Introduction

Pancreatic neuroendocrine tumors(PNETs)are uncommon,representing 1–2%of all pancreatic tumors[1]and are diverse in terms of their physiological and pathological behavior.As well as typical pathological variables,such as tumor size,nodal or distant metastases,there are further variables which can affect outcome,including Ki67 index and whether tumors are functional or nonfunctional,based on their ability to produce biological active peptides[2,3].

The World Health Organization(WHO)2010 classi fied PNETs into three groups according to the tumor cell proliferation activity.In addition,the European Neuroendocrine Tumor Society(ENETS)2006 and the American Joint Committee on Cancer(AJCC)2010 7th edition staging systems stratify disease severity[4].Several reports have correlated prognostic factors such as size,grade,functional status,distant metastases with survival and tumor recurrence[5–8].

Due to the heterogeneity of these tumors,limited data are available on the indications,the surgical approach and the extent of surgical resection upon survival.Surgery is the choice of treatment for localized primary lesions.Although multi-visceral resection for locally advanced disease,or debulking of the majority of the tumor mass for limited metastatic disease may be recommended,data from the literature are unclear as to the impact of these strategies upon survival[2,3,9,10].

随着城市化进程的推进，高楼的建设和道路的修建占用了越来越多的土地资源，改变了下垫面的性质，从而对城市气候造成影响。合理规划城市土地，适当增加绿地面积，可以助推打造生态宜居的沪宁杭城市群。

This study aimed to review the patient,surgical and pathological variables that are related to survival following surgical resection of PNETs.We also compared the ENETS and AJCC systems on the same patient cohort.

阿里噘着嘴翻着眼睛望着他，眼眶露出大大的眼白，算是同意。罗四强牵着他的手，走近哀乐声里。他们越过悼念的人们，以不经意的方式靠近躺在花丛中的遗体。这是个老太太。罗四强低声说：“这是个婆婆，那么老，又不好看，不是你姆妈吧？”

Methods

All patients(n=143)with functioning or non-functioning PNETs undergoing surgical exploration with intention to treat in our unit between January 1988 and December 2013 were identified from a prospectively maintained database.The study was approved by the institutional ethics committee.Demographic details,surgical treatment(enucleation,anatomic resection,multi-visceral resection plus metastasectomy and debulking surgery),pathological variables and follow-up information regarding survival and disease status were evaluated.Functional and hereditary lesions were diagnosed on the basis of the distinct clinical syndromes,serum elevation and positive immunohistochemistry of the relevant hormones.Computed tomography,magnetic resonance imaging,somatostatin receptor scintigraphy and endoscopic ultrasound were used for preoperative assessment of tumor location as deemed necessary.

Scoring/staging systems

华杨大队位于广西东南部的高县，一个被称为“八山一水一田”的边远城镇，而该大队的第十生产队就是典型的山区生产队。据1975年统计，华杨大队总人口1837人，耕地面积为1813.9亩，其中水田为1683.7亩，旱地为130.2亩，山地则有22000多亩，人均耕地0.99亩；十队总人口是142人，总耕地面积为171亩，其中水田有165亩，旱地仅为6亩，人均耕地1.20亩。[注]华杨大队：《一九七五年农业统计年报表》，高县档案馆藏，71/1/75/52。 与其他生产队相比，并没有临近河流，只有一条小水沟供其灌溉！并且四面环山，交通非常不便。

Outcomes

Pancreatic fistula were graded according to the International Study Group on Pancreatic Fistula criteria(ISGPF)[15].The above criteria were applied retrospectively for the diagnosis of fistula in the pre-ISGPF era on the basis of clinical or biochemical findings.All previous pancreatic fistula de finitions applied to our study group were reviewed and re-classi fied accordingly.All patients had regular clinical follow-up and cross sectional imaging according to the clinical condition of each patient.

70年前，和平的阳光再次普照大地，中国人民在经历十四年的苦难、战斗和牺牲之后，终于赢得了抗日战争的胜利。这山河重光的日子，从此永载史册，昭示着中华民族对正义、和平与人民的庄严信念。刚刚过去的9月3日，在天安门广场，习近平总书记发表重要讲话，深刻揭示了这一伟大胜利的重大意义，表达了中华民族铭记历史、缅怀先烈、珍爱和平、开创未来的坚定决心。无论在电视机前，还是身在广场，每一个中国人都强烈地感觉到把我们从根本上、从血液里团结在一起的力量，我们从前辈们气壮山河的业绩中，深刻地体认到我们共同的命运和坚定不移的方向。

Statistical analysis

This study aimed to review prognostic factors related survival following surgical resection of PNETs.Due to the relative infrequency of this disease,the variable pathological behavior of the tumors and evolution of medical therapies available to treat metastatic NETs,the cohort was not subdivided,in line with other similar studies[12,13].Multivariate analyses,however,were performed in an attempt to identify those variables that were independently related to survival.

The accuracy of the staging systems to predict 5-year OS and PFS was assessed using ROC curves.Patients who did not have the potential of 5-year follow-up,namely those who were lost to follow-up after less than five years,or those who died and had follow-up starting less than five years from the end of the study period,were excluded from this analysis.In each case,comparisons between the areas under the ROC(AUROC)curves for the systems were performed using the “roccomp”command in Stata[16].

Major resectional surgery has been advocated as a treatment strategy for PNETs previously[17–24].In the present study,there was no evidence of a signi ficant difference in survival between patients undergoing minor pancreatic resection such as enucleation when compared to those undergoing anatomical segmental resection.However,multi-visceral resection was associated with poor survival,which was similar to patients undergoing debulking surgery.

Table 1 Demography and clinical characteristics of the entire cohort(n=143).

Data were presented as mean±SD or number(percentage)or median(range). ∗margins exclude patients undergoing debulking surgery.MEN 1:multiple endocrine neoplasial 1;VHL:Von Hippel–Lindau;DP:distal pancreatectomy;TP:total pancreatectomy;CP:central pancreatectomy.

Characteristics Data Age(yr)53.5±16.0 Male 57(39.9%)Functioning tumor 54(37.8%)Insulinoma 43(30.1%)Gastrinoma 4(2.8%)Somatostatinoma 3(2.1%)VIPoma 2(1.4%)Glucagonoma 1(0.7%)PP 1(0.7%)Tumor site Head 76(53.1%)Body-tail 61(42.7%)Multicentric 6(4.2%)Hereditary status Sporadic 127(88.8%)MEN 1 10(7.0%)VHL 5(3.5%)Neuro fibrimatosis 1(0.7%)Operation Anatomic resection(Whipple-DP-TP-CP) 78(54.5%)Enucleation 33(23.1%)Multi-visceral 19(13.3%)Debulking 13(9.1%)Size(cm) 2.8(0.6–21.0)＜2 48(33.6%)2–4 45(31.5%)＞4 50(35.0%)Lymph nodes status Negative 94(65.7%)Positive 49(34.3%)Positive margins∗ 18/130(13.8%)Follow-up(mon) 72(0–290)Hospital stay(d) 8(1–96)

Results

Patient demographics and clinical characteristics

Of the 143 patients analyzed,the age at surgery was 53.5±16.0 years,and 60.1%(86/143)were female.Key demographics,surgical and pathological characteristics are summarized in Table 1.Nineteen patients(13.3%)underwent multi-visceral resections,including liver resection for metastases(n=14),hemicolectomy(n=5),adrenalectomy(n=1),small bowel resection(n=1)and subtotal gastrectomy(n=1).Thirteen patients(9.1%)underwent noncurative debulking surgery of the primary tumor.

Predictors of survival:patient factors

The median follow-up,based on the Kaplan–Meier estimate of potential follow-up,was 72 months,with a maximum of 290 months.The actuarial 1-,3-,5-and 10-year OS rates were 90.1%,83.9%,78.1%and 64.9%,respectively.Univariate analysis found increasing age,tumor size,and positive lymph nodes to be associated with signi ficantly shorter OS(Table 2).The type of surgery was also a signi ficant predictor of OS(P＜0.01).The shortest survival was observed in patients undergoing debulking surgery,although post-hoc analysis did not find this to be signi ficantly worse than in multi-visceral resection(P=0.188).However,both of these groups were found to have signi ficantly worse outcomes than those undergoing either anatomic or enucleation surgery(P＜0.01).

The cumulative PFS at 1-,3-,5-,and 10 years was 86.5%,73.5%,65.6%and 46.8%,respectively.Univariate analysis of PFS yielded results consistent with the analysis of OS,with the addition that patients with non-functioning tumors had shorter PFS than those with functional tumors,and patient age became non-signi ficant.Kaplan–Meier curves for selected factors are reproduced in Figs.1 and 2.

The primary end points were the overall survival(OS)and progression free survival(PFS)from the date of surgical resection of the primary lesion.For the PFS,progression was classi fied as radiologically proven recurrence,increase in the tumor size or presence of a new lesion in unresectable disease.

Table 2 Univariate analysis of clinical factors predicting overall survival and progression free survival.

Reported rates are Kaplan–Meier survival estimates at five years,and P values are from log-rank tests on all available follow-up.OS:overall survival;PFS:progression free survival.

5-year OS P value 5-year PFS P value Gender 0.948 0.955 Female 76.5% 65.4%Male 80.5% 65.9%Age(yr) 0.032 0.229＜50 82.6% 67.3%≥50 75.1% 64.4%Hereditary 0.482 0.835 No 77.5% 65.3%Yes 81.3% 68.2%Type of tumor 0.297 0.010 Functioning 85.2% 80.1%Non-functioning 73.9% 57.6%Type of surgery ＜0.001 ＜0.001 Anatomic 87.6% 73.3%Multi-visceral 51.5% 36.8%Enucleation 94.7% 90.9%Debulking 26.9% 7.7%Tumor size(cm) ＜0.001 ＜0.001＜2 94.6% 91.6%2–4 81.0% 64.1%＞4 61.5% 43.9%Site 0.414 0.171 Head 78.8% 67.7%Multicentric 66.7% 33.3%Non-head 77.9% 66.1%Margins(R2 excluded) 0.291 0.291 R0 85.9% 85.9%R1 72.2% 72.2%Lymph nodes ＜0.001 ＜0.001 Negative 91.5% 79.0%Positive 53.7% 40.9%

Predictors of survival:staging systems

贵冶智能工厂的建设是根据贵冶对应用系统功能的需求，结合贵冶生产管理现状、信息化建设现状、两化融合现状为基本出发点，从总体架构和具体实现上进行整体设计，搭建出可实施落地的、符合贵冶的智能工厂框架。

“X”是“确认过眼神X”构式中的唯一变项，变项“X”和“确认过眼神”构成前后承接关系。变项“X”的结构主要有三种：一是“……遇上……”，类似于歌词中的“我遇上对的人”的表达，是由原型构式演化而来；二是“……是……(的)人”，由激活该构式的表达“确认过眼神，你是广东人”及之后推动其蔓延开来的“确认过眼神，是最不想理的人”抽象而来；三是其它：既不是一也不是二的较随意表达的结构，是“确认过眼神X”构式在泛化过程中形成的随意性表达。

In order to compare the predictive accuracy of the scores,a set of ROC curves were produced,considering both OS and PFS for those patients with the potential of five or more years of followup(n=96,Table 4).This analysis found that the WHO,AJCC and ENETS,along with the modi fied versions of these scores,were all signi ficantly predictive of both OS and PFS at five years.Ki67 was not found to be signi ficantly predictive of either outcome(P＞0.05),with the modi fied version of the system only being signi ficantly predictive of OS(AUROC:0.699;P＜0.05).

Pairwise comparisons were then performed,in order to identify any signi ficant difference in accuracy between the scores.This analysis found the modi fied version of Ki67 to be a signi ficant improvement over the original one for the prediction of OS(AUROC:0.699 vs 0.605;P＜0.01).No signi ficant differences were detected between the predictive accuracy of any of the other systems,with AUROCs ranging from 0.701–0.766 for the prediction of 5-year OS,and 0.680–0.753 for 5-year PFS.

Predictors of survival:multivariate analysis

Multivariate analysis of OS and PFS is reported in Table 5.This analysis considered both the staging systems and patient factors,and used a forward stepwise approach to select signi ficant independent predictors of the two outcomes.

The resulting model found OS to be signi ficantly shorter in patients with lymph node metastases and with increasing WHO 2010 grade.The type of surgery was also signi ficant,with the worst outcomes observed in patients undergoing multi-visceral or debulking surgery.The multivariate analysis also found gender to be signi ficant,contrary to the univariate analysis,with survival signi ficantly longer in males.This finding can be explained by the fact that males had signi ficantly different distributions of surgery(P＜0.05),being more likely than females to require palliative surgery(14.0%vs 5.8%)and less likely to receive enucleation(12.3%vs 30.2%).In addition,males had signi ficantly higher WHO grades(P=0.023),with 15.1%of male patients at grade 3,compared to 6.5%of females.Hence,the signi ficance of gender in the multivariate analysis was as a reflection of male patients having similar OS despite having more severe disease.

The multivariate analysis of PFS returned similar results to the analysis of OS,with the exception that the tumor site was also found to be signi ficant,with patients with multicentric tumors having shorter survival than those with tumors in a single site.

Discussion

The data were analyzed by a medical statistician(HJ)using IBM SPSS Statistics 22(IBM Corp.,Armonk,NY)and Stata 14(Stata Corp.,TX).OS and PFS rates were calculated using the Kaplan–Meier methodology,with comparisons across factors made using log-rank tests.

The main findings were that female gender,multi-visceral or debulking surgery,the presence of lymph node metastases and increasing WHO status were independently related to reduced OS and PFS.The association between survival and gender was further explored,as this was not found to be signi ficant in univariate analysis.This revealed that,whilst the crude survival rates for male and female patients were similar,male patients had signi ficantly more advanced disease;hence the difference only became apparent in multivariate analysis,after accounting for confounding factors.

3.1.1 叶缘型。其又称叶枯型，主要为害叶片，是典型的极为常见的病害。起初呈现深绿色短线状病斑，慢慢扩展为短条状，后期沿着叶边的两侧或是叶中脉向上或向下延伸，最后成长条斑状，由深绿变黄，最终转成黄褐色或是灰白色，形成明显的不规则波纹状，与健康部位有明显的界限。

盱眙县明祖陵镇陈某饲养大小猪共150头，其中45日龄猪12头，90日龄猪10头（均为自家母猪繁殖），二窝猪相继出现发病现象。有的猪精神沉郁、食欲减少，出现急性死亡未见体温升高；病程长的病猪，体温在41～42 ℃；有的猪开始出现便秘，1～2 d后有的出现腹泻现象；病猪步态不稳，走路肌肉震颤，四肢无力，倒地四肢呈划水状，叫声嘶哑；病程长的猪身上出现红色斑块、消瘦无力、拉稀、眼脸苍白水肿，1个月左右衰弱死亡（已经死亡4头份）。

Multivariate analysis was then performed using a Cox regression model,with a forward stepwise entry method,to identify independent predictors of OS and PFS.Statistical signi ficance was defined as P values less than 0.05 throughout.

Fig.1.Kaplan–Meier curves for overall survival(A,B)and progression free survival(C,D)by type of surgery and WHO grade respectively.

The multivariate analysis con firms the prognostic importance of lymph node metastases,in line with other studies [1,8,9,12,14,17,25–27].Previous studies have reported an increased OS with biologically active tumors[2,17,18]which may relate to the efficacy of evolving medical therapies available to treat PNETs[18,19,28].

The de finitions of the TNM tumor stages considered in the analysis are shown in Table 3,and the predictive accuracy for outcomes at five years is shown in Table 4.The AJCC and ENETS scores were available for the whole cohort,while the WHO could only be applied to 134 patients,and the Ki67 to 91 patients.Ki67 was not available for patients whose resections were performed prior to the implementation of routine Ki67 testing.OS and PFS were found to differ signi ficantly by tumor stage for all of the staging systems considered(P＜0.05).

根据表1中红枣的姿态概率可知，红枣在落果聚拢区主要受到重力作用充入落果口，因此红枣主要以“平躺”和“侧卧”的姿态充入落果口，两者的概率之和在0. 85以上。故本文设计了一种易于红枣以“平躺”和“侧卧” 的姿态充入落果端口的仿形伞状集果筒。

Multifocal tumors were found to be associated with a reduced PFS in multivariate analysis.The data interpretation is difficult by the low numbers of patients,but may reflect adverse biological behavior of the tumors or of associated genetic diseases such as multiple endocrine neoplasia.The lack of impact of involved resection margins on survival following surgical resection of PNETs is intriguing and has been noted previously[17,29],although some authors have reported an adverse effect upon outcome[5,10,26].This discrepancy may be explained by the untested association between available medical therapies available over the period the patients were observed.It almost certainly demonstrates the indolent nature of PNETs as compared to other malignant tumors.

Surveillance following surgical resection of PNETs is warranted,since the development of metastatic disease is associated with mortality and there are now a number of effective treatments available that have been shown to slow down disease[19,30–33].In fact,selected patients have undergone resection of metastatic disease,and this strategy has been proposed to prolong survival[17,18,21].However,surgery leaving behind macroscopic disease is unlikely to be justi fiable in all cases[22].In the present study,patients who had potentially curative surgery had better OS and PFS compared with the group of patients undergoing noncurative resections.Thus,the extent of surgical resection should be carefully planned preoperatively to minimize the occurrence of this outcome and,therefore,avoid the morbidity of surgery for selected poor prognosis patients[21,23,24].

The secondary aim of the study was to assess the utility of various staging systems for patients undergoing surgical resection of PNETs.Other authors have found differences between the predictive accuracy of the ENETS and AJCC/UICC TNM staging systems[12–14,25].In the present study,their predictive accuracy was very similar.These different prognostic outcomes in patients staged using both systems might be explained by tumor size and lymph node status differences.This may indicate that size is more important factor than the lymph node status.This could be true in ENETS stage IIIa,where a radical resection may not be possible in a large number of patients[12].Hence,the modi fication introduced by Scarpa et al.[14]de fining ENETS T3 only by tumor size and T4 by adjacent organ invasion might have important implications of prognostic signi ficance.This provides a clear separation between potential malignant lesions and invariably malignant lesions with extra-pancreatic invasion.Given the prolonged survival expected in the resected patients on early stages[26],further discrimination of differences in PFS will be required in each stage since the present study includes a small number of patients on the higher stage disease.Nevertheless,the analysis of the different systems and the modi fied version concluded that all of the staging systems are signi ficant predictors of outcome,whilst no one was superior to others at predicting OS or PFS.A large number of series is needed to validate the disparate results of different studies.

The WHO classi fication system,based upon morphological characteristics,was also tested and,although the AUROC was lower than those observed in the AJCC/ENETS scores,none of these comparisons were found to be statistically signi ficant.It has been suggested that the range of Ki67 in WHO grade 2(2–20%)represents a heterogeneous set of tumors and that a different cut-off value of Ki67 should be used for de fining grade 2 tumors[14].The recent proposal of a value of 5%for describing more aggressive PNETs may increase the prognostic capability of the grading system[7,12,14].The above proposal was applied to the study cohort,and was found to signi ficantly improve the predictive accuracy of the score for OS,with the AUROC increasing from 0.605 to 0.699(P＜0.01).

Tumors were classi fied according to WHO 2010 classi fication[4,11]and staged according to ENETS and the American Joint Committee on Cancer/Union for International Cancer Control(AJCC/UICC)TNM systems[12,13].Ki67 grading was also considered,with Ki67≤2%classi fied as grade 1,3–20%as grade 2 and＞20%as grade 3[11].In addition,two modi fied scores were calculated,the “modi fied ENETS”,a four stage version of the score,and “modi fied Ki67”,which included Ki67＜5%in grade 1,5–20%in grade 2 and＞20%to grade 3,as proposed by Scarpa et al.[14].A four stage version of the AJCC,which excluded the letters from the stages,was also considered to make a fairer comparison with the modi fied ENETS staging.

The major limitations of this study are the retrospective nature of data analysis and the duration over which patients were recruited.Over the 25-year study period,not all data was recorded,making it impossible to analyze all factors known to affect outcome such as perineural and lympho-vascular invasion or tumor necrosis.Furthermore,the small sample size and indolent nature of the disease reduces the statistical power,meaning that some of the weaker relationships between factors and survival outcomes,or subtler differences between the predictive accuracy of the staging systems may have been missed(type 2 error).This could be true for some of the stage categories especially on the higher stage disease in which small numbers of patients are included.Clearcut distinctions between early and advance stages have also been missed due to the weakness above.We were also unable to identify postoperative medical therapies provided to the majority of patients.In any case,it is questionable to what degree the inclusion of the above factor would have altered the survival rate in the resected tumors,since most of these modalities are used for locoregional unresectable and metastatic disease[3,34–36].

进入21世纪以来，亚太地区各国签署和实施的RTA数量迅速增长，至2016年末，该地区18个主要国家(包括澳大利亚、文莱、加拿大、智利、中国、印度尼西亚、日本、韩国、马来西亚、墨西哥、新西兰、秘鲁、菲律宾、俄罗斯、新加坡、泰国、美国和越南)已生效实施的双边及多边RTA多达143个，与2000年(26个)相比，增长了457.7%，如表1所示。

Fig.2.Kaplan–Meier curves for progression free survival by tumor type(A)and lymph node positivity(B).

Table 3 Differences in T code and stage de finitions between ENETS,AJCC[12]and modi fied ENETS staging systems as proposed by Scarpa et al.[14].

ENETS:European Neuroendocrine Tumor Society TNM system;AJCC/UICC:American Joint Committee on Cancer/Union for International Cancer Control TNM system.

ENETS Modi fied ENETS AJCC Code T1 Limited to the pancreas,＜2 cm Limited to the pancreas,＜2 cm Limited to the pancreas,≤2 cm T2 Limited to the pancreas,2–4 cm Limited to the pancreas,2–4 cm Limited to the pancreas,＞2 cm T3 Limited to the pancreas,＞4 cm or invading duodenum or bile duct Limited to the pancreas,＞4 cm Beyond the pancreas,no involved arteries T4 Tumor invading adjacent organs or the wall of large vessels Beyond the pancreas Involvement of arteries(unresectable)Stage de finition I a T1,N0,M0–T1,N0,M0 b–T2,N0,M0 II a T2,N0,M0 – T3,N0,M0 b T3,N0,M0 – T1–3,N1,M0 III a T4,N0,M0 – T4,any N,M0 b Any T,N1,M0 –IV Any T,any N,M1 –Any T,any N,M1 Four stage I T1(limited to pancreas,＜2 cm) T1–T2(limited to pancreas,＜4 cm) T1–T2(limited to pancreas,any dimension)II T2–T3(＞4 cm,invading duodenum or bile duct) T3(limited to pancreas,＞4 cm) T3 or N1(outside of the pancreas)III T4 or N1(outside of pancreas,invading large vessels T4 or N1(outside of the pancreas T4(involving large arteries,unresectable)IV M1 M1 M1

In conclusion,the 5-year OS after surgical resection of PNETs is encouraging,even for patients with advanced disease.Multivariate analyses demonstrate that multi-visceral surgical resections do not overcome the effects of advanced pathological staging.The three widely used staging systems,WHO,ENETS and AJCC,all offer good prognostic information following surgical resection of PNETs,with no one system found to have a signi ficant advantage.The modi fication suggested for the proliferation index signi ficantly improved accuracy in the prediction of OS.Because PNETs are a rare subgroup of tumors,careful interpretation of histopathologic data is needed to identify reliable prognostic factors and long-term followup is suggested.

Table 4 Five-year overall survival and progression free survival for the tested staging systems.

The Kaplan–Meier analysis included all available patients(n=143,unless stated otherwise),whilst the ROC analysis considered only those with the potential for five years of follow-up. †Kaplan–Meier estimated rates,based on all available follow-up.Log-rank test found signi ficant differences(P＜0.05)between the stages for all of the classi fications considered.#:No patients left at risk at 5 years.Quoted value is the last calculable rate.OS:overall survival;PFS:progression free survival.

5-year OS† AUROC(SE) P value 5-year PFS† AUROC(SE) P value WHO 2010 0.701(0.069) 0.004 0.680(0.061) 0.005 G1(n=73) 87% 78%G2(n=48) 78% 60%G3(n=13) 29% 23%AJCC 0.766(0.057) ＜0.001 0.753(0.052) ＜0.001 Ia(n=54) 93% 85%Ib(n=26) 92% 74%IIa(n=7) 100% 100%IIb(n=35) 66% 52%III(n=3) 0 0 IV(n=18) 49% 28%Four stage AJCC 0.755(0.058) ＜0.001 0.729(0.055) ＜0.001 I(n=80) 93% 81%II(n=42) 71% 60%III(n=3) 0 0 IV(n=18) 49% 28%ENETS 0.755(0.056) ＜0.001 0.743(0.053) ＜0.001 I(n=46) 94% 85%IIa(n=25) 87% 76%IIb(n=13) 100% 83%IIIa(n=3) 100%# 100%#IIIb(n=38) 60% 47%IV(n=18) 49% 28%Modi fied ENETS 0.735(0.058) ＜0.001 0.720(0.055) ＜0.001 I(n=71) 92% 82%II(n=10) 100% 78%III(n=44) 64% 54%IV(n=18) 49% 28%Ki67(n=91) 0.605(0.097) 0.267 0.642(0.086) 0.104 1(n=46) 82% 78%2(n=41) 81% 62%3(n=4) 25%# 25%#Modi fied Ki67(n=91) 0.699(0.093) 0.035 0.623(0.088) 0.159 1(n=62) 86% 74%2(n=25) 68% 60%3(n=4) 25%# 25%#

Table 5 Multivariate analysis of factors related to overall survival and progression free survival.

Multivariable analysis was performed using Cox regression models,with a forwards stepwise entry method.All of the staging systems in Table 3,and patient factors in Table 2 were considered for entry,with the exception of the original and modified Ki67,which were removed from consideration in order to minimize exclusions due to missing data.HR:hazard ratio;95%CI:95%con fidence interval;OS:overall survival;PFS:progression free survival.

OS PFS HR(95%CI) P value HR(95%CI) P value Gender 0.006 ＜0.001 Female– – – –Male 0.3(0.2–0.7) 0.006 0.3(0.1–0.5) ＜0.001 Type of surgery 0.011 ＜0.001 Anatomic– – – –Multi visceral 4.0(1.4–8.4) 0.008 5.5(2.4–12.7) ＜0.001 Enucleation 0.2(0.0–1.6) 0.135 0.1(0.0–0.6) 0.010 Debulking 2.2(0.9–5.8) 0.098 5.0(1.7–14.2) 0.003 Site 0.017 Head– – – –Multicentric – – 5.4(1.7–17.1) 0.004 Non-head – – 1.2(0.6–2.5) 0.557 Lymph nodes 0.003 0.004 Negative– – – –Positive 4.2(1.6–11.3) 0.003 3.0(1.4–6.4) 0.004 WHO 2010 grade 0.026 0.018 G1– – – –G2 1.0(0.4–2.4) 0.975 2.5(1.3–5.1) 0.010 G3 3.5(1.3–9.4) 0.013 2.8(2.8–7.9) 0.049

Contributors

BN collected data and wrote the paper.HJ analyzed data.MR,SRP and IJR performed operations and revised the paper.AJ performed the research and assisted to the design of the study.ST revised the paper and assisted with the interpretation of the results.RKJ propose the study and critically revised the manuscript.All authors contribute to the interpretation of the study.RKJ is the guarantor.

Funding

None.

Ethical approval

This study was approved by the Clinical Audit Department and Clinical Governance of the Queen Elizabeth Hospital,University Hospitals of Birmingham,UK.

Competing interest

No bene fits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

References

[1]Haugvik SP,Labori KJ,Edwin B,Mathisen Ø,Gladhaug IP.Surgical treatment of sporadic pancreatic neuroendocrine tumors:a state of the art review.Sci World J 2012;2012:357475.

[2]Halfdanarson TR,Rabe KG,Rubin J,Petersen GM.Pancreatic neuroendocrine tumors(PNETs):incidence,prognosis and recent trend toward improved survival.Ann Oncol 2008;19:1727–1733.

[3]Kulke MH,Bendell J,Kvols L,Picus J,Pommier R,Yao J.Evolving diagnostic and treatment strategies for pancreatic neuroendocrine tumors.J Hematol Oncol 2011;4:29.

[4]Klöppel G.Classi fication and pathology of gastroenteropancreatic neuroendocrine neoplasms.Endocr Relat Cancer 2011;18:S1–16.

[5]Gao C,Fu X,Pan Y,Li Q.Surgical treatment of pancreatic neuroendocrine tumors:report of 112 cases.Dig Surg 2010;27:197–204.

[6]Oh TG,Chung MJ,Park JY,Bang SM,Park SW,Chung JB,et al.Prognostic factors and characteristics of pancreatic neuroendocrine tumors:single center experience.Yonsei Med J 2012;53:944–951.

[7]Hamilton NA,Liu TC,Cavatiao A,Mawad K,Chen L,Strasberg SS,et al.Ki-67 predicts disease recurrence and poor prognosis in pancreatic neuroendocrine neoplasms.Surgery 2012;152:107–113.

[8]Krampitz GW,Norton JA,Poultsides GA,Visser BC,Sun L,Jensen RT.Lymph nodes and survival in pancreatic neuroendocrine tumors.Arch Surg 2012;147:820–827.

[9]Tomassetti P,Campana D,Piscitelli L,Casadei R,Santini D,Nori F,et al.Endocrine pancreatic tumors:factors correlated with survival.Ann Oncol 2005;16:1806–1810.

[10]Kim MJ,Choi DW,Choi SH,Heo JS,Park HJ,Choi KK,et al. Surgical strategies for non-functioning pancreatic neuroendocrine tumours.Br J Surg 2012;99:1562–1568.

[11]Klimstra DS,Modlin IR,Coppola D,Lloyd RV,Suster S.The pathologic classi fication of neuroendocrine tumors:a review of nomenclature,grading,and staging systems.Pancreas 2010;39:707–712.

[12]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;104:764–777.

[13]Klöppel G,Rindi G,Perren A,Komminoth P,Klimstra DS.The ENETS and AJCC/UICC TNM classi fications of the neuroendocrine tumors of the gastrointestinal tract and the pancreas:a statement.Virchows Arch 2010;456:595–597.

[14]Scarpa A,Mantovani W,Capelli P,Beghelli S,Boninsegna L,Bettini R,et al.Pancreatic endocrine tumors:improved TNM staging and histopathological grading permit a clinically efficient prognostic strati fication of patients.Mod Pathol 2010;23:824–833.

[15]Bassi C,Dervenis C,Butturini G,Fingerhut A,Yeo C,Izbicki J,et al.Postoperative pancreatic fistula:an international study group(ISGPF)de finition.Surgery 2005;138:8–13.

[16]Cleves MA.From the help desk:comparing areas under receiver operating characteristic curves from two or more probit or logit models.Stata J 2002;2:301–313.Available from http://www.stata-journal.com/article.html?article=st0023.

[17]Bilimoria KY,Talamonti MS,Tomlinson JS,Stewart AK,Winchester DP,Ko CY,et al.Prognostic score predicting survival after resection of pancreatic neuroendocrine tumors:analysis of 3851 patients.Ann Surg 2008;247:490–500.

[18]Wang DS,Zhang DS,Qiu MZ,Wang ZQ,Luo HY,Wang FH,et al.Prognostic factors and survival in patients with neuroendocrine tumors of the pancreas.Tumour Biol 2011;32:697–705.

[19]Bettini R,Partelli S,Boninsegna L,Capelli P,Crippa S,Pederzoli P,et al.Tumor size correlates with malignancy in nonfunctioning pancreatic endocrine tumor.Surgery 2011;150:75–82.

[20]Haynes AB,Deshpande V,Ingkakul T,Vage fiPA,Szymonifka J,Thayer SP,et al. Implications of incidentally discovered,nonfunctioning pancreatic endocrine tumors:short-term and long-term patient outcomes.Arch Surg 2011;146:534–538.

[21]Fendrich V,Langer P,Celik I,Bartsch DK,Zielke A,Ramaswamy A,et al.An aggressive surgical approach leads to long-term survival in patients with pancreatic endocrine tumors.Ann Surg 2006;244:845–853.

[22]Falconi M,Bettini R,Boninsegna L,Crippa S,Butturini G,Pederzoli P.Surgical strategy in the treatment of pancreatic neuroendocrine tumors.JOP 2006;7:150–156.

[23]Abu Hilal M,McPhail MJ,Zeidan BA,Jones CE,Johnson CD,Pearce NW.Aggressive multi-visceral pancreatic resections for locally advanced neuroendocrine tumours.Is it worth it?JOP 2009;10:276–279.

[24]Gu P,Wu J,Newman E,Muggia F.Treatment of liver metastases in patients with neuroendocrine tumors of gastroesophageal and pancreatic origin.Int J Hepatol 2012;2012:13.

[25]Wang SE,Su CH,Kuo YJ,Shyr YM,Li AF,Chen TH,et al.Comparison of functional and nonfunctional neuroendocrine tumors in the pancreas and peripancreatic region.Pancreas 2011;40:253–259.

[26]Ballian N,Loeffler AG,Rajamanickam V,Norstedt PA,Weber SM,Cho CS.A simpli fied prognostic system for resected pancreatic neuroendocrine neoplasms.HPB(Oxford)2009;11:422–428.

[27]Panzuto F,Nasoni S,Falconi M,Corleto VD,Capurso G,Cassetta S,et al.Prognostic factors and survival in endocrine tumor patients:comparison between gastrointestinal and pancreatic localization.Endocr Relat Cancer 2005;12:1083–1092.

[28]Falconi M,Bartsch DK,Eriksson B,Klöppel G,Lopes JM,O’Connor JM,et al.ENETS consensus guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system:well-differentiated pancreatic non-functioning tumors.Neuroendocrinology 2012;95:120–134.

[29]Fischer L,Kleeff J,Esposito I,Hinz U,Zimmermann A,Friess H,et al.Clinical outcome and long-term survival in 118 consecutive patients with neuroendocrine tumours of the pancreas.Br J Surg 2008;95:627–635.

[30]Hill JS,McPhee JT,McDade TP,Zhou Z,Sullivan ME,Whalen GF,et al.Pancreatic neuroendocrine tumors:the impact of surgical resection on survival.Cancer 2009;115:741–751.

[31]Yao JC,Shah MH,Ito T,Bohas CL,Wolin EM,Van Cutsem E,et al.Everolimus for advanced pancreatic neuroendocrine tumors.N Engl J Med 2011;364:514–523.

[32]Raymond E,Dahan L,Raoul JL,Bang YJ,Borbath I,Lombard-Bohas C,et al.Sunitinib malate for the treatment of pancreatic neuroendocrine tumors.N Engl J Med 2011;364:501–513.

[33]Caplin ME,Pavel M,´Cwikła JB,Phan AT,Raderer M,Sedláˇcková E,et al.Lanreotide in metastatic enteropancreatic neuroendocrine tumors.N Engl J Med 2014;371:224–233.

[34]Panzuto F,Boninsegna L,Fazio N,Campana D,Pia Brizzi M,Capurso G,et al.Metastatic and locally advanced pancreatic endocrine carcinomas:analysis of factors associated with disease progression.J Clin Oncol 2011;29:2372–2377.

[35]Ramage JK,Ahmed A,Ardill J,Bax N,Breen DJ,Caplin ME,et al.Guidelines for the management of gastroenteropancreatic neuroendocrine(including carcinoid)tumours(NETs).Gut 2012;61:6–32.

[36]Öberg K,Knigge U,Kwekkeboom D,Perren AESMO Guidelines Working Group..Neuroendocrine gastro-entero-pancreatic tumors:ESMO clinical practice guidelines for diagnosis,treatment and follow-up.Ann Oncol 2012:23 vii124–130.