乳腺癌的临床试验终点事件的时间定义指南 —_2015
中文版NCCN指南(2015_v 2版)

T4:不论肿瘤大小,直接侵犯胸壁和/或皮肤(溃疡或 皮肤结节) N2:同侧腋窝淋巴结转移,临床表现为固定或相互融 合;或缺乏同侧腋窝淋巴结转移的临床证据,但临床 上发现有同侧内乳淋巴结转移 pN2:4~9个腋窝淋巴结转移;或临床上发现内乳淋巴 结转移,但腋窝淋巴结无转移,转移病灶>2.0 mm pN2b:临床上发现内乳淋巴结转移,但腋窝淋巴结无 转移 任何T N3:同侧锁骨下淋巴结转移伴或不伴腋窝淋巴结转移; 或临床上发现同侧内乳淋巴结转移伴腋窝淋巴结转移; 或同侧锁骨上淋巴结转移伴或不伴腋窝或内乳淋巴结 转移 N3a:同侧锁骨下淋巴结转移 N3b:同侧内乳淋巴结及腋窝淋巴结转移 N3c:同侧锁骨上淋巴结转移 pN3:≥10个腋窝淋巴结转移
原发肿瘤≤0.5 cm或 原发肿瘤微浸润 pT1, pT2,或 pT3; pN0或pN1mi(腋窝 淋巴结转移灶≤2 mm)
pN0 PN1mi
组织学类型: 导管癌 小叶癌 混合型癌 化生性癌
BINV-5
仅供内部学习使用
浸润性乳腺癌
激素受体阳性、HER-2阴性的乳腺癌患者的全身辅助治疗
原发肿瘤≤0.5 cm 或原发肿瘤微浸润 pT1, pT2,或 pT3; pN0 或 pN1mi(腋窝 淋巴结转移灶≤2 mm) 组织学类 型: 导管癌 小叶癌 混合型癌 化生性癌 pN0 PN1mi 考虑辅助内分泌治疗(2B类) 辅助内分泌治疗(2B类) ±辅助化疗(2B类)
• • •
专家普遍认为,大于10 mm的切缘属阴性(但此切缘宽度也许过大,而且可能影响美观) 小于1 mm的切缘被认为不足够 对于范围在1~10 mm之间的切缘,一般切缘越宽,局部复发率越低。但是对于位于乳腺纤维-腺分界部 位(如靠近胸壁或皮肤)的肿瘤,手术边缘不足1mm并不一定要进行再次手术,但可以对肿块切除部 DCIS-1 位进行较大剂量推量照射(2B类)。 仅供内部学习使用
2015肿瘤科大事件TOP10

• FDA根据OPTiM III期试验(包含436例患者) 的有效性结果,批准了该药的使用。该研究 纳入了436例患者,相比于直接使用GM-CSF 组的患者,T-Vec组有更多患者获得了持久 应答(16.3% vs 2.1%; P <0.001)。
谢谢大家!
ห้องสมุดไป่ตู้
T-DM1:曲妥珠单抗与细胞毒药物DM1结合
8 Nivolumab免疫疗法的几个适应症获得FDA批准
• 美国食品和药品管理局(FDA)批准Nivolumab用于治疗以铂为基础的 化疗期间或之后出现进展的转移性非小细胞肺癌(NSCLC)。 • 伴有EGFR或ALK肿瘤基因组畸变的患者,在使用Nivolumab之前,用其 他已经批准的治疗方法时很可能出现疾病进展。 • 对于肾细胞癌,Nivolumab获批准用于治 疗经过抗血管生成治疗的晚期患者。 • 对于黑色素瘤,FDA批准Nivolumab联合 Ipilimumab用于治疗BRAF V600野生型、 无法切除的或转移性的黑色素瘤。
3 癌症基因图谱(TCGA)分类把黑色素瘤分为四种亚型
• 癌症基因图谱(TCGA)给出的新数据建议,根据最常见的有意义的突 变基因,把皮肤黑色素瘤分为4个类型:BRAF、RAS、NF1和 TripleWT。 • 来自331例成年的原发性和/或转移性黑色素瘤患者的数据显示,最大 的基因组亚型分型是由BRAF热点突变的存在定义的,52%的病例存在 这种突变。 • 接下来最常见的基因组亚型分型是存在RAS热点突变(28%)和存在 NF1突变(14%)的。 • 最后,研究人员把Triple-WT型定义为没有BRAF、RAS或 NF1热点突变 的。
经历过乳房切除术的患者进行腋窝评估的概率从 2006年的56.6%上升至2012年的67.4%。
抗肿瘤药物临床试验终点技术指导原则

抗肿瘤药物临床试验终点技术指导原则一、概述临床试验终点(End Point)服务于不同的研究目的。
在传统的肿瘤药物的研发中,早期的临床试验目的是评价安全性以及药物的生物活性,如肿瘤缩小。
后期的有效性研究通常评价药物是否能提供临床获益,例如生存期延长或症状改善等。
用于支持药物批准的临床试验终点通常应当是反映临床获益的指标。
在肿瘤领域,生存期改善被认为是评估某种药物临床获益的合理标准。
在20世纪70年代,通常以影像检查或体检等肿瘤评估方法测得的客观缓解率(Objective Response Rate ,ORR)为依据批准抗肿瘤药物上市。
在随后的数十年里,逐渐认识到抗肿瘤药物的审批应该基于更直接的临床获益证据,如生存期改善、患者生活质量提高、体力状况或肿瘤相关症状减轻等。
这些临床获益很多时候并不能通过客观缓解率或与其相关的指标进行预测。
当某种药物用于治疗严重或威胁生命的疾病、对现有治疗有明显改进、或填补治疗空白时,在一定条件下可采用替代终点(Surrogate End Point)支持该药物的上市申请。
这些替代终点可能不像血压或血清胆固醇这类经过充分验证的指标,但可能能合理预测临床获益。
此种情况下,申请人必须承诺进行上市后临床试验以确证该药物的实际临床获益。
如果上市后研究不能证明该药的临床获益,或者申请人未按要求进行承诺的上市后研究,则国家食品药品监督管理局(以下简称SFDA)可将该药物从市场中撤出。
本指导原则的目的是为申请人开展抗肿瘤药物临床试验终点指标的选择提供参考,以使其符合某种药物上市申请的有效性评价要求。
本指导原则主要适用于国内、外均未上市的抗肿瘤新化合物的临床试验研究,新生物制品也可参考部分内容。
本指导原则中仅讨论用于治疗肿瘤患者的药物的终点,未讨论用于预防或降低肿瘤发生率的药物的终点。
二、关于临床试验终点的一般性考虑本节回顾了抗肿瘤药物研发中的一般性问题。
对常用的抗肿瘤药物临床试验终点进行了探讨,并对采用了这些终点的肿瘤临床试验设计中的相关问题进行了讨论。
肿瘤疗效评价标准

癌症化疗新的疗效评价标准中国医学科学院肿瘤医院冯奉仪一. 实体瘤的疗效评价标准( Response Evaluation Criteria in Solid Tumors 、RECIST ) 细胞毒化疗药是通过肿瘤缩小量耒评价其抗肿瘤作用,1979年WHO ( World Health Organization ) 确定了实体瘤双径测量的疗效评价标准。
20多年来,这个标准被国内外的研究者和研究组普遍采用,但WHO的标准存在如下问题:(1)由WHO确定可评价的和可测量大小病灶的改变混为一体,来判断疗效在各研究组间各不相同。
(2)最小病灶的大小及病灶的数量亦无明确的规定。
(3)PD 的定义在涉及单个病灶还是全部肿瘤( 可测量肿瘤病灶的总和)不明确。
(4)新的诊断病变范围的影像学方法,如CT和MRI己被广泛的应用。
因此,多年来造成了对于单个药物、联合化疗方案及治疗方法各研究组之间疗效评价存在差异而难以比较,往往导致不正确的结论。
针对以上问题,1994年EORTC ( European Organization for Research and Treatment of Cancer )、美国NCI ( National Cancer Institute ) 和加拿大NCI在回顾普遍使用的WHO疗效评价的基础上,进行了充分的交流和讨论,以后又相继的召开了多次会议,讨论和完成尚未解决的问题,直至1998年10月在包括学术界、企业、官方当局的会议上取得了一致的意见。
在WHO疗效评价标准的基础上进行了必要的修改和补充,采用简易精确的单径测量代替传统的双径测量方法,保留了WHO标准中的CP、PR、SD、PD。
RECIST首次在1999年美国的ASCO会议上介绍,并于同年的JNCI杂志上正式发表。
抗癌药物的疗效评价至少包括三个不同的目的:(1)在早期临床试验中,客观肿瘤疗效是试验药物或方案的预期目的,其结果是决定该药物或方案是否值得进一步研究的依据,体现在II期临床研究中。
晚期乳腺癌的治疗策略

2022 CSCO BC三阴性晚期乳腺癌解救治疗策略
分层
I级推荐
Ⅱ级推荐
皿级推荐
紫杉类治疗1.单药紫杉类白蛋白紫杉群 1.单引出疗卡墙他演(2A)长春 奥拉帕利(2A)紫杉群
敏感
(1A)多西他喜(2A)紫杉群 瑞演(2A)吉西他演(2A)依托泊 驱质体(2A)多柔比星
(2A)2.联合油疗TX方案 苷(28)2.联合治疗白蛋白紫杉 脂质体(28)化疗+PD-1
乳腺癌:不同分子亚型组成
晚期乳腺癌的分型治疗选择
·激素受体(ER和/或PR)阳性、疾病发展缓慢、无内脏 转移或无症状的内脏转 移的患者,首选内分泌治疗 ·HER-2阳性复发转移乳腺癌,首选抗HER2联合化疗 ·激素受体阴性、有症状的内脏转移、或激素受体阳性但 对内分泌治疗治疗 无效的患者应考虑化疗
2022中国乳腺癌诊疗指南 芳香化酶抑制剂+CDK4/6(哌柏西 利、阿贝西利)是HR+/HER2-绝经 后(自然绝经或手术 去势)或绝经前 但经药物去势后乳腺癌患者一线内 分泌治疗的优先选择
CDK4/6i的上市是乳腺癌治疗史上最具革命性的事件之一
·CDK4/6的上市是乳腺瘾治疗史上最具革命性的事件之一1 ·美国真实世界研究结果显示:2015年-2018年间将CDK4/6+内分泌治疗为HR+/HER2-转移性乳腺癌一线治疗方 案的比例从22%升高至48.7%?
晚期乳腺癌临床诊治的复杂性
转移时点的差异性 转移部位的广泛性 分子指标的多变性 治疗选择的多样性
综合治疗是晚期乳癌治疗的基本原则
ESO-MBC特别工作组建议
少数MBC患者:单发可切除转移灶,手术可能 使患者获得长 期生存的机会,应充分考虑包含 手术在内的综合治疗措施
抗肿瘤药物临床试验终点技能技术总结指导原则

精心整理一、概述临床试验终点(EndPoint)服务于不同的研究目的。
在传统的肿瘤药物的研发中,早期的临床试验目的是评价安全性以及药物的生物活性,如肿瘤缩小。
后期的有效性研究通常评价药物是否能提供临床获益,例如生存期延长或症状改善等。
用于支持药物批准的临床试验终点通常应当是反映临床获益的指标。
在肿瘤领年代,通常解率(考,以使其符合某种药物上市申请的有效性评价要求。
本指导原则主要适用于国内、外均未上市的抗肿瘤新化合物的临床试验研究,新生物制品也可参考部分内容。
本指导原则中仅讨论用于治疗肿瘤患者的药物的终点,未讨论用于预防或降低肿瘤发生率的药物的终点。
二、关于临床试验终点的一般性考虑本节回顾了抗肿瘤药物研发中的一般性问题。
对常用的抗肿瘤药物临床试验终点进行了探讨,并对采用了这些终点的肿瘤临床试验设计中的相关问题进行了讨论。
本节中将讨论的临床试验终点包括总生存期(OverallSurvival,OS)、基于肿瘤测量的终点如无病生存期(Disease-FreeSurvival,DFS)、ORR、完全缓解(CompleteResponse,CR)、疾病进展时间(TimetoProgression,TTP)、无进展生存期(Progression-FreeSurvival,PFS)和基于症状评价的终点。
抗肿瘤药物审批所用的重要临床试验终点比较见下表。
)的历的上市批准。
生存期研究实施和分析中存在的困难包括大型试验随访期较长,以及随后的抗肿瘤治疗可能会混淆生存期的分析。
(二)基于肿瘤测量的临床试验终点在本节,将讨论几种基于肿瘤测量的临床试验终点。
这些终点包括无病生存期(DFS)、客观缓解率(ORR)、疾病进展时间(TTP)、无进展生存期(PFS)和治疗失败时间(TTF)。
所有时间依赖性终点的数据收集和处理均基于间接的评价、计算或估算(如肿瘤的测量)。
关于无进展生存期数据收集和分析的讨论见附3。
当选择基于肿瘤测量的临床试验终点时,应针对该终点在抗肿瘤药物临床获益评价中的不确定性和偏倚进行评估。
乳腺癌分子分型与病理分期及预后的关系

, r 1 5 ]C h o i Y J , Hu r S Y, P a r k J S , e t a 1 . L a p a r o s c o p i c u p p e r v a g i —
・
论
著 ・
乳 腺癌 分 子分 型 与病 理 分 期及 预 后 的关 系
凌臻 栋 , 徐 红 , 孙 建 , 吕凯 荧 , 张成 刚。
( 1 . 上 海 市第一人 民 医院分 院普 外科 , 上海 2 0 0 0 8 1 ; 2 . 上 海 中医药大 学附属 岳 阳 中西 医结合 医院普 外科 , 上海 2 0 0 4 3 7 )
v a g i n a l l o n g i t u d i n a l mo r c e l l a t i o n o f b u l k y u t e r u s wi t h e n —
[ 2 ] S c h e r D, N g h i e m W, Az i z S , e t a 1 . E n d o me t r i a l s t r o ma l
检验 医 学 与 临床 2 0 1 6年 第 1 3卷 增 刊 1 1 L a b Me d C l i n , 2 0 1 6 , V o 1 . 1 3 S u p p l I I
・7 5・
巧均 会 减 少手 术 时 间 的应 用 “ ] 。本 研 究 A 组 患 者 在 手 术 过 程
s ar c o ma me t a s t a t i c f r om t he u t e r u s t O t he i nf e r i or Ve n a
2015_随机对照临床试验的注册及报告规范_蔡宏伟

减少 发表偏倚
过程
增加 结果信度
透明
减少重 复研究
临床试验注册中心
• 美国
– https:///
• 中国临床试验注册中心(WHO临床试验注册平 台一级注册机构)
– /
• 主要作用
1. 向医学科研人员和机构提供临床试验注册服务; 2. 向患者、医疗卫生人员和社会大众提供临床试验 信息的查询服务。
研究方法
• 样本量
– 7a 样本量是如何计算的
• 假设I 类错误为0.05(单侧),效度为80%,吉非替 尼的中位无进展生存期为4.2 个月,非劣效性界值为 87.5%(HR 1.14),变异系数0.5,两组的样本量比 例为1:1,计算结果为每组156 例患者。考虑到10% 的脱落率,变异系数的实际值可能超过0.5,因此本 研究的最终样本量为每组200 例患者。
Icotinb在注册信息
NCT01040780
Icotinb在 注册信息
ChiCTR-TRC-09000506
试验方案修改历史
试验方案修改记录
试验结果信息
医学是复杂的, 医学的研究对象人也是复杂的
• 在交流医学先进经验时,可能会遇到“格林斯潘式的模糊语言”
• 试验组: (埃克替尼组)
– 患者口服埃克替尼(125 mg,每天3 次),以及吉非替尼模
拟片(外观与吉非替尼片剂一致的安慰剂,每日1 次)
• 对照组: (吉非替尼组)
– 患者口服吉非替尼(250 mg,每日1 次),以及埃克替尼模 拟片(外观与埃克替尼一致的安慰剂,每日3 次)
研究方法
• 结局指标
– 受试者流程
• 13a 每组病人随机了多少例,有多少例接受了拟定 的干预措施,有多少病例纳入了主要结局的分析 • 13b 随机分组后,每组病人脱落的数量和原因
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
1© The Author 2015. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: journals.permissions@. Guidelines for time-to-event endpoint definitions in Breast cancer trials: Results of the DATECAN initiative (Definition for the Assessment of Time-to-event Endpoints in CANcer trials)S. Gourgou-Bourgade 1,*, D. Cameron², P. Poortmans 3, B. Asselain 4, D. Azria 5, F. Cardoso 6, R. A’Hern Roger 7, J. Bliss Judith 7, J. Bogaerts Jan 8, H. Bonnefoi Hervé9, E. Brain Etienne 10, M.J. Cardoso 6, B. Chibaudel 11, R. Coleman 12, T. Cufer 13, L. Dal Lago 14, F. Dalenc 15, E. De Azambuja 14, M. Debled 9, S. Delaloge 16, T. Filleron 15, J. Gligorov 17, M. Gutowski 18, W. Jacot 19, C. Kirkove 20, G. MacGrogan 9, S. Michiels 21,22, I. Negreiros Ida 23, B.V. Offersen 24, F. Penault Llorca 25,26, G. Pruneri 27,28, H. Roche 15, N.S. Russell 29, F. Schmitt 30,31, V. Servent 32, B. Thürlimann 33, M. Untch 34, J.A. van der Hage 35, G. van Tienhoven 36, H. Wildiers 37, J. Yarnold 38, F. Bonnetain 39, S. Mathoulin-Pélissier 40,41, C. Bellera 40,41, T.S. Dabakuyo-Yonli 42 1Biostatistic Unit, Montpellier Cancer Institute, Data Center for Cancer Clinical Trials, CTD-INCa, Montpellier, France2Edinburgh Cancer Research Centre, University of Edinburgh, Western General Hospital, Edinburgh, United Kingdom3Department of Radiation Oncology, Institute Verbeeten, Tilburg, The Netherlands4Institut Curie, Paris France5Department of radiation oncology, Montpellier Cancer Institute, Montpellier, France6Breast Cancer Unit, Champalimaud Cancer Center Lisbon, Portugal7Institute of Cancer Research, London, United Kingdom8EORTC Data Center (European Organization of Research and Treatment of Cancer - Statistics department), Brussels Belgium9Institut Bergonié, Comprehensive Cancer Centre, Bordeaux, France10Departments of Clinical Research & Medical Oncology, Institut Curie - Hôpital René Huguenin Saint-Cloud, France11Department of Medical Oncology - Hôpital Saint-Antoine - Paris, France12FRCP, FRCPE YCR National Institute for Health Research Cancer Research Network (NCRN), Academic Unit of Clinical Oncology, Weston Park Hospital, Sheffield Cancer Research Centre Sheffield, United KingdomAnnals of Oncology Advance Access published February 27, 2015 by guest on March 2, 2015/Downloaded from13University Clinic Golnik, Slovenia14Institut Jules Bordet, University « Libre » of Brussels, Brussels, Belgium15Institut Claudius Régaud Toulouse, France16Breast Cancer Group, Gustave Roussy Institute, Villejuif, France17APHP Tenon – University Cancer Institute – Pierre & Marie Curie, Sorbonne University, Paris, France18Department of surgery, Montpellier Cancer Institute, Montpellier, France19Department of medical oncology, Montpellier Cancer Institute, Montpellier, France20Université catholique Louvain, Belgium21Gustave Roussy, Biostatistic and Epidemiology Unit, Villejuif, France22University of Paris-Sud, Villejuif, France23Breast Unit, Hospital CUF Descobertas, Lisbon, Portugal24Department of Oncology, Aarhus University Hospital, Aarhus, Denmark25Centre Jean Perrin, Clermont-Ferrand, France26University of Clermont-Ferrand 1, ERTICA EA4677, UFR medicine, Clermont-Ferrand, France27European Institute of Oncology, Milan, Italy28University of Milan, School of Medicine, Milan, Italy29Department of Radiotherapy, The Netherlands Cancer Institute - Antoni van Leeuwnhoek Hospital, Amsterdam, The Netherlands30IPATIMUP (Institute of Molecular Pathology and Immunology of the University of Porto), Porto , Portugal31Medical Faculty of Porto University, Porto, Portugal32Oscar Lambret Comprehensive Cancer Center, Lille, France33Kantonsspital St.Gallen, Breast Center ; St.Gallen, Switzerland34Clinic for Gynecology, Gynecologic Oncology and Obstetrics- Interdisciplinary Breast Cancer Center, HELIOS Klinikum Berlin-Buch, Berlin, Germany35Department of Surgical Oncology, The Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital Amsterdam, The Netherlands by guest on March 2, 2015 / Downloaded from36Academic Medical Center Amsterdam, the Netherlands37Department of General Medical Oncology, University Hospitals Leuven, Leuven Cancer Institute, Leuven, and Laboratory of Experimental Oncology (LEO), Department of Oncology, KU Leuven, Belgium38The Institute of Cancer Research: Royal Cancer Hospital, London United Kingdom39Methodological and Quality of Life Unit in Oncology (EA3181), CHU Besançon, France40Clinical and Epidemiological Research Unit, Institut Bergonié, Comprehensive Cancer Centre, Bordeaux, France41Clinical Epidemiology Unit, INSERM CIC 14.01 (Clinical Epidemiology), Bordeaux, France 42Biostatistics and Quality of Life Unit, EA4184, Centre Georges François Leclerc Comprehensive Cancer Centre, Dijon, France*Corresponding author : Gourgou-Bourgade Sophie, MSc, Biostatistic Unit, Montpellier Cancer Institute France; Data Center for Cancer Clinical Trials, CTD-INCa, Montpellier, France; Tel: +33 (0)4 67 61 37 75; fax: +33 (0)4 67 61 37 18, Sophie.gourgou@icm.unicancer.frThis work was presented as a poster communication at the following meetings: ASCO (Chicago,June 2013), ECCO/ESMO (Amsterdam, Sept 2013) and SABCS (San Antonio Breast Cancer symposium, San Antonio, Dec 2013).Key Message: "The DATECAN initiative (Definition for the Assessment of Time‐to‐event Endpoints in CANcer trials) aims to provide recommendations for definitions of TTE endpoints using a formal consensus process for multiple cancer sites: breast; sarcomas/GISTs; pancreas; stomach/esophagus; head and neck; colon/rectum; kidney/bladder and lung cancers. Here, we report guidelines for RCTs in breast cancer." by guest on March 2, 2015 / Downloaded fromAbstractBackground: Using surrogate endpoints for overall survival, such as disease-free-survival(DFS) is increasingly common in randomized controlled trials (RCTs). However, the definitions of several of these time-to-event (TTE) endpoints are imprecisely which limits interpretation and cross-trial comparisons. The estimation of treatment effects may be directly affected by the definitions of endpoints. The DATECAN initiative (Definition for the Assessment of Time-to-event Endpoints in CANcer trials) aims to provide recommendations for definitions of TTE endpoints. We report guidelines for RCTs in breast cancer.Methods: A literature review was performed to identify TTE endpoints (primary or secondary) reported in publications of randomized trials or guidelines. An international multidisciplinary panel of experts proposed recommendations for the definitions of these endpoints, based on a validated consensus method that formalize the degree of agreement among experts.Results: Recommended guidelines for the definitions of TTE endpoints commonly used in RCTs for breast cancer are provided for non-metastatic and metastatic settings.Conclusions: The use of standardized definitions should facilitate comparisons of trial results and improve the quality of trial design and reporting. These guidelines could be of particular interest to those involved in the design, conducting, reporting or assessment of RCTKey words (6): guidelines, randomized controlled trial, time-to-event endpoint, efficacy measure, breast cancer by guest on March 2, 2015/Downloaded fromINTRODUCTIONIn randomized cancer clinical trials (RCTs), the validated and most objectively defined evaluation criterion is overall survival (OS), characterized as the time from randomization to patients’ death (all causes). The development of new cytotoxic agents, the current context of strategic trials and the multiplication of lines of treatment, especially in breast cancer, have significantly reduced mortality in certain contexts. This therapeutic progress has resulted in the need for surrogate endpoints and/or intermediate endpoints for OS. Such endpoints are being increasingly used in cancer RCTs. Thus, disease-free survival (DFS) and progression-free survival (PFS) have been used as surrogate endpoints of OS in non-metastatic and metastatic settings, respectively. These surrogate endpoints are gradually replacing OS (1)and their development has been strongly influenced by the need to reduce the number of patients taking part in RCTs, as well as the duration and, ultimately, the cost of RCTs.As recommended by the International Conference on Harmonisation (ICH) guidelines (2) and by the CONSORT statement (3), each time-to-event (TTE) endpoint should be precisely defined. It implies specifying the date of origin, the list of events to be considered, such as failures, and the censoring process. However, despite their extensive use, most TTE endpointsare often poorly defined, and when a definition is provided, it varies from one publication to another as underlined by recent study published in the Journal of clinical oncology (4) and by the Food and Drug Administration (FDA) (5). As an example, in a review of RCT in oncology, Mathoulin-Pélissier et al. showed that a clear definition of survival end-points was reported for only 52% of cancer RCTs published in major journals (4). The heterogeneity of definitions for TTE endpoints was recently highlighted by the international community, as demonstrated by all the publications recommending the definition of specific criteria and/or the preferred use of certain criteria in specific localizations such as for colorectal cancer in the adjuvant setting (6), hepatocellular carcinoma (7), lymphoma (8) or breast cancer (9). However, most of these recommendations were usually based on experts’ opinions, without formal international consensus process and without representation of academic groups in the selected panels of experts, facts that may explain why they have not been widely accepted in current practice.It is important to distinguish the process of selecting a relevant endpoint from the action of defining this same endpoint. The selection of TTE endpoints to assess a therapeutic strategy by guest on March 2, 2015 / Downloaded fromdepends on the characteristics of a given trial including settings (adjuvant versus metastatic) and treatments (systemic, local, or any combination thereof). As such, the choice of the endpoints is trial-specific. Once the endpoint is identified, it then has to be appropriately defined, ideally using a standardized definition to enable future comparisons.Using a formal consensus process, we set up the international DATECAN initiative(D efinition for the A ssessment of T ime-to-event E ndpoints in CAN cer trials)(10), which aimed to obtain standardized consensus definitions of TTE endpoints for multiple cancer sites: breast; sarcomas/GISTs; pancreas; stomach/esophagus; head and neck; colon/rectum; kidney/bladder and lung cancers. Here, we report guidelines for the definition of TTE endpoints used in breast cancer RCT as primary or secondary endpoints.METHODSThe DATECAN project was launched in 2010 regarding three cancer sites: breast, sarcoma/GIST and pancreatic cancer. The coordinating committee (CC) for the breast cancer part of the project included two experts (SGB, SDY).Consensus processA formal consensus method was used to develop these guidelines(11),(12). Its purposewas to formalize the degree of agreement among experts using iterative ratings with feedbacks to identify and select points on which there was either disagreement or uncertainty. The guidelines were subsequently based on the agreement scores. The formal consensus method involved the following steps (Figure 1): (I) the assessment of the evidence with regards to the research question; (II) the elaboration followed by the pre-testing of the questionnaire before collecting experts’ opinions; (III) the scoring of the questionnaires; (IV) the analysis of the experts’ opinions and the drafting of the final report; (V) the peer-review step; (VI) the diffusion of the recommendations. An overview of these steps is provided in Online Supplementary Material 1. A full description of the methodology of the consensus process has been published in a former study (10). by guest on March 2, 2015 / Downloaded fromLiterature reviewWe conducted literature reviews to assess the development of guidelines for TTE endpoints and listed TTE endpoints commonly reported in RCT, either as primary or secondary endpoints. The research algorithms used are available in the Online Supplementary Material 2.QuestionnairesAll experts had to fill in a first questionnaire: they all received the same questionnaire to score each TTE endpoint on a scale of 1 (totally disagree) to 9 (totally agree), according to whether various clinical events should be regarded as events in the definition of TTE outcomes. After the first round, the second questionnaire was personalized for each expert (Online Supplementary Material 3).Items for which a strong consensus had been reached after the first questionnaire were highlighted. For items with no consensus found, the distributions of scores obtained during the first round were summarized (minimum, maximum, and median scores) and the initial score of the RC expert was indicated. In second questionnaire, experts were asked to re-score the items for which no consensus had been reached in first round.RESULTSSelection of time-to-event endpoints to be defined and clinical events of interestEleven endpoints were selected from the literature, according to the trial setting (non-metastatic (9) or metastatic (2)) and for which no consensus methodology in adjuvant settings had been described (9). Clinical events, which could be included in the definition of these endpoints, were identified (Table 1).Experts for the scoring processThe CC drafted a list of 50 experts from European countries to include in the rating committee, meeting twice. Because of an over-selection of French experts at the first selection step, we opened the scoring process to additional experts from other Europeans countries. Out of the 35 experts who filled-in the first questionnaire, 31 (89%) also answered the second questionnaire. They were specialists in medical oncology (n=14; 45%), radiation oncology (n=4; by guest on March 2, 2015 / Downloaded from13%), surgery (n=4; 13%), pathology (n=4, 13%) and methodology/biostatistics (n=5, 16%) (Online Supplement Material4). These experts worked in institutions from many countries, France, Belgium, United-Kingdom, Netherlands, Portugal, Italy, Germany, Denmark, Slovenia and Switzerland. They belonged to various cooperative groups including breast cancer groups from the EORTC (European Organization for Research and Treatment of Cancer), UNICANCER (French Group of Comprehensive Cancer Centers), the American Society for Radiation Oncology (ASTRO), the French Society for Radiation Oncology (SFRO) and from the National Cancer Research Institute breast cancer group of The United-Kingdom (NCRI), the Dutch breast cancer Trialists’ Group (BOOG) and the Italian breast cancer group.Consensus rates after the two rounds of ratingThe consensus process was constituted of two rounds of rating (first round: January 2011 to January 2012; second round February 2012 to May 2012) and one face-to-face meeting (June 3rd 2012, Chicago). During the first round, experts were asked to rate a total of 150 events pertaining to 11 TTE endpoints. At that step, a strong consensus was reached for six (4%) of the events for two endpoints. After the second round, a strong consensus was reached for 60 events (42%), and strong or relative consensus was reached for 80 events (56%). After the two steps ofrating process, no consensus was reached for 70 events covering the 11 endpoints. Those events were thus discussed at the face-to-face meeting.Face-to-face meetingBefore discussing each event on an individual basis and with the aim of harmonization, experts present at the meeting first took a number of decisions. The aim of this preliminary discussion was triple: to select the items that would be discussed, to maintain the consensus process even though not all experts attended the meeting and some experts were therefore unavailable for discussion, and to decide the procedure to adopt in case of an absence of consensus. by guest on March 2, 2015 / Downloaded fromStandardized definitions of the time-to-event endpointsTable 2 lists the events that need to be included in the definition of each TTE endpoint following the consensus process, depending on the disease setting (non-metastatic, metastatic or both settings).Among the 11 endpoints initially listed, one was considered ambiguous (disease-free survival, DFS) and renamed by the experts as Invasive disease-free survival, iDFS.Two definitions for two different endpoints were identical, distant disease-free survival (DDFS) and distant relapse-free survival (DRFS).The reference date was usually the date of randomization, but it could also be the date of diagnosis or treatment initiation, depending on the study. Endpoints were defined according to the setting. Nine endpoints were specifically defined for the non-metastatic setting: breast cancer specific survival (BCSS), invasive disease-free survival (iDFS), Distant disease-free survival (D-DFS), Distant relapse-free survival (D-RFS), Relapse-free survival (RFS), Loco-regional Relapse-free survival (L-RFS), Recurrence-free interval (RFi), Breast Cancer-free interval (BCFi) and Distant recurrence-free interval (D-RFi). Two endpoints were designed for the metastatic setting, Progression-free survival (PFS), and Time-to-progression (TTP).Validation of the guidelines and peer-reviewThe minutes of the face-to-face meeting, which included the final guidelines, were validated by email by all the 31 participating experts and then submitted to a peer-review group for external comments. This group provided a formal, advisory opinion on the content and form of the initial version of the guidelines in particular their applicability, acceptability, and readability.DISCUSSIONIn RCTs, standardized definitions of the TTE endpoints should be adopted to enable consistent interpretation of trial results and facilitate cross-trial comparisons and meta-analyses. Clinical trial endpoints often refer to efficacy, adverse events or quality of life. In early breast cancer, effective methods of early diagnosis and new treatments have led to a longer expected survival for patients. Therefore it is not convenient to use OS as primary endpoint for many clinical trials, especially those conducted in a non-metastatic setting. Many other endpoints have by guest on March 2, 2015 / Downloaded frombeen used to accelerate the development of news drugs and treatments, but with a lack of consistency between the different definitions used in the studies. It is therefore necessary to standardize endpoints to ensure the uniformity of data collection among the studies. This will make trials more useful and facilitate their implementation.A given endpoint should always encompass the same set of events, as clearly highlighted by Hudis et al.(9). However, though OS has been recognized as the least ambiguous and the most clinically relevant endpoint in cancer clinical trials, the other endpoints often used as secondary endpoints need to be standardized. For example, in two adjuvant clinical trials assesing the efficacy of aromatase inhibitors, the event “second primary invasive non-breast cancer” was included for the same primary endpoint in the first trial (13) but not in the second (14). This raises the possibility that a treatment could be declared effective or inefficient depending on the definition used for the endpoint of the study. Birgisson et al.(15) and Nout et al.(16) showed in colorectal (15) and breast cancer(16) patients, respectively, that defining properly TTE endpoints is a central issue when designing trials since it may affect the estimation of treatment effect, the statistical power and thus the final interpretation of the trial, as highlighted for respectively .For breast cancer, Hudis et al.(9) proposed standardized definitions for many endpointsused in the adjuvant setting. However, these definitions of the TTE endpoints were based on recommendations made by an expert group in the absence of any formal international consensus method which therefore limited their use and acceptability in current practice. Moreover, to our knowledge, no definition of TTE endpoints has so far been proposed for metastatic breast cancer. We thus decided to use a formal consensus methodology for the consensus process that resulted in the elaboration of standardized definitions and recommendations regarding 11 TTE endpoints specifically for breast cancer clinical trials. Our initial list of TTE endpoints was established following selected endpoints from published recommendations by an expert group or a literature review of recent randomized trials.After the first round of rating, which involved 35 international experts from various medical specialties, a strong consensus was reached for only 4% of the items. After the second round of rating, the extremely low initial consensus rate went up to a strong or relative consensus for 80 events (56%). The lack of initial consensus highlighted the disparity of expert opinions and the need for harmonization despite the definitions already available for the adjuvant setting by guest on March 2, 2015 / Downloaded from(Hudis et al. for the STEEP group) (9). The main cause of the improvement in the consensus rates may be related to the design of the consensus process. The formal consensus process aimed to guide experts into taking position, while allowing them to maintain their opinion at each scoring round. Another reason for such an improvement may be the different rules used to define the consensus after each round: rules for the second round were slightly relaxed to ensure that the systematic exclusion of a proposal by a rater would block the consensus process (10;11).Moreover, the initial low consensus rate might probably also be due to misinterpretation of the text and tables. Clarification of the remaining issues at the face-to-face meeting led to a consensus for all the events except one, the “death related to protocol treatment” event, in the definition of the endpoint “breast cancer-specific survival”.The consensus process also underlined that the non-relevance of some of the endpoints. Indeed, during the face-to-face meeting, the experts considered DFS irrelevant in the context of breast cancer. They suggested using “invasive disease-free survival” instead of DFS, from in situ carcinoma that should be excluded, as recommended also by Hudis et al. One surprising result after the face-to-face meeting was the conclusion that the definitions for two different endpoints, distant disease-free survival (DDFS) and distant relapse-free survival (DRFS) should be identical. These results also questioned the -relevance of some endpoints. The list of outcomeswas based on our literature review with the objective to propose a large panel of definitions, so that researchers will find a standardized definition for the outcome that best suits the objective of their study. As DDFS is more often used that DRFS, in future trials, we recommend that DDFS should be preferred to DRFS and that a precise definition of DRFS should be given in trials that will use this endpoint.For “invasive contra-lateral breast cancer” and “second primary invasive cancer (non-breast cancer)” events, the experts adopted a conservative approach and assumed that these two events were new primary cancers unless it was proven that they were a metastasis of the studied disease. These two events were therefore not included in any endpoint except “breast cancer-free interval” for invasive contra-lateral breast cancer and DFS (invasive DFS) for “second primary invasive cancer (non-breast cancer)”.Some recommendations were made during the face-to-face meeting. First, lost-to-follow-up should not be included in any of the previously described endpoints, and should be censored at the time of the status last known. Second, all deaths, whatever the cause, should be considered by guest on March 2, 2015 / Downloaded fromevents for “survival endpoints”, except for cancer-specific survival in which “death from breast cancer” was proposed for inclusion. No consensus was reached for “death related to protocol treatment”, probably due to the difficulty to define precisely a death “related to protocol treatment”. Regarding the lack of consensus for the event “death related to protocol treatment”, a recommendation was made that, in future trials, this endpoint should be clearly defined by referring to all the events that it includes.One possible weakness of the consensus process is that it did not take into account rules relating to the censoring process. In this study, we have chosen to focus on recommendations about the definition of the endpoints and not on the recommendations about the censoring process or on the data collection procedures. When a clinical event is not included in a definition, it can be censored, ignored or accounted for (using competing-risk analysis) in the statistical analysis, and the selected method will be study-specific depending on the objectives. Providing guidelines for events to be censored or ignored at the analysis stage is not straightforward and requires both the censoring / ignoring of events in each trial and the precise definition of the impact of the censoring process on the estimation of the treatment effect(16). We deliberately did not address the issue related to the data collection procedures (evaluation schedules and criteria) to achieve the actual calculations of the endpoints defined in the manuscript with a uniformfashion. However, the measurement tools (such as the surveillance schedule, the imaging techniques, etc.) must be defined by the study while accounting for the evolution of the conceptual elements to be included in the definitions of the endpoints. Just like the development of standardized definitions, these issues could be addressed with consensus processes using different independent trials’ scenarios (disease, setting and treatment).Comparison of the guidelines for the definitions of TTE endpoints commonly used in RCTs for breast cancer with those proposed for sarcoma/GIST and pancreatic cancers showed that similar definitions were used for all the TTE endpoint. For example, the three most commonly used TTE endpoints, PFS, TTP and DFS, share exactly the same definitions across these tumor sites. Some slight differences did appear, however. First, some TTE endpoints can be defined for a specific cancer type. For example, disease-specific survival was proposed only for sarcoma/GIST, while pancreas and breast cancer guidelines defined a cancer-specific-survival (CSS). Similarly, time to deterioration in quality of life was defined for pancreas cancer but not for sarcoma/GIST and breast cancers. Second, a TTE endpoint can include the same by guest on March 2, 2015 / Downloaded fromevents, but be named differently, e.g. “distant metastasis-free survival” wich is called “distant-DFS” in breast cancer.The major strength of our study lies in the formal consensus methodology used, which resulted in the elaboration of standardized definitions and recommendations regarding 11 TTE endpoints designed specifically for breast cancer RCT. The participation of many international experts in the consensus process increased the likelihood that these recommendations for early and metastatic breast cancer would be accepted by the scientific community, and as such contribute to their generalizability, their acceptability and their wide-scale implementation in future research in breast cancer.CONCLUSIONThe DATECAN initiative was set up with the objective to provide guidelines for standardized definitions of TTE endpoints in RCTs for different cancer sites, including pancreatic and sarcoma/GIST cancers, for which guidelines have already been finalized as well. Here, we have provided consensus definitions for the most commonly used endpoints in breast cancer clinical trials in both early and metastatic settings. The availability of these guidelines should improve international comparisons of trial results as well as meta-analyses. Theserecommendations should be disseminated for acquisition and endorsement by researchers and academic groups participating in clinical research. In addition, these guidelines should be of interest to other potential users including reviewers and editors of scientific journals, who have recently shown increased interest in the quality of the reporting of clinical trials (4)(15), regulatory authorities (5), and any research scientist body interested in improving outcome measurements and reporting of clinical trials (17). Progress in this field also depends on the collection and publication of detailed data on the distinct clinical events that contribute to the TTE endpoints.The future perspectives of the DATECAN initiative and ongoing work include extending the consensus procedure to other cancer sites (current project in stomach/oesophagus, kidney, bladder, head and neck, colon and lung cancers), assessing the impact of these definitions on academic cancer RCTs, evaluating the statistical properties of the newly defined intermediate endpoints (in particular PFS and DFS) as surrogates for overall survival and extending the consensus to other countries (North America, India, South-east Asia and Australia) for a by guest on March 2, 2015 / Downloaded from。