Original Paper Received: June 3, 2014 Accepted after revision: November 30, 2014 Published online: January 21, 2015

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

Factors Influencing Lymph Node Metastasis in Patients with Ampullary Adenocarcinoma Keiichi Okano a Eisuke Asano a Yoshio Kushida b Hideki Kamada c Hirohito Mori c Yasuyuki Suzuki a  

 

 

 

 

 

Departments of a Gastroenterological Surgery, b Pathology, and c Gastroenterology and Neurology, Faculty of Medicine, Kagawa University, Miki-cho, Kagawa, Japan  

 

 

Key Words Ampullary adenocarcinoma · Lymph node metastasis · Predictive factor

Abstract Aim: In cases of ampullary carcinoma, lymph node involvement affects the selection of treatment strategies. This study aimed to identify clinicopathologic features of ampullary carcinoma with lymph node metastases. Methods: The records of 74 consecutive patients with ampullary adenocarcinoma who underwent pancreaticoduodenectomy (PD) with regional lymph node dissection were retrospectively analyzed. Results: Twenty-two patients (30%) with lymph node metastasis had significantly worse survival after resection than those without lymph node metastasis (p = 0.017). Univariate analyses revealed that preoperative biliary drainage; elevated serum carbohydrate antigen 19–9 (≥36 U/ml); moderate-to-poor pathologic grade (G2/3); perineural, vascular, lymphatic, pancreas, and duodenal invasion; and T category were significantly associated with lymph node metastasis. In multivariate analysis, only pathologic grade (G2/3) remained significantly associated with lymph node metastasis (hazard ratio, 6.51; p = 0.035). In sub-classified analysis for T category, lymph node metastasis was found in 5 of 22 cases (22.7%) of T1 tumors. Four of five cases with lymph node metastases had a dominant G2/3 component,

© 2015 S. Karger AG, Basel 0253–4886/15/0316–0459$39.50/0 E-Mail [email protected] www.karger.com/dsu

whereas only 2 of 17 cases without lymph node metastases had a G2/3 component in T1 tumors (p = 0.0036). Conclusions: Pathologic grade (G2/3) was significantly and independently associated with lymph node metastasis and was also a significant predictor in T1 tumor cases. © 2015 S. Karger AG, Basel

Introduction

The incidence of ampullary carcinoma is estimated to be 5–7 cases per 1,000,000 person-years at-risk, accounting for 0.2% of gastrointestinal tract cancers [1–3]. A number of recent reports have documented 5-year survival rates ranging from 34% to 67.7% [2, 4–6]. In a previous report, we described the clinical significance of the two major pathological subtypes and other prognostic factors of this disease [7]. Poor survival was associated with the presence of the following factors: preoperative biliary drainage; serum carbohydrate antigen (CA) 19–9 level >36 U/ml; serum total bilirubin level >1.0 mg/dl; pathological grade of moderate to poor differentiation; nodal metastasis; advanced T stage (UICC T3 and T4); pancreatobiliary subtype; and perineural, vascular, duodenal, and pancreatic invasion. Nodal metastasis is a significant prognostic factor and is also quite important when selecting a treatment for ampullary adenocarcinoma. Keiichi Okano, MD Department of Gastroenterological Surgery Faculty of Medicine, Kagawa University 1750–1, Ikenobe, Miki-cho, Kita-gun, Kagawa 761–0793 (Japan) E-Mail kokano @ kms.ac.jp

Although pancreatoduodenectomy (PD) is the standard surgical strategy for ampullary cancer, it is associated with a high operative morbidity rate, which persists despite recent reductions in operative mortality to less than 5% [8]. Ampullectomy has not only been attempted for benign lesions, but also for early cancers. For early cancers, ampullectomy can provide an alternative to PD, especially in elderly patients with complications or patients with a certain risk for PD. Lymph node metastasis is thought to be a critical factor in selecting treatment strategies for this age group. In this study, we sought to evaluate clinicopathologic factors related to lymph node metastasis to provide a reference for clinicians selecting surgical procedures for ampullary adenocarcinoma. For the same reason, we also investigated the location and frequency of lymph node metastases, and the pathological relationship between lymph node metastasis and T1 tumor.

the grade. Well-differentiated (grade 1) lesions exhibit glandular structures in >95% of the tumor, moderately differentiated (grade 2) adenocarcinomas have glandular structures in 50–90% of the tumor, and poorly differentiated (grade 3) adenocarcinomas have glandular structures in 5–50% of the tumor. Mucinous adenocarcinoma and signet-ring cell carcinoma are considered to be poorly differentiated (grade 3). The pathologic characteristics of each tumor were reviewed by a single pathologist (Y.K.), who was blinded to the clinical findings. Slides stained with hematoxylin and eosin were examined to determine the tumor type (e.g., adenoma, adenocarcinoma), size, grade, depth of invasion, margins, and subtype (intestinal or pancreatobiliary, for adenocarcinomas only) [12]. Slides were also assessed to check the presence of lymphovascular or perineural invasion, as well as pancreas and duodenal invasion. To examine the nodal status, the surgeon removed dissected lymph nodes from the node-bearing adipose tissue. Lymph nodes were fixed in 10% formalin and embedded in paraffin. A single paraffin section (3 microns thick) of the maximum cut surface of each lymph node was stained with hematoxylin and eosin. The presence of metastasis was diagnosed by a single pathologist as a component of routine histologic examination. Eleven regional lymph nodes (median) were dissected and examined per patient (range: 4–38 lymph nodes).

Patients and Methods We retrospectively reviewed the medical records of 74 patients who underwent PD with regional lymph node dissection for ampullary adenocarcinoma at 12 principal hospitals that were members of the Kagawa Gastroenterology Forum in Kagawa prefecture, Japan. All patients underwent PD between January 2000 and December 2012. The Kagawa Gastroenterology Forum is organized by board-certified members of the Japanese Society of Gastroenterology, and it conducts detailed district surveys across the prefecture every year. This study was approved by the review board of Kagawa University. Recorded data included patient age, patient sex, symptoms and signs at presentation, laboratory data, endoscopic studies, and imaging studies. Perioperative data were also collected from medical records, and included surgical procedure, extent of lymph node dissection (according to the Japanese Society of Biliary Surgery [JSBS]) [9], mortality, complications, adjuvant chemotherapy, and readmission information. Resection for lymph nodes on the posterior surface of the pancreas head was defined as D1. Resection of lymph nodes on the superior mesenteric artery, on the anterior surface of the pancreas head, and along the lower common bile duct in addition to D1 was defined as D2. Surgery-related mortality was defined as any postoperative mortality within 30 days of the initial procedure. Complications were defined as a pancreatic fistula (following the International Study Group on Pancreatic Fistula definition) [10]; delayed gastric emptying, intra-abdominal fluid collection; pneumonia; cholangitis; any cardiac, pulmonary, or vascular complication; or any diagnosis resulting in readmission. Cancer stage was assessed according to the 7th edition of the TNM classification system for ampullary carcinoma issued by the American Joint Committee on Cancer [11]. The grading of ampullary adenocarcinoma is identical that used in the large bowel [12]. Ampullary adenocarcinomas are graded predominantly on the basis of the extent of glandular appearances. The percentage of the tumor that shows the formation of gland-like structures can define

460

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

Statistical Analysis All statistical analyses were performed using SAS 8.2 (SAS Institute Inc., Cary, N.C., USA). Patient and tumor characteristics were compared using the Mann-Whitney U test for continuous variables and Fisher’s exact test or the chi-squared test for categorical variables. Kaplan-Meier times-to-death, stratified by patient and tumor characteristics, were calculated and compared using the log-rank test. Because it was often difficult to distinguish cancerrelated deaths from deaths from other causes (and without persistent cancer) in this population, we analyzed the overall survival rates in this study. Risk factors that were significantly associated with death in univariate models (p < 0.05) were included in a multivariate logistic regression model. Throughout this study, values of p < 0.05 were considered statistically significant.

Results

Patient and Tumor Characteristics The clinicopathologic characteristics of the 74 patients who underwent PD with regional lymph node dissection for ampullary adenocarcinoma are presented in table 1. The median age was 74 years (range 48–86 years). The number of elderly patients (aged 70 years or older) was 46 (62%). Sixty patients (81%) presented with at least one symptom. The most common presentation was obstructive jaundice, and preoperative biliary drainage was performed in 43 patients (58%). Regarding surgical therapies, 30 patients (41%) underwent conventional PD, 32 (43%) underwent pylorus-preserving PD, and 12 (16%) underwent subtotal stomach-preserving PD. The majority of Okano/Asano/Kushida/Kamada/Mori/ Suzuki

Table 1. Clinicopathologic background of 74 patients with an ampullary adenocarcinoma who underwent pancreaticoduodenectomy

Age, years, median (range) Gender, n (%) Male Female Primary symptoms, n (%) Asymptomatic Symptomatic Preoperative biliary drainage, n (%) Yes No Surgical procedure, n (%) PD PPPD SSPPD Lymph node dissection (JSBS), n (%) D1 D2 D3 Surgical morbidity, n (%) Pancreas fistula, n (%) ISGPF Grade B/C Surgical mortality, n (%) Overall stage (UICC), n (%) I II III IV Lymph node metastasis, n (%) Positive Negative Pathologic grade, n (%) Well (G1) Moderate to poor (G2/3) Pathologic subtype, n (%) Intestinal Pancreatobiliary

74 (46–86) 35 (47) 39 (53) 14 (19) 60 (81) 43 (58) 31 (42) 30 (41) 32 (43) 12 (16) 15 (20) 58 (78) 1 (1) 35 (47) 12 (16) 1 (1) 38 (51) 29 (38) 6 (8) 1 (1) 22 (30) 52 (70) 56 (76) 18 (24) 35 (56) 27 (44)

JSBS = Japanese Society of Biliary Surgery; UICC = Union for International Cancer Control; ISGPF = International Study Group on Pancreatic Fistula; PD = pancreaticoduodenectomy; PPPD = pylorus preserving pancreaticoduodenectomy; SSPPD = subtotal stomach preserving pancreaticoduodenectomy.

patients (78%) received standard lymphadenectomies (JSBS D2). Surgical mortality and morbidity were 1 and 47%, respectively. All patients completed follow-up, which lasted an average of 22 months (range: 1–152 months).

around 1 year. The protocol drugs were S-1 in 9 patients, tegafur–uracil in 5 patients, gemcitabine (GEM) in 5 patients, the combination of S-1 and GEM in 1 patient, 5-fluorouracil (5-FU) in 1 patient, and the combination of 5-FU and cisplatin (CDDP) in 1 patient. The remaining patients received no adjuvant chemotherapy. No patient underwent adjuvant radiotherapy or received neo-adjuvant therapy. Adjuvant chemotherapy did not show any relationship with prognosis or lymph node metastasis. Factors Associated with Lymph Node Metastasis Among patients with lymph node metastasis, actuarial 3- and 5-year survival rates after resection were 65 and 57%, respectively. Among patients without lymph node metastasis, the corresponding rates were 85 and 85% (fig. 1). Patients with lymph node metastasis had significantly worse survival after resection than did patients without lymph node metastasis (p = 0.017). Univariate analyses revealed that preoperative biliary drainage, elevated the serum CA19–9 level (≥36 U/ml), moderate-topoor pathologic grade (G2/3), perineural invasion, vascular invasion, lymphatic invasion, pancreas invasion, duodenal invasion, and T category (Union for International Cancer Control standard) were significantly associated with lymph node metastasis (table  2). However, in our multivariate analysis, only pathologic grade (G2/3) remained significantly associated with lymph node metastasis (hazard ratio 6.51, 95% CI 1.13–56.6, p = 0.035). In the overall study cohort, pathologic grade (G2/3) correlated with histologically proven lymph node metastases, with a sensitivity of 55%, a specificity of 89%, a positive predictive value (PPV) of 67%, and a negative predictive value (NPV) of 82%. In patients with T1 tumor, pathologic grade (G2/3) was related to histologically proven lymph node metastases, with a sensitivity of 80%, a specificity of 88%, a PPV of 67%, and an NPV of 94%. In the analysis of macroscopic appearance [13], 4 of 10 patients (40%) with ulcerative lesions had lymph node metastases. On the other hand, 12 (29%) of the 42 patients with protruding lesions had lymph node metastases. However, the difference between these groups was not statistically significant.

Adjuvant Treatment Adjuvant chemotherapy after resection depended on the surgeon’s preference. A total of 22 patients (30%) received adjuvant chemotherapy, which was provided for

Locations and Appearance Associated with Lymph Node Metastasis The highest incidence of positive nodes was found on the posterior surface of the pancreas head (14 of 22 patients, 63%). In addition, lymph node metastases were found along the superior mesenteric artery (4 patients, 18%), on

Lymph Node Metastasis of Ampullary Adenocarcinoma

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

461

1.0

Without nodal metastasis

Overall survival

0.8 0.6 With nodal metastasis

0.4 p = 0.0047

0.2

0

Fig. 1. Overall survival of patients who un-

derwent radical resection for ampullary adenocarcinoma, stratified by whether the case involved lymph node metastasis.

Without nodal metastasis 52 With nodal metastasis 22

Recurrence Twelve (16%) patients experienced tumor recurrence. Hepatic recurrence (n = 9, 12%) was observed most frequently, followed by lymph node recurrence (n = 2, 2.7%) and lung recurrence (n = 1, 1.3%). The recurrence ratio Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

24

36

48

60

18 5

11 4

Months after operation

the anterior surface of the pancreas head (2 patients, 9%), around the abdominal aorta (2 patients, 9%), and along the common hepatic artery (1 patient, 5%), respectively. Lymph node metastasis was found in 5 patients (23%) who had T1 tumor (table 3; fig. 2). In contrast, lymph node metastasis was found in none of the patients who had tumor invasion within the mucosa (pTis). Of the 5 patients who had T1 tumor involving the lymph nodes, 4 (80%) had tumors with a dominant component of moderately differentiated adenocarcinoma (G2). In contrast, only 28% of all cases in this study had a dominant component of moderately-to-poorly differentiated adenocarcinoma (G2/3). Among 33 patients with Tis and T1, lymph node metastasis was detected pathologically in 1 of 27 (3%) well differentiated adenocarcinoma (G1) and 4 of 6 (67%) moderately to poorly adenocarcinoma (G2/3) cases. Lymph node metastasis was found in 17 patients (41%) with tumor invasion beyond the sphincter of Oddi (pT2–4). Of these 17 patients, 8 (47%) had a dominant component of moderately to poorly differentiated adenocarcinoma (G2/3). In patients with T1 tumor (n = 22), lymph node metastasis was related to pathologic grade (G2/3) (p = 0.0036) and vascular invasion (p = 0.0006) in univariate analyses.

462

12 39 17

34 12

26 6

was significantly (p = 0.011) higher in patients with lymph node metastasis (36.4%), as compared with its value in patients without lymph node metastasis (10.2%). No significant differences were observed between the patients with and without lymph node metastasis in terms of the sites of recurrence.

Discussion

Various prognostic factors have been proposed for ampullary adenocarcinoma: lymph node metastasis [2, 4, 6, 13–15], depth of invasion [16, 17], tumor size [18, 19], histologic grade [4, 6, 18, 19], resection margin [2], perineural invasion [5, 16], lymphovascular invasion [6], gross appearance [13], operative transfusion [4], and histologic subtype [20–24]. Of these factors, lymph node involvement has been considered a particularly critical factor for the survival of patients with ampullary carcinoma, as well as the selection of treatment strategies for this disease. Kang et al. [25] have reported the effect of the number of metastatic lymph nodes on prognosis based on an analysis of a large database of 1,057 patients with ampullary adenocarcinoma. They suggested a new nodal classification of ampullary adenocarcinoma, as follows: N0 (no metastatic lymph node), N1 (1–2 metastatic lymph nodes), and N2 (>3 metastatic lymph nodes). In a study of patients with ampullary adenocarcinoma, Sakata et al. [26] have also reported that the number of metastatic lymph nodes is a better predictor of outcomes after resecOkano/Asano/Kushida/Kamada/Mori/ Suzuki

Table 2. Clinicopathologic factor influencing lymph node metastasis

Gender Male Female Age 70 Biliary drainage No Yes CEA, ng/ml 5.0 CA19-9, U/ml 36 T-Bil, mg/dl 1.0 Tumor size, mm 20 UICC T category Tis T1 T2–4 Pathologic grade Well (G1) Moderate to poor (G2/3) Perineural invasion No Yes Vascular invasion No Yes Lymphatic invasion No Yes Pancreas invasion No Yes Duodenal invasion No Yes Pathologic subtype Intestinal Pancreatobiliary Gross appearance Protruding Nonexposed Exposed Mixed Ulcerative Other Adjuvant chemotherapy No Yes

Node negative (n = 52)

Node positive (n = 22)

Univariate p value

23 (66%) 29 (74%)

12 (34%) 10 (26%)

0.41

23 (82%) 29 (62%)

5 (18%) 17 (38%)

0.064

26 (87%) 26 (60%)

4 (13%) 18 (40%)

0.036

40 (70%) 5 (71%)

17 (30%) 2 (29%)

0.94

27 (81%) 14 (56%)

6 (18%) 11 (44%)

0.032

31 (76%) 16 (62%)

10 (24%) 10 (38%)

0.22

20 (71%) 25 (68%)

8 (29%) 12 (32%)

0.98

11 (100%) 17 (87%) 24 (59%)

0 5 (23%) 17 (41%)

46 (82%) 6 (33%)

Multivariate analysis odds ratio (95% CI)

p value

1.23 (0.1–14.2)

0.86

2.65 (0.48–18.07)

0.26

0.001

5.37 (0.4–108.6)

0.2

10 (18%) 12 (67%)

0.0001

6.51 (1.13–56.6)

0.035

39 (81%) 6 (38%)

9 (19%) 10 (62%)

0.0013

5.07 (0.66–53.3)

0.12

33 (87%) 18 (53%)

5 (13%) 16 (47%)

0.0013

1.69 (0.15–19.7)

0.65

26 (90%) 25 (58%)

3 (10%) 18 (42%)

0.0025

1.24 (0.17–10.2)

0.82

37 (86%) 14 (46%)

6 (14%) 16 (54%)

0.0003

1.65 (0.12–29.7)

0.12

20 (91%) 30 (61%)

2 (9%) 19 (39%)

0.0064

1.2 (0.41–42.9)

0.88

27 (77%) 17 (63%)

8 (23%) 10 (37%)

0.22

30 (71%) 6 (75%) 24 (71%) 6 (67%) 6 (60%) 10 (77%)

12 (29%) 2 (25%) 10 (29%) 3 (33%) 4 (40%) 3 (23%)

0.8

41 (79%) 11 (50%)

11 (21%) 11 (50%)

0.51

Lymph Node Metastasis of Ampullary Adenocarcinoma

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

463

Table 3. Relationship between T factor and lymph node metastasis

UICC

Depth of invasion

Node negative

Node positive pathologic grade

Tis T1 T2–4

Mucosa Sphincter of Oddi Beyond Oddi

p pathologic grade

n = 52

G1

G2/3

n = 22

G1

G2/3

11 17 24

11 (100%) 15 (88%) 20 (83%)

0 2 (12%) 4 (17%)

0 5 17

0 1 (20%) 9 (53%)

0 4 (80%) 8 (47%)

0.0036 0.035

tion than is the lymph node ratio. Although the number of lymph node metastases is thought to be a very important indicator of prognosis after the radical treatment of ampullary adenocarcinoma [27], the clinicopathologic factors that are involved in lymph node metastasis have not been well characterized. Kayahara et al. [13] have reported that the gross appearance of the tumor is the most important prognostic factor for lymph node metastasis in ampullary adenocarcinoma. In the current study, lymph node involvement was more frequently seen in ulcerative population, but the difference did not reach statistical significance. On the other hand, we have identified several clinicopathologic factors that are related to lymph node metastasis. Indeed, univariate analyses revealed that 2 clinical factors and 7 pathological factors were associated with lymph node metastasis. In multivariate analysis, only pathologic grade remained significantly associated with lymph node metastasis. Zhou et al. [27] also confirmed the presence of a strong relationship between tumor differentiation (moderately to poorly differentiated) and lymph node metastases in patients with ampullary adenocarcinoma. These results support our findings regarding the factors that are related to lymph node metastases. Pathologic grade can be predicted through a preoperative biopsy. However, preoperative biopsies are limited in their ability to identify the actual dominant pathological grade. Accordingly, we recommend a biopsy that obtains multiple samples from different sites within the primary tumor. A pathologic grade of moderate-to-poor differentiation is an independent predictive factor for lymph node metastasis. An audit of lymph node dissection techniques revealed that detailed knowledge of anatomical distribution of lymph nodes can provide major surgical advantages. According to the results of a nationwide survey, the Japanese Society of Biliary Surgery suggests that the extent of 464

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

Color version available online

G1 = Well differentiated adenocarcinoma; G2 = moderately differentiated adenocarcinoma; G3 = poorly differentiated adenocarcinoma.

Fig. 2. Photomicrograph showing a representative ampullary can-

cer that extended into the sphincter of Oddi (T1 tumor). The arrow indicates the extent of tumor infiltration, and the arrow heads indicate the layer of the sphincter of Oddi.

lymph node dissection follows the standard D2 procedure [9]. It has been reported that radical PD with extended lymph node dissection, an especially aggressive treatment, does not confer a survival benefit for patients with pancreatic or periampullary adenocarcinomas, but does lead to increased morbidity [28]. The present study has shown that over 60% of patients with lymph node metastases have an involved node located on the posterior surface of the pancreas head (JSBS N1). Several studies based on lymphatic mapping have previously demonstrated that nodes on the posterior surface of the pancreas head are the first echelon nodes in ampullary cancers, and that other stations are involved much less frequently Okano/Asano/Kushida/Kamada/Mori/ Suzuki

[13, 17, 29]. In the present study, metastatic lymph nodes around the superior mesenteric artery (JSBS N2) were found in 18% of patients. Our results on the incidence of lymph node involvement in individual stations are similar to previous reports in the literature [13, 17, 29]. It has been reported that the location of metastatic nodes is significantly associated with patient survival. Indeed, reported 5-year survival rates are 53% for N1, 26% for N2, and 0% for N3 metastases (p < 0.001) [30]. Early ampullary cancer (defined as pTis and pT1 ampullary cancer) has a fairly good prognosis. In a previous study, these tumors were found to be radically resectable in all cases, and patients with these tumors had a 5-year survival rate of 83.7% [17]. To achieve a good prognosis in early ampullary cancer, complete resection is mandatory. In terms of complete resection, PD should be the standard operation for ampullary cancer. However, as has been confirmed by the present study, morbidity after PD remains significant. Indeed, we observed a complication rate of 47%, and there was 1 death (postoperative hemorrhage associated with pancreatic fistula). As a consequence of PDrelated morbidity and mortality, local resection has been attempted for early ampullary cancers. Yet, an accurate understanding of lymph node involvement is necessary to determine the indication of local resection, and is therefore essential knowledge when deciding whether to pursue curative ampullectomy. Nevertheless, few studies have provided adequate pathologic details on lymph node involvement in early ampullary cancers. In the present study, we analyzed the PPV, NPV, sensitivity, and specificity of pathologic grade for lymph node metastases. Interestingly, sensitivity (80%) and NPV (94%) were higher in patients with T1 tumors than in the overall study cohort. Although PD with lymph node dissection is considered to be a standard procedure for malignant tumor, ampullectomy might be acceptable for patients who are at surgical risk and who have a T1 tumor with well-differentiated adenocarcinoma. We hope that our study provides details that may be of use to research regarding ampullectomy. Despite the opportunities presented by ampullectomy, de Castro et al. [31] reported that PD should be recommended even for pT1 carcinoma, because 28% of patients with T1 disease have lymph node metastasis. This high prevalence of lymph node involvement is supported by a study of Terasawa et al. [17], who found that 20% of T1 stage tumor cases had lymph node metastases. In our study, none of the 11 patients with pTis tumors had lymph node metastases. On the other hand, lymph node metastasis was identified in 5 (23%) of the 22 patients with pT1 (invasion within the sphincter of Oddi) tumors. Further-

more, 4 (80%) of these 5 patients, had tumors with a dominant component of moderately differentiated adenocarcinoma. These results suggest that local resection without lymph node resection may be inappropriate for T1 tumors, especially those with a component of moderately to poorly differentiated adenocarcinoma proven by preoperative biopsy. Accordingly, ampullectomy without lymph node dissection may be considered for patients who have some risk for PD and diagnosed with T1 tumors that are well-differentiated adenocarcinomas. To achieve curative local resection, it is essential to first identify ampullary adenocarcinoma without lymph node metastasis. The relatively new development of endoscopic ultrasonography (EUS) and multidetector computed tomography (MDCT) have been expected to provide convincing imaging for preoperative staging and the detection of lymph node metastases. Chen et al. [32] reported that EUS provided superior detection of lymph node metastasis (47%), as compared with routine ultrasonography (7%) and computed tomography (CT; 33%). Shoup et al. [33] reported that neither EUS nor CT is sensitive for detecting lymph node metastases (21 and 42%, respectively), though they have similar specificities (80 and 73%, respectively). To date, preoperative imaging modalities have limited abilities to diagnose regional lymph node metastases from ampullary adenocarcinoma. Exact and reliable results are only obtained after precise pathologic examination of the surgical specimen. On the other hand, EUS can accurately define the depth of invasion in 75–83% of cases [34, 35]. Our results concerning the relationships between clinicopathologic factors and nodal status should provide useful evidence when selecting surgical strategies in the absence of definitive data on lymph node metastases. However, our study has several limitations. First, although the pathologic grade was significantly and independently associated with lymph node metastasis, the preoperative diagnosis for pathologic grade using biopsy specimens sometimes differ from the final pathologic diagnosis of the surgical specimen. Thus, improvement of accuracy in preoperative biopsy is extremely important. The behavior of tumors that are heterogeneously composed of various grades is less predictable than the behavior of tumors that are homogeneous. On the other hand, the rate of concordance between preoperative biopsy grade and final diagnosis has been reported to be 80–90% in gastric cancer (abstract P-352 from JDDW 2013). These results might support the idea that the information derived from preoperative biopsy is certainly useful in cases of ampullary adenocarcinoma. Second, the study has a retrospective design and is multi-institutional, be-

Lymph Node Metastasis of Ampullary Adenocarcinoma

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

465

cause of the rarity of ampullary adenocarcinoma. Although the board-certified members of the Japanese Society of Gastroenterology and Gastroenterological Surgery provided JSBS and TNM classifications in their survey responses, the data for several cases were incomplete, and there was naturally some variation in the clinicians’ evaluations. We suggest that a large prospective study of factors related to lymph node metastases may be warranted, particularly to validate the association with pathologic grade and the importance of invasion to the sphincter of Oddi that we have reported. Further prospective confirmation is necessary regarding the survival benefits that are conferred by adequate lymph node dissection and the diagnosis of lymph node metastasis using preoperative imaging.

Acknowledgments The authors are especially grateful to the following members of Kagawa Gastroenterology Forum who willingly took part in the survey: Hideyuki Inoue, MD (Kagawa Rosai Hospital), Masaki Wato, MD (Kagawa Prefectural Central Hospital), Yo Fukuda, MD (Takamatsu Municipal Hospital), Tomohiko Nishihira, MD (Takamatsu Red Cross Hospital), Takashi Maeba, MD (Social Insurance Ritsurin Hospital), Tsutomu Masaki, MD (Faculty of Medicine, Kagawa University), Aiichiro Kajikawa, MD (Zentsuji National Hospital), Mikihiro Kawada, MD (Kagawa Saiseikai Hospital), Fumihiko Kinekawa, MD (Sanuki Municipal Hospital), Sumi Yoshino, MD (Takamatsu Municipal Hospital), Hidenori Hata, MD (Mitoyo General Hospital), Minoru Horikita, MD (Kagawa Prefectural Shirotori Hospital), Makoto Oryu, MD (KKR Takamatsu Hospital, Kagawa).

Funding Sources

Conclusions

Patients who had ampullary adenocarcinoma with lymph node metastasis had significantly worse survival after resection than those without lymph node metastasis. Pathologic grade (moderate-to-poor differentiation) was significantly and independently associated with lymph node metastasis and was also a significant predictor in T1 tumor cases.

Supported by a Grant-in-Aid for Scientific Research (25462115) from the Ministry of Education, Science and Culture of Japan, and grants from the Pancreas Research Foundation of Japan.

Disclosure Statement The authors have no conflicts of interest to disclose.

References 1 Kim RD, Kundhal PS, McGilvray ID, Cattral MS, Taylor B, Langer B, Grant DR, Zogopoulos G, Shah SA, Greig PD, Gallinger S: Predictors of failure after pancreaticoduodenectomy for ampullary carcinoma. J Am Coll Surg 2006;202:112–119. 2 Howe JR, Klimstra DS, Moccia RD, Conlon KC, Brennan MF: Factors predictive of survival in ampullary carcinoma. Ann Surg 1998; 228:87–94. 3 Fischer HP, Zhou H: Pathogenesis of carcinoma of the papilla of Vater. J Hepatobiliary Pancreat Surg 2004;11:301–309. 4 Talamini MA, Moesinger RC, Pitt HA, Sohn TA, Hruban RH, Lillemoe KD, Yeo CJ, Cameron JL: Adenocarcinoma of the ampulla of Vater. A 28-year experience. Ann Surg 1997; 225:590–599; discussion 599–600. 5 Duffy JP, Hines OJ, Liu JH, Ko CY, Cortina G, Isacoff WH, Nguyen H, Leonardi M, Tompkins RK, Reber HA: Improved survival for adenocarcinoma of the ampulla of Vater: fiftyfive consecutive resections. Arch Surg 2003; 138:941–948; discussion 948–950. 6 Monson JR, Donohue JH, McEntee GP, McIlrath DC, van Heerden JA, Shorter RG, Nagorney DM, Ilstrup DM: Radical resection for

466

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

7

8

9 10

11 12

carcinoma of the ampulla of Vater. Arch Surg 1991;126:353–357. Okano K, Oshima M, Yachida S, Kushida Y, Kato K, Kamada H, Wato M, Nishihira T, Fukuda Y, Maeba T, Inoue H, Masaki T, Suzuki Y: Factors predicting survival and pathological subtype in patients with ampullary adenocarcinoma. J Surg Oncol 2014;110:156–162. Yeo CJ, Cameron JL, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, Hruban RH, Ord SE, Sauter PK, Coleman J, Zahurak ML, Grochow LB, Abrams RA: Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226:248–257; discussion 257–260. Japanese Society of Biliary Surgery: Classification of Biliary Tract Carcinoma, ed 2. Tokyo, Kanehara, 2004. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J, Neoptolemos J, Sarr M, Traverso W, Buchler M: Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005;138:8–13. AJCC Cancer Staging Manual, ed 7. New York, Springer-Verlag, 2010. Albores-Saavedra J, Henson DE, Klimstra DS: Tumors of the gallbladder, extrahepatic bile

13

14

15

16

ducts, and ampulla of Vater; in Rosai J, Sobin L (eds): Tumor Pathology Third Series, Fascicle 27. Washington, DC, Armed Forces Institute of Pathology, 2000, pp 259–316. Kayahara M, Nagakawa T, Ohta T, Kitagawa H, Miyazaki I: Surgical strategy for carcinoma of the papilla of Vater on the basis of lymphatic spread and mode of recurrence. Surgery 1997;121:611–617. Roder JD, Schneider PM, Stein HJ, Siewert JR: Number of lymph node metastases is significantly associated with survival in patients with radically resected carcinoma of the ampulla of Vater. Br J Surg 1995; 82: 1693–1696. Park SJ, Kim SW, Jang JY, Lee KU, Park YH: Intraoperative transfusion: is it a real prognostic factor of periampullary cancer following pancreatoduodenectomy? World J Surg 2002;26:487–492. Chan C, Herrera MF, de la Garza L, Quintanilla-Martinez L, Vargas-Vorackova F, Richaud-Patin Y, Llorente L, Uscanga L, Robles-Diaz G, Leon E, et al: Clinical behavior and prognostic factors of periampullary adenocarcinoma. Ann Surg 1995; 222: 632–637.

Okano/Asano/Kushida/Kamada/Mori/ Suzuki

17 Terasawa H, Uchiyama K, Tani M, Kawai M, Tsuji T, Tabuse K, Kobayashi Y, Taniguchi K, Yamaue H: Impact of lymph node metastasis on survival in patients with pathological T1 carcinoma of the ampulla of Vater. J Gastrointest Surg 2006;10:823–828. 18 Klempnauer J, Ridder GJ, Maschek H, Pichlmayr R: Carcinoma of the ampulla of Vater: determinants of long-term survival in 94 resected patients. HPB Surg 1998;11:1–11. 19 Su CH, Shyr YM, Lui WY, P’Eng FK: Factors affecting morbidity, mortality and survival after pancreaticoduodenectomy for carcinoma of the ampulla of Vater. Hepatogastroenterology 1999;46:1973–1979. 20 Carter JT, Grenert JP, Rubenstein L, Stewart L, Way LW: Tumors of the ampulla of Vater: histopathologic classification and predictors of survival. J Am Coll Surg 2008;207:210–218. 21 Kim WS, Choi DW, Choi SH, Heo JS, You DD, Lee HG: Clinical significance of pathologic subtype in curatively resected ampulla of Vater cancer. J Surg Oncol 2012;105:266–272. 22 Kimura W, Ohtsubo K: Incidence, sites of origin, and immunohistochemical and histochemical characteristics of atypical epithelium and minute carcinoma of the papilla of Vater. Cancer 1988;61:1394–1402. 23 Zhou H, Schaefer N, Wolff M, Fischer HP: Carcinoma of the ampulla of Vater: comparative histologic/immunohistochemical classification and follow-up. Am J Surg Pathol 2004;28:875–882. 24 Kimura W, Futakawa N, Zhao B: Neoplastic diseases of the papilla of Vater. J Hepatobiliary Pancreat Surg 2004;11:223–231.

Lymph Node Metastasis of Ampullary Adenocarcinoma

25 Kang HJ, Eo SH, Kim SC, Park KM, Lee YJ, Lee SK, Yu E, Cho H, Hong SM: Increased number of metastatic lymph nodes in adenocarcinoma of the ampulla of vater as a prognostic factor: a proposal of new nodal classification. Surgery 2014;155:74–84. 26 Sakata J, Shirai Y, Wakai T, Ajioka Y, Akazawa K, Hatakeyama K: Assessment of the nodal status in ampullary carcinoma: the number of positive lymph nodes versus the lymph node ratio. World J Surg 2011;35:2118–2124. 27 Zhou J, Zhang Q, Li P, Shan Y, Zhao D, Cai J: Prognostic relevance of number and ratio of metastatic lymph nodes in resected carcinoma of the ampulla of Vater. Chin J Cancer Res 2013;25:735–742. 28 Farnell MB, Pearson RK, Sarr MG, DiMagno EP, Burgart LJ, Dahl TR, Foster N, Sargent DJ; Pancreas Cancer Working Group: A prospective randomized trial comparing standard pancreatoduodenectomy with pancreatoduodenectomy with extended lymphadenectomy in resectable pancreatic head adenocarcinoma. Surgery 2005; 138: 618–628; discussion 628–630. 29 Shirai Y, Ohtani T, Tsukada K, Hatakeyama K: Patterns of lymphatic spread of carcinoma of the ampulla of Vater. Br J Surg 1997; 84: 1012–1016.

Dig Surg 2014;31:459–467 DOI: 10.1159/000370251

30 Sakata J, Shirai Y, Wakai T, Yokoyama N, Sakata E, Akazawa K, Hatakeyama K: Number of positive lymph nodes independently affects long-term survival after resection in patients with ampullary carcinoma. Eur J Surg Oncol 2007;33:346–351. 31 de Castro SM, van Heek NT, Kuhlmann KF, Busch OR, Offerhaus GJ, van Gulik TM, Obertop H, Gouma DJ: Surgical management  of neoplasms of the ampulla of vater: local resection or pancreatoduodenectomy and prognostic factors for survival. Surgery 2004;136:994–1002. 32 Chen CH, Tseng LJ, Yang CC, Yeh YH, Mo LR: The accuracy of endoscopic ultrasound, endoscopic retrograde cholangiopancreatography, computed tomography, and transabdominal ultrasound in the detection and staging of primary ampullary tumors. Hepatogastroenterology 2001;48:1750–1753. 33 Shoup M, Hodul P, Aranha GV, Choe D, Olson M, Leya J, Losurdo J: Defining a role for endoscopic ultrasound in staging periampullary tumors. Am J Surg 2000;179:453–456. 34 Tio TL, Tytgat GN, Cikot RJ, Houthoff HJ, Sars PR: Ampullopancreatic carcinoma: preoperative TNM classification with endosonography. Radiology 1990;175:455–461. 35 Cannon ME, Carpenter SL, Elta GH, Nostrant TT, Kochman ML, Ginsberg GG, Stotland B, Rosato EF, Morris JB, Eckhauser F, Scheiman JM: EUS compared with CT, magnetic resonance imaging, and angiography and the influence of biliary stenting on staging accuracy of ampullary neoplasms. Gastrointest Endosc 1999;50:27–33.

467

Copyright: S. Karger AG, Basel 2015. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.