive tracer. Dis Colon Rectum 2002;45:1476 – 80. 26. 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;8:86S– 89S. 27. Hayashi H, Ochiai T, Mori M, et al. Sentinel lymph node mapping for gastric cancer using a dual procedure with dye- and gamma probe-guided techniques. J Am Coll Surg 2003;196:68 –74. 28. Ellner SJ, Hoh CK, Vera DR, Darrah DD, Schulteis G, Wallace AM. Dose-dependent biodistribution of [99mTc]DTPA-mannosyldextran for breast cancer sentinel node mapping. Nucl Med Biol 2003;30:805–10. 29. Sorenson JA, Phelps ME, eds. Physics in Nuclear Medicine. 2nd ed. Orlando, FL: Grune & Stratton, 1987. 30. Cserni G, Vajda K, Tarjan M, Bori R, Svebis M, Baltas B. Nodal staging of colorectal carcinomas from quantitative and qualitative aspects. Can lymphatic mapping help staging? Pathol Oncol Res 1999;5:291– 6. 31. Joosten J, Strobbe LJ, Wauters CA, Pruszczynski M, Woobbes T, Ruers TJ. Intraoperative lymphatic mapping and the sentinel node concept in colorectal carcinoma. Br J Cancer 1999;86:482– 6.

681

32. Mulsow J, Winter DC, O’Keane JC, O’Connell PR. Sentinel lymph node mapping in colorectal cancer. Br J Surg 2003;90:659 – 67. 33. Wood TF, Saha S, Morton DL, et al. Validation of lymphatic mapping in colorectal cancer: in vivo, ex vivo, and laparoscopic techniques. Ann Surg Oncol 2001;8:150 –7. 34. Wood TF, Spirt M, Rangel D, et al. Lymphatic mapping improves staging during laparoscopic colectomy for cancer. Surg Endosc 2001;15:715–9. 35. Guiliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 1994;220:391– 401. 36. Krag DN, Weaver DL, Alex JC, Fairbank JT. Surgical resection and radiolocalization of the sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993;2:335–9. 37. Albertini J, Lyman GH, Cox C, et al. Lymphatic mapping and sentinel node biopsy and lymphatic mapping of patients with breast cancer. JAMA 1996;276:1818 –22. 38. McMasters KM, Tuttle T, Carlson D, et al. Sentinel lymph node biopsy for breast cancer: a suitable alternative to routine axillary dissection in multi-institutional practice when optimal technique is used. J Clin Oncol 2000;18:2560 – 6. 39. Hansen NM, Giuliano AE. Lymphatic mapping in breast cancer. In: Nieweg OE, Essner R, Reintgen DS, Thompson JF, eds. Lymphatic Mapping and Probe Applications in Oncology. New York: Marcel Dekker, 2000:167– 83.

Ann Surg Oncol, Vol. 11, No. 7, 2004