Locoregional Dissemination and Extended Lymphadenectomy in Pancreatic Cancer

Transcription

1 Review Dig Surg 1999;16: Locoregional Dissemination and Extended Lymphadenectomy in Pancreatic Cancer Laureano Fernández-Cruz a Colin Johnson b Christos Dervenis c a Department of Surgery, Hospital Clínic, Barcelona, Spain; b Department of Surgery, Southampton General Hospital, Southampton, UK, and c First Department of Surgery, Agia Olga Hospital Konstantoponlion, Athens, Greece Key Words Pancreatic cancer W Pancreatic resection W Pancreatoduodenectomy W Lymphadenectomy W Extrapancreatic plexus invasion W Extended pancreatectomy W Pancreatic cancer spread Summary Carcinoma of the pancreas is characterized by the high frequency of intrapancreatic (from 75 to 100%) and extrapancreatic neural invasion (from 64 to 69%). Even smallsized tumors (T 1 ) show plexus invasion. Carcinoma of the pancreas is also associated with a high incidence (76%) of lymph node metastasis. The knowledge of local and regional tumor spread is mandatory in the planning of rational surgical treatment with the intention to cure. At present, it does not seem possible to predict the direction of lymph drainage leading to nodal involvement in different anatomical areas. However, the anterior and posterior pancreaticoduodenal areas are generally involved at first and nodes farther away from the primary tumor mostly show metastases only after involvement of the nearer nodes. We believe, radical pancreatoduodenectomy should be based on three aspects: wide lymph node dissection; radical retroperitoneal dissection, and pancreatectomy with an extirpation line left of the coeliac axis for tumors of the head and left pancreatectomy for tumors of the body and tail of the pancreas. Conclusions: Cure or long-term palliation of pancreatic cancer is generally possible only after complete erradication of the primary tumor, including its local and regional extensions. Despite the recent advances in oncological therapies, surgery remains the only possibility to achieve long-termsurvival for patients with pancreatic cancer. Current surgical procedures for resection of pancreatic head tumors include the classical Whipple operation, pylorus-preserving pancreatoduodenectomy and total pancreatectomy. When the tumor is in the body or tail of the pancreas, the choice is distal pancreatectomy. According to the radicality of the procedure the alternatives are the standard, radical or extended radical operation. These terms have recently been clarified by a group of expert pancreatologists at the Castelfranco Veneto Meeting [1]. At present, long-term survival after pancreatectomy is disappointing due to the high rate of local recurrence after resection. Lymph node involvement, spread of the cancer cells along the lymphatic vessels, peripancreatic nerves and connective tissues, as well as invasion of adjacent structures such as major vessels, seem to be the main causes for local recurrence after pancreatic resection. These facts gave rise to doubts on the oncological principles regarding standard pancreaticoduodenectomy, the most widely performed operation around the world. Based on the local behavior of pancreatic carcinoma, ABC Fax karger@karger.ch S. Karger AG, Basel /99/ $17.50/0 Accessible online at: Dr. Laureano Fernández-Cruz Department of Surgery Hospital Clínic Villarroel 170 E Barcelona (Spain)

2 Table 1. Pancreatic head carcinoma: intra- and extra-pancreatic neural invasion Intra reference % Extra reference % Nagakawa et al. [14, 15] 97 Nagakawa et al. [14, 15] 64 Fortner et al. [20] 82 Matsuda and Nimura [16] 64 Nakao et al. [17] 90 Nakao et al. [17] 69 Nagai et al. [18] 75 Fortner [2] and lately other Japanese surgeons [3 7] advocated a more extended operation including wide dissection of lymph nodes including soft tissue, plexus nerve resection, and vascular removal and reconstruction. The clinical results in some individual series have suggested that more extensive resection improved survival compared with the standard Whipple resection [3, 7, 8], but these results have not been reproducible in other surgical centers [9, 10]. Therefore, the debate still continues today on which is the best oncologically sound operation in patients with pancreatic cancer awaiting more efficient adjuvant therapy [11]. The operation of choice should control the local extension of the tumor to prevent recurrence and should be associated with an acceptable morbidity [11]. Knowledge of the local and regional spread via neural invasion and lymph node metastasis is mandatory in the planning of rational surgical treatment with the intention to cure. Neural and Soft Tissue Invasion: Vascular Involvement Anatomical studies of the extrahepatic plexus have been done by Yoshioka and Wakabayashi [12] who described the division into two main portions: the first is the plexus pancreaticus capitalis which extends from the right celiac ganglion to the upper medial margin of the uncinate process of the pancreas, and the second is the plexus pancreaticus capitalis which extends from the superior mesenteric artery to the medial margin of the uncinate process. Other extrapancreatic plexus have been described around the superior mesenteric artery and the hepatoduodenal ligament. In specimens from patients with pancreatic cancer, examination of pancreatic tumor spread revealed a high incidence of perineural invasion regardless of the size of the tumor or the histological type (fig. 1). The perineurum of the nerves represents a barrier for cancer cells. Studies by Nagakawa et al. [13] suggest that pancreatic cancer cells use the weak area of the perineurum at the site of blood vessel entries to invade nerve plexus. In several clinical studies extrapancreatic nerve plexus invasion was observed in 64 69% of cancers of the head of the pancreas [14 17] (table 1). Even in patients with small tumors (T 1 ), a frequent involvement of the extrapancreatic plexus has been found [18]. Nakao et al. [19] found histopathological extrapancreatic nerve plexus invasion in 14 of 30 patients with carcinoma of the body and tail of the pancreas. Pancreatic cancer is also characterized by the high frequency of intrapancreatic neural invasion (75 100%) [14, 15, 17, 18, 20]. Therefore to erradicate pancreatic cancer a complete plexus dissection must be performed around the superior mesenteric artery, common hepatic artery and coeliac axis including soft tissue removal. This wide retroperitoneal dissection with skeletonization of the major vascular structures is not generally included in the commonly used standard pancreaticoduodenectomy, and could result in the high incidence (20 40%) of apparently resectable carcinomas of the pancreas which were histologically noncurative as tumor cells were observed in the resection margin (fig. 2). Fortner et al. [21] reported that 11 of 12 patients with primary tumors of 2.5 cm or less in diameter had peripancreatic soft tissue invasion microscopically. Therefore even in small tumors, removal of a wide margin is obviously a necessity for anatomical control of the disease. Tumors adherent to the superior mesenteric-portal vein confluence are a major cause of positive margin resection. In a recent report of 23 patients [22] with en bloc vein resection for tumors in this particular area, microscopically positive margins were found in only 17% of patients, therefore in some patients venous involvement is a function of tumor location rather than an indicator of aggressive tumor biology (fig. 3). However, bilateral narrowing of the superior mesenteric vein or occlusion of the superior mesenteric-portal vein confluence has been associated with poor patient survival compared with unilateral (right lateral) venous abnormalities [6]. 314 Dig Surg 1999;16: Fernández-Cruz/Johnson/Dervenis

3 Fig. 1. Close relation between the pancreatic head tumor and the plexus pancreaticus capitalis and extrapancreatic plexus around the superior mesenteric artery. Fig. 3. Pancreatic tumor adherent to the superior mesenteric-portal vein confluence. In most patients venous involvement is a function of tumor location and en bloc vein resection is recommended. Lymph Node Metastasis Pancreatic head carcinoma Intrapancreatic head carcinoma Extrapancreatic nerve plexus invasion Invasion pancreatic margin Fig. 2. Routes of neural invasion. Lymphatic metastasis of pancreatic cancer was initially studied by Cubilla et al. [23] with the use of Fortner s regional pancreatectomy specimen. They reported that 22 cases of ductal cell carcinoma of the pancreas often metastasized to the superior head (45%), posterior pancreaticoduodenal (45%), superior body (27%) and inferior head (23%) lymph node groups. The lymph node involvement in the periaortic area was not mentioned in this study. Hagihara [24], using a radioisotope study, described two routes of lymphatic drainage: a pathway from the pancreas head to lymph nodes around the celiac axis, and another pathway to the lymph nodes around the superior mesenteric artery. Deki and Sato [25] also observed lymphatics from the pancreas head draining to the area Locoregional Dissemination and Extended Lymphadenectomy in Pancreatic Cancer Dig Surg 1999;16:

4 Fig. 4. Lymph node groups corresponding to lymphatic areas draining from the pancreas. Routes of lymphatic drainage from the head of the pancreas lead to: the anterior pancreaticoduodenal group; the posterior pancreaticoduodenal group; the mesenteric artery group; the coeliac trunk group; the hepatoduodenal ligament group; the aorta group, and the inter aorta-cava group. Routes from the body and tail of the pancreas lead to the coeliac trunk group, the splenic artery group, the superior mesenteric artery group, the aorta group, and inferior border of the pancreas group. Table 2. Lymph node involvement in 3 Japanese studies Area of involvement References Nagakawa et al. [28] Kayahara et al. [27] Nakao et al. [30] Posterior pancreaticoduodenal nodes (13), % Anterior pancreaticoduodenal nodes (17), % Superior mesenteric artery nodes (14), % Paraaortic nodes (16), % Common hepatic artery nodes (8), % Hepatoduodenal ligament nodes (12), % The figures in parentheses are the Japanese nomenclature for the areas. between the aorta and vena cava upward and downward from the right renal vein and through the lymphatics around the superior mesenteric artery. These investigations stressed the importance of the lymph nodes around the superior mesenteric artery and aorta in the lymphatic drainage of pancreatic cancer (fig. 4). Based on several histopathological studies, in 1986 the Japanese Pancreatic Society proposed a precise classification of lymph node involvement in pancreatic cancer [26]. According to this classification several Japanese surgeons have extensively studied the frequency of nodal involvement in different anatomical areas. The overall nodal involvement in pancreatic head carcinoma is 76% [27 31]. The areas most frequently involved in three different Japanese studies are detailed in table 2. These studies suggest that the lymph nodes near the primary tumor are generally involved at first (posterior and anterior pancreati- 316 Dig Surg 1999;16: Fernández-Cruz/Johnson/Dervenis

5 Pancreatic head carcinoma Posterior pancreaticoduodenal nodes Hepatoduodenal ligament nodes Anterior pancreaticoduodenal nodes Hepatoduodenal ligament nodes Body and tail carcinoma Splenic artery nodes Coeliac trunk nodes Superior mesenteric artery nodes Aorta nodes Superior mesenteric artery nodes Inferior pancreatic nodes Paraaortic nodes Fig. 5. Routes of lymphatic node spread. Fig. 6. Routes of lymphatic node spread. coduodenal nodes) and nodes farther away from the primary tumor (superior mesenteric artery nodes, paraaortic nodes, common hepatic artery nodes and hepatoduodenal ligament nodes) mostly show metastasis only after involvement of the near nodes (fig. 5). Based on these histopathological studies, regional lymph node dissection is needed in patients with pancreatic head carcinoma. Even in patients with small tumors (2.5 cm or less), Fortner et al. [21] have found 42% lymph node metastases making wide lymphadenectomy necessary in these potentially curable tumors. In a recent clinicopathologic analysis of 12 patients with carcinoma of the head of the pancreas who survived 5 years after surgery, extrapancreatic and perineural invasion was present in 75 and 83%, respectively [32]. Furthermore, lymph node involvement was noted in 42% of cases [32]. It should be noted that 5-year survival cannot be equated to cure because the majority of survivors died of recurrent or metastatic disease. The lymph node metastatic involvement in carcinoma of the body and tail of the pancreas has been investigated by Nakao et al. [19] in 30 specimens obtained from patients who underwent pancreatic resection. Fourteen of 30 patients had lymph node involvement, and the areas most frequently involved were around the splenic artery (5 of 30 patients), aorta (4 of 30 patients), and coeliac trunk (4 of 30 patients). These results support a wide lymphadenectomy including superior mesenteric and paraaortic node dissection in patients with carcinoma of the body and tail of the pancreas (fig. 6). Further advances in adjuvant therapies, perhaps combining chemotherapy, radiotherapy and immunotherapy, may lead to improvements in disease-free and overall survival. But these improvements would have a major impact on outcome after a rational oncological surgical therapy. At present, long-term survival can be expected only after complete tumor resection (Ro, no residual tumor). Lymphadenectomy in Radical Pancreaticoduodenectomy Because long-term survival is generally possible only in cases of complete erradication of the primary tumor including its local and regional extensions, radical pancreaticoduodenectomy for tumors of the pancreatic head and radical left pancreatectomy for tumors of the body and tail should be attempted. The rationale for these procedures is the following: (1) The high incidence of lymph node metastasis, 75% in pancreatic head carcinoma and 46% in body and tail carcinoma; (2) the high incidence (65%) of extrapancreatic neural invasion, even small tumors (T 1 ) show plexus invasion and extrapancreatic neural invasion is a major cause of tumor presence in the margins (resected specimen), and (3) unilateral superior mesenteric portal vein abnormalities are not always a sign of the aggressiveness of the tumor and en bloc vein resection should be performed. In tumors of the head of the pancreas lymphadenectomy starts from the coeliac axis area following along the common hepatic artery and the right and left branches of the hepatic artery. All these vessels and the portal vein in the hepatoduodenal ligament are skeletonized. In this way the first and second portions of the plexus pancreaticus capitalis and the extrapancreatic plexus around the vessels are resected. The pancreas is transected at the level of Locoregional Dissemination and Extended Lymphadenectomy in Pancreatic Cancer Dig Surg 1999;16:

6 the coeliac axis, and pancreatoduodenectomy is performed including a wide dissection and removal of the anterior and posterior pancreaticoduodenal lymph nodes and the skeletonization of the superior mesenteric artery with perivascular neural and soft tissue removal (fig. 7). Lymphadenectomy is completed with removal of the lymph nodes and soft tissue in the anteolateral aorta until the origin of the inferior mesenteric artery, and also the clearance of nodes in the space between the aorta and vena cava above and below the renal vein. Gerota s fascia on the right kidney is also removed. In pancreatic tumors of the body and tail, the pancreas is transected to the right of the portal vein. At this point, lymphadenectomy starts along the common hepatic artery, following the skeletonization of the coeliac trunk (with removal of the right and left ganglions). The splenic artery is ligated at the origin of the coeliac trunk. The Fig. 7. Wide dissection of the space between the aorta and vena cava, and resection including soft tissue, lymph nodes and the neurovascular bundles along the aorta, superior mesenteric artery, coeliac trunk and hepatoduodenal ligament. Fig. 8. Modes of spread from tumors of the body and tail of the pancreas. Lymphatic spread along the common hepatic artery, coeliac trunk, splenic artery, superior mesenteric artery, aorta and inferior border of the pancreas. Extrapancreatic neural spread along the plexus pancreaticus capitalis and extrapancreatic plexus around the superior mesenteric artery. 318 Dig Surg 1999;16: Fernández-Cruz/Johnson/Dervenis

7 spleen and tail of the pancreas are mobilized to the midline and removed with wide dissection of the inferior border of the pancreas including soft tissue and lymph nodes. At this point, the lymph node group around the anterior surface of the aorta and the nodes along the superior mesenteric artery with the neurovascular bundles are resected. The area of resection of lymph nodes and soft tissue including extrapancreatic plexus is detailed in figure 8. Conclusion Cure or long-term palliation of pancreatic cancer is generally possible only after complete erradication of the primary tumor, including its local and regional extensions. References 1 Pedrazoli S, Beger H, Obertop H, Andren- Sandberg A, Fernández-Cruz L, Henne-Bruns D, Lüttges J, Neoptolemos JP: Conference report: A surgical and pathological based classification of resective treatment: Summary of an international workshop on surgical procedures in pancreatic cancer. Dig Surg 1999;16: Fortner JG: Regional resection of cancer of the pancreas. A new surgical approach. Surgery 1973;73: Ishikawa O, Ohigashi H, Sasaki Y, Kabuto T, Fukuda Y, Furukawa H, Imoaka S, Iwanage T: Practical usefulness of lymphatic and connective tissue clearance for the carcinoma of the pancreas head. Ann Surg 1988;208: Tsuchiya R, Tsunoda T, Yamaguchi T: Operation of choice for resectable carcinoma of the head of the pancreas. Int J Pancreatol 1990;6: Satake K, Nishiwaki H, Yokomatsu H, Kawazoe Y, Kim K, Haku A, Umeyama K, Miyazaki I: Surgical curability and prognosis for standard versus extended resection for T1 carcinoma of the pancreas. Surg Gynecol Obstet 1992; 175: Ishikawa O, Ohigashi H, Imaoka S, Furukawa H, Sasaki Y, Fujita M, Kuroda Ch, Iwanaga T: Preoperative indications for extended pancreatectomy for locally advanced pancreas cancer involving the portal vein. Ann Surg 1992;215: Nakao A, Harada A, Nonami T, Keneko T, Inoue S, Takagi H: Clinical significance of portal invasion by pancreatic head carcinoma. Surgery 1995;117: Henne-Bruns D, Kremer B, Meyer-Pannwitt U, Schroeder US: Partial duodenopancreatectomy with radical lymphadenectomy in patients with pancreatic and periampullary carcinomas: Initial results. Hepatogastoenterology 1993;40: Yeo CJ, Cameron JL: Arguments against radical (extended) resection for adenocarcinoma of the pancreas. Adv Surg 1994;27: Mukaiya M, Hirata K, Satoh T, Kimura M, Yamashiro K, Ura H, Oikawa I, Denno R: Lack of survival benefit of extended lymph node dissection for ductal adenocarcinoma of the head of the pancreas: Retrospective multiinstitutional analysis in Japan. World J Surg 1998;22: Pedrazoli S, Di Carlo V, Dionigi R, Michelassi F, Mosca F, Pederzoli P, Pasquali C, Kloppel G, Dhaena K, and the Lymphoadenectomy Study Group: Standard versus extended lymphadenectomy associated with pancreatoduodenectomy in surgical treatment of adenocarcinoma of the head of the pancreas: A multicenter, prospective, randomized trial. Ann Surg 1998;228: Yoshioka H, Wakabayashi T: Therapeutic neurotomy on head of pancreas for relief of pain due to chronic pancreatitis. Arch Surg 1958;76: Nagakawa T, Kayahara M, Veno K: A clinopathological study of neural invasion in cancer of the pancreatic head. Cancer 1992;69: Nagakawa T, Kayahara M, Ohta T, Ueno K, Konishi Y, Miyazaki I: Patterns of neural and plexus invasion of human pancreatic cancer and experimental cancer. Int J Pancreatol 1991;10: Nagakawa T, Mori K, Nakano T, Kadoya M, Kobayashi H, Akiyama T, Kayahara M, Ohta T, Ueno K, Higashino Y: Perineural invasion of carcinoma of the pancreas and biliary tract. Br J Surg 1993;80: Matsuda M, Nimura Y: Perineural invasion of pancreas head carcinoma. Jp Surg Soc 1983;84: Nakao A, Harada A, Nonami T, Kaneko T, Takagi H: Clinical significance of carcinoma invasion of the extrapancreatic nerve plexus in pancreatic cancer. Pancreas 1996;12: Nagai H, Kuroda A, Morioka Y: Lymphatic and local spread of T1 and T2 pancreatic cancer. A study of autopsy material. Ann Surg 1986;204: Nakao A, Harada T, Nonami T, Kaneko S, Nomoto H, Koyama N, Kanazumi N, Nakashima N, Takagi H: Lymph node metastasis in carcinoma of the body and tail of the pancreas. Br J Surg 1997;84: Fortner JG, Kim DK, Cubilla A, Turnbull A, Pahnke LD, Shils ME: Regional pancreatectomy: En bloc pancreatic, portal vein and lymph node resection. Ann Surg 1977;186: Fortner JC, Klimstra D, Senie RT, Maclean BJ: Tumor size is the primary prognosticator for pancreatic cancer after regional pancreatectomy. Ann Surg 1996;223: Fuhrman GM, Leach SD, Stately CA, Cusak JC, Charnasangavej Ch, Cleary Kr, El-Naggar AK, Fenoglio CJ, Lee JE, Evans DB: Rationale for en bloc vein resection in the treatment of pancreatic adenocarcinoma adherent to the superior mesenteric-portal vein confluence. Ann Surg 1996;223: Cubilla AL, Fortner J, Fitzgerald PJ: Lymph node involvement in carcinoma of the head of the pancreas area. Cancer 1978;41: Hagiwara K: Experimental and clinicopathological studies on lymphatic flow of the pancreas. Igaku Kenkiu 1990;52: Deki H, Sato Y: An anatomical study of the peripancreatic lymphatics. Surg Radiol Anat 1988;10: Japanese Pancreatic Society: General Rules for Surgery and Pathological Studies on Cancer of the Pancreas, ed 3. Tokyo, Kanehara, Kayahara M, Nagakawa T, Kobayashi H, Mori K, Nakano T, Kadoya N, Ohta T, Ueno K, Miyazaki I: Lymphatic flow in carcinoma of the pancreas. Cancer 1992;70: Nagakawa T, Konishi I, Ueno K, Ohta T, Kayahara M: A clinical study on lymphatic flow in carcinoma of the pancreatic head area: Peripancreatic regional lymph node grouping. Hepatogastroenterology 1993;40: Nagakawa T, Kobayashi H, Ueno K, Ohta T, Kayahara M, Mori K, Nakano T, Takeda T, Konishi Y, Miyazaki I: The pattern of lymph node involvement in carcinoma of the head of the pancreas. A histologic study of the surgical findings in patients undergoing extensive nodal dissections. Int J Pancreatol 1993;13: Nakao A, Harada T, Nonami T, Kaneko S, Nomoto H, Koyama N, Kanazumi N, Nakashima N, Takagi H: Lymph node metastasis in carcinoma of the body and tail of the pancreas. Br J Surg 1997;84: Kayahara M, Nagakawa T, Ueno K, Ohta T, Tsukioka Y, Miyazaki Y: Surgical strategy for carcinoma of the head area based on clinicopathologic analysis of nodal involvement and plexus invasion. Surgery 1995;117: Colon K, Klimstra D, Brennan MF: Long-term survival after curative resection for pancreatic ductal adenocarcinoma. Ann Surg 1996;223: Locoregional Dissemination and Extended Lymphadenectomy in Pancreatic Cancer Dig Surg 1999;16: