Pharmacoepidemiology used for pharmacovigilance

Transcription

1 Pharmacoepidemiology used for pharmacovigilance Stéphanie Tcherny-Lessenot, MD MSc MPH Pharmacoepidemiologist Global Pharmacovigilance & Epidemiology, Sanofi R&D 23 January 2013 Châtenay-Malabry January

2 Objectives Objectives of the training are to: Provide some background on Pharmacoepidemiology (PE) Provide the current European Regulatory context relative to Pharmacovigilance (PV) Connect Pharmacoepidemiology to Pharmacovigilance 2

3 Agenda Background on pharmacoepidemiology Current European PV regulations Place of pharmacoepidemiology in pharmacovigilance Illustrative cases Conclusion 3

4 What is pharmacoepidemiology (PE)? Epidemiology is the study of the distribution and determinants of diseases in populations Pharmacoepidemiology is the study of the use of and the effects of drugs in large numbers of people Genetic Epidemiology Pharmacoepidemiology Cardiovascular Epidemiology Cancer Epidemiology Nutrition Epidemiology Environmental Epidemiology Ref: Textbook of Pharmacoepidemiology Brian L. Strom and Stephen E. Kimmel- John Wiley & Sons, Ltd 4

5 What are Epidemiological Studies? = Observational Studies = Non-interventional Studies = No treatment assignment by a third party (decision only by patient and physician) = In real life setting Main types Prospective new data collection Retrospective use of existing data Cross-sectional one shot data Epidemiologic study designs: Cohort study (nested) Case-control study Case-crossover study Main data sources Field studies: specifically performed to address specific needs, and study participants recruited for the purpose of the study Databases studies: Study subjects are identified in the existing databases where data are usually collected for administrative purpose; thus, not specific to the study 5

6 Which Pharmaco-epidemiologic study designs? Study design Descriptive (non-comparative) Analytic (comparative) Case report Case series (uncontrolled, open) Ecologic (analyses of secular trends) Non-experimental (observational) Cross-sectional * Survey Cohort Case-control * Experimental RCT ** Interventional *** ** *** Randomized Control Trial 6

7 Main designs in (pharmaco-)epidemiology Exposure cohort Outcome? + Rx Cross-sectional Exposure? Case-control Outcome 7

8 Cohort study A group or groups of individuals to be followed forward and defined on Status of exposure to a specified risk factor for a disease (e.g., drug) Status of disease presence Exposure is defined at time of cohort entry and is assessed before outcome Cohort (prospective) Outcome? + Rx Exposure Time retrospective (if existing data) Drug registry : is a cohort of patients defined by exposure to the drug Provide information about: Incidence data of outcome(s) of interest Disease course between exposed and non-exposed patients Possible delayed drug effect of newly marketed drug 8

9 Cross-sectional survey Exposure and disease status are assessed simultaneously at a specific time point among individuals in a well-defined population Cross-sectional (one snapshot) Cross-sectional Provide information about: Frequency and characteristics of a disease Health status and health care needs Prevalence of disease or other health outcomes in certain occupations Conditions of use of a drug (i.e. at treatment initiation) 9

10 Case-control study Goal is to compare cases with a disease to controls without the disease, looking for differences in antecedent exposures Exposed? Case-control (retrospective) Cases Controls Exposed? Provide information about: Multiple causes (number of exposures) of a single disease Allow to study rare diseases 10

11 Advantages & Disadvantages of PE study designs Design Advantages Disadvantages Cross-sectional Logistically easier & faster Less expensive Not always possible to distinguish whether the exposure precedes or follows the disease since exposure and disease status are measured at the same point in time No longitudinal data Cohort Can study multiple outcomes Can study uncommon exposures Selection bias less likely Unbiased exposure data Possibly biased outcome data More expensive If done prospectively, may take years to complete Case-control Can study multiple exposures Can study uncommon diseases Logistically easier & faster Less expensive Control selection problematic Possibly biased exposure data 11

12 Different databases available Medical claims or hospital administrative databases US: PharMetrics, Premier, LabRx Population-based registries Danish and Swedish registry Automated medical records UK: THIN, GPRD US: Department of Defense, GE 12

13 Agenda Background on Pharmacoepidemiology Current European PV regulations Place of Pharmacoepidemiology in Pharmacovigilance Illustrative cases Conclusion 13

14 PV & RMP Regulations are linked to crises! 14

15 Pharmaceutical regulation evolution since 50 years: toward risk management ICH E2E CIOMS VI: Management of safety information from Clinical Trials (not regulatory) All EU-RMP regulations now in VOL9A, Part I, Section 3 Regulation n 726/ art 6 Directive 2004/27/EC - art 8(3) EU - RMP Template Guidelines on Risk Management Systems Guideline on Risk Management ATMPs New EU Pharmacovigilance legislation PDUFA III Premarketing Risk Assessment Good PV Practices & PE Development and Use of Risk Minimization Action Plans FDAAA PDUFA IV (REMS) Draft REMS guidance ICH: International Conference on Harmonisation; CIOMS: Council for international organizations of Medical Sciences; ATMP: Advanced Therapy Medicinal Product; REMS: Risk Evaluation and Mitigation Strategy; PDUFA: Prescription Drug User Fee Act; FDAAA: Food and Drug Administration Amendments Act 15

16 European regulations Mar 2004: EU directive 2001/83/CE amended by the EU directive 2004/27/CE (art( 8(3)) Introduction of the inclusion of A detailed description of the pharmacovigilance and, where appropriate, of the risk management system within a Market Authorization (MA) application Dec 2004: ICH E2E Pharmacovigilance Planning (CHMP/ICH/5716/3) Intended to aid in planning pharmacovigilance activities (e.g., preparation for early post-marketing period of a new drug) Main focus on Safety Specifications and Pharmacovigilance Plan that might be submitted at the time of MA application Nov 2005: Guideline on Risk Management Systems (CHMP/96268/2005) Form of the risk management system, a RMP, should be presented to CAs Jan 2007: Vol 9A Rules Governing Medicinal Products in the European Union Legal Basis of the MAH s Obligations for Human Pharmacovigilance Nov 2008: Vol 9A updated with details on Post-Authorization Safety Studies (PASS), defined as: pharmacoepidemiological study or clinical trial carried out in accordance with the terms of a marketing authorization [ ] This includes all company-sponsored studies conducted within the EU, and those conducted outside the EU as part of a Risk Management Plan, where the investigation of safety is among study objectives. 16

17 European regulations Situations when an EU-RMP is required EU-RMP should be submitted with the application for a new marketing authorisation for Any product containing a new active substance A similar biological medicinal product A generic/hybrid medicinal product where a safety concern requiring additional risk minimisation activities has been identified with the reference medicinal product with an application for a pediatric use marketing authorization with an application involving a significant change in a marketing authorisation (e.g. new dosage form, new route of administration, new manufacturing process of a biotechnologically-derived product, significant change in indication including pediatric indication) unless it has been agreed with the Competent Authority that submission is not required On request from a Competent Authority (both pre-and post- authorisation) On the initiative of a MAA/MAH when they identify a safety concern with a medicinal product at any stage of its life cycle Refs: Guideline on Risk Management Systems (CHMP/96268/2005) incorporated in Vol 9A, Part I, Section 3 in Mar 2007; Updated Vol9A in Sep

18 European regulations New! Dec 2010: New EU PV legislation in force in July 2012 Conditional Marketing Authorisation granted with conditions and deadlines Penalties in case of non-compliance (Suspension or revocation of MA or Financial penalties) Revised & broader definition of PASS Any study relating to an authorised medicinal product conducted with the aim of identifying, characterising or quantifying a safety hazard, confirming the safety profile of the medicinal product, or of measuring the effectiveness of risk management measures MAH can be required to conduct also Post-Authorisation Efficacy (PAES) studies*: To collect data in everyday medical practice (observational) Request at any time in the product life cycle from initial MA * Ref : Amendment EU Directive 2010/84/E- 31 Dec

19 New EU PV legislation in force in July 2012 PhV Risk Assessment Committee (PRAC) New! PRAC = New scientific committee at the EMA with a key role in the pharmacovigilance assessments Scope: All medicinal products (RMP, PSUR, PASS, signal detection, Urgent Union Procedure) Replaces the CHMP Pharmacovigilance Working Party (PhWP) At the same level as the CHMP Composition: Representatives from MS with competence in PhV and risk assessment 6 independent scientific experts appointed by the EU Commission Representatives from Healthcare Professionals (1+1) and Patient associations (1+1) Decision making Final responsibility for opinion on the risk-benefit remains with CHMP (centralised products) or CMDh (other products); CHMP or CMDh should rely on the recommendations of the PRAC CHMP: Committe for human medicinal products; CMDh: Committe for mature products 19

20 New EU PV legislation Impact on Post-Marketing activities (PSURs, RMPs, PASS, PAES ) Systematic development of post-authorization non interventional (real-life) life) studies For safety (risk) as well as efficacy (benefit) => effectiveness Pharmacoepidemiology methods (databases, cohorts, registries ) Link with the ENCePP program from the EMA New regulatory process for endorsement Repetitive Benefit Evaluation Medical efficacy expertise to be maintained and resourced during the whole life-cycle Development of methods for assessing effectiveness of minimization Accurate and detailed ongoing evaluation of exposure Develop methods for drug utilization studies Proactivity in signal detection and labeling updates 20

21 Agenda Background on Pharmacoepidemiology Current European PV regulations Place of Pharmacoepidemiology in Pharmacovigilance Illustrative cases Conclusion 21

22 Safety obtained during Clinical Development is only the tip of the iceberg At the time of clinical development Benefits Risks Intrinsic Toxicity Tolerance Post-marketing surveillance Efficacy Benefits Efficacy + Usage Risks Intrinsic Toxicity Product defects Medication errors Poor compliance Risk perception Wrong population Tolerance + Condition of use 22

23 Limited safety data at time of drug approval Limited number of people exposed in RCTs Some type of patients may have never been enrolled Lack of detection of uncommon/rare or delayed adverse drug reactions (ADRs) ADRs specific to high risk populations due to misuse of the drugs by prescribers or patients Causal inference may be difficult to draw Pharmacoepidemiology can contribute information about drug safety and effectiveness in real life setting that is not available from premarketing studies 23

24 Post approval Safety data Pharmacovigilance Plan [Risk assessment] Continuous characterization of safety profile & detection of newly emerging safety signals or risks Routine pharmacovigilance Collection of ADRs Specific forms for documentation Expedited Reporting Signal detection Analysis PSURs DSURs (if applicable) Additional PV activities Additional PV activities Active surveillance (Sentinel sites (eg, OMOP), intensive monitoring schemes or ) PE studies (Case control, Cohort ) Clinical trials Nonclinical studies 24

25 Post approval Risk Minimization Activities Prevent or minimize safety risks Evaluation of the need for risk minimization Routine risk minimization: Legal status Pack size Package leaflet (PIL) Labeling (SPC, USPI, ) Additional Risk Minimization activities: Additional Risk Minimization activities: Communication plan Reminder tools Restricted prescription Drug registry 25

26 Risk Management is an iterative process! Risk assessment, Risk minimization & its effectiveness Routine: pharmacovigilance activities including PS(U)Rs : Safety Risk Assessment Additional: PV plan to identify and characterize specific risks, e.g.: PASS (i.e., pharmacoepidemiology or non-clinical investigations) Clinical studies Others Minimization Activities Drug utilization studies (e.g., using existing databases or prescription survey) Routine: labeling Additional: Communication and education (e.g., Continuous Medical Education) Reminder tools (e.g. physician brochure or patient card) Restricted prescription (e.g., specialists only prescribers) Drug registry Effectiveness Assessment 26

27 Effectiveness of minimization measures 1. Implementation of measures - Measures by increasing level of constraints - Communication plan - Reminder tools (i.e. prescriber checklist) - Restricted prescription - Mandatory drug registry - 1st assessment: success of implementation 2. Assessment of effectiveness what is the IMPACT? Final outcomes indicators Process indicators Clinical knowledge Score Clinical actions Appropriate use ADR occurrence or severity Frequency of events 27

28 In Summary: Main features of post-marketing studies Objectives Safety (risk assessment and characterization) Causal relationship for a potential risk Quantification (incidence) Preventability: risk factors, early detection, markers & surrogates Other objectives Efficacy (benefit) Use (drug utilization studies) Cost/effectiveness Real-life Real conditions of use: per labeling, distribution system, and reimbursement determined by the local health care organization Observational setting, opposed to controlled/randomized clinical trial situations Risk assessment in conditions of minimization Methods = (pharmaco)-epidemiology Epidemiology designs and methods to be developed 28

29 Some other practical aspects/questions When do we have to start studies?? For how long? Quickly after launch Usually several years, until there is enough evidence on drug use Where do we have to implement studies? A diversified set of countries (i.e. at least two European countries), and where requested or needed as per local country What is the reasonable goal to reach? Practical goals to be set, based on labeling, past experience, and literature (perfect use probaly never observed) Goals to be agreed with regulators before starting the study What do we do with study results? Submission to regulatory agencies (i.e. at time of RMP updates and/or PSUR updates) Results to be interpreted cautiously, considering limitations of data sources & bias 29

30 Agenda Background on Pharmacoepidemiology Current European PV regulations Place of Pharmacoepidemiology in Pharmacovigilance Illustrative cases Conclusion 30

31 Example of specific PV plan to investigate risk of hepatic injury and Risk Minimization Plan to minimize this risk Specific PV plan Elevation of transaminases enzymes (hepatotoxicity?) Standard PV: ADRs Enhanced PV: use of specific form for fast capture and best evaluation of severe/serious hepatic cases Pharmaco-epi studies: In databases (Pharmo in NL, Saskatchewan in Canada, GPRD in the UK, HMOs in the US) to assess risk of liver injury Clinical studies: to increase patient exposure enhancing profile understanding /Real-life pragmatic CT Mechanistic studies (in-vitro, pharmaco-genetic, Proteonomics/ metabonomics study) Hepatic surveillance programme (HSP): Voluntary although target all patients on Drug X to assess risk of liver injury Specific Minimization plan Labeling: monthly monitoring of liver enzymes (stopping rules) Contraindication Legal status (prescription by specialists only) Limitation of prescription size (30 days) Educational program: Introductory letter to prescribers, prescriber s guide, patient information brochure, prescriber s pocket card, patient card Reminder tools: prescriber check list for LFT monitoring Controlled dispensing: patients with specific genotyping not authorized to receive drug (if applicable) Effectiveness of tools (PE studies) Cross-sectionnal surveys (market research) Drug utilization study using databases Drug utilization data through HSP 31

32 Example of Nested Case Control Study Using US claims databases to evaluate severe liver injury associated with drug x Cases (N = 414) Controls (N = 40,572) Age (mean±sd), years 65.1 (11.3) 65.6 (10.8) Cases Controls Crude OR Adjusted OR (95%CI) Sex (women) 134 (32.4%) 12,887 (31.8%) CHF 187 (45.2%) 11,143 (27.5%) Drug X vs. not 184 /230 11,782 /28, [1.3, 2.2] Diabetes 139 (33.6%) 9,620 (23.7%) Hypertension 314 (75.8%) 29,187 (71.9%) Stroke 43 (10.4%) 3,210 (7.9%) AMI 37 (8.9%) 2,833 (7.0%) Tx duration of drug x (median, min, max), days 44 (0, 2274) 0 (0, 3272) Patients treated with drug X have 1.7 times higher risk of severe liver injury than those not treated with drug X Tx duration, others in class 0 (0, 2475) 90 (0, 3826) 32

33 Example of retrospective cohort using databases to evaluate severe hepatic injury associated with telithromycin Data sources: Ingenix Proprietary Research Database PHARMetrics Integrated Outcome Database Design: retrospective cohort Applications: The studies were part of the telithromycin Risk Management Plan commitment The results were presented to the FDA Advisory Committee in December

34 Crude and Adjusted Risk Ratios of Severe Hepatic Injury Crude Adjusted* Risk ratio 95% C.I. Risk ratio 95% C.I. AUG ** 1.00 N/A 1.00 N/A CLA MOX TEL * Covariates age, sex, prior history of liver disease, and Charlson Index were adjusted in the GEE model ** Augmentin (AUG) was used as a reference group in the GEE model CLA=clarithromycin; MOX= moxifloxacin; TEL= telithromycin 34

35 Example of evaluation of drug utilization using crosssectional survey Cross-sectional surveys to measure prescribers understanding of labeling recommendations and subsequent compliance to contraindications and laboratory monitoring of Drug X ,7 98,5 98,9 99, No use in CHF class IV/unstable III % ,1 49,1 43,1 64,2 No concomitant use of CYP3A4 inhibitors Creatinine planned within one week Creatinine planned within one month 10 0 Switzerland (N=273) Germany (N=369) 35

36 Example of evaluation of drug utilization using US claims database Prevalence of AF/AFL and cardiovascular risk factors (N = 1,820) N % [95%CI] Patient profile at initiation of Drug X Diagnosis of AF/AFL 1, [92.9, 95.1] Diagnosis of hypertension 1, [72.2, 79.6] Diagnosis of diabetes [18.0, 21.7] Diagnosis of stroke [5.4, 7.7] Diagnosis of AMI [3.7, 5.7] 1 risk factors* 1, [78.0, 81.7] *including hypertension, diabetes, stroke, AMI, or aged > 70 years Contraindication Prevalence of worsening / hospitalized CHF within 0-30 days before drug x prescription (N =1,820) Worsening /hospitalized CHF # of cases % [95%CI] Within 0-30 days prior to drug x prescription [3.0, 4.8] 36

37 Agenda Background on Pharmacoepidemiology Current European PV regulations Place of Pharmacoepidemiology in Pharmacovigilance Illustrative cases Conclusion 37

38 Conclusion Post-marketing evaluation methods are needed in the domain of Pharmacovigilance: To increase knowledge on drug safety in real life setting To increase knowledge on drug utilization in real life setting To address new PV regulations and increasing requirements regarding PV assessment using pharmacoepidemiology methods A specific competency in PharmacoEpidemiology is required The European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) Project led by the European Medicines Agency (EMA) Goal: strengthen the post-authorisation monitoring of medicinal products in Europe by facilitating the conduct of multi-centre, independent, post-authorisation studies focusing on safety and on benefit/risk 38

39 Thank you! Merci! Gracias! Danke! 謝謝! ありがとう! 39

40 Back-ups Châtenay-Malabry January

41 Advantages & Disadvantages of data sources Data sources Advantages Disadvantages Field studies All data specific to the study needs may be collected and recorded in a consistent way for each patients (usually based on a study protocol) General population based Timely information available for patients Regulatory requirements of general AE collection, not specific to the study Participation rates may be low, questionable representativeness Limited number of study subjects may be pratically planned for a study Selection bias: (study subjects enrolled may be different from the target population) voluntary patients patient most frequently seen in consultation «welltreated» patients More expensive and timeconsuming 41

42 Advantages & Disadvantages of data sources Data sources Advantages Disadvantages Database studies Data already collected, avoid recall bias Usually large population based data All patients in the database are included (limited selection bias) More cost-efficient and less time-consuming Lag time of data availability Data limited to what s available in the database Difficult to get a sufficient number of study subjects, if the exposure of interest is uncommon Data specific to the database population, may not be generalizable 42

43 ICH and CIOMS ICH E2E - Pharmacovigilance Planning (PVP) Tripartite harmonized guideline finalized (step 4) in November 2004 Intended to aid in planning pharmacovigilance activities (eg, preparation for early post-marketing period of a new drug) Main focus on Safety Specifications and Pharmacovigilance Plan that might be submitted at the time of license application CIOMS VI - Management of Safety Information from Clinical Trials (2005) Basis of Risk Management during development ICH: International Conference on Harmonisation; CIOMS: Council for international organizations of Medical Sciences 43

44 New EU PV legislation Extract from the original text New Article 22a After the granting of a marketing authorisation, the national competent authority may impose an obligation on the marketing authorisation holder: (a) to conduct a post-authorisation safety study if there are concerns about the risks of an authorised medicinal product. If the same concerns apply to more than one medicinal product, the national competent authority shall, following consultation with the Pharmacovigilance Risk Assessment Committee, encourage the marketing authorisation holders concerned to conduct a joint post-authorisation safety study; (b) to conduct a post-authorisation efficacy study when the understanding of the disease or the clinical methodology indicate that previous efficacy evaluations might have to be revised significantly. The obligation to conduct the postauthorisation efficacy study shall be based on the delegated acts adopted pursuant to Article 22b while taking into account the scientific guidance referred to in Article 108a. The imposition of such an obligation shall be duly justified, notified in writing, and shall include the objectives and timeframe for submission and conduct of the study.» * Ref : Amendment EU Directive 2010/84/E- 31 Dec