07Session 6 理解臨床試驗的風(fēng)險管理understanding the risk managment of clinical trials MM CN 醫(yī)療器械臨床試驗管理與統(tǒng)計分析培訓(xùn)班講義

1、第六局部：理解臨床試驗的風(fēng)險管理Institute of Executive Development State Food and Drug Administration of ChinaTraining Program of Clinical Trial Management in Medical Device DevelopmentNanjing, ChinaSeptember 28-29, 2021Mary Ann MorrisAssociate Consultant 助理參謀Global Medical Qualtiy, Eli Lilly & Company, USA 全球醫(yī)藥質(zhì)量部

2、，美國禮來制藥公司Session 6： Understanding the Risk Management of Clinical TrialsWhat is Quality?什么是質(zhì)量？ Information/Data sufficient to support the decision making process throughout the clinical development and post-market approval足以支持臨床研發(fā)全過程和上市批準(zhǔn)后決策的信息/數(shù)據(jù)Collecting data, generating information, enabling dec

3、ision making by:收集數(shù)據(jù)，產(chǎn)生信息，使得以下人員可以做出決策：Sponsors 申辦者Ethics Committees 倫理委員會Regulators 監(jiān)管者Investigators 研究者Healthcare Professionals 衛(wèi)生保健專業(yè)人員Study Subjects 研究受試者2Clinical Quality How its evolving臨床質(zhì)量 如何演變而來IDENTIFYING THE PROBLEMS鑒別問題Enforcement actions with clear messages on deficiencies缺陷清晰信息的強(qiáng)制行動Lac

4、k of confidence in data due to poor quality由于質(zhì)量不佳缺乏對數(shù)據(jù)的信心Insufficient monitoring of investigation sites不充分的研究機(jī)構(gòu)監(jiān)查Questioning if sponsors have CROs in control疑慮申辦者是否有CRO在管理Corrections failed to be systemic or effective糾正不成功系統(tǒng)或無效DISCUSSIONS ABOUT THE SOLUTIONS解決方案討論Discussions on quality and how to im

5、prove the deficiencies討論質(zhì)量和如何改善缺陷Build quality into the program構(gòu)建質(zhì)量于程序中Apply risk management to target resources appropriately應(yīng)用適當(dāng)?shù)娘L(fēng)險管理到目標(biāo)資源中Define controls to prevent errors明確監(jiān)控來預(yù)防錯誤Identify problems and intervene before issues become endemic鑒別問題并在變得嚴(yán)重前解決它3Parallel to ICH Q10, Q9, & Q8相關(guān)ICH Q10、Q9和

6、Q8Quality Management Systems (QMS Q10)質(zhì)量管理體系Not a GCP requirement but, if you have one, you are more likely to meet expectations不是GCP要求，但如果您已經(jīng)做的化，那么更可能符合預(yù)期Defines the required controls明確所要求的監(jiān)控“Say what you do, do what you say, prove it, improve it“說了您要做什么，做您所說的，證明它，改善它Quality Risk Management (QRM Q9

7、)質(zhì)量風(fēng)險管理Target resources to highest risk: data quality, data integrity, and human research subject protection高度風(fēng)險的目標(biāo)資源：數(shù)據(jù)質(zhì)量，數(shù)據(jù)完整性和人體研究受試者保護(hù)Initial efforts apply to monitoring最初的努力應(yīng)用于監(jiān)查Quality by Design (QbD Q8)設(shè)計質(zhì)量Well designed studies reduce systemic errors設(shè)計良好的研究減少系統(tǒng)錯誤Focus on protocols and case re

8、port forms重點在試驗方案和病例報告表4Principles to Assure Quality保證質(zhì)量的原那么Determine the factors that are critical to quality確定關(guān)鍵性的質(zhì)量要素Use a risk-based approach to determine where quality should be improved (i.e., where does quality matter)運用風(fēng)險管理方法來確定什么地方的質(zhì)量應(yīng)當(dāng)改善Build in quality up front, rather than inspect at the

9、 end構(gòu)建質(zhì)量在先，而不是在最后進(jìn)行視察Develop a “closed loop system to manage quality, including a feedback mechanism to check that the mitigating plans are working, and modify the risks and plans as necessary開展“封閉循環(huán)系統(tǒng)來管理質(zhì)量，包括反響機(jī)制來檢查減緩方案的實施，并在必要時改善風(fēng)險和方案5Determining Factors Critical to Quality確定關(guān)鍵的質(zhì)量要素Voice of the Cu

10、stomer客戶的聲音Quality Objectives質(zhì)量目標(biāo)Patient Safety病人平安性Data Quality / Integrity數(shù)據(jù)質(zhì)量完整性Protocol Compliance試驗方案依從性6Example: Critical to Quality例證：關(guān)鍵性質(zhì)量7Identifying & Managing Quality Risks鑒別和管理質(zhì)量風(fēng)險8Quality Risk Management in the GCP Environment環(huán)境中質(zhì)量風(fēng)險管理 Say what you do說您要做的 Do what you say做您所說的 Prove it證

11、明它 Improve it改善它9PLAN方案Prospectively define processes & responsibilities for key clinical trial activities, starting with protocol development從試驗方案開始，前瞻性地規(guī)定關(guān)鍵臨床試驗活動的程序和責(zé)任Pre-identify the procedures and data critical to reliability of study findings預(yù)先確定規(guī)程和對研究結(jié)果可靠性有關(guān)鍵性意義的數(shù)據(jù)Identify critical measuremen

12、ts 鑒別關(guān)鍵的量化點Determine how errors and deviations will be detected and prevented確定如何發(fā)現(xiàn)和預(yù)防錯誤和偏離10Identifying Risks - Questions to Ask鑒別風(fēng)險提問What could go wrong?什么可能出錯？What must go right for us to succeed?我們要成功的話什么必須做對？Where are we vulnerable?我們的弱點在哪里？How do we know whether we are achieving our objectives

13、?如何知道我們是否正在到達(dá)我們的目標(biāo)What activities are most complex?什么活動最復(fù)雜？Is there a predicate rule?有現(xiàn)有的條例嗎？11Using the Failure Modes and Effects Analysis (FMEA) to Identify and Manage Risks采用失效模式和效應(yīng)分析來鑒別和管理風(fēng)險Process for analysis of potential failure modes within a system for classification by the severity and lik

14、elihood of the failures. 在分類失敗的嚴(yán)重性和可能性系統(tǒng)中潛在失效模式分析的程序A successful FMEA activity helps a team to identify potential failure modes based on past experience, enabling the team to design those failures out of the system with the minimum of effort and resource expenditure, thereby reducing development tim

15、e and costs成功的活動有助于團(tuán)隊依據(jù)過去的經(jīng)驗鑒別潛在的失效模式，使之能用最小的努力和資源本錢排除那些失敗因素在系統(tǒng)之外，因而減少研發(fā)的時間和費用What is FMEA?什么是？12Quality Risk Identification and Mitigation Planning : FMEA質(zhì)量風(fēng)險鑒別和預(yù)防方案：Identify鑒別Failure Mode What could go wrong?失效模式什么可能出錯？Failure Effect What could happen if it did go wrong失效效應(yīng)如果出錯什么可能發(fā)生？Potential Cau

16、se What are the possible causes of the issues潛在原因問題可能的原因是什么？Assessment & Prioritization評價和優(yōu)化Severity How bad would it be if the risk happened嚴(yán)重性如果風(fēng)險發(fā)生它會糟糕到什么程度？Occurrence How frequently does the cause occur發(fā)生起因發(fā)生的頻率怎樣？Detection How easy is it to detect the issue if it occurs檢出如果發(fā)生的發(fā)現(xiàn)它的容易程度如何？13Poten

17、tial Risks Examples潛在風(fēng)險案例System Level系統(tǒng)水平Organization structures and responsibilities體制機(jī)構(gòu)和職責(zé)Quality systems of organizations體制的質(zhì)量系統(tǒng)Computer systems計算機(jī)系統(tǒng)Human resources including personnel qualifications包括人員資質(zhì)的人力資源Compliance metrics, quality audits and/or inspection outcomes依從性指標(biāo)，質(zhì)量稽查和或視察結(jié)果Project Le

18、vel工程水平Device properties器械性能Investigational product manufacturing process研究產(chǎn)品生產(chǎn)程序Safety and clinical trials平安性和臨床試驗Study budget, clinical trial site selection and management, CRO involvement, laboratory setup, study databases, safety data monitoring etc.研究預(yù)算、臨床試驗機(jī)構(gòu)選擇和管理、合同研究組織介入、實驗室建立、研究數(shù)據(jù)庫、平安性數(shù)據(jù)監(jiān)查等

19、Complexity of trial design, subject population etc.實驗設(shè)計復(fù)雜性，受試者群等14“Traditional Methods Dont Work傳統(tǒng)方法無效Whos in the room 誰在其中Experience/ “Gut 經(jīng)驗/“膽量Memory and “filters 記憶和“過濾Gaps / selective recall 差異/選擇性回憶Optimism bias 樂觀主義偏愛Isolated fixes or “one-off solutions孤立地修正或“一次性解決方案Traditional methods depend

20、 on:傳統(tǒng)方法取決于 Challenges 挑戰(zhàn): 15A New Approach.新的方法Organizational analysis method: structured multidimensional input & interdependencies體制分析方法：有方案的多方位投入和相互關(guān)聯(lián)Risk-find techniques 尋求風(fēng)險技術(shù)Risk assessment (FMEA) 風(fēng)險評價Systematic reviews of sources of risk 來源或風(fēng)險的系統(tǒng)審閱Explores unfamiliar territory 探索不熟悉的領(lǐng)域Identi

21、fies and challenges assumptions 鑒別和挑戰(zhàn)設(shè)想Integrated “holistic approach:整合“整體方法 Benefits 效益: 16Techniques for Identifying Risks鑒別風(fēng)險的技術(shù)Asking questions and questioning assumptions “5 WHYS提問和質(zhì)問設(shè)想 5個為什么“Comparison with reference standards/history與參照標(biāo)準(zhǔn)/史料比照Checklists or questionnaires 清單或問答卷Structure & pro

22、cess mapping/mindmapping構(gòu)架和程序?qū)τ?想法對映Asking questions RACI charts, governance documents提問RACI圖標(biāo)，管理文件Algorithms/ decision trees 推算/決策樹Virtual walkthrough 有效預(yù)排17Risk Identification Biggest Challenge風(fēng)險鑒別 最大的挑戰(zhàn)Quality by Design depends on reliable identification of risks in advance質(zhì)量設(shè)計取決于預(yù)先可靠的風(fēng)險識別Greates

23、t challenge in areas with little precedent of risk detection, such as:在沒有發(fā)現(xiàn)風(fēng)險先例的領(lǐng)域中的最大風(fēng)險，例如：New or reorganized structures and processes新的或重組機(jī)構(gòu)和程序Innovative projects and clinical programs創(chuàng)新工程和臨床規(guī)劃Innovative methodology 創(chuàng)新方法Changes in external environment 外部環(huán)境的變化18Quality Risk Identification and Mitig

24、ation Planning : FMEA質(zhì)量風(fēng)險識別和預(yù)防方案：FEMA Identify 識別Failure Mode What could go wrong?失效模式 什么出錯？Failure Effect What could happen if it did go wrong失效效應(yīng) 如果出錯什么可能發(fā)生？Potential Cause What are the possible causes of the issues 可能原因 問題的可能根源是什么？Assessment & Prioritization 評價和優(yōu)先排序Severity How bad would it be if

25、 the risk happened嚴(yán)重性 如果風(fēng)險發(fā)生它將會糟糕到什么程度Occurrence How frequently does the cause occur發(fā)生 原因出現(xiàn)的頻率會怎樣Detection How easy is it to detect the issue if it occurs探測 如果問題出現(xiàn)發(fā)現(xiàn)的容易程序如何19Risk Prioritization 風(fēng)險優(yōu)先排序Whats the likelihood the risk will occur?風(fēng)險會出現(xiàn)的可能性是什么？If the risk does occur, how easy can it be det

26、ected?如果風(fēng)險出現(xiàn)，它能被發(fā)現(xiàn)的容易程度怎樣？If the risk occurs, how severe is the impact?如果風(fēng)險出現(xiàn)，影響的嚴(yán)重性會怎樣？20Risk Assignment and Prioritization風(fēng)險劃分和優(yōu)先排序Tier 22級Tier 11級Tier 11級Tier 33級Tier 22級Tier 11級Tier 33級Tier 33級Tier 22級H高M(jìn)中L低 L 低 M 中 H 高Impact影響度Probability概率 Activity has risk assigned & quantified by probability

27、 of occurrence and impact if it occurs通過出現(xiàn)概率和出現(xiàn)的影響度來劃分和量化活動的風(fēng)險 Prioritization to focus on Tier 1 events first優(yōu)先排序先集中在第一級事件上21DO 進(jìn)行 Specific requirements, and responsibilities are communicated prospectively to:特殊要求和責(zé)任要預(yù)先交流給：Affected sponsor staff 相關(guān)申辦者人員CRO and service provider personnel, and合同研究組織和效勞提供商人員和Clinical investigators and their staff臨床研究者和他們的雇員Execute the plan 實施方案Protocol 試驗方案Monitoring Plan 監(jiān)查方案Other 其它22CHECK/ACT 檢查/行動Risk-based monitoring 基于風(fēng)險的監(jiān)查Verify critical activities, including QC, are