- Email: [email protected]
Evaluation of colloid size for sentinel nodes detection using radioisotope in early gastric cancer
Cancer Letters 200 (2003) 19–24 www.elsevier.com/locate/canlet
Evaluation of colloid size for sentinel nodes detection using radioisotope in early gastric cancer Yoshikazu Uenosono*, Shoji Natsugoe, Hiroshi Higashi, Katsuhiko Ehi, Futoshi Miyazono, Sumiya Ishigami, Shuichi Hokita, Takashi Aikou First Department of Surgery, Kagoshima University School of Medicine, 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan Received 13 March 2003; received in revised form 19 May 2003; accepted 26 May 2003
Abstract The purpose of this study was to investigate the relationship between colloid size and the detection of sentinel nodes (SN) in early gastric cancer. Three size of 99mTechnetium – tin colloids (500, 100 and 50 nm) were preoperatively injected into the submucosa under endoscopic control. Lymph node metastasis and micrometastasis was examined. RI-uptake in the hottest nodes and the total RI-uptake in the hot nodes were highest in the size of 100 nm. At least one lymph node metastasis, including micrometastasis, was included in the hot nodes. RI-labeled colloid size was one of the important factors to detect SN in early gastric cancer. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Sentinel node navigation surgery; Early gastric cancer; Lymph node metastasis; Micrometastasis; Radioisotope
1. Introduction In attempts to improve quality of life of patients with gastrointestinal cancer, less invasive procedures such as endoscopic mucosal resection and laparoscopic surgery have been developed [1,2]. Sentinel node navigation surgery (SNNS) is an example of a less invasive approach for the purpose of reducing lymphadenectomies. SNNS has been already applied to carcinoma of the breast and melanoma [3,4] and is currently under consideration for carcinoma of the gastrointestinal tract [5,6]. Dyes and radioisotopes * Corresponding author. Tel.: þ81-99-275-5361; fax: þ 81-99265-7426. E-mail address: [email protected] (Y. Uenosono).
(RI) of 99mTechnetium (99mTc) are used as tracers for detecting SNs. 99mTc – tin colloid, 99mTc – sulfur colloid, 99mTc –serum albumin and 99mTc –phytate have been used as tracers in labeled colloids [7 – 10]. Since the size of these RI-labeled colloids varies, we suggest that RI uptake in lymph nodes may also vary. Colloid size may be an important factor in determining how certain tracers move from injection site to lymph nodes via lymphatics [11]. However, few reports have described the relationship between colloid size and detection of sentinel lymph nodes. In the fundamental study, we investigated whether the particle size of tin colloid was controllable by changing the conditions under which the colloid was formed. In brief, isotonic sodium chloride and tin solutions were mixed ratios of 1:4, 1:2, 1:1, 2:1 and
0304-3835/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0304-3835(03)00389-6
20
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
4:1. Particle size was measured using a Coulter Model N4SD Sub-micron Particle Analyzer. When isotonic sodium chloride and tin solutions were mixed at ratios of 1:4, 1:2, 1:1, 2:1 and 4:1, mean particle size was approximately 50, 100, 500, 1000 and 1500 nm, respectively. When isotonic sodium chloride and tin solutions were mixed at a ratio of 1:4, particle size was inhomogeneous at 15 min. As the ratio of tin solution decreased, colloid particle size increased. Accordingly, we ascertained that the particle size of tin colloids is controllable by manipulating the conditions under which the colloids form. In the present study, RI-labeled colloids of various sizes were injected into the submucosa of patients with early gastric cancer, followed by lymphatic mapping. We then examined not only overt metastases by routine histological examination, but also micrometastases by immunohistochemistry.
2. Patients and methods 2.1. Patients Between February 1999 and January 2001, 36 patients with early gastric cancer were enrolled in this study. Informed consent was obtained from 27 male and 9 female patients who ranged in age from 37 to 85 years of age (mean, 66.3 years). Sixteen patients underwent distal partial gastrectomy, 7 underwent total gastrectomy, and 13 underwent partial resection of the stomach. The number of patients classified with D0, D1, and D2 lymphadenectomy was 5, 21, and 10, respectively. The number of dissected lymph nodes per patients ranged from 1 to 69 (median, 19.8). According to the Japanese Classification of Gastric Carcinoma [12], the stomach was divided into three regions: U (upper third), M (middle third), and L (lower third). Four tumors occurred in the U region, 26 in the M region, and 6 in the L region. Sixteen patients had mucosal cancer and 20 had submucosal cancer. All lesions were classified histopathologically into two types: 23 lesions were classified as differentiated adenocarcinoma (papillary and tubular) and 13 lesions were classified as undifferentiated adenocarcinoma (poorly differentiated, signet ring cell, mucinous, and miscellaneous).
2.2. RI injection The 99mTc– tin colloid solution was divided into three categories according to the mixture ratio between the technetium and the anhydrous stannous chloride solution (1 mM anhydrous stannous chloride solution, Nihon Medi-physics Co. Ltd, Nishinomiya, Japan) as follows: group A, mixture ratio of 1:1, group B, mixture ratio of 1:2 and group C, mixture ratio of 1:4. Mean colloid size of A, B, and C were 500, 100, and 50 nm, respectively. The number of patients in groups A, B, and C were 12, 13, and 11, respectively. After each solution was inverted five times to agitate gently, it was allowed to reach room temperature for 30 minutes, and injected. One day prior to surgery, approximately 222 MBq (2 ml) of 99mTc– tin colloid was endoscopically injected into the submucosa of the stomach at 4 sites (0.5 ml each) near the tumor using a disposable 23-gauge needle (MAJ-75, Olympus, Japan). During surgery, RI uptake in each lymph node was measured using Navigator GPS (TYCO HEALTHCARE, Ltd, Tokyo, Japan) (window setting above 95 keV, measuring time 1 s). After the resection, the absence of residual radioactivity in the abdomen was confirmed intraoperatively with Navigator GPS. RI uptake in all dissected lymph nodes was measured again after surgery. The maximum value of RI uptake was recorded for each lymph node. When counts for individual lymph nodes were 10 times greater than background levels, and RI-uptake was more than onetenth of the value for hottest node which indicated the highest value of RI-uptake in each case, the lymph node was identified as a hot node. The amount of RIuptake of all hot nodes in each case was defined as a term ‘total RI-uptake’. In this study, hot nodes were defined as SNs. 2.3. Assessment of lymph node metastasis After each lymph node was cut at the plane of largest dimension, it was fixed in 10% formaldehyde and embedded in paraffin. Paraffin sections (3 mm each) were stained with hematoxylin-eosin, and additional sections were used for immunohistochemical analysis. Lymph nodes were stained using AE1/AE3 (20:1 mixture of AE1 to AE3; Boehringer Mannheim, Germany), a monoclonal antibody
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
cocktail which reacts to a broad spectrum of human cytokeratins (CK) [13,14]. All sections were incubated at 60 8C overnight. The tissue sections were then deparaffinized in xylene and rehydrated through a series of graded ethanols. After incubating the slides in citrate buffer solution (pH 6.0) for 6 min in a pressure cooker, the sections were treated with CK monoclonal antibody at a 1:100 dilution. CK reactivity was developed using an alkaline phosphatase technique (APAAP method) [15]. 2.4. Statistical evaluation Statistical analysis was performed by the Mann – Whitney test. A value of p , 0:05 was considered statistically significant.
3. Results The number of SNs per patient ranged from 0 to 7 (median, 2.7 nodes). SNs were identified in 32 of 36 patients (88.9%). When examined according to colloid size, the number of hot nodes ranged from 0 to 4 (mean, 1.8) in group A, from 1 to 6 (mean, 3.8) in group B, and from 1 to 7 (mean, 3.0) in group C. SNs were not detected in 4 of 12 patients (33.3%) in group A, although SNs were identified in all patients from groups B and C. The number of SNs was significantly greater in group B than in group A ðp ¼ 0:013Þ: RI-uptake of the lymph node with highest value in each case ranged from 17 to 111 count per second
Table 1 Number of hot nodes and RI-uptake count according to particle size
21
(cps) (mean, 35.0 cps) in group A, from 47 to 478 cps (mean, 136.4 cps) in group B, and from 23 to 251 cps (mean, 107.3 cps) in group C. The RI-uptake of smaller colloid size (50 and 100 nm) was significantly higher than that of larger colloid size (500 nm) (p ¼ 0:0021 and p ¼ 0:0046; respectively). The mean values for total RI-uptake in groups A, B, and C were 68.9, 264.9, and 223.7 cps, respectively. Significant differences were found between group A and group B, and group A and group C (p ¼ 0:0025 and p ¼ 0:0081; respectively) (Table 1). Lymph node metastasis was found in 3 patients by routine histological examination using hematoxylin and eosin staining. Lymph node metastases were found not only in SNs but also in other nodes. In the 33 patients without histological evidence of nodal involvement, lymph node micrometastases were found in 3 patients by immunohistochemical examination. These micrometastases were only found within the SNs (Table 2). In all 6 patients with histologically confirmed lymph node metastasis, tumors were located in the middle third of the stomach and were macroscopically depressed type with submucosal invasion. Although lymph node micrometastasis were included in SNs, nodal involvement in other regions was found in two patients with lymph node metastasis detected by routine histological examination (cases No. 2, 3) (Table 3). Three lymph node metastases, including two micrometastases were found in group A, one lymph node metastasis in group B and two lymph node metastases, including one micrometastasis in group C.
22
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
Table 2 Lymph node metastasis and micrometastasis in the sentinel and other nodes
In this series, sensitivity and specificity were 100% in each group, because no lymph node metastasis, including micrometastasis was found in four patients in whom SNs were not detected.
4. Discussion Sentinel lymph node navigation surgery has been introduced for breast cancer and melanoma [3,4,16]. The accuracy rate in breast cancer is reported to be in the range of 96– 100% with radio-guided SN detection [17 – 20]. Although SNNS has been considered in patients with carcinoma of the gastrointestinal tract, its clinical applicability remains controversial [6,7]. Certain issues must be addressed in order to develop SNNS in the gastrointestinal tract as a clinically relevant tool. What kind of RI tracer is suitable? What colloid size is easily moved to the lymph nodes and remains in the SNs? How many SNs are proper in
SNNS? In the present study, we focused on colloid size for detecting SNs in early gastric cancer. At present, a 99mTc –tin colloid with 200 – 700 nm in size was generally used to detect SNs [21]. However, few reports have examined a suitable colloid size for detecting SNs. We used a 99mTc –tin colloid in this study because the colloid size can be changed easily by varying the ratio between isotonic sodium chloride and the tin solution. When small size of colloids was used, the number of hot nodes per patient and the accumulated doses of RI increased. As judged from the accumulated RI in the lymph nodes, small colloid particles moved easily from the lymphatics to the lymph nodes. Colloids of 500 nm in size appeared to be hampered in their ability to move into lymph nodes via the lymphatics, and no hot nodes could be detected in 33.3% of cases. On the other hand, SNs could be detected in all cases when smaller colloids (50 and 100 nm in size) were used. We would suggest that larger colloids are not
Table 3 Clinicopathological findings in 6 patients with lymph node metastasis Case
Age/sex
Tumor location
Macroscopic type
Tumor size (cm)
Histologic type
Tumor depth
Site of hot nodes
Metastatic site
1 2 3 4 5 6
50/M 41/F 50/F 77/F 51/M 79/M
M M M M M M
Depressed Depressed Depressed Depressed Depressed Depressed
2.4 5.0 2.7 2.0 6.2 2.2
Undiff. Undiff. Undiff. Undiff. Diff. Diff.
sm sm sm sm sm sm
4d,6 3 3,4d 3 3 3
4d 1,3 3,4d,7 3(CK) 3(CK) 3(CK)
M—Middle third of the stomach, Undiff—undifferentiated carcinoma, Diff—differentiated carcinoma, sm—submucosa.
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
clinically applicable. In addition, the number of SNs varied widely for the smallest colloids, and this may be because particle size was inhomogeneous when isotonic sodium chloride and tin solutions were mixed at a ratio of 1:4. Based on our results, we recommend 100 nm as a suitable colloid size for detecting SNs in gastric cancer. The number of SNs that may be detectable by RI injection in gastric cancer is still unclear. In the present study, we created a lymphatic mapping, including SNs, to try to elucidate this problem. Furthermore, the presence or absence of lymph node micrometastasis is one of the important problems when performing SNNS [22]. The clinical significance of lymph node micrometastasis is controversial in gastric cancer [23 – 25]. At present, metastatic lymph nodes are typically removed, especially in early stage gastric cancer, because these patients can be expected to have a good prognosis. In this series, at least one metastatic lymph node, including micrometastasis was involved in SNs. D2 lymphadenectomy was performed in patients with lymph node metastasis after confirming the metastasis with frozen sections taken during surgery. Recently, Kitagawa et al. reported that SNs by radio-guided detection in T1 and T2 gastric cancer were detected in 138 (95.2%) of 145 patients and SNs were positive in 22 of 24 patients with lymph node metastases [26]. In the present study, SNs were not detected in some patients receiving large colloid injections and lymph node metastasis was not fortunately found in these cases; this highlights the need to select proper colloid size for successful detection of SNs. In conclusion, we demonstrated the relationship between RI colloid size and detection of SNs. Our results support the use of a colloid particle size of approximately 100 nm for detecting SNs in early gastric cancer. It is useful for patients with early gastric cancer to establish the SN concept, because they will benefit by less invasive surgery such as laparoscopic surgery and reduction of lymphadenectomy.
References [1] K. Takeshita, M. Tani, H. Inoue, I. Saeki, S. Hayashi, T. Honda, et al., Endoscopic treatment of early oesophageal or gastric cancer, Gut 40 (1997) 123– 127.
23
[2] M. Ohgami, Y. Otani, K. Kumai, T. Kubota, Y.I. Kim, M. Kitajima, Curative laparoscopic surgery for early gastric cancer: five years experience, World J. Surg. 23 (1999) 187 –192. [3] D.L. Morton, D.R. Wen, J.H. Wong, J.S. Economou, L.A. Cagle, F.K. Storm, et al., Technical details of intraoperative lymphatic mapping for early stage melanoma, Arch. Surg. 127 (1992) 392 –399. [4] D.N. Krag, D.L. Weaver, J.C. Alex, J.T. Fairbank, Surgical resection and radiolocalization of the sentinel lymph node in breast cancer using a gamma probe, Surg. Oncol. 2 (1993) 335 –339. [5] Y. Kitagawa, M. Ohgami, H. Fujii, M. Mukai, T. Kubota, N. Ando, et al., Laparoscopic detection of sentinel lymph nodes in gastrointestinal cancer: a novel and minimally invasive approach, Ann. Surg. Oncol. 8 (9) (2001) 86–89. [6] T. Aikou, H. Higashi, S. Natsugoe, S. Hokita, M. Baba, S. Takao, Can sentinel node navigation surgery reduce the extent of lymph node dissection in gastric cancer?, Ann. Surg. Oncol. 8 (9) (2001) 90– 93. [7] A.J. Wilhelm, G.S. Mijnhout, E.J. Franssen, Radiopharmaceuticals in sentinel lymph-node detection—an overview, Eur. J. Nucl. Med. 26 (4) (1999) 36–42. [8] C. Tsopelas, Particle size analysis of 99mTc-labeled and unlabeled antimony trisulfide and rhenium sulfide colloids intended for lymphoscintigraphic application, J. Nucl. Med. 42 (2001) 460–466. [9] M.A. Davis, A.G. Jones, H. Trindade, A rapid and accurate method for sizing radiocolloids, J. Nucl. Med. 15 (1974) 923 –928. [10] A. Warbick, G.N. Ege, R.M. Henkelman, G. Maier, D.M. Lyster, An evaluation of radiocolloid sizing techniques, J. Nucl. Med. 18 (1977) 827 –834. [11] L. Bergqvist, S.E. Strand, B.R. Persson, Particle sizing and biokinetics of intestitial lymphoscintigraphic agents, Semin. Nucl. Med. 13 (1983) 9–19. [12] Japanese Gastric Cancer Association, Japanese classification of gastric carcinoma 2nd ed, Gastric Cancer 1 (1998) 10–24. [13] M.J. Woodcock, R. Eichner, W.G. Nelson, T.T. Sun, Immunolocalization of keratin polypeptide in human epidermis using monoclonal antibodies, J. Cell Biol. 95 (1982) 580 –588. [14] M.B. Listrom, L.W. Dalton, Comparison of keratin monoclonal antibodies MAK-6, AE1:AE3, and CAM-5.2, Am. J. Clin. Pathol. 88 (1987) 297–301. [15] J.L. Cordell, B. Falini, W.N. Erber, A.K. Ghosh, Z. Abdulaziz, S. MacDonald, et al., Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes), J. Histochem. Cytochem. 32 (1984) 219 –229. [16] A.E. Giuliano, D.M. Kirgan, J.M. Guenther, D.L. Morton, Lymphatic mapping and sentinel lymphadenectomy for breast cancer, Ann. Surg. 220 (1994) 391–398. [17] G. Canavese, M. Gipponi, A. Catturich, C. Vecchio, D. Tomei, G. Nicolo, et al., Technical issues and pathologic implications
24
[18]
[19]
[20]
[21]
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24 of sentinel lymph node biopsy in early-stage breast cancer patients, J. Surg. Oncol. 77 (2001) 81 –87. M.S. Sabel, P. Zhang, J.M. Barnwell, J.S. Winston, T.C. Hurd, S.B. Edge, Accuracy of Sentinel node biopsy in predicting nodal status in patients with breast carcinoma, J. Surg. Oncol. 77 (2001) 243– 246. K. Maruyama, M. Sasako, T. Kinoshita, T. Sano, H. Katai, Can sentinel node biopsy indicate rational extent of lymphadenectomy in gastric cancer surgery? Fundamental and new information on lymph-node dissection, Langenbecks Arch. Surg. 384 (1999) 149–157. S. Imoto, H. Fukukita, K. Murakami, H. Ikeda, N. Moriyama, Pilot study on sentinel node biopsy in breast cancer, J. Surg. Oncol. 73 (2000) 130 –133. Y. Kitagawa, M. Ohgami, H. Fujii, M. Mukai, N. Ando, T. Kubota, et al., Sentinel node mapping in gastrointestinal cancer: Radio-guided method and dye method, Gastroenterol. Surg. 23 (2000) 1611–1615.
[22] J.R. Siewert, R. Kestlmeier, R. Busch, K. Bottcher, J.D. Roder, J. Muller, C. Fellbaum, H. Hofler, Benefits of D2 lymph node dissection for patients with gastric cancer and pN0 and pN1 lymph node metastases, Br. J. Surg. 83 (1996) 1144–1147. [23] K. Ishida, T. Katsuyama, A. Sugiyama, S. Kawasaki, Immunohistochemical evaluation of lymph node micrometastases from gastric carcinomas, Cancer 79 (1997) 1069–1076. [24] T. Fukagawa, M. Sasako, G.B. Mann, T. Sano, H. Katai, K. Maruyama, Y. Nakanishi, T. Shimoda, Immunohistochemically detected micrometastases of the lymph nodes in patients with gastric carcinoma, Cancer 92 (2001) 753–760. [25] S. Natsugoe, T. Aikou, Y. Higashi, H. Sakita, M. Matsumoto, T. Nakajo, et al., The concept of lymph node micrometastasis for sentinel node navigation surgery, J. Clin. Surg. 55 (2000) 317 –321. [26] Y. Kitagawa, H. Fujii, M. Mukai, T. Kubota, Y. Otani, M. Kitajima, Radio-guided sentinel node detection for gastric cancer, Br. J. Surg. 89 (2002) 604–608.
Evaluation of colloid size for sentinel nodes detection using radioisotope in early gastric cancer Yoshikazu Uenosono*, Shoji Natsugoe, Hiroshi Higashi, Katsuhiko Ehi, Futoshi Miyazono, Sumiya Ishigami, Shuichi Hokita, Takashi Aikou First Department of Surgery, Kagoshima University School of Medicine, 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan Received 13 March 2003; received in revised form 19 May 2003; accepted 26 May 2003
Abstract The purpose of this study was to investigate the relationship between colloid size and the detection of sentinel nodes (SN) in early gastric cancer. Three size of 99mTechnetium – tin colloids (500, 100 and 50 nm) were preoperatively injected into the submucosa under endoscopic control. Lymph node metastasis and micrometastasis was examined. RI-uptake in the hottest nodes and the total RI-uptake in the hot nodes were highest in the size of 100 nm. At least one lymph node metastasis, including micrometastasis, was included in the hot nodes. RI-labeled colloid size was one of the important factors to detect SN in early gastric cancer. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Sentinel node navigation surgery; Early gastric cancer; Lymph node metastasis; Micrometastasis; Radioisotope
1. Introduction In attempts to improve quality of life of patients with gastrointestinal cancer, less invasive procedures such as endoscopic mucosal resection and laparoscopic surgery have been developed [1,2]. Sentinel node navigation surgery (SNNS) is an example of a less invasive approach for the purpose of reducing lymphadenectomies. SNNS has been already applied to carcinoma of the breast and melanoma [3,4] and is currently under consideration for carcinoma of the gastrointestinal tract [5,6]. Dyes and radioisotopes * Corresponding author. Tel.: þ81-99-275-5361; fax: þ 81-99265-7426. E-mail address: [email protected] (Y. Uenosono).
(RI) of 99mTechnetium (99mTc) are used as tracers for detecting SNs. 99mTc – tin colloid, 99mTc – sulfur colloid, 99mTc –serum albumin and 99mTc –phytate have been used as tracers in labeled colloids [7 – 10]. Since the size of these RI-labeled colloids varies, we suggest that RI uptake in lymph nodes may also vary. Colloid size may be an important factor in determining how certain tracers move from injection site to lymph nodes via lymphatics [11]. However, few reports have described the relationship between colloid size and detection of sentinel lymph nodes. In the fundamental study, we investigated whether the particle size of tin colloid was controllable by changing the conditions under which the colloid was formed. In brief, isotonic sodium chloride and tin solutions were mixed ratios of 1:4, 1:2, 1:1, 2:1 and
0304-3835/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0304-3835(03)00389-6
20
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
4:1. Particle size was measured using a Coulter Model N4SD Sub-micron Particle Analyzer. When isotonic sodium chloride and tin solutions were mixed at ratios of 1:4, 1:2, 1:1, 2:1 and 4:1, mean particle size was approximately 50, 100, 500, 1000 and 1500 nm, respectively. When isotonic sodium chloride and tin solutions were mixed at a ratio of 1:4, particle size was inhomogeneous at 15 min. As the ratio of tin solution decreased, colloid particle size increased. Accordingly, we ascertained that the particle size of tin colloids is controllable by manipulating the conditions under which the colloids form. In the present study, RI-labeled colloids of various sizes were injected into the submucosa of patients with early gastric cancer, followed by lymphatic mapping. We then examined not only overt metastases by routine histological examination, but also micrometastases by immunohistochemistry.
2. Patients and methods 2.1. Patients Between February 1999 and January 2001, 36 patients with early gastric cancer were enrolled in this study. Informed consent was obtained from 27 male and 9 female patients who ranged in age from 37 to 85 years of age (mean, 66.3 years). Sixteen patients underwent distal partial gastrectomy, 7 underwent total gastrectomy, and 13 underwent partial resection of the stomach. The number of patients classified with D0, D1, and D2 lymphadenectomy was 5, 21, and 10, respectively. The number of dissected lymph nodes per patients ranged from 1 to 69 (median, 19.8). According to the Japanese Classification of Gastric Carcinoma [12], the stomach was divided into three regions: U (upper third), M (middle third), and L (lower third). Four tumors occurred in the U region, 26 in the M region, and 6 in the L region. Sixteen patients had mucosal cancer and 20 had submucosal cancer. All lesions were classified histopathologically into two types: 23 lesions were classified as differentiated adenocarcinoma (papillary and tubular) and 13 lesions were classified as undifferentiated adenocarcinoma (poorly differentiated, signet ring cell, mucinous, and miscellaneous).
2.2. RI injection The 99mTc– tin colloid solution was divided into three categories according to the mixture ratio between the technetium and the anhydrous stannous chloride solution (1 mM anhydrous stannous chloride solution, Nihon Medi-physics Co. Ltd, Nishinomiya, Japan) as follows: group A, mixture ratio of 1:1, group B, mixture ratio of 1:2 and group C, mixture ratio of 1:4. Mean colloid size of A, B, and C were 500, 100, and 50 nm, respectively. The number of patients in groups A, B, and C were 12, 13, and 11, respectively. After each solution was inverted five times to agitate gently, it was allowed to reach room temperature for 30 minutes, and injected. One day prior to surgery, approximately 222 MBq (2 ml) of 99mTc– tin colloid was endoscopically injected into the submucosa of the stomach at 4 sites (0.5 ml each) near the tumor using a disposable 23-gauge needle (MAJ-75, Olympus, Japan). During surgery, RI uptake in each lymph node was measured using Navigator GPS (TYCO HEALTHCARE, Ltd, Tokyo, Japan) (window setting above 95 keV, measuring time 1 s). After the resection, the absence of residual radioactivity in the abdomen was confirmed intraoperatively with Navigator GPS. RI uptake in all dissected lymph nodes was measured again after surgery. The maximum value of RI uptake was recorded for each lymph node. When counts for individual lymph nodes were 10 times greater than background levels, and RI-uptake was more than onetenth of the value for hottest node which indicated the highest value of RI-uptake in each case, the lymph node was identified as a hot node. The amount of RIuptake of all hot nodes in each case was defined as a term ‘total RI-uptake’. In this study, hot nodes were defined as SNs. 2.3. Assessment of lymph node metastasis After each lymph node was cut at the plane of largest dimension, it was fixed in 10% formaldehyde and embedded in paraffin. Paraffin sections (3 mm each) were stained with hematoxylin-eosin, and additional sections were used for immunohistochemical analysis. Lymph nodes were stained using AE1/AE3 (20:1 mixture of AE1 to AE3; Boehringer Mannheim, Germany), a monoclonal antibody
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
cocktail which reacts to a broad spectrum of human cytokeratins (CK) [13,14]. All sections were incubated at 60 8C overnight. The tissue sections were then deparaffinized in xylene and rehydrated through a series of graded ethanols. After incubating the slides in citrate buffer solution (pH 6.0) for 6 min in a pressure cooker, the sections were treated with CK monoclonal antibody at a 1:100 dilution. CK reactivity was developed using an alkaline phosphatase technique (APAAP method) [15]. 2.4. Statistical evaluation Statistical analysis was performed by the Mann – Whitney test. A value of p , 0:05 was considered statistically significant.
3. Results The number of SNs per patient ranged from 0 to 7 (median, 2.7 nodes). SNs were identified in 32 of 36 patients (88.9%). When examined according to colloid size, the number of hot nodes ranged from 0 to 4 (mean, 1.8) in group A, from 1 to 6 (mean, 3.8) in group B, and from 1 to 7 (mean, 3.0) in group C. SNs were not detected in 4 of 12 patients (33.3%) in group A, although SNs were identified in all patients from groups B and C. The number of SNs was significantly greater in group B than in group A ðp ¼ 0:013Þ: RI-uptake of the lymph node with highest value in each case ranged from 17 to 111 count per second
Table 1 Number of hot nodes and RI-uptake count according to particle size
21
(cps) (mean, 35.0 cps) in group A, from 47 to 478 cps (mean, 136.4 cps) in group B, and from 23 to 251 cps (mean, 107.3 cps) in group C. The RI-uptake of smaller colloid size (50 and 100 nm) was significantly higher than that of larger colloid size (500 nm) (p ¼ 0:0021 and p ¼ 0:0046; respectively). The mean values for total RI-uptake in groups A, B, and C were 68.9, 264.9, and 223.7 cps, respectively. Significant differences were found between group A and group B, and group A and group C (p ¼ 0:0025 and p ¼ 0:0081; respectively) (Table 1). Lymph node metastasis was found in 3 patients by routine histological examination using hematoxylin and eosin staining. Lymph node metastases were found not only in SNs but also in other nodes. In the 33 patients without histological evidence of nodal involvement, lymph node micrometastases were found in 3 patients by immunohistochemical examination. These micrometastases were only found within the SNs (Table 2). In all 6 patients with histologically confirmed lymph node metastasis, tumors were located in the middle third of the stomach and were macroscopically depressed type with submucosal invasion. Although lymph node micrometastasis were included in SNs, nodal involvement in other regions was found in two patients with lymph node metastasis detected by routine histological examination (cases No. 2, 3) (Table 3). Three lymph node metastases, including two micrometastases were found in group A, one lymph node metastasis in group B and two lymph node metastases, including one micrometastasis in group C.
22
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
Table 2 Lymph node metastasis and micrometastasis in the sentinel and other nodes
In this series, sensitivity and specificity were 100% in each group, because no lymph node metastasis, including micrometastasis was found in four patients in whom SNs were not detected.
4. Discussion Sentinel lymph node navigation surgery has been introduced for breast cancer and melanoma [3,4,16]. The accuracy rate in breast cancer is reported to be in the range of 96– 100% with radio-guided SN detection [17 – 20]. Although SNNS has been considered in patients with carcinoma of the gastrointestinal tract, its clinical applicability remains controversial [6,7]. Certain issues must be addressed in order to develop SNNS in the gastrointestinal tract as a clinically relevant tool. What kind of RI tracer is suitable? What colloid size is easily moved to the lymph nodes and remains in the SNs? How many SNs are proper in
SNNS? In the present study, we focused on colloid size for detecting SNs in early gastric cancer. At present, a 99mTc –tin colloid with 200 – 700 nm in size was generally used to detect SNs [21]. However, few reports have examined a suitable colloid size for detecting SNs. We used a 99mTc –tin colloid in this study because the colloid size can be changed easily by varying the ratio between isotonic sodium chloride and the tin solution. When small size of colloids was used, the number of hot nodes per patient and the accumulated doses of RI increased. As judged from the accumulated RI in the lymph nodes, small colloid particles moved easily from the lymphatics to the lymph nodes. Colloids of 500 nm in size appeared to be hampered in their ability to move into lymph nodes via the lymphatics, and no hot nodes could be detected in 33.3% of cases. On the other hand, SNs could be detected in all cases when smaller colloids (50 and 100 nm in size) were used. We would suggest that larger colloids are not
Table 3 Clinicopathological findings in 6 patients with lymph node metastasis Case
Age/sex
Tumor location
Macroscopic type
Tumor size (cm)
Histologic type
Tumor depth
Site of hot nodes
Metastatic site
1 2 3 4 5 6
50/M 41/F 50/F 77/F 51/M 79/M
M M M M M M
Depressed Depressed Depressed Depressed Depressed Depressed
2.4 5.0 2.7 2.0 6.2 2.2
Undiff. Undiff. Undiff. Undiff. Diff. Diff.
sm sm sm sm sm sm
4d,6 3 3,4d 3 3 3
4d 1,3 3,4d,7 3(CK) 3(CK) 3(CK)
M—Middle third of the stomach, Undiff—undifferentiated carcinoma, Diff—differentiated carcinoma, sm—submucosa.
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24
clinically applicable. In addition, the number of SNs varied widely for the smallest colloids, and this may be because particle size was inhomogeneous when isotonic sodium chloride and tin solutions were mixed at a ratio of 1:4. Based on our results, we recommend 100 nm as a suitable colloid size for detecting SNs in gastric cancer. The number of SNs that may be detectable by RI injection in gastric cancer is still unclear. In the present study, we created a lymphatic mapping, including SNs, to try to elucidate this problem. Furthermore, the presence or absence of lymph node micrometastasis is one of the important problems when performing SNNS [22]. The clinical significance of lymph node micrometastasis is controversial in gastric cancer [23 – 25]. At present, metastatic lymph nodes are typically removed, especially in early stage gastric cancer, because these patients can be expected to have a good prognosis. In this series, at least one metastatic lymph node, including micrometastasis was involved in SNs. D2 lymphadenectomy was performed in patients with lymph node metastasis after confirming the metastasis with frozen sections taken during surgery. Recently, Kitagawa et al. reported that SNs by radio-guided detection in T1 and T2 gastric cancer were detected in 138 (95.2%) of 145 patients and SNs were positive in 22 of 24 patients with lymph node metastases [26]. In the present study, SNs were not detected in some patients receiving large colloid injections and lymph node metastasis was not fortunately found in these cases; this highlights the need to select proper colloid size for successful detection of SNs. In conclusion, we demonstrated the relationship between RI colloid size and detection of SNs. Our results support the use of a colloid particle size of approximately 100 nm for detecting SNs in early gastric cancer. It is useful for patients with early gastric cancer to establish the SN concept, because they will benefit by less invasive surgery such as laparoscopic surgery and reduction of lymphadenectomy.
References [1] K. Takeshita, M. Tani, H. Inoue, I. Saeki, S. Hayashi, T. Honda, et al., Endoscopic treatment of early oesophageal or gastric cancer, Gut 40 (1997) 123– 127.
23
[2] M. Ohgami, Y. Otani, K. Kumai, T. Kubota, Y.I. Kim, M. Kitajima, Curative laparoscopic surgery for early gastric cancer: five years experience, World J. Surg. 23 (1999) 187 –192. [3] D.L. Morton, D.R. Wen, J.H. Wong, J.S. Economou, L.A. Cagle, F.K. Storm, et al., Technical details of intraoperative lymphatic mapping for early stage melanoma, Arch. Surg. 127 (1992) 392 –399. [4] D.N. Krag, D.L. Weaver, J.C. Alex, J.T. Fairbank, Surgical resection and radiolocalization of the sentinel lymph node in breast cancer using a gamma probe, Surg. Oncol. 2 (1993) 335 –339. [5] Y. Kitagawa, M. Ohgami, H. Fujii, M. Mukai, T. Kubota, N. Ando, et al., Laparoscopic detection of sentinel lymph nodes in gastrointestinal cancer: a novel and minimally invasive approach, Ann. Surg. Oncol. 8 (9) (2001) 86–89. [6] T. Aikou, H. Higashi, S. Natsugoe, S. Hokita, M. Baba, S. Takao, Can sentinel node navigation surgery reduce the extent of lymph node dissection in gastric cancer?, Ann. Surg. Oncol. 8 (9) (2001) 90– 93. [7] A.J. Wilhelm, G.S. Mijnhout, E.J. Franssen, Radiopharmaceuticals in sentinel lymph-node detection—an overview, Eur. J. Nucl. Med. 26 (4) (1999) 36–42. [8] C. Tsopelas, Particle size analysis of 99mTc-labeled and unlabeled antimony trisulfide and rhenium sulfide colloids intended for lymphoscintigraphic application, J. Nucl. Med. 42 (2001) 460–466. [9] M.A. Davis, A.G. Jones, H. Trindade, A rapid and accurate method for sizing radiocolloids, J. Nucl. Med. 15 (1974) 923 –928. [10] A. Warbick, G.N. Ege, R.M. Henkelman, G. Maier, D.M. Lyster, An evaluation of radiocolloid sizing techniques, J. Nucl. Med. 18 (1977) 827 –834. [11] L. Bergqvist, S.E. Strand, B.R. Persson, Particle sizing and biokinetics of intestitial lymphoscintigraphic agents, Semin. Nucl. Med. 13 (1983) 9–19. [12] Japanese Gastric Cancer Association, Japanese classification of gastric carcinoma 2nd ed, Gastric Cancer 1 (1998) 10–24. [13] M.J. Woodcock, R. Eichner, W.G. Nelson, T.T. Sun, Immunolocalization of keratin polypeptide in human epidermis using monoclonal antibodies, J. Cell Biol. 95 (1982) 580 –588. [14] M.B. Listrom, L.W. Dalton, Comparison of keratin monoclonal antibodies MAK-6, AE1:AE3, and CAM-5.2, Am. J. Clin. Pathol. 88 (1987) 297–301. [15] J.L. Cordell, B. Falini, W.N. Erber, A.K. Ghosh, Z. Abdulaziz, S. MacDonald, et al., Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes), J. Histochem. Cytochem. 32 (1984) 219 –229. [16] A.E. Giuliano, D.M. Kirgan, J.M. Guenther, D.L. Morton, Lymphatic mapping and sentinel lymphadenectomy for breast cancer, Ann. Surg. 220 (1994) 391–398. [17] G. Canavese, M. Gipponi, A. Catturich, C. Vecchio, D. Tomei, G. Nicolo, et al., Technical issues and pathologic implications
24
[18]
[19]
[20]
[21]
Y. Uenosono et al. / Cancer Letters 200 (2003) 19–24 of sentinel lymph node biopsy in early-stage breast cancer patients, J. Surg. Oncol. 77 (2001) 81 –87. M.S. Sabel, P. Zhang, J.M. Barnwell, J.S. Winston, T.C. Hurd, S.B. Edge, Accuracy of Sentinel node biopsy in predicting nodal status in patients with breast carcinoma, J. Surg. Oncol. 77 (2001) 243– 246. K. Maruyama, M. Sasako, T. Kinoshita, T. Sano, H. Katai, Can sentinel node biopsy indicate rational extent of lymphadenectomy in gastric cancer surgery? Fundamental and new information on lymph-node dissection, Langenbecks Arch. Surg. 384 (1999) 149–157. S. Imoto, H. Fukukita, K. Murakami, H. Ikeda, N. Moriyama, Pilot study on sentinel node biopsy in breast cancer, J. Surg. Oncol. 73 (2000) 130 –133. Y. Kitagawa, M. Ohgami, H. Fujii, M. Mukai, N. Ando, T. Kubota, et al., Sentinel node mapping in gastrointestinal cancer: Radio-guided method and dye method, Gastroenterol. Surg. 23 (2000) 1611–1615.
[22] J.R. Siewert, R. Kestlmeier, R. Busch, K. Bottcher, J.D. Roder, J. Muller, C. Fellbaum, H. Hofler, Benefits of D2 lymph node dissection for patients with gastric cancer and pN0 and pN1 lymph node metastases, Br. J. Surg. 83 (1996) 1144–1147. [23] K. Ishida, T. Katsuyama, A. Sugiyama, S. Kawasaki, Immunohistochemical evaluation of lymph node micrometastases from gastric carcinomas, Cancer 79 (1997) 1069–1076. [24] T. Fukagawa, M. Sasako, G.B. Mann, T. Sano, H. Katai, K. Maruyama, Y. Nakanishi, T. Shimoda, Immunohistochemically detected micrometastases of the lymph nodes in patients with gastric carcinoma, Cancer 92 (2001) 753–760. [25] S. Natsugoe, T. Aikou, Y. Higashi, H. Sakita, M. Matsumoto, T. Nakajo, et al., The concept of lymph node micrometastasis for sentinel node navigation surgery, J. Clin. Surg. 55 (2000) 317 –321. [26] Y. Kitagawa, H. Fujii, M. Mukai, T. Kubota, Y. Otani, M. Kitajima, Radio-guided sentinel node detection for gastric cancer, Br. J. Surg. 89 (2002) 604–608.