Hyaluronic acid is a useful tool for intraoperative sentinel node detection in gastric cancer surgery

Surgical technique Hyaluronic acid is a useful tool for intraoperative sentinel node detection in gastric cancer surgery Joji Kitayama, MD, Hironori Ishigami, MD, Makoto Ishikawa, MD, Hiroharu Yamashita, MD, Daisuke Soma, MD, Hideyo Miyato, MD, and Hirokazu Nagawa, MD, Tokyo, Japan

Background. We assessed whether a mixture of hyaluronic acid (HA) and dye can facilitate dye-guided sentinel node (SN) mapping in gastric surgery. Although dye-guided, SN-navigated surgery is clinically applied for the treatment of early gastric cancer, there are still some practical problems. Because dyes are carried out from the SN within 20 to 30 minutes, it is sometimes difficult to detect SNs accurately, especially when they are located in a deep area in obese patients. Methods. Patent blue or ferumoxides, superparamagnetic iron nanocolloids, with or without HA, were injected into the gastrointestinal tract of the pig, and the time course of dye transfer through the lymphatic system of the pig mesentery was assessed. Results. When a mixture of HA and patent blue at a volume ratio of 1:4 was injected into the submucosal layer, the time to stain the SN did not differ from that with patent blue alone; however, HA markedly prolonged the time the blue dye was retained in the SN. Patent blue alone stained the efferent lymphatics of the SN and spread to other lymph nodes within 20 minutes after submucosal injection. At the same time point, in contrast, blue stain was restricted to a part of the SN, and the efferent lymphatics were not stained for 2 hours when patent blue was mixed with HA. When a mixture of HA and ferumoxides was used as the tracer, the ferumoxides were still observed in the mesenteric SN even at 2 days after injection. Iron staining showed that Fe was trapped primarily in cells in the peripheral sinus of the SN, suggesting that the iron nanoparticles were mostly incorporated by phagocytic macrophages in the SN within a few hours. Conclusions. Our data indicate that a mixture with HA prolongs the stay of a dye tracer in the SN and thus enables easy and accurate detection of the SN. HA may be a useful tool to develop a more sophisticated SN mapping technique. (Surgery 2007;141:815-20.) From the Department of Surgical Oncology, University of Tokyo Graduate School of Medicine, Japan

Sentinel node (SN) biopsy has been adopted widely in the surgical treatment of cutaneous melanoma and breast cancer.1,2 Recently, sentinel lymph node mapping has been introduced Supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by a grant from the Ministry of Health, Labour and Welfare of Japan. Accepted for publication January 12, 2007. Reprint requests: Joji Kitayama, MD, Department of Surgical Oncology, University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail: [email protected]. 0039-6060/$ - see front matter © 2007 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2007.01.026

for other solid cancers including gastrointestinal malignancies.3-6 Gastric cancer is a significant health problem throughout the world.7,8 In Japan, because of the high prevalence of gastric cancer, nationwide screening programs to assess the stomach were established a few decades ago, which have enabled the detection of many gastric cancer patients with early stage disease. The incidence of lymph node metastasis in early gastric cancer (EGC) is low, and the prognosis good with a 5-year survival rate more than 90%.9 Although controversial, gastrectomy with D2 lymph node dissection is considered the standard treatment for EGC in Japan and other countries.10,11 The controversy on this issue strongly encourages the introduction of the sentinel node concept to gastric cancer surgery. SURGERY 815

816 Kitayama et al

Although the clinical application of SN biopsy in gastric cancer was considered to be technically difficult initially because of the complicated and multidirectional lymphatic flow, many successive feasibility studies have demonstrated that sentinel lymph node mapping and biopsy is a reliable and clinically applicable method with high diagnostic accuracy.12-18 The optimal procedure for the detection of sentinel nodes, however, remains unclear. Thus far, 2 materials, dyes and radioisotopes (RI), have been used as tracers. In contrast to the RIguided method, dye-guided mapping is technically easy, less expensive, and more widely acceptable; however, dyes are not visible if the SN is embedded in a deep, fatty area of the body. Moreover, because dyes pass easily through the SN and so do not remain in the SN for a long time, mistaking a non–sentinel node for a SN is a possibility. Therefore, dye-guided identification of the SN is sometimes difficult when the SN is located in deep sites in obese patients. In this study, therefore, we focused on hyaluronic acid (HA) to overcome this limitation of the dye-guided SN mapping method. HA is a large molecule composed of 2 carbohydrates with high molecular weight and forms a transparent viscous liquid at room temperature.19 In this study, we hypothesized that mixture with HA may delay the passage of the dye through the SN because of its high viscosity as well as its molecular structure. We examined whether the addition of HA could be advantageous for dye-guided SN mapping. MATERIAL AND METHODS Reagents and animal experiment. A 1900-2500 kDa HA preparation (Suvenyl) was donated by Chugai Pharmaceutical Company (Tokyo, Japan). Patent blue dye with 5% concentration was purchased from Wako Pure Chemical (Tokyo, Japan), and ferumoxides, superparamagnetic iron oxide, were obtained from Eiken (Tokyo, Japan). A Sangen-strain pig (Saitama Experimental Animal Supply, Saitama, Japan) weighing approximately 10 kg was anesthetized with ketamine (60 mg/kg) and underwent laparotomy via a midline incision. Dye with or without HA (0.5 ml) was injected into the submucosal layer of the intestine or stomach with a 27-gauge fine needle, and lymphatic passage of the dye was observed continuously for 3 hours at laparotomy. Dye transfer was evaluated every minute after submucosal injection, and the time when dye staining was initially identified in SN or efferent lymphatics was measured. Then, the abdominal wound was closed, and the pig underwent laparotomy at 1 or 2 days after dye injection. These exper-

Surgery June 2007

iments were performed according to the guidelines of the Animal Care Committee of the University of Tokyo. Histochemical staining of iron. SNs that were stained with ferumoxides were fixed and sectioned at 5-␮m thickness. One set of sections was stained with hematoxylin and eosin and another set was used for histochemical staining for ferric or ferrous iron. Turnbull’s blue staining for ferrous iron was carried out following a protocol described previously.20 In brief, sections were washed with distilled water and immersed in a potassium ferricyanide mixture (containing 1 part 2% potassium ferricyanide [Sigma-Aldrich, St Louis, Mo] and 1 part 2% hydrochloric acid [Sigma]) for 30 minutes, followed by washing with distilled water 3 times for 5 minutes each. Statistical analysis. All statistical calculations were carried out using JUMP 6.0.0 statistical software (SAS Institute, Cary, NC). The mean of values was analyzed with the Aspin-Welch-Satterthwaite t test and differences with a P value less than .05 were considered to be statistically significant. RESULTS Mixture of HA prolongs retention of blue dye in SN. The time course of visualization of the lymphatic system was observed in the pig mesentery. When 5% w/v patent blue was injected into the submucosal area of the intestine or stomach, the dye stained the regional lymphatics immediately after injection and reached the mesenteric nodes (SN) in 4 ⫾ 2 minutes (n ⫽ 8), and immediately stained all the area of the SNs. Within the following 20 minutes, the dye spread out of the SNs into the neighboring nodes. The average time to detect dye staining of the efferent lymphatics was 13 ⫾ 5 minutes (n ⫽ 8). Thus, many nodes were detected as blue nodes at 1 hour after injection in each experiment. The addition of HA to patent blue solution increased the viscosity while decreasing the brightness of the blue color of the dye. When the ratio of HA is more than 33%, the solution is too viscous to be injected through a fine needle. Thus, to consider the practical usefulness, we used a ratio of HA to patent blue of 1:4 (final concentration of patent blue, 10 mg/ml) for the following in vivo experiments. The mesenteric lymphatics were stained blue within a few minutes after submucosal injection, and the staining reached the draining SN within 10 minutes in all experiments. The time required for staining the SN (5 ⫾ 1 minutes, n ⫽ 6) did not differ with patent blue alone (4 ⫾ 2 minutes, n ⫽ 8). Interestingly, however, the dye did not

Kitayama et al 817

Surgery Volume 141, Number 6

Table I. The time required to stain the sentinal node (SN) and efferent lymphatics of the SN after submucosal injection of dye solutuion Dye solution

n

Patent Blue Patent Blue ⫹HA Ferumoxides Ferumoxides⫹HA

8 6 4 4

Time to stain Time to stain efferent SN (min) lymphatics of SN (min) 4⫾2 5⫾1 7⫾4 11 ⫾ 3

13 ⫾ 5 140 ⫾ 22⬍* 22 ⫾ 4 180⬍**

The dye transfer was evaluated at every minute after submucosal injection and the time when the dye staining was initially identified in SN or efferent lymphatics was described. *; The blue dye stain was restricted in the SN and did not stain efferent lymphatics for 3 hours in one experiment. **; The dark brown dye stain was restricted in the SN for 3 hours in all of the 4 experiments and still stayed there even 2 days after injection.

Fig 1. SN staining at 20 minutes after injection of patent blue alone (A) or HA and patent blue mixture (B). A, Injection of 0.5 ml 5% patent blue alone stained the whole SN (arrow) and also showed the spread of blue dye through the efferent lymphatics at the same time point (triangle). B, HA and patent blue were mixed at a ratio of 1:4 and the solution was injected into the submucosal layer of pig intestine. Blue stain was restricted to a part of SN (arrow).

spread so quickly in the SNs. At 20 minutes after injection, the blue staining was still restricted to a part of the SNs as well as afferent lymphatics (Fig 1). The blue part gradually enlarged in the SNs, but dye staining of efferent lymphatics of the SNs could not be observed at least 2 hours after submucosal injection in all experiments (Table I). The efferent lymphatics of SNs was stained clearly with patent blue alone at 13 ⫾ 5 minutes (n ⫽ 8) after dye injection. Thus, the mixture with HA made the blue dye stay in the SNs for a longer duration (P ⬍ .0001). Repeated laparotomy on the following day

revealed that the SNs were still stained faint blue although most of the blue dye had spread to other nodes. These observations indicate that HA reduced the speed of dye transfer in the SNs. Mixture of HA enables ferumoxides to stay in SN for days. We also tested another material, ferumoxides. Ferumoxides are superparamagnetic iron oxide coated with a dextran membrane and form microaggregates with an average particle size of approximately 35 to 50 nm.21 The microparticles are incorporated in the reticuloendothelial system, and this agent has been used as a safe and efficacious contrast agent for the detection of focal liver lesions in magnetic resonance imaging (MRI) diagnosis. Therefore, we considered that ferumoxides might be another useful tracer for SN mapping. As observed with patent blue, a mixed solution of HA with ferumoxides at a ratio of 1:4 also reached the SN within 11 ⫾ 3 minutes (n ⫽ 4) after submucosal injection in the jejunum; this was not different from that of ferumoxides alone (7 ⫾ 4 minutes, n ⫽ 4). Ferumoxides alone stained efferent lymphatics of SNs at 22 ⫾ 4 minutes after submucosal injection; however, when used with HA, the ferumoxides did not stain the efferent lymphatics and stayed in the SN up to 3 hours (P ⬍ .0001). Surprisingly, when injected with HA, the ferumoxides stayed in the nodes and did not spread to other nodes even 2 days after injection (Fig 2). Most of the ferumoxides were phagocytosed by cells in peripheral sinus in SN. Iron staining of the nodes showed that Fe was trapped mostly in cells in the sinusoidal area in the peripheral sinus of lymph nodes (Fig 3). This suggests that the ferumoxides molecules were incorporated actively by phagocytic activity within hours, which hampered the transfer of ferumoxides through the efferent lymphatics.

818 Kitayama et al

Surgery June 2007

Fig 2. SN was stained by a ferumoxides and HA mixture even at 2 days after injection. HA and ferumoxides were mixed at a ratio of 1:4, and 0.5 ml of the solution was injected into the submucosal layer of pig intestine.

The nodes that were mapped with ferumoxides alone without HA also showed some phagocytic cells containing iron molecules, but the number of such cells was much less. This suggests that a certain time period is required for endocytosis of ferumoxides molecules, and that most of the ferumoxides particles passed through the SN when transferred without HA. As observed with patent blue, however, a mixture of HA and ferumoxides was retained in the SN for a longer time, at least for 20 to 30 minutes, which may be sufficient for endocytosis by sinusoidal phagocytes. DISCUSSION SN biopsy based on the sentinel lymph node concept may replace conventional lymph node dissection in the treatment of certain cancers such as cutaneous melanoma and breast cancer. Recent technical advances in endoscopic submucosal dissection (ESD) and laparoscopic surgery have enabled safe and less invasive treatment of early cancer lesions.22,23 Therefore, the detection of SN should be a great advantage for many patients with EGC, because the frequency of nodal metastasis is relatively low in EGC. Many feasibility studies have demonstrated promising results with respect to the detection rate and accuracy of SN in gastric cancer.5,12-18 However, there are still some practical problems in the clinical application of SN biopsy in EGC. The RIguided method requires special equipment such as a gamma probe and collimator to avoid background radioactivity, as well as adherence to troublesome regulations for handling radioactive

Fig 3. Iron staining of SN stained by ferumoxides. A solution containing a mixture of ferumoxides and HA solution was injected as described in the legend of Fig 2, and the SN was sectioned and stained by Turnbull’s blue method. Low magnification showed that iron was mostly located in the peripheral sinus (A), and high magnification revealed massive incorporation of iron in phagocytes in the peripheral sinus (B).

agents, which hampers its widespread clinical use. In contrast, dyes are generally carried out from the SN through efferent lymphatics, and so the staining of SN and primary lymph vessels diminishes with time. Therefore, it is sometimes difficult to accurately detect the SNs, especially when they are located in a deep area in obese patients. In other cases, too many blue nodes including non-SNs can be diagnosed as SNs when it takes a long time to detect SNs. In fact, many clinical studies have recommended that SNs should be detected within 5 to 15 minutes after dye injection.12,13,16 Our in vivo results clearly indicate that a mixture of HA with patent blue significantly prolonged the time that the dye was retained in the SN, although it did not significantly delay staining of the afferent lymphatics. The exact reason why the dye can be

Kitayama et al 819

Surgery Volume 141, Number 6

retained for a long time in SNs is still unclear; however, it can be speculated that the patent blue forms a large molecular complex with HA that is not dissociated unless the HA is degraded into small pieces in SNs. HA is present physiologically in the stroma of various tissues including cancers and the abdominal cavity.24 Moreover, stromal HA is delivered preferentially through the lymphatic system and degraded primarily in the draining lymph nodes.25 Our data on the time course of dye migration are consistent with these facts and support the usefulness of HA for SN detection. In the pig mesentery, if used with HA, staining with patent blue was restricted to mesenteric SNs up to 2 hours, suggesting that surgeons can spend at least 2 hours for intraoperative detection of SNs. This seems to be long enough to search for SNs in any patient. In fact, in 10 patients (excluding 1 case with failure of dye injection), we succeeded in detecting 1 to 7 perigastric SNs within 30 minutes without any difficulty (data not shown). This result, although preliminary, indicates that HA also facilitates accurate intraoperative detection in humans. It also suggests that the gastric wall contains many lymphatics with several SNs, because the blue dye was never drawn out from the SN at this time point. We also evaluated a mixture of HA with ferumoxides. Because ferumoxides are widely used as a negative contrast in MRI, it may be applied for the preoperative detection of sentinel nodes by MRI in the future. In fact, ferumoxides have been used for the detection of SN in lung cancer.26,27 It has been reported that the migration time to the SN was 17 to 47 minutes when injected around a lung tumor. When ferumoxides were injected into the submucosal layer of pig intestine, the lymphatic system also was visualized more quickly, but the time to stain SN was slightly later than with blue dye, possibly due to the larger size and higher specific gravity of the molecule. Interestingly, when injected with HA into the gastrointestinal submucosal layer, the ferumoxides stayed in the SN without spreading to other sites even after 2 days. Histologic study revealed that most of the ferumoxides were incorporated in immune cells located in the peripheral sinus of SN. Similar findings also were observed with patent blue. From these findings, it is possible that endocytosis of these nanomolecules in the reticuloendothelial system prevents the spread of these tracers. This is a big advantage clinically for SN detection, because the tracer can be endoscopically injected a few days before surgery. These findings encouraged us to use ferumoxides for SN mapping in gastric cancer surgery. Unfortunately, however, unlike the case of the pig mesentery, the lymphatic visualization of dark brown ferumoxides

was much less impressive than with blue dye in humans, and thus it was not suitable for clinical use. Our results, however, suggest that a mixture of HA and another nanomaterial with a blue color and heavy weight would be an ideal tracer for SN mapping in the stomach. In humans, mesenteric fat often hampers the easy detection of blue nodes, which is a disadvantage of the dye method of sentinel node mapping in abdominal surgery. The addition of HA, however, increased the viscosity of the dye fluid and prolonged the retention of the tracer in the SN, where a local immunoreaction could be induced. This is favorable for accurate identification of SNs, especially in obese patients. HA is a unique and useful tool for SN detection, and the discovery of suitable tracers to mix with HA could lead to the development of a more sophisticated SN mapping technique in the future. REFERENCES 1. Morton DL, Wen DR, Wong JH, Economou JS, Cagle LA, Storm FK, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992;127:392-9. 2. Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 1994;220:391-8.; discussion 398-401. 3. Kitagawa Y, Fujii H, Mukai M, Kubota T, Ando N, Watanabe M, et al. The role of the sentinel lymph node in gastrointestinal cancer. Surg Clin North Am 2000;80:1799-1809. 4. Saha S, Nora D, Wong JH, Weise D. Sentinel lymph node mapping in colorectal cancer—a review. Surg Clin North Am 2000;80:1811-9. 5. Keshtgar MR, Amin A, Taylor I. Intraoperative lymphatic mapping and the sentinel node concept in colorectal carcinoma. Br J Surg 1999;86:1225-6. 6. Yasuda S, Shimada H, Chino O, Tanaka H, Kenmochi T, Takechi M, et al. Sentinel lymph node detection with Tc99m tin colloids in patients with esophagogastric cancer. Jpn J Clin Oncol 2003;33:68-72. 7. Allum WH, Fielding JW. Gastric cancer: a 25-year review. Br J Surg 1990;77:353. 8. Harrison JD, Fielding JW. Prognostic factors for gastric cancer influencing clinical practice. World J Surg 1995;19: 496-500. 9. Itoh H, Oohata Y, Nakamura K, Nagata T, Mibu R, Nakayama F. Complete ten-year postgastrectomy follow-up of early gastric cancer. Am J Surg 1989;158:14-6. 10. Siewert JR, Kestlmeier R, Busch R, Bottcher K, Roder JD, Muller J, et al. Benefits of D2 lymph node dissection for patients with gastric cancer and pN0 and pN1 lymph node metastases. Br J Surg 1996;83:1144-7. 11. Shimada, Y. JGCA (The Japan Gastric Cancer Association). Gastric cancer treatment guidelines. Jpn J Clin Oncol 2004;34:58. 12. Hiratsuka M, Miyashiro I, Ishikawa O, Furukawa H, Motomura K, Ohigashi H, et al. Application of sentinel node biopsy to gastric cancer surgery. Surgery 2001;129:335-40. 13. Ichikura T, Morita D, Uchida T, Okura E, Majima T, Ogawa T, et al. Sentinel node concept in gastric carcinoma. World J Surg 2002;26:318-22.

820 Kitayama et al

14. Kitagawa Y, Fujii H, Mukai M, Kubota T, Otani Y, Kitajima M. Radio-guided sentinel node detection for gastric cancer. Br J Surg 2002;89604-8. 15. Miwa K, Kinami S, Taniguchi K, Fushida S, Fujimura T, Nonomura A. Mapping sentinel nodes in patients with earlystage gastric carcinoma. Br J Surg 2003;90:178-82. 16. Ryu KW, Lee JH, Kim HS, Kim YW, Choi IJ, Bae JM. Prediction of lymph nodes metastasis by sentinel node biopsy in gastric cancer. Eur J Surg Oncol 2003;29:895-9. 17. Zulfikaroglu B, Koc M, Ozmen MM, Kucuk NO, Ozalp N, Aras G. Intraoperative lymphatic mapping and sentinel lymph node biopsy using radioactive tracer in gastric cancer. Surgery 2005;138:899-904. 18. Park DJ, Lee HJ, Lee HS, Kim WH, Kim HH, Lee KU, et al. Sentinel node biopsy for cT1 and cT2a gastric cancer. Eur J Surg Oncol 2006;32:48-54. 19. Fraser, J. R., Laurent, T. C., and Laurent, U. B. Hyaluronan: its nature, distribution, functions and turnover. J Intern Med 1997;242:27-33. 20. Morris CM, Candy JM, Oakley AE, Bloxham CA, Edwardson JA. Histochemical distribution of non-haem iron in the human brain. Acta Anat (Basel) 1992;144:235-57.

Surgery June 2007

21. Pouliquen D, Le Jeune JJ, Perdrisot R, Ermias A, Jallet P. Iron oxide nanoparticles for use as an MRI contrast agent: pharmacokinetics and metabolism. Magn Reson Imaging 1991;9:275-83. 22. Yahagi N, Fujishiro M, Kakushima N, Odashima S, Nakamura M, Yamamichi N, et al. The current status of endoscopic therapy of early stage gastric cancer. Nippon Naika Gakkai Zasshi 2005;94:56-62. 23. Kitano S, Shiraishi N. Minimally invasive surgery for gastric tumors. Surg Clin North Am 2005;85:151-64. 24. Toole BP. Hyaluronan: from extracellular glue to pericellular cue. Nat Rev Cancer 2004;4:528-39. 25. Weigel PH. Functional characteristics and catalytic mechanisms of the bacterial hyaluronan synthases. IUBMB Life 2002;54:201-11. 26. Nakagawa T, Minamiya Y, Katayose Y, Saito H, Taguchi K, Imano H, et al. A novel method for sentinel lymph node mapping using magnetite in patients with non-small cell lung cancer. J Thorac Cardiovasc Surg 2003;126:563-7. 27. Minamiya Y, Ito M, Katayose Y, Saito H, Imai K, Sato Y, Ogawa J. Intraoperative sentinel lymph node mapping using a new sterilizable magnetometer in patients with nonsmall cell lung cancer. Ann Thorac Surg 2006;81:327-30.