Pancreatology 13 (2013) 589e593

Contents lists available at ScienceDirect

Pancreatology journal homepage: www.elsevier.com/locate/pan

Original article

The role of lymph node ratio in recurrence after curative surgery for pancreatic endocrine tumours Claudio Ricci a, *, Riccardo Casadei a, Giovanni Taffurelli a, Salvatore Buscemi a, Marielda D’Ambra a, Francesco Monari a, Donatella Santini b, Davide Campana a, Paola Tomassetti a, Francesco Minni a a b

Department of Internal Medicine, Emergency and Surgery (DIMES), Alma Mater Studiorum, University of Bologna, S.Orsola-Malpighi Hospital, Italy Department of Haematology and Oncology, Alma Mater Studiorum, University of Bologna, S. Orsola-Malpighi Hospital, Italy

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 August 2013 Received in revised form 4 September 2013 Accepted 11 September 2013

Background: The prognostic role of lymph nodes metastasis in pancreatic neuroendocrine tumours is unclear. Methods: Retrospective study of 53 patients who underwent a curative standard resection for pancreatic neuroendocrine tumours. The endpoint was to define the role of the lymph nodes ratio in recurrence after curative surgery. The following data were considered as possible factors for predicting the risk of recurrence: gender, age, presence of symptoms, hormonal status, site of tumours, type of resection, size of the tumours, radical resection, pathological T, N and M stage, the Ki67 index, the number of lymph nodes harvested, the number of metastatic lymph nodes and the lymph node ratio. Recurrence rate and time of recurrence were evaluated. Results: Twelve (26.4%) patients developed a recurrence with a median time of 42.8 (1e305) months. At multivariate analysis, the only factors related to recurrence were: size of lesions (HR 1.1, C.I. 95% 1.0e1.1, P ¼ 0.011), Ki67  5% (HR 3.6, C.I. 95% 1.3e10, P ¼ 0.014) and LNR > 0.07 (HR 5.2, C.I. 95% 1.1e25, P ¼ 0.045). Conclusions: Our study confirmed that the lymph nodes ratio played an important role in the recurrence rate and suggested that a low number of metastatic lymph nodes reduced the disease free survival. Copyright Ó 2013, IAP and EPC. Published by Elsevier India, a division of Reed Elsevier India Pvt. Ltd. All rights reserved.

Keywords: Pancreatic neoplasm Neuroendocrine tumours Lymph nodes

1. Introduction The prognostic role of lymph node metastasis is well known in pancreatic ductal adenocarcinoma [1] and, in recent years, the impact of the lymph node ratio (LNR) on survival has been clarified in patients who have undergone pancreatic resection for epithelial malignant disease [2e4]. Regarding pancreatic neuroendocrine tumours (PETs), only a few reports have shown that lymph node (LN) metastasis could play a role in the prognosis [5e9] and only one study [10] reported an LNR > 0.20 as a predicting factor for recurrence after curative surgery. The aim of the present study was to evaluate the role of the LNR in the recurrence after curative

* Corresponding author. Dipartimento di Scienze Mediche e Chirurgiche (DIMEC), Chirurgia Generale e Minni, Alma Mater Studiorum, Università di Bologna, Policlinico S. Orsola-Malpighi, Via Massarenti n. 9, 40138 Bologna, Italy. Tel.: þ39 051 341541; fax: þ39 051 341483. E-mail address: [email protected] (C. Ricci).

surgery in patients with PETs in a single high volume pancreatic centre. 2. Materials and methods This was a retrospective study of a prospective database regarding 107 consecutive patients affected by sporadic PETs, surgically treated from January 1980 to December 2012. Patients with an R2 resection (n ¼ 10), those with an atypical resection (enucleation and middle pancreatectomy) (n ¼ 26) and those without information regarding the number of Ki67-positive cells (n ¼ 18) were excluded. The remaining 53 patients, who underwent a curative standard pancreatic resection, were retrospectively studied. The endpoint of this study was to define the role of the LNR in recurrence after curative surgery. The following data were considered as possible factors for predicting the risk of recurrence: gender, age ( 5%) [14] .The number of lymph nodes harvested, the number of metastatic lymph nodes and the lymph node ratio were included in the study. The LNR was calculated as the ratio of the number of metastatic lymph nodes to the total number of lymph nodes harvested. The LNR was first analysed in the three categories (LNR ¼ 0; LNR between 0 and 0.2; LNR > 0.2) proposed by Boninsegna et al. [10] and it was then analysed in two categories (LNR  0.07 and LNR > 0.07), calculated using a ROC curve (AUC ¼ 0.768; P ¼ 0.009) for our population (Fig. 1). All specimens were examined by the same high volume pathologist (DS) [15]. Informed consent was obtained for each patient. The postoperative results were also reported but were not included in the analysis. Postoperative mortality was defined as in hospital or 30-day death. Follow-up examinations were conducted every 6 months for the first 2 years and annually thereafter with clinical examination, serum CgA with enzyme-linked immunosorbent assay (ELISA) > 21 U/l and immunoradiometric assay (IRMA) > 87 ng/ml and computed tomography (CT). Somatostatin receptor scintigraphic scanning (Octreoscan) was used from 2000 to 2006 and positron emission tomography with 68 Ga-DOTANOC was used from 2006 to 2012 [16] to confirm clinical, laboratory or radiological suspicion of recurrence. Recurrence rate, time and site of recurrence were evaluated. Disease-free survival (DFS) was calculated from the date of surgery to the date of the diagnosis of

recurrence. Overall survival (OS) was calculated from the date of surgery to death (both disease related or from other causes). Means or median, standard deviations or range and frequencies were used to describe the data. OS and DFS were estimated using the KaplaneMeier method. Risk of recurrence was reported as a Hazard Ratio (HR) with a confidence interval of 95% (C.I. 95%). Cox regression stepwise analyses were used to identify the factors related to recurrence in univariate and multivariate analyses. All variables with P values 0.20 [10] Calculated LNR cut-off (>0.07)

1.1 3.2 0.7 0.7 0.6 0.5 1.1 0.2 1.6 6.3 6.8 4.5 1.2 6.4 4.3 1.1 0.9 1 4.3 9.1 8.1

(0.4e4.5) (0.9e11.9) (0.2e2.7) (0.2e2.7) (0.2e2.1) (0.1e1.4) (1.0e1.1) (0.1e250) (0.2e13) (1.3e28) (0.7e61) (1.5e13.5) (1.1e1.3) (1.9e20) (1.9e9.3) (0.9e1.2) (0.9e1.1) e (0.8e23) (1.8e45) (2.2e30)

0.845 0.074b 0.596 0.587 0.410 0.208 0.009b 0.657 0.652 0.018b 0.087b 0.006b 0.001b 0.002b 0.2. According to the LNR observed by our experience, thirtyfour (64.2%) patients had an LNR  0.07 and 19 (35.8%) an LNR > 0.07. No postoperative mortality was observed. Postoperative morbidity occurred in 47.2% of the patients and the rate of pancreatic fistula was 28.3% (15/53 cases). Median postoperative length of hospital stay was 15 (7e98) days. After a median follow-up of 50.7 (1e305) months, 7 patients (13.2%) had died, 3 from disease progression. Twelve (26.4%) patients developed a recurrence of the disease with a mean time of 42.8 (1e305) months. Recurrences were located in the liver in 8 cases (66.6%) and were local in 4 (33.4%). All patients with recurrences were unresectable and they were treated with somatostatin analogues in 10 (83.3%) cases and systemic chemotherapy in 2 (17.7%) cases with a Ki67 index >20%. Peptide receptor radionuclide therapy (PPRT) was carried out in 5 (41.6%) patients as a first line therapy after failure of medical therapy. Mean OS were 238  22 and median DFS was 134  18 months. At univariate analysis (Tables 3 and 4), sex, presence of symptoms, hormonal status, tumour site, type of resection, number of harvested and metastatic lymph nodes, R status and T status did not influence the risk of recurrence. Age 55 years, an increase in tumour size, metastatic lymph nodes, liver metastases and TNM ENETS stage significantly increased the risk of recurrence. The Ki67 index (as a continuous variable), the WHO 2010 classification and the Ki67 index modified by Scarpa et al. [14]. were significantly related to DFS. Regarding the LNR, patients with an LNR between 0 and 0.20 and with LNR > 0.20 showed a significantly increased risk of recurrence (HR 4.3; CI 95% 0.8e23; P ¼ 0.089; HR 9.1; CI 95% 1.8e 45; P ¼ 0.007, respectively). When an LNR cut-off of 0.07 was applied, the DFS was significantly lower in patients with an LNR > 0.07 (HR 8.1; CI 95% 2.2e30; P ¼ 0.002) than in those with an LNR  0.07 (Fig. 2). At multivariate analysis, the only factors significantly related to recurrence were: size of lesions, Ki67  5% and LNR > 0.07. In particular, increased tumour size and a Ki67 index 5% were independent factors related to short DFS (HR 1.1, C.I. 95% 1.0e1.1, P ¼ 0.011 and HR 3.6, C.I. 95% 1.3e10, P ¼ 0.014, respectively) and, finally, an LNR > 0.07 also presented a DFS lower than those with an LNR  0.07 (HR 5.2, C.I. 95% 1.1e25, P ¼ 0.045). 4. Discussion Many factors, such as age, gender [17e19], positive margins [20], poorly differentiated tumours [6,17,20,21] and distant metastases [5,6,18,19,21e23] have been reported as factors influencing long term survival in patients having PETs who underwent surgical curative resection. In our largest single-institution series we reported a rate of recurrence that resulted of 26.4% with a mean DFS of about fifteen years. The choice of DFS as a parameter for evaluating the role of nodal involvement seems to be more appropriate with respect to OS for two reasons. First, PETs are characterized by Table 4 Multivariate analysis regarding the factors influencing recurrence in patients having pancreatic endocrine tumours (PETs) after curative surgery. Factors

HR

(C.I. 95%)

P value

Size of lesions (mm)a Ki 67  5% [14] Calculated LNR cut-off (>0.07)

1.1 3.6 5.2

(1.0e1.1) (1.3e10) (1.1e25)

0.011 0.014 0.045

a

Calculated as a continuous value.

592

C. Ricci et al. / Pancreatology 13 (2013) 589e593

Fig. 2. Disease-free survival (DFS) after resection for pancreatic neuroendocrine tumours comparing patients with LNR  0.07 (mean DFS 229 months) and LNR > 0.07 (mean DFS 83 months).

long term survival, even if a recurrence of disease is present. Second, to our knowledge, the only three studies [9,10,24] which reported factors related to a recurrence of PETs after curative resection showed that nodal involvement seemed to play an important role. In fact, in a study [24] previously conducted by our group, the WHO 2000 classification was an independent risk factor in reducing DFS. This may also be because the WHO 2000 classification uses lymph node metastasis as one of the pathological parameters to differentiate benign from malignant PETs. Moreover, Kramptiz et al. reported that lymph node involvement influenced DFS in a large cohort of patients [9]. Boninsegna et al., clearly demonstrated the relationship between lymph node involvement and recurrence [10] using the LNR. The authors applied the LNR cutoff proposed for epithelial peri-ampullary malignancy and demonstrated that all patients with an LNR > 0.20 presented an increased risk of recurrence when compared with those who had an LNR ¼ 0 while patients with an LNR between 0 and 0.20 did not present a significantly increased risk. In our series, at univariate analysis, we found similar results with a significantly increased risk only for the last category (LNR > 0.20; P ¼ 0.007). This result was not confirmed in the multivariate model (P ¼ 0.494) whereas an LNR cut-off value of 0.07 was significantly associated with recurrence in both the univariate (P ¼ 0.002) and multivariate (P ¼ 0.045) models, with a mean DFS in patients with an LNR  0.07 superior to those with an LNR > 0.07 (229 vs 83 months, respectively). Our low LNR cut-off could be explained by considering the role of the single parameters which made up the LNR: the number of metastatic lymph nodes and the total number of lymph nodes sampled. Regarding the number of metastatic lymph nodes, some studies have demonstrated the prognostic role in peri-ampullary cancer [25]. Regarding PETs, only the study of Krampitz et al. [9], reported that patients with even only 1 or 2 metastatic lymph nodes presented a time to liver progression of the disease (P ¼ 0.001) and DFS (P ¼ 0.004) shorter than those without metastatic lymph nodes, suggesting that a low number of metastatic lymph nodes also puts patients at a high risk of recurrence. In our experience, the mean number of metastatic lymph nodes was low, and it was a significant prognostic factor at univariate analysis (P ¼ 0.018). However, the quality of a lymphadenectomy also plays an important role in the evaluation of the LNR as has been demonstrated for pancreatic or peri-ampullary cancer [25,26]. Parekh et al. [26], in a large cohort of patients having PETs, analysed the quality

of lymphadenectomy and reported a low number of sampled lymph nodes either in enucleation, with a median of 5 lymph nodes (range 0e5), or in typical resection, with a median of 6 lymph nodes in a pancreatico-duodenectomy (0e17) and 4 lymph nodes (0e7) in a distal pancreatectomy. Furthermore, the authors demonstrated that a low number of lymph nodes harvested depended on the pathologist, on an inadequate lymphadenectomy and on the type of resection. In our series, in order to reduce any overestimation of the LNR, we excluded all patients with atypical resection (enucleation or middle pancreatectomy), and all the pathological specimens were reviewed by the same high volume pathologist. In fact, the median number of lymph nodes harvested was higher than that reported by other authors [26]. Tumour size also increased the risk of recurrence both in uni- (P ¼ 0.009) and multivariate (P ¼ 0.011) analyses. This is probably due to a high risk of malignancy as reported by Rindi et al. [27] in the largest multicentric study. The Ki67 index could be one of the most important independent predicting factors not only for OS but also for DFS [10]. Our results confirmed that a Ki67 index >5% was an independent risk factor for recurrence (P ¼ 0.014) with a mean DFS lower than those with a lower Ki67 index (209 vs 38 months). However, also in our series, the WHO 2010 classification, which is based on a different Ki67 cutoff (20%), did not independently predict DFS. In fact, the Ki67 index range which defines NET G2 seems to be too large, causing some ‘well-differentiated tumours’ with a high Ki67 index to show rapid disease progression, as has been reported in some studies [28]. In conclusion, our results confirmed that the LNR played an important role in the recurrence rate and suggested that a low number of metastatic lymph nodes reduced the DFS. Tumour size, the LNR and the Ki67 index seemed to be useful parameters for improving risk stratification in patients undergoing surgical curative resection for PETs. These parameters could permit the identification of patients who would rapidly develop a recurrence after curative surgery and, in the future, they could be important in selecting patients for studies regarding possible adjuvant therapies.

References [1] Katz MH, Wang H, Fleming JB, Sun CC, Hwang RF, Wolff RA, et al. Long-term survival after multidisciplinarym management of resected pancreatic adenocarcinoma. Ann Surg Oncol 2009;16(4):836e47. [2] Pawlik TM, Gleisner AL, Cameron JL, Winter JM, Assumpcao L, Lillemoe KD, et al. Prognostic relevance of lymph node ratio following pancreaticoduodenectomy for pancreatic cancer. Surgery 2007;141(5): 610e8. [3] Slidell MB, Chang DC, Cameron JL, Wolfgang C, Herman JM, Schulick RD, et al. Impact of total lymph node count and lymph node ratio on staging and survival after pancreatectomy for pancreatic adenocarcinoma: a large, population based analysis. Ann Surg Oncol 2008;15(1):165e74. [4] La Torre M, Cavallini M, Ramacciato G, Cosenza G, Rossi Del Monte S, Nigri G, et al. Role of the lymph node ratio in pancreatic ductal adenocarcinoma. Impact on patient stratification and prognosis. J Surg Oncol 2011 Nov 1;104(6):629e33. [5] Tomassetti P, Campana D, Piscitelli L, Casadei R, Santini D, Nori F, et al. Endocrine pancreatic tumours: factors correlated with survival. Ann Oncol 2005;16:1806e10. [6] Hochwald SN, Zee S, Conlon KC, Colleoni R, Louie O, Brennan MF, et al. Prognostic factors in pancreatic endocrine neoplasms: an analysis of 136 cases with a proposal for low-grade and intermediate-grade groups. J Clin Oncol 2002;20:2633e42. [7] Casadei R, Ricci C, Pezzilli R, Campana D, Tomassetti P, Calculli L, et al. Value of both WHO and TNM classification systems for patients with pancreatic endocrine tumours: results of a single-center series. World J Surg 2009;33(11):2458e63. [8] Bettini R, Boninsegna L, Mantovani W, Capelli P, Bassi C, Pederzoli P, et al. Prognostic factors at diagnosis and value of WHO classification in a monoinstitutional series of 180 non-functioning pancreatic endocrine tumours. Ann Oncol 2008;19:903e8. [9] Krampitz GW, Norton JA, Poultsides GA, Visser BC, Sun L, Jensen RT. Lymph nodes and survival in pancreatic neuroendocrine tumours. Arch Surg 2012 Sep;147(9):820e7.

C. Ricci et al. / Pancreatology 13 (2013) 589e593 [10] Boninsegna L, Panzuto F, Partelli S, Capelli P, Delle Fave G, Bettini R, et al. Malignant pancreatic neuroendocrine tumour: lymph node ratio and Ki67 are predictors of recurrence after curative resections. Eur J Cancer 2012;48(11): 1608e15. [11] American Joint Committee on Cancer. AJCC cancer staging manual. 6th ed. New York, NY: Springer; 2002. p. 157e64. [12] Rindi G, Klöppel G, Couvelard A, Caplin M, Couvelard A, de Herder WW, et al. TNM staging of foregut (neuro)endocrine tumours: a consensus proposal including a grading system. Virchows Arch 2006;449(4):395e401. [13] Bosman F. WHO classification of tumours of the digestive system. Lyon, France: IARC Press; 2010. [14] Scarpa A, Mantovani W, Capelli P, Beghelli S, Boninsegna L, Bettini R, et al. Pancreatic endocrine tumours: improved TNM staging and histopathological grading permit a clinically efficient prognostic stratification of patients. Mod Pathol 2010;23(6):824e33. [15] Gullo L, Migliori M, Falconi M, Pederzoli P, Bettini R, Casadei R, et al. Nonfunctioning pancreatic endocrine tumours: a multicenter clinical study. Am J Gastroenterol 2003;98(11):2435e9. [16] Ambrosini V, Campana D, Bodei L, Nanni C, Castellucci P, Allegri V, et al. 68GaDOTANOC PET/CT clinical impact in patients with neuroendocrine tumours. J Nucl Med 2010;51(5):669e73. [17] Venkatesh S, Ordonez NG, Ajani J, Schultz PN, Hickey RC, Johnston DA, et al. Islet cell carcinoma of the pancreas. A study of 98 patients. Cancer 1990;65: 354e7. [18] Solorzano CC, Lee JE, Pisters PW, Vauthey JN, Ayers GD, Jean ME, et al. Nonfunctioning islet cell carcinoma of the pancreas: survival results in a contemporary series of 163 patients. Surgery 2001;130:1078e85. [19] Bilimoria KY, Talamonti MS, Tomlinson JS, Stewart AK, Winchester DP, Ko CY, et al. Prognostic score predicting survival after resection of pancreatic neuroendocrine tumours: analysis of 3851 patients. Ann Surg 2008;247(3): 490e500.

593

[20] Madeira I, Terris B, Voss M, Denys A, Sauvanet A, Flejou JF, et al. Prognostic factors in patients with endocrine tumours of the duodenopancreatic area. Gut 1998;43:422e7. [21] Panzuto F, Nasoni S, Falconi M, Corleto VD, Capurso G, Cassetta S, et al. Prognostic factors and survival in endocrine tumour patients: comparison between gastrointestinal and pancreatic localization. Endocr Relat Cancer 2005;12:1083e92. [22] Jarufe NP, Coldham C, Orug T, Mayer AD, Mirza DF, Buckels JA, et al. Neuroendocrine tumours of the pancreas: predictors of survival after surgical treatment. Dig Surg 2005;22:157e62. [23] Casadei R, Ricci C, Tomassetti P, Campana D, Minni F. Factors related to longterm survival in patients affected by well-differentiated endocrine tumors of the pancreas. ISRN Surg 2012;2012:389385. [24] Casadei R, Ricci C, Pezzilli R, Campana D, Tomassetti P, Calculli L, et al. Are there prognostic factors related to recurrence in pancreatic endocrine tumours? Pancreatology 2010;10(1):33e8. [25] Valsangkar NP, Bush DM, Michaelson JS, Ferrone CR, Wargo JA, Lillemoe KD, et al. N0/N1, PNL, or LNR? The effect of lymph node number on accurate survival prediction in pancreatic ductal adenocarcinoma. J Gastrointest Surg 2013;17(2):257e66. [26] Parekh JR, Wang SC, Bergsland EK, Venook AP, Warren RS, Kim GE, et al. Lymph node sampling rates and predictors of nodal metastasis in pancreatic neuroendocrine tumour resections: the UCSF experience with 149 patients. Pancreas 2012;41(6):840e4. [27] Rindi G, Falconi M, Klersy C, Albarello L, Boninsegna L, Buchler MW, et al. TNM staging of neoplasms of the endocrine pancreas: results from a large international cohort study. J Natl Cancer Inst 2012 May 16;104(10):764e77. [28] Raymond E, Dahan L, Raoul JL, Bang YJ, Borbath I, Lombard-Bohas C, et al. Sunitinib malate for the treatment of pancreatic neuroendocrine tumours. N Engl J Med 2011;364(6):501e13.