《(7.2)--消化系统肿瘤消化系统肿瘤（术前吉西他滨同步放化+手术+辅助化疗VS手.pdf》由会员分享，可在线阅读，更多相关《(7.2)--消化系统肿瘤消化系统肿瘤（术前吉西他滨同步放化+手术+辅助化疗VS手.pdf（12页珍藏版）》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、original reportsPreoperative Chemoradiotherapy VersusImmediate Surgery for Resectable and BorderlineResectable Pancreatic Cancer:Results of theDutch Randomized Phase III PREOPANC TrialEva Versteijne,MD1;Mustafa Suker,MD,PhD2;Karin Groothuis,MSc3;Janine M.Akkermans-Vogelaar,BSc3;Marc G.Besselink,MD,P
2、hD4;Bert A.Bonsing,MD,PhD5;Jeroen Buijsen,MD,PhD6;Olivier R.Busch,MD,PhD4;Geert-Jan M.Creemers,MD,PhD7;Ronald M.van Dam,MD,PhD8;Ferry A.L.M.Eskens,MD,PhD9;Sebastiaan Festen,MD,PhD10;Jan Willem B.de Groot,MD,PhD11;Bas Groot Koerkamp,MD,PhD2;Ignace H.de Hingh,MD,PhD12;Marjolein Y.V.Homs,MD,PhD9;Jeanin
3、 E.van Hooft,MD,PhD13;Emile D.Kerver,MD14;Saskia A.C.Luelmo,MD15;Karen J.Neelis,MD,PhD16;Joost Nuyttens,MD,PhD17;Gabriel M.R.M.Paardekooper,MD18;Gijs A.Patijn,MD,PhD19;Maurice J.C.van der Sangen,MD,PhD20;Judith de Vos-Geelen,MD21;Johanna W.Wilmink,MD,PhD22;Aeilko H.Zwinderman,PhD22;Cornelis J.Punt,M
4、D,PhD22;Casper H.van Eijck,MD,PhD2;and Geertjan van Tienhoven,MD,PhD1for the Dutch Pancreatic Cancer GroupabstractPURPOSE Preoperative chemoradiotherapy may improve the radical resection rate for resectable or borderlineresectable pancreatic cancer,but the overall benefit is unproven.PATIENTS AND ME
5、THODS In this randomized phase III trial in 16 centers,patients with resectable or borderlineresectable pancreatic cancer were randomly assigned to receive preoperative chemoradiotherapy,whichconsisted of 3 coursesof gemcitabine,the secondcombined with 15 3 2.4 Gy radiotherapy,followedby surgeryand
6、4 courses of adjuvant gemcitabine or to immediate surgery and 6 courses of adjuvant gemcitabine.Theprimary end point was overall survival by intention to treat.RESULTS Between April 2013 and July 2017,246 eligible patients were randomly assigned;119 were assignedto preoperative chemoradiotherapy and
7、 127 to immediate surgery.Median overall survival by intention to treatwas 16.0 months with preoperative chemoradiotherapy and 14.3 months with immediate surgery(hazard ratio,0.78;95%CI,0.58 to 1.05;P=.096).The resection rate was 61%and 72%(P=.058).The R0 resection ratewas 71%(51 of 72)in patients w
8、ho received preoperative chemoradiotherapy and 40%(37 of 92)in patientsassigned to immediate surgery(P,.001).Preoperative chemoradiotherapy was associated with significantlybetter disease-free survival and locoregional failure-free interval as well as with significantly lower rates ofpathologic lymp
9、h nodes,perineural invasion,and venous invasion.Survival analysis of patients who underwenttumor resection and started adjuvant chemotherapy showed improved survival with preoperative chemo-radiotherapy(35.2v19.8months;P=.029).Theproportionofpatientswhosufferedseriousadverseeventswas52%versus 41%(P=
10、.096).CONCLUSION Preoperative chemoradiotherapy for resectable or borderline resectable pancreatic cancer did notshow a significant overall survival benefit.Although the outcomes of the secondary end points and predefinedsubgroup analyses suggest an advantage of the neoadjuvant approach,additional e
11、vidence is required.J Clin Oncol 38:1763-1773.2020 by American Society of Clinical OncologyINTRODUCTIONApproximately 20%of patients with pancreatic ductaladenocarcinoma(PDAC)have resectable or border-line resectable disease.Standard treatment is re-section followed by adjuvant chemotherapy.1,2Onlyap
12、proximately half of the patients who undergo tumorresection actually receive adjuvant chemotherapy.3-5Furthermore,approximately half of the resectionsare microscopically incomplete(R1)6,7;one quarter ofpatients will develop disease recurrence within6 months.8Preoperative(neoadjuvant)chemoradiotherap
13、yinpatients with resectable or borderline resectable PDAChas not yet been proven superior,although it isstandard of care for many other cancers.9,10Pre-operative chemoradiotherapy,by inducing down-staging of the tumor,might increase the R0 resectionrate.R0 resection rate is an important prognosticfa
14、ctor,diminishing local and distant recurrence rates,hence improving survival.In addition,compliancewith preoperative chemoradiotherapy is better com-pared with postoperative chemotherapy.PotentialASSOCIATEDCONTENTSee accompanyingEditorial onpage 1757Data SupplementProtocolAuthor affiliationsand supp
15、ortinformation(ifapplicable)appearat the end of thisarticle.Accepted onDecember 17,2019and published atascopubs.org/journal/jco on February 27,2020:DOI https:/doi.org/10.1200/JCO.19.02274Written on behalf ofthe Dutch PancreaticCancer Group.Volume 38,Issue 16 1763disadvantages of preoperative chemora
16、diotherapy are morecomplicated surgery by radiation toxicity and fewer resec-tions because of early tumor progression.Whether the latteris a disadvantage in the long run is not completely certain.Most of the studies that advocate preoperative chemo-radiotherapyarenonrandomizedstudieswithselectionbia
17、sby reporting survival after resection rather than by intentionto treat(ITT).Interpretation and comparison of thesestudies are difficult,if not impossible.Surgical radicalityafter preoperative chemoradiotherapy has never beenstudied in a multicenter randomized trial.A recent meta-analysis of the eff
18、ect of preoperative chemo(radio)therapyin resectable or borderline resectable PDAC showeda prolonged overall survival(OS)when compared withimmediate surgery(18.8 v 14.8 months).7This evidence,however,is weak because most studies were observational.Apart from ours,11 trials arereported,4of which are
19、phaseIII trials.Three trials were completed,of which 2 reportedresults(1 in abstract only)11,12;1 was prematurely closedbecause of positive results at interim analysis,13and 2 wereprematurely closed because of poor accrual.14,15Four areactive or recruiting,and 1 is of unknown status16-20(DataSupplem
20、ent,online only).The Dutch Pancreatic CancerGroup(DPCG)initiated the PREOPANC trial with the aim toinvestigate whether preoperative chemoradiotherapy pro-vides better OS than immediate surgery in patients withresectable or borderline resectable PDAC.PATIENTS AND METHODSPatients and Study DesignThis
21、randomized phase III study was performed in 16 high-volume pancreatic surgery centers from the DPCG.Theprotocol was centrally approved by the Erasmus MC ethicscommittee(MEC-2012-249;December 11,2012).Eligible patients had pathologically confirmed resectable orborderline resectable PDAC,without dista
22、nt metastases(M0),according to the Union for International CancerControl classification(TNM 7th edition).21A multiphasecomputed tomography(CT)scan of the abdomen,in-cluding noncontrast enhanced,arterial,venous,and portalcontrast phase axial scans,were required within 4 weeksbefore randomization.Tumo
23、r size,location,and relation tothe celiac axis,superior mesenteric artery(SMA)and su-perior mesenteric vein(SMV),common hepatic artery,andportal vein were reported.A tumor without arterial in-volvement and with venous involvement,90 was con-sidered resectable;a tumor with arterial involvement,90and/
24、or venous involvement between 90 and 270 withoutocclusion was considered borderline resectable.Otherinclusion criteria were a WHO performance status of#1and adequate hematologic,renal,and hepatic function.ExclusioncriteriawerecT1tumor(,2cm,withoutvascularinvolvement),history of malignancy within 5 y
25、ears,andprevious radiotherapy or chemotherapy that precludedtreatment.22Eligible patients provided written informedconsentandwererandomlyassignedbeforebiliarydrainage,which carried a risk of dropout but optimally reflects clinicalpractice,whereinimmediate surgeryispreferablyperformedbefore biliary d
26、rainage.23TreatmentPatients were randomly assigned 1:1 to preoperative che-moradiotherapy or immediate surgery.Patients assigned topreoperative chemoradiotherapy underwent a staging lap-aroscopy to rule out occult metastases.Chemoradiotherapywas to start within 4 weeks after random assignment.Pa-tie
27、nts with jaundice underwent biliary drainage,preferablywith a self-expandable metal stent;bilirubin level had tobe,1.5 times the normal limit before chemotherapywas started.Radiotherapy consisted of 15 fractions of2.4 Gy in 3 weeks to the pancreatic tumor and suspiciouslymph nodes,combined with 1,00
28、0 mg/m2gemcitabine ondays 1,8,and 15 of 4 weeks,preceded and followed bya modified course of gemcitabine(1,000 mg/m2gemci-tabine on days 1 and 8 of 3 weeks).24-26Within 4 weeksthereafter,CT evaluation was performed.Explorative lap-arotomy,with subsequent resection,if possible,was con-ductedbetween14
29、and18weeksafterrandomassignment.After resection and confirmation of PDAC,the remaininggemcitabinewasadministeredat1,000mg/m2ondays1,8,and 15 in 4 courses of 4 weeks.For patients randomly assigned to immediate surgery,preoperative biliary drainage was recommended for bili-rubin levels.250 mmol/L.Surg
30、ery was to be performedwithin 4 weeks after random assignment;staging lapa-roscopy was at the surgeons discretion.After resection andconfirmation of PDAC,patients received 6 courses of ad-juvant gemcitabine at 1,000 mg/m2on days 1,8,and 15 of4 weeks.In both groups,resection was performed according t
31、o theconsensus statement of the International Study Group onPancreatic Surgery(ISGPS).27A classic or a pylorus-preservingpancreatoduodenectomy with locoregional lymph nodedissection was performed for pancreatic head tumors.Fortumors that involved the pancreatic body or tail,pancreasbody and tail res
32、ection with splenectomy was performed.Reconstruction after pancreatoduodenectomy was left tothe surgeons preference.A standardized pathology pro-cedure,on the basis of the Leeds Pathology Protocol,wasapplied,28including description of the tumor origin,ex-tension,lymph node metastases,vascular and/or
33、 peri-neural invasion,and resection margins.Margins wereconsidered microscopically positive(R1)if vital tumor waspresentat#1mmfromthetransectionmargins(pancreas,bile duct,stomach,and/or duodenum)or the circumfer-ential dissection(the anterior and posterior sides of thepancreas,the SMA,and the SMV).2
34、9All patients underwent follow-up assessment with CT scansand serum cancer antigen 19-9(CA19.9)at 6,12,18,and1764 2020 by American Society of Clinical OncologyVolume 38,Issue 16Versteijne et al24 months after random assignment and yearly thereafter.WHO performance status,weight,disease status(locore
35、-gional and distant),death,and cause of death wereassessed at follow-up.End PointsThe primary end point was OS,defined as time fromrandom assignment to death as a result of any cause.Secondary end points were disease-free survival(DFS),locoregional failurefree interval(LFFI),distant meta-stasisfreei
36、nterval(DMFI),resection rate,R0 resection rate(per protocol),and toxicity of both surgery and pre-andpostoperative treatment.In case of missing follow-up data,patients were censored when last known to be alive anddisease free.Subgroup analyses were prespecified forresectable and borderline resectabl
37、e PDAC separatelyand for patients who underwent resection and startedadjuvant chemotherapy.Post hoc,a per-protocol analysiswas added,including data of patients who appeared tohave no distant metastases and started the intendedtreatment.In addition,a per-protocol analysis was addedthat investigated t
38、he prognostic value of R0 resection onOS.Postoperative mortality was defined as mortality asa result of any cause within 30 days after resection orduring the index hospitalization if.30 days.Postoperativecomplications were registered and graded according toISGPS guidelines.Toxicity was scored accord
39、ing toCommon Terminology Criteria for Adverse Events(ver-sion 4.0).30Statistical AnalysisThe trial was designed to have 80%power to detect a6-month difference in median OS by ITT between bothtreatment groups(17 months with preoperative chemo-radiotherapy and 11 months with immediate surgery).Atleast
40、 176 events were required to detect this(2-sided test;a-level,0.05;b-level,0.20).Assuming a 10%dropout rate,122 patients in each treatment arm were required.Primaryanalyses were performed by ITT,irrespective of any pro-tocol deviations or violations.The Kaplan-Meier curves forOS,DFS,LFFI,and DMFI(in
41、cluding the hazard ratio HRand95%CI)werecompared betweenthe2groupswiththelog-rank test(stratified for resectability resectable v bor-derline resectable).We tested differences between theresectable and borderline resectable groups with the in-teraction test of hazard rates.Moreover,for theper-protoco
42、lanalyses and the predefined subgroup analyses,the out-comes were presented as Kaplan-Meier curves and com-pared with the log-rank test(stratified for resectability).Theresectionrate,R0 resectionrate,and toxicitywerequantifiedby proportions and odds ratios and associated 95%CIs;Fishers exact test wa
43、s used to test for differences.All testswere 2-sided and performed at the 5%significance level.Allstatisticalanalyseswereperformedusingversion3.5.2oftheR statistical package(R Foundation for Statistical Com-puting,Vienna,Austria).This trial was registered withEudraCT(2012-003181-40)and the Netherlan
44、ds TrialRegister(3709).RESULTSFromApril 2013 to July 2017,248 patients from 16 centerswere randomly assigned:120 were assigned to pre-operative chemoradiotherapy and 128 to immediate sur-gery.Two patients were excluded from the analysis forwithdrawal of informed consent,which left 119 and 127patient
45、s,respectively,for the ITT analysis(Fig 1).Baselinecharacteristics were well balanced between both groups(Table 1).Seven patients in the preoperative chemo-radiotherapy group did not receive preoperative treatment,3 of whom had an urgent indication for surgery(DataSupplement).In the preoperative che
46、moradiotherapy group,5 patients(4%)had no staging laparoscopy;in 13 patients(11%),metastatic disease was found at laparoscopy(Fig 1).Afterlaparoscopy,91 patients(91 of 119;76%by ITT)startedpreoperative chemoradiotherapy,which in 10 patients waspostponed because of persistent high bilirubin levels.Ei
47、ghty-one patients(89%)completed chemoradiotherapy.Reasons for not completing chemoradiotherapy were dis-ease progression(3 patients)and toxicity(5 patients).Twopatients died as a result of a myocardial infarction duringpreoperative treatment.CT evaluation revealed diseaseprogression in 10 patients(F
48、ig 1).Explorative laparotomywas performed in 82 patients(including 7 patients whounderwent immediate surgery for several reasons),ofwhom 72 underwent a resection(72 of 119;61%by ITT).TheR0resectionratewas71%(51of72patients).Ofthese72 patients,24(33%)had pathologic lymph nodes,28(39%)perineural invas
49、ion,and 14(19%)venous invasion(Data Supplement).In the immediate surgery group,6 patients(5%)did notundergo surgery(Fig 1).Staging laparoscopy or laparotomyrevealed metastatic disease in 14 patients(12%)andunexpected locally advanced disease in 15 patients(12%).Resectionwasperformed in92patients(92o
50、f127;72%byITT).The R0 resection rate was 40%(37 of 92 patients).Ofthese 92 patients,72(78%)had pathologic lymph nodes,67(73%)perineural invasion,and 33(36%)venous in-vasion(Data Supplement).The resection rate was not significantly different betweenthe preoperative chemoradiotherapy and the immediate