evaluation of the efficacy and cost of two monitoring ... - protoc… · for public clinical...

Evaluation of the efficacy and cost of two monitoring strategies for public clinical research

© Bordeaux CHU of 50

Evaluation of the efficacy and cost of two monitori ng strategies

for public clinical research. OPTIMON study: OPTImisation of MONitoring

Financed by the French national hospital clinical research programme (PHRC) – project accepted 10/10/2005

Version dated 8 February 2008

Sponsor Centre Hospitalier Universitaire de Bordeaux 12 rue Dubernat 33400 Talence, France Tel: 05 56 79 47 98 - Fax: 05 56 79 49 26

Coordinating investigator – Coordinating centre Geneviève Chêne USMR, CHU de Bordeaux - INSERM CIE 7 et U593 bâtiment ISPED – case 11 Université Victor Segalen Bordeaux 2 146, rue Léo Saignat 33076 Bordeaux cedex, France

Editorial committee C. Alberti INSERM CIE 4 E. Bellissant Unité de Biostatistique, CHU de Rennes - Université de Rennes 1 - INSERM CIC 0203 J. Bénichou Unité de Biostatistique, CHU de Rouen - INSERM U657 F. Carrat INSERM U707 G. Chatellier URC Hôpital Européen Georges Pompidou, AP-HP - INSERM CIE 5 D. Costagliola INSERM U720 J. Demotes-Mainard INSERM CIC 05 - European Clinical Trials Infrastructures Network F. Guillemin INSERM CIE 6 A. Leizorovicz EA 3736, Faculté RTH Laënnec, Université de Lyon 1 J.P. Pignon Service de Biostatistique & Epidémiologie, Institut Gustave Roussy P.M. Preux UF Recherche Clinique & Biostatistique, CHU de Limoges - EA 3174, Université de Limoges O. Rascol INSERM CIC 9302 P. Ravaud URC Bichat Hospital, AP-HP, Paris - INSERM U738 J.M. Tréluyer URC Cochin Hospital, AP-HP, Paris

Suggested citation

Chêne G, Alberti C, Bellissant E, Bénichou J, Carrat F, Chatellier G, Costagliola D, Demotes-Mainard J, Guillemin F, Leizorovicz A, Moore N, Pignon JP, Preux PM, Ravaud P, Tréluyer JM, for the OPTIMON study group. Evaluation of the efficacy and cost of two monitoring strategies for public clinical research. OPTIMON trial: OPTImisation of MONitoring of clinical research studies [protocol not published]. Bordeaux CHU (University Hospitals) France: OPTIMON study group; 2005.



MAIN CORRESPONDENTS

Coordinating investigator Prof. Geneviève Chêne USMR bâtiment ISPED – case 11 146 rue Léo Saignat 33076 BORDEAUX Cedex, France Tel. : 05 57 57 12 57 Fax: 05 57 57 11 72 [email protected]

Coordinating centre USMR bâtiment ISPED – case 11 146 rue Léo Saignat 33076 BORDEAUX Cedex, France Tel: 05 57 57 11 29 Fax: 05 57 57 15 78 [email protected]

Web site:

https://ssl2.isped.u-bordeaux2.fr/optimon/

Sponsor Centre Hospitalo-Universitaire de Bordeaux 12 rue Dubernat 33400 Talence, France Tel: 05 56 79 47 98 Fax: 05 56 79 49 26

- Project leader - Statistician Valérie Journot Tel. : 05 57 57 92 45

- Clinical Research Associate Laure Lallemand Tel. : 05 57 57 14 39

- Data Manager Angéline André Tel. : 05 57 57 11 35

- Secretary Karine Surlanne Tel: 05 57 57 11 29



CONTENTS 1. Summary .................................................................................................................................5 2. Justification .............................................................................................................................6

2.1. The nature of the problem .......................................................................................................... 6 2.2. Current state of knowledge......................................................................................................... 7 2.3. Hypothesis of the study................................................................................................................ 8 2.4. Justification of choices made for this study............................................................................... 8

2.4.1. Procedures compared ............................................................................................................................ 9 2.4.2. Judgement criterion for the comparison ................................................................................................ 9 2.4.3. Outline of the study............................................................................................................................... 9 2.4.4. Unit of randomisation............................................................................................................................ 9

3. Objectives ..............................................................................................................................10 3.1. Main objective............................................................................................................................ 10 3.2. Secondary objectives.................................................................................................................. 10

4. Study plan and randomisation ............................................................................................10 4.1. Study plan................................................................................................................................... 10 4.2. Randomisation ........................................................................................................................... 11

5. Eligibility criteria..................................................................................................................11 6. Methods compared ...............................................................................................................13

6.1. Classic monitoring strategy....................................................................................................... 13 6.2. Optimised monitoring strategy................................................................................................. 14

6.2.1. Definition of the optimised monitoring strategy by the methodology and management centre.......... 15 6.2.2. Validation of the optimised monitoring strategy by the OPTIMON validation committee ................ 15 6.2.3. Optimised monitoring strategy............................................................................................................ 15

7. Collecting OPTIMON data..................................................................................................17 7.1. Updating data from the clinical research study ...................................................................... 17 7.2. Transmission of the database from the clinical research study to the OPTIMON team .... 17 7.3. OPTIMON on-site visits............................................................................................................ 17 7.4. Central monitoring procedures in the methodology and management centres ................... 17

8. Judgement criteria................................................................................................................18 8.1. Main judgement criterion ......................................................................................................... 18 8.2. Secondary judgement criteria................................................................................................... 18

9. Statistical aspects ..................................................................................................................19 9.1. Size of the study.......................................................................................................................... 19

9.1.1. Review of the information concerning the proportion of serious errors in clinical research studies... 19 9.1.2. Calculation in a situation of non-inferiority: principles, hypotheses and results................................. 19 9.1.3. Calculation taking into account the clusters ........................................................................................ 20

9.2. Interim analysis.......................................................................................................................... 20 9.3. Analysis strategy ........................................................................................................................ 20 9.4. Statistical methods ..................................................................................................................... 21

9.4.1. Analysis of the proportion of patients whose observation contain no serious error ............................ 21 9.4.2. Cost ..................................................................................................................................................... 21

10. Surveillance of the trial ........................................................................................................21 10.1. Scientific committee ............................................................................................................... 21 10.2. Independent Surveillance Committee .................................................................................. 21 10.3. OPTIMON Validation Committee ....................................................................................... 22

11. Practical organisation of the trial........................................................................................22 11.1. General organisation.............................................................................................................. 22 11.2. Web Site and Instructions ..................................................................................................... 22 11.3. Confidentiality of data ........................................................................................................... 23

12. Ethical and regulatory aspects ............................................................................................23 13. Expected benefits ..................................................................................................................23 14. Références..............................................................................................................................24 15. Annexes..................................................................................................................................26



Annexe 1. Grille OPTIMON évaluant le risque patient dans les études de recherche clinique. 27 Annexe 2. Plan de monitorage général OPTIMON. ...........................................................33 Annexe 3. Accord de participation à OPTIMON Promoteur d'une étude de recherche clinique. 38 Annexe 4. Accord de participation à OPTIMON Investigateur Coordonnateur d'une étude de recherche clinique. .......................................................................................................41 Annexe 5. Accord de participation à OPTIMON Investigateur pri ncipal d'un site investigateur. 44 Annexe 6. ......................................................................................................................................47 Décision du Comité de Protection des Personnes Sud Ouest Outre-Mer. .............................47 Annexe 7. Note d’information OPTIMON..........................................................................48

NOTE D’INFORMATION OPTIMON.............................................................................................. 48 Annexe 8. Liste des Centres de Méthodologie et de Gestion volontaires pour participer à l'essai OPTIMON. .......................................................................................................................49 Annexe 9. Procédure d'anonymisation complète de l'identité du patient ........................50



1. Summary

Title: Evaluation of the efficacy and cost of two monitoring strategies for public clinical research. OPTIMON trial: OPTImisation of MONitoring of clinical research studies.

Coordinating investigator: Prof. Geneviève Chêne, Bordeaux University Hospitals, INSERM CIE 7 and U593.

Context: The choice of monitoring strategy for a clinical research study is usually a compromise which takes into account the resources available, the potential benefit/risk ratio for all those taking part in the study and the anticipated use of the results. Under the regulations in force, the sponsor is responsible for monitoring but often delegates it to methodology and management centres. The current heterogeneity of practice, the amount of finance devoted to each study and European discussions on the creation of similar quality standards for industrial and academic trials has led to questions being asked about the effectiveness of monitoring activities in an academic context.

Main objective: The aim of our study is to compare the efficacy of two monitoring strategies: one based on classic quality assurance standards, the other "optimised" a priori on the major scientific and regulatory principles.

Outline: A non-inferiority study comparing two monitoring strategies, where the unit of randomisation will be the investigation site for multicentre clinical research studies. The studies will be stratified according to the level of risk run by the patient, assessed according to the risk scale.

Main eligibility criteria: • Methodology and Management Centre: Must come under the auspices of an institution (INSERM

[French National Institute for Health and Medical Research], university hospital, French National Federation of Anti-Cancer Centres, Institut Pasteur, etc.), have at least 2 years experience in multicentre studies, standard operating procedures finalised before the start of the OPTIMON trial, and volunteer to take part in the OPTIMON trial.

• Clinical research studies: Must be classified in level A, B or C by the scale, whether or not requiring a sponsor and the approval of an ethics committee, must be in the active inclusion period or beginning the inclusion of patients during the year, must be clinical research studies for which the data of patients included in OPTIMON are entered and checked regularly throughout their duration and for which the data of patients included in OPTIMON will be ready to be definitively frozen within three years from the start of the OPTIMON study, studies which have planned to include at least 20 patients, are multicentre studies including at least 4 investigation sites, have planned to collect data in an observation file on paper or in electronic form, and have obtained the agreement of the coordinating investigator, and if appropriate of their sponsors, to be included in OPTIMON.

• Investigation sites: Must be participating in a clinical research study included in OPTIMON, have been suggested by the Methodology and Management Centre to take part in OPTIMON, have planned to include at least 5 patients in the study, and have principal investigators who have agreed to take part in OPTIMON.

Monitoring strategies compared: • Classic strategy: According to the current standards in the pharmaceutical industry • Optimised strategy: Concerning the level of risk (classified according to the scale), establishing in

advance scientific and regulatory priorities. The main differences from the classic strategy relate to the number of monitoring visits to the investigation sites, the proportion of data checked against source documents and the implementation of a blocking measure concerning consent.

Main judgement criterion: Proportion of patients whose observation includes at least one serious error. A serious error is an error concerning the conformity of the patient's consent, the declaration of serious or unexpected adverse events, compliance with eligibility criteria and the exactitude of the main judgement criterion.

OPTIMON Validation Committee : Will validate the classification of the studies according to risk level; will validate, completely independently, the components of each monitoring strategy for a given study.

Secondary judgement criteria : Anomalies other than serious errors, delays in obtaining results, costs.

Size of the study : OPTIMON plans to include at least 1,800 patients.



2. Justification

How data should be monitored in academic studies is open to question not only in France but throughout the European Community. Directive 2001/20/EC (the field of application of which only concerns clinical trials on medicinal products) lays down a priori the same obligations for commercial and non-commercial trials. However, the cost of exhaustive (all patients) and systematic (all data) monitoring is an obstacle to the application of these principles in the academic field. The decision on 24th November 2006 setting the rules of good clinical practice for biomedical research studies concerning medicinal products for human use states that "the sponsor ... will determine the extent and appropriate modalitie s of monitoring based on aspects such as the objective, the experimental plan, the complexity of the research, blinding, the number of people taking part in the research, the assessment criteria and the length of the research study. In general, monitoring on the research site before, during and at the end of the research study is suitable. However, in certain exceptional cases, the sponsor can decide that centralised monitoring accompanied by procedures such as training investig ators, meetings and detailed written instructions can guarantee that the research study is carried out appropriately in accordance with good clinical practice ....The data to be verified can be selected using statistical sampling."

The sponsor must not however introduce a double standard: transparency and credibility of data is an essential requirement both for industrial and acade mic sponsors .

The ECRIN (European Clinical Research Infrastructures Network, http://www.ecrin.org) programme brings together national clinical research infrastructure networks with the aim of harmonising the tools and practices of academic clinical research, improving quality, and offering a certain number of services facilitating transnational studies in Europe and Canada1. At present ECRIN covers Sweden, Denmark, Germany, Italy, Spain and France, as well as Quebec, which allows transnational studies to be extended to the North American continent. After primary work financed by the 6th Framework Programme, ECRIN is developing transnational workgroups, one of which concerns monitoring data and is coordinated by Geneviève Chêne. Its object is to define and then apply harmonised procedures for monitoring transnational (and also national) clinical studies, not only for drug trials but for any clinical resea rch study . This project is thus a pilot study for this workgroup, and forms the French contribution, destined to be extended to other participating countries. France is in fact quite advanced in this area2 in as far as it has academic sponsors , and monitoring strategies adapted to risk have already been defined and applied by some of them.

2.1. The nature of the problem

In this context, it is fundamental that methodology and management centres which provide the methodological and logistic support for clinical research studies for an academic sponsor should adopt reliable, standardised approaches for monitoring research data. Indeed, in clinical research, it is not only the rigour of methodological choices that determines the quality of a study's results; carrying out the study with precision is also very important because it is essential both for the safety of people taking part and the credibility of the results 3,4. It is important, for example, that patients' consent is given complying with the requirements of the regulations, or that there should be a measurement for the main judgement criterion for all the patients included so that the results of the study are valid 5-7.

In France, for more than ten years, the political will to structure academic clinical research of a very high scientific level has led to financing and setting up hundreds of studies, many of which have been multicentre. This has occurred particularly through the Programme Hospitalier de Recherche Clinique (PHRC - the Hospital Clinical Research Programme). The proportion of public financing devoted to monitoring is estimated to be between 20 and 40% depending on the type of study or the sponsor. The size of funds devoted to monitoring the progress of a clinical research study thus leads to questions being asked concerning the effectiveness of the activities involved in this monitoring for which the sponsor is responsible , whether he delegates them or not to an appropriate structure, in particular to a methodology and management centre.

The choice of a monitoring strategy for a study is usually a compromise taking into account the resources available in terms of time and staff, the potential benefit/risk ratio involved for all those participating in the study and the use anticipated for the results4. Important initiatives, such as that of the Assistance Publique – Hôpitaux de Paris (AP-HP), have resulted in definition of a level of monitoring depending on the risk for those taking part in the study (appendix 1). Even though it has never been formally validated, this classification is widely used by academic sponsors such as AP-HP or INSERM for their sponsoring activities. It focuses on optimised monitoring concentrating on preventing the most serious regulatory or scientific errors, i.e. those throwing the validity of the results into question 8,9.



The use of this classification very frequently leads methodology and management centres to concentrate on the essentials: patients' consent and serious or unexpected adverse events, completeness of data concerning eligibility criteria and the main judgement criterion. This compromise does not seem to affect the credibility of the results or the safety of patients. Nevertheless, these "intelligently optimised" monitoring strategies have not been sufficiently formalised to be easily reproducible and have never formally demonstrated their effectiveness in terms of quality assurance or their advantage in terms of cost. Consequently, the academic methodology and management centres find it difficult to argue that such an optimised strategy has no harmful results on the safety of patients or the integrity of data, or that they save time and expense.

Several methodology and management centres from different institutions have worked together to produce this protocol which aims to evaluate the usefulness of such "optimised" monitoring strategies, which they regularly use. The choice of an experimental study, comparing the two strategies in real clinical research studies , taking the investigation sites as the unit of randomisation, aims to answer the following question: In the context of academic clinical research, can a classic monitoring strategy be replaced by an optimised strategy without increasing the frequency of serious errors endangering the quality of care given to patients or the quality of the study from the point of view of compliance with regulations and the validity of the main result?

2.2. Current state of knowledge

Academic clinical research concerns the evaluation of therapeutic, diagnostic or prognostic strategies, and may they involve drugs or health products, usually after they have been put on the market, surgical procedures, radiotherapy, dietary regimes or even gene therapy, strategies for the use of diagnostic tests or the construction of prognostic scores 10. The priority in this research is to produce information of use in improving the management of patients; its benefit is essentially scientific before being possibly of commercial use.

This academic clinical research is complementary to and not in competition with industrial clinical research which essentially consists of the clinical development of drugs and health products before their registration (phases I to III in humans) and after they have been put on the market (phase IV). The main aim of industrial clinical research is to gather the necessary information and knowledge of a health product according to a standardised development plan before the marketing phase.

In the whole field of clinical research, clinical trials are the area where the procedures involved have up until now been most precisely described according to international quality assurance standards (International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use, Good Clinical Practice). During a trial, monitoring concerns all the activities aimed at controlling the integrity of data from a scientific (reliability, validity) and regulatory (compliance with the legislation) point of view 7. These activities thus consist of observing that a trial is carried out complying both with the protocol and the legislation. The objective is to prevent, detect, correct and ensure the traceability of errors so that the trial is carried out correctly and produces valid results 4. To achieve this objective, the sponsor has an obligation for performance and not just of best efforts. In France, his role has been strengthened by European Directive 2001/20/EC transposed into the new Public Health Law for which the application orders came into force in 2006-2007.

The very long experience of industry and the standardisation of trials in the private sector have led to the description of models of monitoring with a very high level of quality assura nce , though resulting in considerable cost. Certain non-commercial research groups, such as the European Organisation for Research and Treatment of Cancer (EORTC), set up an advanced system of reflection on quality assurance in the context of trials on cancer more than twelve years ago, in order to ensure a certain level of quality assurance but with a lower cost 11-14. Among other things, this move allowed EORTC to become a recognised partner of North American clinical research groups.

Other experiments saw the light of day in the large American study, MrFIT 15, and extended reflection on quality assurance for clinical trials into the wider field of acquiring epidemiological data. As to the clinical research field, while it is wider than that of clinical trials, it is still relatively specific when compared with the collection of epidemiological data in general, as two aspects must be taken into account simultaneously in the monitoring activities: observing the rights and safety of those taking part, and the integrity of data.

This debate on compromises for monitoring clinical research studies has led to certain general principles being accepted:



• Rapidity of collecting data and thus of working in real-time are important for detecting existing errors quickly and preventing their recurrence, for detecting fraud and for detecting particular signs concerning toxicity or efficacy 16-19;

• Training of personnel is in part responsible for the quality of the information collected and of monitoring the progress of a study and must therefore be a preliminary step for clinical research studies 15;

• Double entry of data does not minimise errors to any great extent and represents a considerable cost 15,20,21;

• Methodologists and the existence of methodology and management centres for coordinating monitoring and data management play a central role in the quality of a study, particularly for multicentre trials 9,16,22-24.

Decisions taken on the level of monitoring for an academic clinical research study are essentially based on the following three points 4:

• The risk level anticipated by the sponsor: A phase I clinical trial during which a drug is administered for the first time to a limited number of healthy subjects or patients has a higher risk level than a phase IV study or the comparison of two vitamin combinations for the long-term prevention of cancer;

• Obtaining the resources : Different academic sponsors do not devote the same resources to monitoring studies at similar risk levels;

• The use of the data which it is possible to anticipate: The monitoring level is higher for a clinical trial from which the information will be used to make up a file for agencies registering health products than for a study which aims to construct a prognostic score, even if this has the merit of being published in the best scientific journals.

A classic monitoring strategy includes a certain number of visits to the investigation sites before the start of the study, and, once it has started, intense verification of data, training of personnel and double entry of data.

This strategy has the advantage of combining the systems of reference known within the industry. It has the disadvantage of involving high cost which is difficult for an academic sponsor to meet. Moreover, there are other disadvantages: by aiming to cover everything, it does not concentrate sufficiently on the essential aspect of errors which affect the validity of the results. It is not unusual for trials by the industry to be criticised for the lack of a significant proportion of data or other bias even though they have been carried out according to high quality assurance standards 25.

Academic methodology and management centres often propose an optimised monitoring strategy : clinical research studies are often very different from one another, tailored to answer a scientific question. The resources available are more limited, and by concentrating on the essentials, investigators are trained in priorities and often to improve the overall quality of academic studies.

Practice depends on experience, the relationship between investigator and sponsor and on the resources available, the type of intervention being assessed, the phase of development, and the importance of the project for the Sponsor. These monitoring strategies are still not standardised enough to be reproducible .

2.3. Hypothesis of the study

The OPTIMON trial proposes the hypothesis that it is possible to define an optimised monitoring strategy from the risk level and the major scientific principles of a clinical research study, concentrating effort on the data really influencing the result of the trial and on the regulatory aspect, in particular collection of patient consent and serious or unexpected adverse events. This study also puts forward the hypothesis that the process is instructive for the staff involved in a clinical centre and that by making the priorities clear, this strategy may also improve the overall quality of a study. Consequently, while not increasing the frequency of all the errors or anomalies, it would be less onerous and less costly than an exhaustive, systematic, classic monitoring strategy and would allow valid results to be obtained in a shorter period of time. The OPTIMON trial does not therefore only aim to examine a main judgement criterion concerning the essential items (and which forms the basis for calculating the size of study), but lists many other points of comparison in the secondary judgement criteria.

2.4. Justification of choices made for this study



2.4.1. Procedures compared

The OPTIMON trial is an experimental study which aims to compare an optimised monitoring strategy with a classic strategy, in real clinical research studies.

We have chosen to apply it to studies using paper or electronic case records as data acquisition methods. Paper records are handled by all the methodology and management centres and require no special technology. They are still frequently used in an academic context. Electronic recording via the Internet26-28 is being used more and more often. In studies after the OPTIMON trial, other acquisition methods may be compared with paper and electronic records: tablet PCs, electronic pens.

The optimised strategy is based using a scale evaluating the risk (see appendix 1), the validity and reproducibility of which was evaluated in 2006 in the Pre-OPTIMON study 29,30. The OPTIMON trial is the first time that it has been used in real situations.

This is also the case for the general monitoring plan (see appendix 2), which sets out the different components of monitoring for each of the classic or optimised monitoring strategies. This general plan was defined in 2006, following Pre-OPTIMON, in a process to find a consensus among French academic clinical research professionals.

2.4.2. Judgement criterion for the comparison

This is a composite criterion , measured at the individual (patient) level, including the sum of errors seriously calling into question observance of the regulations, or the credibility of the results of the study 4,6,7,13,14 for complete observation of the clinical research study. The formulation of errors comes from a procedure used for trials by the Agence Nationale de Recherches sur le Sida (ANRS - National Agency for Research on AIDS) 31 which was agreed by consensus between the 4 methodology and management centres involved in handling 95% of ANRS trials.

The list consists of all the consent clauses which must comply with the regulations and the principles for declaring serious adverse or unexpected events and includes two important elements for validity of the results of the study: compliance with eligibility criteria and presence of the data necessary for analysis of the main judgement criterion of the clinical research study.

2.4.3. Outline of the study

It seems to us important to validate an optimised monitoring strategy and therefore to compare it experimentally with a classic strategy in the framework of real clinical research studies . Another approach would have been to use simulations from data collected. However, one of the components of the optimised monitoring strategy is to include a group of pre-determined principles and instructions which will influence how the study is carried out. Interaction between the strategy and how investigators work cannot be captured in simulations. In addition, monitoring is not concerned only with epidemiological data but also with regulatory aspects (obtaining patients' consent, the occurrence of serious or unexpected adverse events) for which simulation seems to be impossible. Finally, formalising the different strategies in a real situation will allow future academic clinical research reference systems to be solidly constructed.

We anticipate that the activities evaluated are likely to produce better results in such a study, in a context where the methodology and management centres volunteering to participate are the most motivated, than in real life (the Hawthorne effect). If the optimised monitoring strategy demonstrates that it is not inferior to the classic strategy and moreover provides advantages, the next step will be to formally generalise it in concrete terms and follow up its application. If this strategy does not show its value, it could be anticipated that its more generalised application would lead to still worse results and that it should either be improved or abandoned.

2.4.4. Unit of randomisation

A unit of randomisation could be:

• A person taking part in a study : The data of some individuals would be monitored according to the classic strategy and of others according to the optimised strategy. We considered that this choice had an important limitation: it would result in a certain progressive uniformity of behaviour in a given investigation site, e.g. if the same investigator followed the patients.

• The investigation site (defined as the equivalent of a hospital department): In a given clinical research study, some investigation sites would be monitored according to the classic strategy and others according to the optimised strategy. Indeed, the heterogeneity of the application of procedures and regulations within the framework of a monitoring strategy depends more often on the organisation



at investigation site level than at patient level. Moreover, this choice limits the risk of contamination of the strategies compared. An experimental study is then a study randomised in clusters, with the comparison of errors nevertheless being made by taking the patient as the statistical unit 32. Investigation sites will be chosen which envisage including at least 4 patients, given that running a study in investigation sites which are too small poses problems which require special monitoring 22.

• The clinical research study : Some studies would be monitored using a classic strategy and others with an optimised strategy. Nevertheless, it is difficult to compare monitoring from one study to another given that the proportion of errors may vary depending on the type and area of the study.

We have therefore decided to perform a trial randomised in clusters , the cluster being the investigation site. Each methodology and management centre must therefore organise two types of monitoring and take precautions to avoid internal contamination. A given investigation site may only be randomised once, for its participation in a first clinical research study included in the OPTIMON trial. If it is to participate in other clinical research studies included in the OPTIMON trial, the monitoring procedure determined for the first study will be applied to this site. This trial requires that the coordinating investigator of the clinical research study, the principal investigator of the investigation site, and if appropriate the sponsor, accept the principle of the OPTIMON trial. For this reason there is an informed consent form for them (see appendices 3, 4, 5). The "calendar time" effect will be tested at the time of analysis. Specific methodological and organisational aspects have been taken into account particularly when drawing up this protocol to comply with the recently published recommendations for this type of study 33. The main judgement criterion will be the proportion of patients whose observation contains no serious error for a person taking part in research and will be presented at the individual level as well as at the investigation site level.

In the following text, two types of study will be considered:

• The experimental study comparing two monitoring strategies, called the "OPTIMON trial";

• The clinical research studies in which the two monitoring strategies will be evaluated; these studies will be systematically called "clinical research studies".

3. Objectives

3.1. Main objective

The objective of our trial is to compare the efficacy of two monitoring strategies for clinical research studies in an academic context: one strategy based on classic quality assurance standards, the other optimised a priori on the main scientific and regulatory principles, the frequency of serious errors concerning patient eligibility criteria and measurement of the main judgement criterion (validity of the study's principal results), and on compliance with requirements for consent and the declaration of serious adverse events (observing regulations).

3.2. Secondary objectives

The secondary objectives consist of comparing the monitoring strategies for:

• The various serious errors presented separately and not grouped together as in the main objective: eligibility criteria, main judgement criterion of the clinical research study, compliance with consent requirements and those for serious adverse events;

• Anomalies other than serious errors;

• The length of time taken to obtain the main results of the clinical research study;

• The direct and indirect costs.

4. Study plan and randomisation

4.1. Study plan

OPTIMON is a non-inferiority study to see if it is possible to replace classic monitoring by optimised monitoring based on the risk level of the clinical research study, without altering the quality of care



provided to patients, the quality of the results or compliance of the clinical research study with the regulations.

It is to be a randomised, comparative, experimental study on a no n-blinded monitoring strategy comparing two monitoring strategies within the same clinical research study:

• A classic monitoring strategy: frequent on-site monitoring visits, verification of 100% of the data for 100% of the patients throughout the study,

• An optimised monitoring strategy: determination a priori of monitoring priorities according to the scientific objectives of the protocol and the risk level, limited on-site monitoring visits and verification of data limited to key data, a priori verification procedures and training of investigator and site personnel focused on the scientific aspect of the protocol.

The unit of randomisation is the investigation site. This is thus a cluster sampling study 33 (figure 1).

Figure 1. Diagram of the OPTIMON study comparing th e efficacy and cost of two monitoring strategies for public clinical research. MMC: Methodology and Management Centres.

Both strategies will be followed by monitoring visits to the investigation sites . These visits will be performed subsequently by the OPTIMON team's clinical research associate to check that the clinical research study has been carried out according to its protocol thus to confirm the validity of the trial data. This clinical research associate will monitor all the patients of all investigation sites for OPTIMON's main judgement criterion (See § 6.1).

On this basis, there will be direct comparison between the two monitoring strategies (classic v. optimised). With stratification according to risk level it will be possible to study the heterogeneity of the efficacy of an optimised monitoring strategy in line with the risk level of the study.

It is unlikely that a single investigation site would be asked to take part in several clinical research studies coming within our trial comparing monitoring strategies. If this were to be the case, monitoring the study in the investigation site would be carried out using the same monitoring strategy as that previously allocated to the site by randomisation.

4.2. Randomisation

The list allocating monitoring strategies to investiga tion sites has been established in advance by randomisation by the OPTIMON team's statistician and validated by an independent statistician. Randomisation was carried out per level in line with the A, B or C risk levels of the clinical research studies. The numbers in the two strategy groups (classic and optimised monitoring) are equal. A complete document describing the randomisation procedure (methods, block size, program used) is kept confidentially by the OPTIMON team's statistician.

The list is on the web server linked to the OPTIMON site by the OPTIMON team's computer scientist. A methodology and management centre can include and randomise an investigation site by connecting to the OPTIMON site using a secure connection. The site's eligibility criteria are verified and whether it has been previously included in OPTIMON via another study. The result of the randomisation is automatically sent to the methodology and management centre by fax. It is planned to set up an emergency procedure before the start of the OPTIMON trial in case the server malfunctions.

5. Eligibility criteria

Methodology and management centres envisaging including a clinical research study in OPTIMON are asked to contact the OPTIMON team to discuss in detail the various eligibility criteria.

The following are considered as eligible:

Selection Monitoring plan stratification

Randomisation of sites

Evaluation of patients

Eligible MMCs Eligible studies

A B C

Classic monitoring

Optimised monitoring On-site visits

Selection

Eligible sites



• Methodology and Management Centres

− coming under the auspices of an institution (INSERM [French National Institute for Health and Medical Research], university hospital, French National Federation of Anti-Cancer Centres, Institut Pasteur, etc.);

− with at least 2 years experience in multicentre clinical research studies;

− which have standard operating procedures finalised before inclusion in OPTIMON; and

− volunteer to apply the OPTIMON trial in the clinical research studies they coordinate.

• Clinical research studies

− classified in levels A, B or C according to the OPTIMON scale (see appendix 1);

− whether or not a sponsor is required;

− whether or not approval by an ethics committee is required;

− which are multicentre, with at least 4 investigation sites selected for the OPTIMON trial;

− the period of inclusion of which is compatible with OPTIMON (starting within the year, or already active with at least 20 patients still to be included);

− with a main judgement criterion (possibly on part of study) for which the database can be frozen within a period compatible with OPTIMON (3 years)

− that have planned to record data in an observation file on paper or in electronic form;

− that enter data with regular computerised controls throughout the study

− that have obtained the coordinating investigator’s, and if applicable the sponsor’s (see appendix 3) approval to participate in the OPTIMON trial (see appendix 4).

• Investigation sites (within clinical research studies)

− participating in a clinical research study included in OPTIMON;

− suggested by a methodology and management centre to take part in OPTIMON;

− that have planned to include at least 5 patients in the clinical research study concerned;

− where the principal investigator for the clinical research study concerned has agreed to take part in the OPTIMON trial (see appendix 5).

Comments:

• Sequential trials are not eligible because of aspects particular to monitoring them which make it difficult to compare the two monitoring strategies to be evaluated.

• So that a large variety of clinical research studies are included in the OPTIMON trial, the number of subjects from a given study included in OPTIMON is limited to a maximum of 100. The investigation sites to be included in OPTIMON are put forward by a methodology and management centre.

• To allow long-term studies to be included, which would not be compatible with the duration of OPTIMON (evaluated after 3 years), the period of monitoring taken into consideration by OPTIMON could be reduced. Thus, in a study planned to last 5 years, only data from the first year of monitoring could be retained for transfer to OPTIMON. The duration will be chosen according to the study schedule or intermediate analyses.

• An investigation site included in the OPTIMON trial for a clinical research study may participate again in the OPTIMON trial for a new research study if it fulfils the eligibility criteria for this new study, but it will not be randomised again: the patient files for the new study will be monitored according to the strategy allocated by the randomisation performed on inclusion of this site for the first study.



6. Methods compared

Ideally, the monitoring process for an investigation site, the method of which we wish to compare with the method used elsewhere, will start when the latter is recruited and end when the database is frozen for final analysis of the study.

We are describing the main stages of this monitoring of the classic strategy and of the optimised strategy. These stages will systematically be very precisely described for each study by the methodology and management centre in charge of monitoring each clinical research study before it starts.

In a given methodology and management centre, the clinical research associate or associates in charge of monitoring the study included must organise the work to avoid any contamination between the randomisation arms:

- either two different CRAs will monitor the investigation sites randomised to each monitoring strategy,

- or one CRA will be in charge of monitoring all the investigation sites whatever their randomisation. In this case, he or she must comply with procedures set up jointly with the OPTIMON team (use of OPTIMON tools, creation of memory aids, identifying labels, etc.)

The time budgeted in this study for clinical research associates will be made available to the sponsor of each study to strengthen the research personnel budget. Travel costs for the additional visits to investigation sites have also been budgeted.

After describing the components of each strategy, a summary document will show the different stages and highlight the major differences: the number of visits, the type a nd proportion of data verified with the source documents, the procedure blocking i nclusion if there is non-compliance concerning a patient's consent .

All the stages of the classic or optimised monitoring strategies are also set out in appendix 2.

6.1. Classic monitoring strategy

The classic monitoring strategy is similar in method to the monitoring a service provider would adopt for industry.

• Initial contact : The initial contact is a telephone conversation, an exchange of e-mails or a short meeting between the methodology and management centre and the investigation site explaining the aim of the trial or the study to the investigator. Investigators receive a summary of the project and a form for agreement to participate. This contact is the subject of a standardised report.

• Verification that the investigation site is adequat e: The methodology and management centre must check using a standard questionnaire that the investigation site has the material resources and the skills necessary to set up and run the study: qualification, availability and motivation of the investigator, previous clinical research experience, commitment to the rules of good practice (in particular, the existence of source documents), availability of equipment which may be needed, etc. After this check, the site may be accepted or not for the study. This questionnaire will be completed during an on-site visit, which could be combined with the setting-up visit. However, this questionnaire may be completed from a distance (by letter, e-mail, or better still by telephone) if the investigation site is experienced and known to the study's methodology centre, i.e. if it has all the following characteristics:

− is experienced in clinical research in multicentre studies (with remote monitoring);

− is experienced in the pathology being studied in the protocol;

− is experienced in working with the methodology and management centre designated to manage the study or with another methodology and management centre;

− is experienced in the processes involved in the study (interventions, investigations etc.).

• Setting-up : During an on-site visit, the methodology and management centre will present the scientific and practical aspects of the protocol and the regulations to the investigation site. During this meeting, the key points of the protocol will be set out, with details of the study interventions and investigations and of the observation file and how to complete it. Such a meeting could last between 2 and 4 hours and should take place in the presence of all members of the investigation site team taking part in the study. A complete standardised report will be compiled. After this visit, the investigation site is authorised to include patients in the study. If the site is experienced and known to the methodology



and management centre, this "visit" can be made by telephone (teleconference with oral comments on previously downloaded documents or which are visible via Internet).

• Verification of the compliance of observation file and database data with the source file and verification that procedures have been observed : This is verified by frequent, systematic visits to all sites, from the time of inclusion of the first patient. The frequency of visits to sites depends on the rate of patient recruitment and the volume of data collected. It will be defined in advance at the start of each protocol by the methodology and management centre and is validated by the OPTIMON validation committee. An additional visit can be made if there is a major problem, i.e. endangering the patient's safety (a dispensing error, an at-risk intervention/investigation outside the protocol, an unfulfilled eligibility criterion concerning the legitimacy of including the patient etc.), the quality of results of the study (allocation of the randomisation treatment, unblinding, bias in selection, information or indication, suspicion of fraud etc.), the rights of the patient (consent, respect of anonymity etc.), or possibly regulatory aspects (failure to declare an investigator, investigator file etc.). Each visit will last for a whole day. It will concern all the source documents for 2 to 3 patients. For each visit, a monitoring plan will be established in advance. It will include a minimum of: an interview with the centre's principal investigator, examination of data, verification of the compliance of consent forms, compliance with the protocol and its amendments, compliance with source documents (100% of observation file data in 100% patients), accounting for the trial treatment if applicable. The investigation file (trial documents) will be systematically checked. An intermediate visit report will always be produced.

• Verification of understanding of the investigation site's internal and external circuits : During the first monitoring visit to the investigation site (for all sites once the first patient has been included), the clinical research associate will systematically ensure that personnel participating in the study have understood its practical requirements: how to fill in source files and observation files, the data circuits, study interventions, sampling, etc. He or she will also check that the study is being carried out according to the protocol so that corrective action or training can be carried out.

• Consent : At each on-site visit following a new inclusion, the clinical research associate will check the existence of the patient and that the consent form has been completed correctly.

• Searching out undeclared serious adverse events : At each on-site visit, the clinical research associate will look for serious adverse events which have not been declared spontaneously by the investigator. If necessary, he will ask the investigator to make an immediate declaration of the event according to the declaration procedure set out in the study protocol.

• On-site corrections : At each on-site visit, necessary corrections will be made to the observation file with the principal investigator, explaining the errors to him to avoid their being made again.

• Verification of data forms received : As the observation file is filled in, the investigation site will send the methodology and management centre copies of the forms completed. For forms received which have not yet been verified on-site, the methodology and management centre's clinical research associate checks he has received all the forms expected and the overall quality of their completion before sending them to be entered.

• Remote or centralised monitoring : Exhaustive computerised controls will be made on the data entered to check their completeness and consistency. Following these controls, the clinical research associate will send the investigator requests for clarification or correction of any inconsistent data. These controls will be performed on all the data from all patients in all investigation sites. Regular contact will be maintained by telephone, fax or e-mail with the key people in the investigation site to ensure that procedures are observed, and a report will be compiled in the form of a standardised contact form.

• Administrative closure of the site : The administrative closure of the site will be performed during an on-site visit. Material remaining on the investigation site (observation files etc.) will be collected, depending on the sponsor's wishes consent forms may be collected in a sealed envelope, observation files will be validated and closed, the number of treatment units will be determined, and authorisation for destruction given. A report will be compiled, with a quantitative account of documents and verified material left in situ or recovered. The investigation site can then be definitively closed.

6.2. Optimised monitoring strategy

The optimised monitoring strategy concentrates on the essential regulatory aspects, patients' consent and serious or unexpected adverse events, and on the essential scientific aspects concerning the



judgement criterion and the eligibility criteria. In particular, this strategy includes verification of the compliance of consent forms very early in the sequence of events during inclusion followed by a blocking measure in the event of non-compliance.

The specific features of optimised monitoring are as follows:

6.2.1. Definition of the optimised monitoring strat egy by the methodology and management centre

• Risk level of the study: Starting from the scale and the study protocol, the methodology and management centre assesses the protocol and determines the level of risk run by the patient (A, B or C).

• Monitoring plan specific to the study: From the study protocol and the general optimised monitoring plan, the methodology and management centre sets out the practical modalities of monitoring in a monitoring plan specifically for the study.

• Components specific to studying the OPTIMON judgement criterion: From the study protocol and the OPTIMON protocol, the methodology and management centre determines which specific components should be used for studying the main OPTIMON judgement criterion.

6.2.2. Validation of the optimised monitoring strat egy by the OPTIMON validation committee

Using the scale, the general optimised monitoring plan, the study protocol and the OPTIMON protocol, the OPTIMON validation committee validates the risk level, the specific monitoring plan for the study and the specific components for studying the OPTIMON judgement criterion. Particular care will be taken to see that the decisions taken are homogeneous.

6.2.3. Optimised monitoring strategy

N.B. The following stages are identical to those described for the classic monitoring strategy: initial contact, verification that the investigation site is adequate, setting up, remote monitoring, verification of data forms received.

• Initial contact : The initial contact is a telephone conversation, an exchange of e-mails or a short meeting between the methodology and management centre and the investigation site, explaining the aim of the trial or the study to the investigator. Investigators receive a summary of the project and a form for agreement to participate. This contact is the subject of a standardised report.

• Verification that the investigation site is adequat e: The methodology and management centre must check using a standard questionnaire that the investigation site has the material resources and the skills necessary to set up and run the study: qualification, availability and motivation of the investigator, previous clinical research experience, commitment to the rules of good practice (in particular, the existence of source documents), availability of equipment which may be needed, etc. After this check, the site may be accepted or not for the study. This questionnaire will be completed during an on-site visit, which could be combined with the setting-up visit. However, this questionnaire may be completed from a distance (by letter, e-mail, or better still by telephone) if the investigation site is experienced and known to the study's methodology centre, i.e. if it has all the following characteristics:

− is experienced in clinical research in multicentre studies (with remote monitoring);

− is experienced in the pathology being studied in the protocol;

− is experienced in working with the methodology and management centre designated to manage the study or with another methodology and management centre;

− is experienced in the processes involved in the study (interventions, investigations etc.).

• Setting-up: During an on-site visit, the methodology and management centre will present the scientific and practical aspects of the protocol and the regulations to the investigation site. During this meeting, the key points of the protocol will be set out, with details of the study interventions and investigations and of the observation file and how to complete it. Such a meeting could last between 2 and 4 hours and should take place in the presence of all members of the investigation site team taking part in the study. A complete standardised report will be compiled. After this visit, the investigation site is authorised to include patients in the study. If the site is experienced and known to the methodology



and management centre, this "visit" can be made by telephone (teleconference with oral comments on previously downloaded documents or which are visible via Internet)..

• Verification of compliance of the observation file and database data with the source file and verification that procedures have been observed : This verification is made by on-site visits. The frequency of on-site visits depends on the risk level:

− Risk level A : No on-site visit is planned.

− Risk level B : One on-site visit, with verification of 100% of key data, is carried out for 10% of patients, all sites considered together, according to a sampling plan determined before the start of the study by the methodology and management centre and validated by the OPTIMON validation committee. An additional visit may be performed in the event of a major problem (see definition § 5.1). For each visit, a monitoring plan will be established in advance. It will include at least an interview with the site's principal investigator, verification of key data, and verification that procedures have been observed. An intermediate visit report will be systematically produced.

− Risk level C : An on-site visit, with verification of 100% of key information, is carried out for each site on a percentage of patients corresponding to one day's monitoring. An additional visit may be performed in the event of a major problem (see definition § 5.1). For each visit, a monitoring plan will be established in advance. It will include at least an interview with the site's principal investigator, verification of key data, and verification that procedures have been observed. An intermediate visit report will be systematically produced.

• Verification of understanding of the investigation site's internal and external circuits :, After the inclusion of the first patient and reception of the corresponding forms by the methodology and management centre, the clinical research associate will systematically ensure by telephone that the personnel participating in the study have understood its practical requirements: how to fill in source files and observation files, data circuits, study interventions, sampling, etc. He or she will also check that the study is being carried out according to the protocol so that corrective action or training can be carried out.

• Consent form : The consent form has an additional sheet with a part masked at the places for the surname and first name of the patient and his signature. This sheet must be faxed to the methodology and management centre on pre-inclusion of the patient. The methodology and management centre will refuse inclusion/randomisation if the visible parts do not appear to comply with the requirements. During the next on-site visit, the clinical research associate will verify the existence of the patient and that this document has been correctly completed.

• Searching out undeclared serious adverse events : At each on-site visit, the clinical research associate will look for SAEs which have not been declared spontaneously by the investigator. If necessary, he will ask the investigator to make an immediate declaration of the event according to the declaration procedure set out in the study protocol. If no on-site visit is anticipated, this information will be sought from a distance by comparing the various forms received by the methodology and management centre. If a serious adverse event is detected which has not been spontaneously declared, the clinical research associate will ask the investigator to declare the event immediately according to the declaration procedure set out in the study protocol.

• On-site corrections : For level B and C studies, at each on-site visit, any corrections necessary will be made to the observation file with the principal investigator, explaining the errors to him to avoid their being made again. For level A studies, or if no on-site visit is anticipated, the investigator is asked for the corrections using correction requests.

• Verification of data forms received : As the observation file is filled in, the investigation site will send the methodology and management centre copies of the forms completed. For forms received which have not yet been verified on-site, the methodology and management centre's clinical research associate checks he has received all the forms expected and the overall quality of their completion before sending them to be entered.

• Remote or centralised monitoring : Exhaustive computerised controls will be made on the data entered to check their completeness and consistency. Following these controls, the clinical research associate will send the investigator requests for clarification or correction of any inconsistent data. These controls will be performed on all the data from all patients in all investigation sites. Regular contact will be maintained by telephone, fax or e-mail with the key people in the investigation site to ensure that procedures are observed, and a report will be compiled in the form of a standardised contact form..



• Administration closure of the site : The administrative closure of the site is performed by letter for level A and B studies, and during an on-site visit for level C studies. Material remaining on the investigation site (observation files etc.) will be collected, depending on the sponsor's wishes consent forms may be collected in a sealed envelope, observation files will be validated and closed, the number of treatment units will be determined, and authorisation for destruction given. A report will be compiled, with a quantitative account of documents and verified material left in situ or recovered. The investigation site can then be definitively closed.

7. Collecting OPTIMON data

7.1. Updating data from the clinical research study

Whatever the monitoring strategy, data from the clinical research study are entered, checked and corrected as they are collected. When the methodology and management centre considers that the time is right, the database is frozen for final or intermediate statistical analysis. If the period of monitoring of the clinical research study is too long with regard to the eligibility criteria of the OPTIMON trial (see § 5), part of the data could be frozen (complete monitoring of the first X patients included) specifically to be transmitted to OPTIMON.

7.2. Transmission of the database from the clinical research study to the OPTIMON team

Part of the database is then forwarded in a specific format to the OPTIMON team. Only the variables necessary for calculation of the judgement criteria of the OPTIMON trial are transmitted. Details of the practical modalities of transmission and the format are given in appendix 7.

7.3. OPTIMON on-site visits

These visits will subsequently be made by the OPTIMON team's clinical research associate to collect the OPTIMON study data . He or she will compare the data in the database from the methodology and management centre with the patient's source documents. Information will be collected from all the investigation sites taking part in OPTIMON, for all patients, for the main OPTIMON judgement criterion (see § 8.1).

7.4. Central monitoring procedures in the methodolo gy and management centres

All the methodology and management centres have at least 2 years’ experience of multicentre clinical research studies. They have standard operating procedures, in particular for monitoring studies (list to be given to the OPTIMON Coordination Centre when a study is included).

The following aspects are particularly harmonised:

• Compiling the protocol and observation file

• The form of the information leaflet and consent form

• Notification of inclusions and monitoring the rhythm of inclusions

• The project team meeting with a predefined agenda, examination of warning signals and taking corrective action;

• Computer checks, after entry, of 100% of data;

• Management of error correction requests.

This harmonisation work uses the current reference systems, in particular those of the Réseau Français des Unités d'Essais Cliniques (French Network of Clinical Trial Units).

Each methodology and management centre will have to record the internal organisation preventing contamination between data concerning th e OPTIMON classic and optimised monitoring strategies (to be given to the OPTIMON C oordination Centre when a study is included).



8. Judgement criteria

8.1. Main judgement criterion

The main judgement criterion is the proportion of patients whose observation for the cl inical research study contained no serious errors .

It is a composite criterion, measured at the individual (patient) level, including errors which seriously challenge observation of the regulations, or credibility of the results of the study. The way errors are formulated comes from a procedure used for ANRS trials (ANRS = French National Agency for AIDS Research) 31.

The errors concern the following two regulatory aspects - consent and serious or unexpected adverse events - and the following two aspects concerning the scientific integrity of the data - failure to respect eligibility criteria without prior dispensation, and incorrect value or data missing for the main judgement criterion.

• Non-compliance of the patient's consent form for whatever reason: The consent form cannot be found on site; the patient’s name is illegible or absent; the patient's signature is missing; the date of the patient's signature is later than the date at which it should have been signed or it is illegible or absent; one of the items which had to be filled in by the investigator is missing or illegible or the date is later than the visit when it was supposed to shown; the name, date and the patient's signature are visibly not in his handwriting.

• Serious or unexpected adverse event not declared in a way which complies with the regulations in force, while it has been known to the investigator for more than 48 hours;

• Failure to comply with at least one eligibility cri terion (inclusion or exclusion) without prior dispensation . A request for dispensation is a request, made by the investigator of the investigation site to the methodology and management centre, to include a patient for whom an eligibility criterion is not observed. A procedure for dispensation requests is systematically set up before the start of the trial, and the investigation sites are informed about it during the initial explanatory meeting and monitoring visits.

• Value missing for the main judgement criterion (possibly calculated on part of the monitoring period: see comment 3 § 5 eligibility criteria), whatever the reason, including not updating a survival criterion. Each file will be reviewed by the OPTIMON validation committee (see § 10.4) which will confirm and document the error without knowing the monitoring strategy applied.

8.2. Secondary judgement criteria

There are a great many secondary judgement criteria. They underline the advantages and disadvantages of the optimised strategy that condition its non-inferiority relative to the classic strategy for the principal criterion. Certain indicators should allow the impact of monitoring procedures on later stages of the clinical research to be assessed (indicators concerning the speed of carrying out the project, or of the total number of anomalies). Their evaluation is very important in the context of a study where the result of the main judgement criterion is in part conditioned by the definition of the new intervention studied (optimised monitoring).

• Breakdown of the main judgement criterion according to the type of serious error:

− Proportion of errors related to consent

− Proportion of errors relating to serious or unexpected adverse events

− Proportion of errors relating to eligibility criteria

− Proportion of errors relating to the main judgement criterion of the clinical research study

• Indicators of the speed of carrying out the study:

− Length of the inclusion period

− Length of time between the last collection of data for the last patient and the availability of a validated frozen database

− Length of time between the last collection of data and the availability of the first results



• Indicators concerning the number of anomalies relative to the number of patients:

− Number of requests for correction or confirmation

− Total number of missing pieces of data relative to the expected number of pieces of data

− Number of residual errors after freezing the database

• Specific indicator concerning regulatory aspects:

− Serious adverse effects not declared spontaneously

− Length of time for declaration of serious adverse events declared spontaneously

• Indicators of direct and indirect costs. The costs directly related to applying each strategy should be taken into account stating:

− Investments necessary in material and training and costs of maintenance, which thus provides the cost of acquisition. Investments are classified as redeployable or not, i.e. whether or not limiting the possibility of doing other things in the future and therefore the cost of abandoning.

− Costs related to carrying out the study (if possible, individual per patient);

− Cost of the detection of errors;

− Cost of the consequences of detected and undetected errors;

− Cost of the surveillance of the monitoring strategies.

The cost indicators will be defined more precisely in the near future and the protocol will consequently be modified.

9. Statistical aspects

9.1. Size of the study

There are two difficulties in calculating the number of patients needed for this study:

• The first is the hypothesis of non-inferiority in comparing the effect of the strategies in the study. In this context, the size of the study depends on a margin of non-inferiority for the main judgement criterion, here the frequency of serious errors, which must be defined in advance. The calculation anticipates the results being presented in the form of a confidence interval for the difference between the strategies compared rather than a statistical test. Interpreting the risks of error (α and β) is consequently reversed 34,35.

• The second difficulty concerns calculation for a trial where the randomisation unit is a cluster . Compared with calculation in a trial where the randomisation unit is the patient, a similar power necessitates a greater study size given the resemblance inside the clusters. The calculation must, in principle, take into account the intracluster correlation and the size of the clusters 33.

To our knowledge, there is no consensus on a margin of serious errors for a study such as ours. Moreover, there is little literature on the precise situation for calculating size for a non-inferiority study in clusters 36.

9.1.1. Review of the information concerning the pro portion of serious errors in clinical research studies

In three studies which had the specific objective of identifying the reasons for these errors in an academic context, the proportion of patients whose observation contained no serious errors defined in a similar way to that proposed in our project was 80% in a study in the EORTC 22, 90% in an American Cancer and Leukemia Group9 study, and 95% in a study by the Eastern Cooperative Oncology Group (ECOG)24. These studies date from the 1980s to the 1990s and it may be thought that the expected proportion of patients whose observation contains no serious error with classic monitoring could currently be about 95%. Nevertheless this value is quite uncertain and our project will provide more up-to-date information on this subject.

9.1.2. Calculation in a situation of non-inferiorit y: principles, hypotheses and results



Calculating study size in a comparative trial of superiority is based on the statistical test used for the main judgement criterion. In a comparative trial of non-inferiority, a conventional statistical test cannot be used because non-inferiority cannot be concluded from the absence of difference and the detection of difference may have no clinical relevance and may, in practice, correspond to non-inferiority. It is preferable to analyse the main results using a confidence interval of the difference between the strategies compared, presenting a range with which the observed point value for the difference is compatible. If each point inside this range corresponds to an unimportant difference, then the strategy on trial may be considered as not inferior to the reference strategy.

This reasoning is the basis for choosing the non-inferiority margin . In the present case, the investigators have agreed to consider that a strategy would not be of interest in which the results in terms of observations without serious error were too low compared with a classic strategy. A margin of 5% seems to be the greatest extent remaining acceptable.

The confidence interval for the difference between the proportions of patients whose observation includes no serious error in each of the monitoring strategies (classic strategy – optimised strategy) will therefore be calculated from the results, and its lower limit compared to the –5% value. If the confidence interval of the difference contains values lower than -5%, the optimised strategy will be concluded to be inferior.

This procedure, which is fairly intuitive in the case of non-inferiority trials, corresponds in reality to a statistical test procedure which, nevertheless, inverts the null and alternative hypotheses . Indeed, the null hypothesis is that there is a difference of at least 5% and the trial is designed to reject this hypothesis in favour of an alternative hypothesis that a smaller difference exists. A 95% one-sided confidence interval is calculated. Choosing a power of 80% means that the total number of patients included will allow an 80% probability of showing that the optimised monitoring strategy is not inferior to the classic monitoring strategy if this is actually the case in reality.

The calculation with nQuery Advisor (version 5) shows that with 260 patients per strategy compared, i.e. a total of 520 in all for a 1:1 ratio, there would be a power of 81% 35,37.

9.1.3. Calculation taking into account the clusters

In a comparative study with clusters as the randomisation unit, an inflation factor for the size of the study must be used to take into account the resemblance between clusters, in order to maintain a sufficient power.

The inflation factor (FI) takes into account the size of the clusters (n) and the resemblance of patient observations within each cluster for the main judgement criterion (intracluster correlation coefficient, ρ):

ρ)1(1 −+= nFI .

The intracluster correlation coefficient varies from 0 (no resemblance) to 1 (perfect resemblance).

In the case of OPTIMON, a likely situation was chosen following investigators’ experience where the number of patients included per investigation site would be 5 and the intracluster correlation coefficient 0.6: the total number obtained for the patients to be included is thus 1,768. Thus 1,800 patients will be included, which will represent on average 360 investigation sites.

Recruitment will be carried out in the French methodology and management centres volunteering to take part and may possibly be widened to foreign centres in the ECRIN network.

9.2. Interim analysis

Subject to amendment by the Independent Surveillance Committee during its first meeting, an interim analysis will be performed when half the data to be collected, i.e. the data for 900 patients, is available in the OPTIMON database. The interim analysis will focus on the main judgement criterion of the OPTIMON trial.

9.3. Analysis strategy

Each criterion, except those related to cost, will be shown at patient level and at the methodology and management centre level.

The analysis will be systematically performed in line with classification of the risk level: A, B or C. The heterogeneity of the effect of the optimised strategy compared with the classic strategy will be systematically tested. If it is not heterogeneous, a global result will be presented.



The calendar effect will also be systematically tested to subsequently verify a learning effect or contamination between the two strategies which might occur with time.

9.4. Statistical methods

9.4.1. Analysis of the proportion of patients whose observation contain no serious error

The proportion of patients whose observation contains no serious error will be shown for each monitoring strategy. It will be calculated by comparing the total number of patients whose observation has at least one serious error validated by the OPTIMON validation committee with the total number of patients included in the OPTIMON trial. The confidence interval of the difference of these proportions (optimised strategy – classic strategy) will be calculated. If the confidence interval of the difference contains values lower than -5%, it will be possible to conclude that the optimised strategy is inferior relative to the classic strategy. If this is not the case, the conclusion will be that it is not inferiority.

The analysis therefore depends on the method for calculating the confidence interval. The weighted Woolf method will be used. If there is an imbalance related to randomisation by cluster, the confidence interval will be estimated using a marginal model (Generalised Estimating Equations, GEE). The estimates will be compared with those obtained by a logistic model with random effects, so as to verify the stability of the results with the hypotheses of the model.

The following co-variables are taken into account in the model: risk level, type of study, area of research, level of complexity of collecting data, number of items of data collected, size of the investigation site, methodology and management centre, sponsor, occurrence of at least one interim analysis in the study. Modelling will take into account the clusters as a random factor.

Finally, a detailed analysis of the monitoring reports due in the classic monitoring strategy will be performed to distinguish qualitatively between the excesses and the actions which lead to a real gain in quality. This analysis will be carried out separately from a standardised reading scale which will be finalised a priori with the OPTIMON validation committee.

9.4.2. Cost

The differences in cost in terms of mean cost will be compared and if there are differences for a particular criterion, the expense for one less error or for a unit of time less will be calculated.

10. Surveillance of the trial

10.1. Scientific committee

This is composed of the OPTIMON correspondent of each methodology and management centre, the OPTIMON team project leader, representatives of the sponsors of the clinical research studies included, and representatives of regulatory bodies (Ministry of Health, AFSSAPS [French Agency for the Safety of Health Products], etc.).

The Scientific Committee for the OPTIMON trial will meet before the start of the trial then at least every 6 months. Its purpose is to take all important decisions concerning the good organisation of the trial and compliance with the protocol. It will decide on any relevant modification to the protocol necessary for continuation of the trial.

10.2. Independent Surveillance Committee

This is composed of personnel who are independent of the methodology and management centres taking part in the study and of the study's Coordination Centre: (members appointed during the 1st meeting of the SC on 1st April 2008).

It is appointed by the Scientific Committee during the OPTIMON start-up meeting. It will meet once when OPTIMON starts, then once a year, and during the interim analysis of the OPTIMON trial.

The task of the Independent Surveillance Committee is to oversee respect of patients and maintain the scientific and ethical integrity of the study; this committee may be consulted by the Scientific Committee.

During the study it may help take decisions for which an independent judgement is desirable, particularly concerning stopping the study prematurely, major modifications to the protocol which become necessary due to recruitment or monitoring, or modification to the statistical analysis criteria.



It will carry out the interim analysis and will re-evaluate the size of the study at that time (see § 9.2 and 9.3).

10.3. OPTIMON Validation Committee

This is composed of: Fabrice Bonnet, Julie Boussuge, Gilles Chatellier, Bruno Giraudeau, Jean-Pierre Pignon. They will meet regularly according to the rhythm of studies being included and the rhythm that data is acquired on the judgement criteria.

The role of this committee is to:

• validate the risk level (A, B or C.) of the study put forward by the methodology and management centre;

• validate the components of each classic or optimised monitoring strategy proposed by the methodology and management centre before inclusion of the clinical research study in OPTIMON;

• validate the components of OPTIMON's main judgement criterion proposed by the methodology and management centre;

• validate OPTIMON's main judgement criterion. (Each file will be reviewed by the committee which will confirm and document each value missing for the main judgement criterion blinded for the monitoring strategy applied).

11. Practical organisation of the trial

11.1. General organisation

A Coordination Centre, based in Bordeaux, will organise the whole of the study procedure. A project leader, a statistician, will be responsible for scheduling the study and for logistics diagnosis. A clinical research associate will be responsible for the technical and administrative management of the project (practical organisation, meetings, reports, distribution etc.) and for direct contact with methodology and management centres for setting up and running the project. A data manager will prepare the database needed for the study. This database will be comprised of the information collected to allow comparison between strategies and will require that data files from the different methodology and management centres be received in an appropriate format. The data manager will prepare the forms for data collection during on-site visits after OPTIMON. These items of data will be entered in the OPTIMON database.

11.2. Web Site and Instructions

A web site devoted to the OPTIMON trial is in place at the address https://ssl2.isped.u-bordeaux2.fr/optimon/. The documents of the OPTIMON study (protocol, evaluation scale for the risks of studies, information leaflet) and various tools available for methodology and management centres for managing their studies are freely accessible on this site:

Suggested method for allocating patient code letters

Producing a patient calendar

Memorandum on OPTIMON monitoring strategies applied to a study

Identification labels for patients included in OPTIMON

Set-up visit or intermediate visit report form

Examples of data collection forms and associated computer checks

The OPTIMON correspondents of MMCs and investigation sites can also connect via a reserved access, which gives them specific rights, allows studies participating in OPTIMON to be declared and each patient to be included.

A set of instructions has also been produced by the OPTIMON Coordination Centre and is available via the OPTIMON trial web site in the form of a practical guide to OPTIMON:

Process for the inclusion of a study in OPTIMON

Procedure for reception/management of data collection forms



Procedure for the 4th sheet of the patient informed consent form

…

11.3. Confidentiality of data

All patient data will be sent in a file to the OPTIMON team in an anonymous form: the identifier for the patient will be composed of the identifying acronym of the methodology and management centre, the identifying acronym of the clinical research study, and the identifier of the patient used in the study (investigation site numerical code, patient numerical code and patient letter code string).

All information collected concerning patients will remain strictly confidential. Only people authorised by the sponsor and involved in running the trial and a representative of the health authorities may consult the medical file of the patient, after obtaining the agreement of the investigating doctor of the investigation site, and for the sole aim of verifying the accuracy of the data collected. These people are themselves subject to professional secrecy.

The data will be entered at the Coordination Centre of the study. The file will be subject to a declaration to the CNIL (National Commission for Computing and Liberty) (reference methodology 001) and to the CCTIRS (Consultative Committee on the Treatment of Information in Research in the Area of Health).

12. Ethical and regulatory aspects

The agreement of the sponsor, the coordinating investigator of each clinical research study and the principal investigator of the investigation site concerned must be obtained.

The Sud Ouest Outre Mer Ethics Committee has been asked whether it is relevant to submit the OPTIMON trial to it for approval, and its answer was negative (see appendix 6). The information leaflet of the clinical research studies included in the OPTIMON trial will not be modified, but an information leaflet specifically concerning OPTIMON will be given to the patient, on transferring his data to the OPTIMON database (see appendix 9: information leaflet).

13. Expected benefits

The OPTIMON trial could result in the following benefits:

• The description and construction of a reference system for optimised monitoring for academic clinical research at French national or European level;

• Improvement in the quality of data from academic clinical research studies, in particular in protocols produced within the PHRC (French national hospital clinical research programme);

• A reduction in the time needed to obtain exact data thus improving preparation for meetings of independent clinical trial monitoring committees or other clinical research studies necessitating the review of interim data;

• Construction of a web site available to sponsors, investigators and methodology and management centres grouping together the necessary information for classifying the type of study and components of monitoring.



14. Références

1. Demotes-Mainard J, Ohmann C. European Clinical Research Infrastructures Network: promoting harmonisation and quality in European clinical research. Lancet 2005;365:107-108.

2. Demotes-Mainard J, Chêne G, Libersa C, Pignon J. Clinical research infrastructures and networks in France: report on the French ECRIN workshop. Thérapie 2005(in press).

3. Hill A. The clinical trial. Br Med J 1951;7:278-82.

4. Spriet A, Dupin-Spriet T. Bonne pratique des essais cliniques des médicaments. 3rd ed. Bâle: Karger, 2004.

5. Pocock S. Clinical trials: a practical approach. Chichester: Wiley, 1983.

6. Meinert C. Clinical trials: design, conduct, and analysis. New York: Oxford University Press, 1986.

7. Spilker B. Guide to clinical trials. Philadelphia: Lippincott Williams & Wilkins, 2000.

8. De Pauw M. Quality control in data monitoring of clinical trials. Acta Urol Belgica 1994;62:31-5.

9. Weiss R, Vogelzang N, Peterson B, Panasci L, Carpenter J, Gavigan M, et al. S. Successful system of scientific data audits for clinical trials. A report from the cancer and leukemia group B. JAMA 1993;270:459-64.

10. Chêne G, Bégaud B. Méthodologie de la recherche clinique. Apprentissage de l'exercice médical. Paris: Ellipses, 2004: 36-59.

11. Marinus A. Quality assurance in EORTC clinical trials. Eur J Cancer 2002;38:S159–S161.

12. Steward W, et al. Chemotherapy administration and data collection in an EORTC Collaborative Group-can we trust the results. Eur J Cancer 1993;29A:943-7.

13. Vantongelen K, Rotmensz N, Van der Schueren E. Quality control of validity of data collected in clinical trials. Eur J Cancer Clin Oncol 1989;8:1241-7.

14. Vantongelen K, Steward W, Blackledge G, Verweij J, Van Oostrom A. EORTC joint ventures in quality control: treatment related variables and data acquisition in chemotherapy trials. Eur J Cancer 1991;27:201-7.

15. Neaton J, Duchene A, Svendsen K, Wentworth D. An examination of the efficiency of some quality assurance methods commonly employed in clinical trials. Stat Med 1990;9:115-24.

16. Glicksman A, Reinstein L, Brotman R, McShan D. Quality assurance programs in clinical trial. Cancer Treat Rep 1980;64:425-33.

17. Pollock B. Quality assurance for interventions in clinical trials. Cancer 1994;74:2647-52.

18. Ottevanger P, Therasse P, Van de Velde C, et al. Quality assurance in clinical trials. Crit Rev Oncol/Hematol 2003;47:213-35.

19. Schaake-Koning C, Kirkpatrick A, Kroger R, Van Zandwijk N, H B. The need for immediate monitoring of treatment parameters and uniform assessment of patient data in clinical trials. A quality control study of the EORTC radiotherapy and lung cancer cooperative groups. Eur J Cancer 1991;27:615-9.

20. Gibson D, Harvey A, Everett V, Parmar M. Is double data entry necessary? The CHART trials. CHART steering committee. Continuous, hyperfractionated, accelerated radiotherapy. Control Clin Trials 1994;15:482-8.

21. Day B, Fayers B, Harvey B. Double data entry: what value, what price? Control Clin Trials 1998;19:15-24.

22. Sylvester R, Pinedo H, De Pauw M, Staquet M, Buyse M, Renard Jea. Quality of institutional participation in multicenter clinical trials. N Engl J Med 1981;8:852-5.

23. Knatterud G, Rockhold F, George S, Barton F, Davis C, Fairweather W, et al. Guidelines for quality assurance in multicenter trials: a position paper. Control Clin Trials 1998;19:477-93.



24. Begg C, Carbone P, Elson P, Zelen M. Participation of community hospitals in clinical trials: analysis of five years of experience in the eastern cooperative oncology group. N Engl J Med 1982;306:1076-80.

25. Chan A, Hrobjartsson A, Haahr M, Gotzsche P, Altman D. Empirical evidence for selective reporting of outcomes in randomized trials: comparison of protocols to published articles. JAMA 2004;291:2457-65.

26. Christiansen D, Hoslcing J, Dannenberg A, Williams O. Computer-assisted data collection in multicenter epidemiologic research the atherosclerosis risk in communities study. Control Clin Trials 1990;11:101-115.

27. Krischer J, Hurley C, Pillalamati M, Pant S, Bleichfeld C, Opel M. An automated patient registration and treatment randomization systems for multicenter clinical trials. Control Clin Trials 1991;12:367-77.

28. Santoro E, Nicolis E, Franzosi M, Tognoni G. Internet for Clinical Trials: Past, Present, and Future. Control Clin Trials 1999;19:194-201.

29. Journot V, Pignon J-P, Gaultier C, Bouxin-Métro A, Ravaud P, Chêne G. Validity and Reproducibility of a Risk Scale in Academic Clinical Research Studies. 28th Annual Conference of the ISCB 2007, Alexandroupolis.

30. Journot V, Pignon J-P, Gaultier C, Bouxin-Métro A, Ravaud P, Chêne G. Validité et Reproductibilité d'une échelle de risque dans les études de recherche clinique institutionnelles. 1ère Conférence Francophone d'Epidémiologie Clinique 2007, Bordeaux.

31. Chêne G, Aboulker J, Chermak A, et al. Indicateurs de qualité des sites cliniques de l'ANRS. 6ème Séminaire de recherche clinique 2004, Paris.

32. Litchfield J, Freeman J, Schou H, Elsley M, Fuller R, Chubb B. Is the future for clinical trials internet-based? A cluster randomized clinical trial. Clin Trials 2005;2:72-9.

33. Campbell M, Elbourne D, Altman D. CONSORT statement: extension to cluster randomised trials. Br Med J 2004;328:702-708.

34. Jones B, Jarvis P, Lewis J, Ebbutt A. Trials to assess equivalence: the importance of rigorous methods. Br Med J 1996;313:36-39.

35. Machin D, Campbell M. Statistical tables for the design of clinical trials. Oxford: Backwell Scientific Publications, 1987.

36. Piaggio G, Carroli G, Villar J, et al. Methodological considerations on the design and analysis of an equivalence stratified cluster randomization trial. Stat Med 2001;20:401-16.

37. Newcombe R. Interval estimation for the difference between independent proportions: comparison of eleven methods. Stat Med 1988;17:873-890.



15. Annexes

• Annexe 1. Grille de classification du niveau de risque des études de recherche clinique selon l'échelle OPTIMON.

• Annexe 2. Plan de monitorage général OPTIMON.

• Annexe 3. Accord de participation à OPTIMON – Promoteur d'une étude de recherche clinique incluse.

• Annexe 4. Accord de participation à OPTIMON – Investigateur Coordonnateur d'une étude de recherche clinique incluse.

• Annexe 5. Accord de participation à OPTIMON – Investigateur principal d'un site investigateur inclus.

• Annexe 6. Décision du Comité de Protection des Personnes Sud ouest Outre Mer.

• Annexe 7. Note d’information Optimon

• Annexe 8. Liste des Centres de Méthodologie et de Gestion volontaires pour participer à l'essai OPTIMON.

• Annexe 9. Procédure d'anonymisation complète de l'identité du patient



Annexe 1. Grille OPTIMON évaluant le risque patient dans les études de recherche clinique.



MODE D'EMPLOI

• ETAPE I : Déterminer le ou les thèmes de l'étude, et définir ses caractéristiques. Expliciter. • ETAPE II : Reprérer les conditions d'augmentation du risque. Expliciter. • ETAPE III : Déterminer le niveau de risque en cheminer dans la grille selon les informations définies aux étapes I et II : – sous étape 1 : Entrer dans la grille par le thème d'étude, défini à l'étape I. – sous étape 2 : Classer l'étude à un niveau de risque intermédiaire selon ses caractéristiques, définies à l'étape I. – sous étape 3 : Pour les colonnes � à �, augmenter le risque d'un niveau si l'étude porte sur une intervention à risque, une méthode d'investigation à risque, ou une population cible définie par un état aggravant le risque lié aux interventions ou aux investigations. Augmenter éventuellement le risque de deux niveaux si la présence de plusieurs conditions accroît fortement le risque. Pour la colonne �, augmenter le niveau de risque d'un niveau si l'état définissant la population cible aggrave le risque lié aux interventions ou aux investigations. remarque : Si l'étude peut être classée dans deux thèmes différents (exemple : comparaison d'une technique chirurgicale et d'un médicament), réaliser deux classements et retenir le risque final le plus élevé.

QUELQUES EXEMPLES

objectif / méthode / commentaires cheminement dans la grille protocole multicentrique de traitement des leucémies aiguës myéloblastiques de l'enfant et de l'adolescent (0 à 18 ans) par interleukine 2 (pas d'AMM chez l'enfant) ; risque de morbidité sévère liée à l'interleukine 2

essai de médicament → colonne � étude confirmatoire sur produits sans autorisation → risque intermédiaire C intervention et population à risque → risque final D

comparaison de l’efficacité (nb selles / jour à 8 mois post opératoires) de la réalisation d’un néo-rectum obtenu par 2 techniques chirurgicales classiques sur la fonction sphinctérienne : coloplasty - pouch vs réservoir en J

étude de chirurgie → colonne � techniques sur organe interne → risque intermédiaire B aucune condition à risque supplémentaire → risque final B

évaluation des orthèses de repos dans la rhizarthrose : étude prospective randomisée contrôlée

dispositif médical → colonne � dispositif médical avec marquage CE de classe I → risque intermédiaire A aucune condition à risque supplémentaire → risque final A

étude de physiopathologie : recherche d’anomalies de la voie sensorielle afférente primaire chez les schizophrènes par enregistrements de stimulations de nerfs de la cheville (examen douloureux)

étude de physiopathologie → colonne � acte invasif → risque intermédiaire B aucune condition à risque supplémentaire → risque final B

étude de physiopathologie : établir les valeurs de référence de la différence de potentiel nasal et du courant court circuit sur biopsie rectale chez l'enfant sain (3 mois à 3 ans) et atteint de mucoviscidose (risque d'inhalation et de déshydratation)

étude de physiopathologie → colonne � acte invasif → risque intermédiaire B population à risque à risque supplémentaire → risque final C

cohorte psychiatrique périnatale : étude des troubles anxieux de la grossesse et du post-partum, conséquences sur la mère, le nouveau né et le développement du jeune enfant

questionnaire → colonne � questionnaire sans difficultés particulières → risque intermédiaire A aucune condition à risque supplémentaire → risque final A



GLOSSAIRE • autorisation : autorisation de mise sur le marché pour un médicament, autorisation d'utilisation pour un radio-isotope, dans une indication précise ;

l'autorisation n'est valable que si l'utilisation se fait dans cette indication. • chirurgie

− généralisation d’une nouvelle technique : technique en cours de diffusion dans un grand nombre de sites. − mise au point d’une nouvelle technique : technique en cours de mise au point dans un petit nombre de sites.

• dispositif médical (Article L5211-1 du Code de la Santé Publique – Ordonnance n° 2001-198 du 1 mars 2001 art. 1 Journal officiel du 3 mars 2001) − dispositif médical : On entend par dispositif médical tout instrument, appareil, équipement, matière, produit, à l’exception des produits d’origine humaine,

ou autre article utilisé seul ou en association, y compris les accessoires et logiciels intervenant dans son fonctionnement, destiné par le fabricant à être utilisé chez l’homme à des fins médicales et dont l’action principale voulue n’est pas obtenue par des moyens pharmacologiques ou immunologiques ni par métabolisme, mais dont la fonction peut être assistée par de tels moyens. Les dispositifs médicaux qui sont conçus pour être implantés en totalité ou en partie dans le corps humain ou placés dans un orifice naturel, et qui dépendent pour leur bon fonctionnement d’une source d’énergie électrique ou de toute source d’énergie autre que celle générée directement par le corps humain ou la pesanteur, sont dénommés dispositifs médicaux implantables actifs. Ces dispositifs médicaux sont destinés à être utilisés à des fins : . de diagnostic, de prévention, de contrôle de traitement ou d’atténuation d’une maladie ; . de diagnostic, de prévention, de traitement, d’atténuation ou de compensation d’une blessure ou d’un handicap ; . d’étude, remplacement ou modification de l’anatomie ou d’un processus physiologique . . de maîtrise de la conception

− marquage CE : symbole visuel CE attestant de la conformité du dispositif médical à une liste d’items appelés "exigences essentielles", obligatoire depuis le 14 juin 1998, valable en France et dans tout autre état membre de l’Union européenne ou partie à l’accord sur l’Espace économique européen, d'une validité de 5 ans.

− classe de marquage CE : Il existe 4 classes (I, IIa, IIb, III) qui correspondent à un niveau de risque croissant. Les modalités de classification sont définies par 18 règles réparties dans 4 groupes (Annexe IX du Livre Vbis du Code de la Santé Publique – édition 2003) :

. règles applicables aux dispositifs non invasifs ; . règles applicables aux dispositifs invasifs ; . autres règles applicables aux dispositifs actifs ; . règles spéciales.

critères utilisés pour la classification des dispositifs médicaux : . durée d’utilisation (temporaire, court terme, long terme) ; . degré invasif du dispositif et le type d’invasion ; . possibilité de réutilisation ou de non réutilisation du dispositif ; . destination à visée thérapeutique ou diagnostique du dispositif ; . dépendance d’une source d’énergie ; . partie du corps en contact avec le dispositif médical. − orifice du corps : tout ouverture naturelle du corps, ainsi que la surface externe du globe oculaire, ou toute ouverture artificielle permanente, par exemple

une stomie. − dispositif médical invasif : dispositif qui pénètre partiellement ou entièrement à l'intérieur du corps, soit par un orifice, soit à travers la surface du corps.



− dispositif médical actif : tout dispositif médical dépendant pour son fonctionnement d'une source d'énergie électrique ou de toute source d'énergie autre que celle générée directement par le corps humain ou par la pesanteur et agissant par conversion de cette énergie. Les dispositifs médicaux destinés à transmettre de l'énergie, des substances ou d'autres éléments, sans modification significative, entre un dispositif médical actif et le patient ne sont pas considérés comme des dispositifs médicaux actifs.

• étude confirmatoire : étude dont l'objectif est l'efficacité ; se déroule après des études exploratoires ; essai de phase II ou III dans le développement d'un médicament.

• étude exploratoire : étude dont les objectifs sont la sécurité, la tolérance, la faisabilité, l'exploration de doses, l'activité ; se déroule après les premières études sur l'homme ; essai de phase I ou II dans le développement d'un médicament.

• intervention : procédure, produit ou objet à l'étude. − autres interventions : par exemple, examen clinique, prise en charge du patient, régime alimentaire, activité physique…

• investigation : tout moyen d'évaluation de l'état du patient, de la pathologie à l'étude ou des effets de l'intervention à l'étude. • morbidité sévère : aggravation de l'état basal dans la pathologie.


© CHU Bordeaux OPTIMON, Comparaison stratégies de monitorage, version finalisée pour soumission PHRC national, 29 avril 2005 sur 50

Annexe 2. Plan de monitorage général OPTIMON.

Tableau 1. Plan de monitorage général OPTIMON, selo n la stratégie de monitorage. STRATEGIE DE MONITORAGE ACTIONS CLASSIQUE OPTIMISEE

contact initial prise de contact entre le CMG et le site investigateur

-systématique (rencontre, téléphone, courriel…) et tracé

vérification de l’adéquation du site investigateur vérification des moyens techniques et des compétences du site investigateur

-systématique -si site investigateur connu et expérimenté : questionnaire standard administré par courrier / courriel / téléphone -si site investigateur inconnu ou non expérimenté : sur site sur questionnaire standard (peut être couplé avec la mise en place)

MIS

E EN

PLA

CE

mise en place rappel des aspects scientifiques du protocole et de la réglementation et discussion des aspects pratiques

-systématique -avant l'inclusion du premier patient -si site investigateur connu et expérimenté : par téléphone -si site investigateur inconnu ou non expérimenté : sur site

vérification de la conformité des données du cahier d’observation et de la base par rapport au dossier source et vérification du respect des procédures

-sur site -dès le 1er patient inclus puis fréquence à définir au début de l'étude par le CVO, et en cas de problème majeur -100% des patients -100% de toutes les données + respect des procédures -100% des sites investigateurs

-niveau A : pas de visite sur site -niveau B : 10% des patients (plan de sondage établi en début d’étude) sur 100% des points clef visite supplémentaire si problème majeur - niveau C : au moins une visite par site (% patients ⇔ 1 jour de monitorage) sur 100% des points clef visite supplémentaire si problème majeur

vérification de la compréhension des circuits internes et externes au site investigateur

-systématique -lors de la 1ère visite de monitorage (dès 1ère inclusion) -sur site

-systématique -après l’inclusion du premier patient et réception des fiches de recueil de données au CMG -par téléphone

consentement vérification de l’existence du patient et de la conformité du remplissage

-sur site à la visite suivante -systématique à chaque nouveau patient

-copie masquée du consentement à l’inclusion -sur site à la visite suivante ou à la clôture du site

recherche des EIG non déclarés détection des EIG non déclarés spontanément

-systématique sur site -systématique sur site ou à distance

corrections -à chaque visite, sur toutes les données -demandes de correction gérées à distance

-niveaux B et C : à chaque visite sur les points clef -demandes de correction gérées à distance

vérification des fiches de recueil de données vérification de la réception des toutes les fiches attendues et de la qualité globale du remplissage

-systématique avant envoi à la saisie pour les fiches non vérifiées sur site

MO

NIT

OR

AGE

monitorage à distance ou centralisé -contrôles informatisés sur les données saisies vérifiant l’intégrité et la cohérence des données -demandes de correction (DDC) -contacts avec les personnes clef du site investigateur pour assurer le respect des procédures

-systématique -100% des patients -100% de toutes les données + respect des procédures -100% des sites investigateurs

CLÔ

TUR

E clôture administrative du site -récupération du consentement (si souhait du promoteur) -validation/clôture du cahier d’observation -décompte/destruction des traitements -clôture du site investigateur

-systématique -sur site -100% des patients -100% des sites investigateurs

-systématique -niveaux A et B : par courrier / niveau C : sur site -100% des patients -100 % des sites investigateurs



Tableau 2. Définition des points clef.

DONNEES CLEF

existence du patient : vérification sur site existence du document source attestant l'existence du patient consentement vérification sur site

présence d'un consentement au nom du patient conformité de la version du formulaire de consentement présence et lisibilité du nom du patient présence de la signature du patient présence et lisibilité de la date de signature du patient conformité de la date de signature par le patient unicité de l'écriture entre date de signature et signature du patient présence et lisibilité du nom du médecin conformité de l'habilitation du médecin présence de la signature du médecin présence et lisibilité de la date de signature du médecin conformité de la date de signature par le médecin unicité de l'écriture entre date de signature et signature du médecin

consentement vérification de la copie masquée à distance si pas de vérification sur site du consentement

présence d'un consentement attribué au patient conformité de la version du formulaire de consentement présence du nom du patient présence de la signature du patient présence et lisibilité de la date de signature du patient conformité de la date de signature par le patient présence et lisibilité du nom du médecin conformité de l'habilitation du médecin présence de la signature du médecin présence et lisibilité de la date de signature du médecin conformité de la date de signature par le médecin unicité de l'écriture entre date de signature et signature du médecin

critères d'éligibilité majeurs présence et exactitude des données ou présence des prélèvements nécessaires à leur calcul examens et visites majeurs vérification de la conformité des dates de visite au calendrier théorique critère de jugement principal présence et exactitude des données ou présence des prélèvements nécessaires à son calcul critères de jugement secondaires majeurs présence et exactitude des données ou présence des prélèvements nécessaires à leur calcul

PO

INT

S C

LEF

RESPECT DES PROCEDURES



respect des procédures vérification sur site et/ou à distance

liste de randomisation et procédure d’allocation des traitements insu (suivi patient et mesure des critères de jugement) processus de validation des critères de jugement circuit du traitement (schéma d’administration, à la pharmacie et/ou dans le service clinique) déclaration des EIG (recherche d'événements non déclarés) circuit des prélèvements (prélèvement, transport, conservation, dosage) mise à jour du dossier investigateur critères d’éligibilité (éventuellement, vérification des pré-inclus non inclus et des patients non sollicités)

Tableau 3. Glossaire.

TERME DEFINITION

Centre de Méthodologie et de Gestion (CMG)

Un Centre de Méthodologie et de Gestion est une équipe à laquelle peut faire appel le promoteur d’une étude de recherche clinique pour assurer un soutien méthodologique et logistique à cette étude. Le centre prend en charge la méthodologie de mise en place pratique et la gestion de l’étude. Il peut s’agir d’URC des hôpitaux, d’unités INSERM ou d’autres structures. Il peut s'agir d'une structure assurant également la promotion de l'étude. Au sein d’OPTIMON, elles sont toutes appelées "Centre de Méthodologie et de Gestion".

Comité de Validation de l’Etude (CVE)

Le Comité de Validation de l’Etude a pour rôle la validation des événements majeurs de l’étude : critère de jugement principal, toxicité grave…

Comité de Validation OPTIMON (CVO)

Un Comité de Validation composé de professionnels expérimentés en recherche clinique est constitué avant le démarrage d'OPTIMON. Il a pour rôle la définition d’un plan de monitorage spécifique à chaque étude à partir du cadre défini dans le protocole OPTIMON. Il fixe les modalités pratiques des monitorages classique et optimisé, les critères d’éligibilité majeurs, les critères secondaires majeurs et les examens majeurs, en fonction des caractéristiques de l’étude.

Copie masquée du consentement (cas d'une inclusion / randomisation centralisée)

Le formulaire de consentement comporte un feuillet autocopiant supplémentaire. Ce feuillet comporte deux zones fortement quadrillées, donc presque complètement masquées, aux emplacements où le patient renseigne son nom et sa signature, de sorte que l'on peut vérifier que le formulaire a été rempli, mais pas déchiffrer le nom. Ce feuillet est faxé au CMG lors de la demande d'inclusion d'un patient. L'inclusion d'un patient sans copie masquée du consentement renseignée et reçue par le CMG est interdite. La vérification sur site postérieure permet de contrôler l'identité du signataire du formulaire. S'il n'y a pas de visite sur site, seuls certains points seront vérifiés (Cf. Tableau 2).

Critères d’éligibilité majeurs Parmi les critères d’éligibilité de l’étude, le Comité de Validation choisit ceux qu’il semble pertinent de considérer comme faisant partie des points clef. Il s’agit de critères permettant de vérifier que le patient a bien l’indication souhaitée dans l’étude, qu’il n’est pas mis en danger du fait de sa participation à l’étude, et qu’il satisfait aux contraintes réglementaires de l’étude (âge en particulier).

Critères de jugement secondaires majeurs

Parmi les critères de jugement secondaires de l’étude, le Comité de Validation choisit ceux qu’il semble pertinent de considérer comme faisant partie des points clef. Il s’agit de critères importants du point de vue des résultats scientifiques et de la sécurité des patients (toxicité en particulier).



TERME DEFINITION

Données clef Parmi les points clef, les données clef sont les composantes du critère de jugement d'OPTIMON.

Evénement Indésirable Grave (EIG) L’investigateur notifie immédiatement au promoteur tous les événements indésirables graves, à l’exception de ceux qui sont recensés dans le protocole ou dans la brochure investigateur comme ne nécessitant pas une notification immédiate. Les événements indésirables et/ou les résultats d’analyse anormaux définis dans le protocole comme déterminants pour les évaluations de la sécurité sont notifiés au promoteur, conformément aux exigences de notification et dans les délais spécifiés dans le protocole.

Examens majeurs Parmi les examens cliniques, biologiques,... définis dans le calendrier de suivi d'un patient, le Comité de Validation d'OPTIMON choisit ceux qu’il semble pertinent de considérer comme faisant partie des points clef. Il s'agit d'examens présentant une plus grande importance pour la sécurité du patient et/ou les résultats de l'étude.

Points clef Ce sont les aspects de l'étude particulièrement importants pour le respect de la réglementation, la sécurité du patient, la qualité des résultats de l'étude. Certains sont définis a priori, d'autres doivent être précisés en début d'étude par le Comité de Validation d'OPTIMON en fonction des spécificités de l'étude. Les points clef se décomposent en données clef et respect des procédures.

Problème majeur Il s’agit d’un problème : - qui met en jeu la sécurité du patient : erreur de dispensation (dose, nature du médicament), intervention/investigation à risque hors protocole, critère d'éligibilité portant sur la légitimité de l'inclusion du patient non rempli,... - qui met en jeu la qualité des résultats de l'étude : allocation du traitement de randomisation, levée d'insu, biais de sélection, d'information, d'indication, suspicion de fraude,... - qui met en jeu les droits du patient : consentement, anonymisation,… - éventuellement, qui met en jeu les aspects réglementaires : non déclaration d’un investigateur, dossier investigateur,…

Conduite à tenir en cas de problème : - tenir compte de l'origine du problème : mauvaise intention du site, mauvaise organisation du site, erreur passagère, erreur systématique,… - problème mineur : action correctrice à distance pour le site concerné - problème majeur : nouvelle visite et action correctrice sur site pour le site concerné ; vérifier la présence du même problème et éventuellement action correctrice à distance pour les autres sites - dans certains cas, la détection d'un problème peut conduire à un amendement au protocole.

Processus de validation des critères de jugement

Dans certaines études, la mesure du critère de jugement principal peut être influencée par la connaissance du groupe de traitement. Un processus de validation de la mesure en insu du groupe de traitement par un Comité de Validation peut donc être mis en place.

Site investigateur Un site investigateur comprend :



TERME DEFINITION

- le(s) service(s) clinique(s) où se déroule l’étude clinique, - le service des prélèvements (éventuellement inclus au sein du service clinique), - les laboratoires biologiques où sont analysés ou congelés les prélèvements, - la pharmacie où sont dispensés les traitements de l’étude.

Site investigateur connu et expérimenté

Un site investigateur est considéré comme connu et expérimenté s’il présente toutes les caractéristiques suivantes : - expérience dans la recherche clinique sur des études multicentriques (avec monitorage à distance), - expérience dans la pathologie à l’étude dans le protocole, - expérience avec le CMG désigné pour la gestion de l’étude ou un autre CMG, - expérience avec les procédures de l’étude (type d’intervention, produit à l’essai,…)



Annexe 3. Accord de participation à OPTIMON Promoteur d'une étude de recherche clinique.

Essai OPTIMON : OPTImisation du MONitorage des étud es de recherche clinique

Evaluation de l’efficacité et du coût de deux stratégies de monitorage pour la recherche clinique institutionnelle.

Financé par le PHRC national – acceptation du projet le 10/10/2005.

Cher Collègue,

Nous vous proposons d’inclure votre étude de recherche clinique "NOM DE L'ETUDE" dans une étude de comparaison de stratégies de monitorage dont nous vous présentons le schéma et les conditions de réalisation.

Nous sommes à votre entière disposition pour répondre à toutes vos questions, en particulier celles que vous pourriez vous poser sur les conséquences attendues pour le déroulement de votre propre étude.

Nous espérons que vous accepterez de nous aider à établir ainsi les standards de monitorage de la recherche clinique institutionnelle.

Pourquoi l’essai OPTIMON ?

Le choix d’une stratégie de monitorage d’une étude de recherche clinique est le plus souvent un compromis prenant en compte les moyens disponibles, le ratio bénéfice/risque potentiel pour toutes les personnes qui s’y prêtent et l’usage anticipé des résultats. Selon la réglementation, le promoteur est responsable du monitorage mais le délègue souvent à des Centres de Méthodologie et de Gestion. L’hétérogénéité actuelle des pratiques, l’importance des financements consacrés par étude et les discussions européennes sur la création de standards de qualité similaires pour les essais industriels et académiques conduisent à s’interroger sur l’efficacité des activités du monitorage dans le contexte institutionnel.

Quel est l’objectif principal de l’essai OPTIMON ?

Notre étude a pour objectif de comparer l’efficacité de deux stratégies de monitorage : l’une fondée sur des standards classiques d’assurance qualité, l’autre optimisée a priori sur des grands principes scientifiques et réglementaires.

Quelles méthodes seront utilisées pour atteindre ce t objectif ?

OPTIMON est une étude de non infériorité, comparant deux stratégies de monitorage, dont l’unité de randomisation sera le site investigateur au sein d’études de recherche clinique multicentriques. Il s’agit d’une étude dite randomisée en cluster. Les études seront stratifiées selon le niveau de risque de la grille OPTIMON.

Quelles sont les stratégies de monitorage comparées ?

Certains sites investigateurs seront donc monitorés de façon classique et les autres d’une façon optimisée en fonction du niveau de risque. Le type de monitorage sera attribué par tirage au sort. Les deux stratégies comparées sont :

1) Une stratégie classique selon les standards en vigueur dans l’industrie pharmaceutique,

2) Une stratégie optimisée : sur le niveau de risque (classification de la grille), établissant à l’avance des priorités scientifiques et réglementaires. Les différences principales avec la stratégie classique portent sur le nombre de visites de monitorage dans les sites investigateurs, la proportion de données vérifiées par rapport aux documents sources et sur l’implémentation d’une mesure bloquante pour le consentement.

Les procédures seront définies à l’avance par le Centre de Méthodologie et de Gestion coordonnant les aspects méthodologiques et le niveau de risque, le contenu de chaque stratégie sera validé par un Comité de Validation indépendant des équipes réalisant la recherche.

Ci-joint vous disposez du protocole complet de l’essai OPTIMON. Quel que soit la stratégie de monitorage attribuée, la qualité de la recherche devrait être garantie par une visite a posteriori permettant de vérifier les données. L’ensemble du surcoût de monitorage et d’analyse sera pris en charge par l’essai OPTIMON.



Quels sont les études et les centres cliniques conc ernés ?

Les études de recherche clinique de niveau A, B ou C de la classification de la grille, nécessitant ou non un promoteur et l’avis d’un Comité de Protection des Personnes (CPP), démarrant après le démarrage d'OPTIMON et se terminant dans les 3 ans, ayant prévu d’inclure au moins 20 patients, multicentriques, comprenant au moins 4 sites investigateurs, ayant prévu d’effectuer le recueil de données sur un cahier d’observation papier, ayant obtenu l’accord de l’investigateur coordonnateur et, le cas échéant, du promoteur pour comparer deux stratégies de monitorage au sein de leur étude de recherche clinique pourront se voir proposer de participer à l’essai OPTIMON.

Les sites investigateurs (au sein des études de recherche clinique) ayant prévu d’inclure au moins 5 patients dans l’étude de recherche clinique concernée, dont l’investigateur principal a donné son accord pour participer, n’ayant pas déjà été inclus dans OPTIMON dans le cadre d’une autre étude de recherche clinique pourront être concernés.

Comment seront comparées les stratégies de monitora ge ?

La proportion de patients dont l’observation comporte au moins une erreur grave sera comparée entre les deux stratégies. Une erreur grave est une erreur portant sur la conformité du consentement du patient et de la déclaration des événements indésirables graves ou inattendus, sur la conformité des critères d’éligibilité et sur l’exactitude du critère de jugement principal.

La comparaison portera également sur d’autres critères comme les anomalies autres que les erreurs graves, les délais d’obtention des résultats et les coûts.

Quelle sera la taille de l’essai OPTIMON ?

On prévoit d’inclure au moins 1800 patients. On fait l’hypothèse que la proportion de patients dont l’observation ne comporte pas d’erreur grave avec la stratégie de monitorage classique sera de 95%. La marge de non infériorité est fixée à 5%. Avec 260 patients inclus dans chacune des stratégies, on a une probabilité de 81% de mettre en évidence une non infériorité si elle existe vraiment. Le risque d’erreur est de 5%. Le réajustement prenant en compte le schéma en clusters fait l’hypothèse que 5 patients seraient inclus par site investigateur et que la corrélation intra-site investigateur pour le critère de jugement principal serait de 0,6. Le facteur d’inflation est de 3,4. Le nombre total de patients à inclure dans cette étude est de 1800 dans 360 sites investigateurs, en moyenne.

Pour réaliser cette étude, le Centre Coordonnateur devra recevoir de la part du Centre de Méthodologie et de Gestion coordonnant votre étude de recherche clinique certaines données. Néanmoins, il s’engage formellement à ce que ces données soient strictement utilisées aux fins définies pour l’essai OPTIMON et ne puissent interférer avec votre propre agenda scientifique.

Il est très important que les investigateurs coordonnateurs des études concernées acceptent volontairement de participer et montrent leur intérêt. C’est pourquoi nous vous sollicitons pour signer cet engagement.

Nous espérons que vous comprendrez l’intérêt de l’essai OPTIMON et que vous donnerez votre accord pour que votre étude de recherche clinique "NOM DE L'ETUDE" puisse y être incluse.



ACCORD DE PARTICIPATION A OPTIMON

PROMOTEUR

Promoteur ...................................................................................

Etude de recherche clinique ...................................................................................

Représentant ...................................................................................

Fonction ...................................................................................

J’accepte librement que mon étude de recherche clinique citée ci-dessus soit incluse

dans l’essai OPTIMON visant à comparer deux stratégies de monitorage.

J’ai compris que la participation à l’essai OPTIMON nécessiterait des aménagements

organisationnels pour le Centre de Méthodologie et de Gestion en charge de la

coordination de l’étude, pour l'Investigateur Coordonnateur de l'étude et les

investigateurs des sites qui seront sollicités. Ceux-ci seront informés et devront

accepter librement de participer.

Le surcoût sera supporté par le budget de l’essai OPTIMON.

La participation à l’essai OPTIMON ne me dégage en rien de mes obligations et de

mes droits en tant que représentant du promoteur pour l’étude de recherche clinique

citée ci-dessus.

J’ai compris que les responsables de l’essai OPTIMON n'utiliseront pas les données

de l'étude de recherche clinique citée ci-dessus en dehors des stricts objectifs de

l’essai OPTIMON.

Nom Prénom Date d'accord Signature

......................................... └─┴─┘└─┴─┘└─┴─┴─┴─┘



Annexe 4. Accord de participation à OPTIMON Investigateur Coordonnateur d'une étude de recherch e clinique.




Cher Collègue,

Nous vous proposons d’inclure votre étude de recherche clinique "NOM DE L'ETUDE" dans un essai de comparaison de stratégies de monitorage dont nous vous présentons le schéma et les conditions de réalisation.


















Les études de recherche clinique de niveau A, B ou C de la classification de la grille, nécessitant ou non un promoteur et l’avis d’un Comité de Protection des Personnes, démarrant après le démarrage d'OPTIMON et se terminant dans les 3 ans, ayant prévu d’inclure au moins 20 patients, multicentriques, comprenant au moins 4 sites investigateurs, ayant prévu d’effectuer le recueil de données sur un cahier d’observation papier, ayant obtenu l’accord de l’investigateur coordonnateur et, le cas échéant, du promoteur pour comparer deux stratégies de monitorage au sein de leur étude de recherche clinique pourront se voir proposer de participer à l’essai OPTIMON.













INVESTIGATEUR COORDONNATEUR


Investigateur Coordonateur ...................................................................................

Fonction ...................................................................................

J’accepte librement que mon étude de recherche clinique citée ci-dessus soit incluse

dans l’essai OPTIMON visant à comparer deux stratégies de monitorage.



coordination de l’étude et pour les investigateurs des sites investigateurs qui seront

sollicités. Ceux-ci seront informés et devront accepter librement de participer.



mes droits en tant qu’investigateur coordonnateur pour l’étude de recherche clinique

citée ci-dessus.



l’essai OPTIMON.

Nom Prénom Date d'accord Signature

......................................... └─┴─┘└─┴─┘└─┴─┴─┴─┘



Annexe 5. Accord de participation à OPTIMON Investigateur principal d'un site investigateur.




Cher Collègue,

Nous vous proposons d’inclure votre étude de recherche clinique "NOM DE L'ETUDE" dans un essai de comparaison de stratégies de monitorage dont nous vous présentons le schéma et les conditions de réalisation.


















Les études de recherche clinique de niveau A, B ou C de la classification de la grille, nécessitant ou non un promoteur et l’avis d’un Comité de Protection des Personnes, démarrant après le démarrage d'OPTIMON et se terminant dans les 3 ans, ayant prévu d’inclure au moins 20 patients, multicentriques, comprenant au moins 4 sites investigateurs, ayant prévu d’effectuer le recueil de données sur un cahier d’observation papier, ayant obtenu l’accord de l’investigateur coordonnateur et, le cas échéant, du promoteur pour comparer deux stratégies de monitorage au sein de leur étude de recherche clinique pourront se voir proposer de participer à l’essai OPTIMON.













INVESTIGATEUR PRINCIPAL

Site investigateur ...................................................................................


Investigateur Principal ...................................................................................

Fonction ...................................................................................

J’accepte librement que mon site investigateur participant à l'étude de recherche

clinique citée ci-dessus soit inclus dans l’essai OPTIMON visant à comparer deux

stratégies de monitorage.



coordination de l’étude et pour les investigateurs de mon site qui seront sollicités.



mes droits en tant qu’investigateur pour l’étude de recherche clinique citée ci-dessus.



l’essai OPTIMON.

Nom Prénom Date de l'accord Signature

......................................... └─┴─┘└─┴─┘└─┴─┴─┴─┘



Annexe 6. Décision du Comité de Protection des Personnes Sud Ouest Outre-Mer.



Annexe 7. Note d’information OPTIMON

NOTE D’INFORMATION OPTIMON Evaluation de l’efficacité et du coût de deux strat égies de monitorage

pour la recherche clinique institutionnelle étude OPTIMON : OPTImisation du MONitorage

Promoteur de l’étude : Centre Hospitalier Universitaire de BORDEAUX

Madame, Monsieur, Vous avez accepté de participer à l'étude que vous a proposée votre médecin. Cette étude est associée à une autre étude appelée OPTIMON. Vous trouverez ci-dessous des informations sur l'étude OPTIMON. N’hésitez pas à poser toutes les questions que vous jugerez utiles à votre médecin.

Optimon, c'est quoi ?

Dans Optimon, on cherche à déterminer la meilleure façon de gérer les études de recherche biomédicale. On va donc comparer deux méthodes de gestion des données :

Dans la première méthode de gestion des données, l'équipe qui gère l'étude de recherche biomédicale se rend très souvent dans l'hôpital où le patient est suivi. Elle contrôle toutes les données de toutes les façons possibles.

Dans la seconde méthode de gestion des données, l'équipe qui gère l'étude de recherche biomédicale se rend plus ou moins souvent dans l'hôpital selon que la recherche présente plus ou moins d'inconvénients pour le patient. Elle contrôle toujours beaucoup les données importantes, c'est-à-dire celles garantissant la sécurité du patient, le respect de la loi et la qualité des résultats de la recherche, mais elle contrôle moins les autres données.

Que se passera-t-il pour vous ?

Il n'y aura aucune conséquence pour vous, et vous ne vous rendrez compte de rien. Vous participerez à l'étude comme prévu. Vos données seront collectées par votre médecin et transmises à l'équipe qui gère l'étude . Vos données seront rentrées au fur et à mesure dans la base informatique, de façon anonyme. Elles seront vérifiées et corrigées si besoin, en utilisant l'une des deux méthodes de gestion des données.

Le moment venu, vos données seront transmises à l'équipe d’OPTIMON, de façon anonyme. L'équipe OPTIMON viendra alors dans l'hôpital vérifier si les données présentes dans la base informatique sont bien exactes. Puis l'équipe OPTIMON calculera la méthode de gestion des données qui a produit les données les plus exactes. Ce résultat indiquera comment mieux gérer les études de recherche biomédicales dans l'avenir.

Pourquoi vous ?

Une partie seulement des patients inclus dans l'étude dans votre hôpital entreront dans OPTIMON.Au total, les données de 1800 personnes, qui, comme vous, participent à une étude de recherche biomédicale dans un hôpital français, entreront dans OPTIMON.

Quels sont vos droits ?

A la fin d'OPTIMON, quand la comparaison des méthod es de gestion des données sera terminée, vous serez informé des résultats, par l'i ntermédiaire de votre médecin (loi n° 2004-806 du 9 août 2004 relative à la Politique de Santé Publique - art L1122-1 du Code de la Santé Publique). Vous disposez d’un droit d’accès, de rectification et d'opposition (loi n° 2004-801 du 6 août 2004 modifiant la loi n° 78-17 du 6 janvier 1978 relativ e à l'informatique, aux fichiers et aux libertés - article L1111-7 du Code de la Santé Publique). Cela signifie que vous pouvez refuser que vos données soient transmises à l'équipe OPTIMON, maintenant ou plus tard. A tout moment, vous pouvez demander à prendre connaissance de vos données entrées dans la base informatique OPTIMON, et vous pouvez demander leur correction si besoin, et même leur retrait. Pour cela, vous pouvez contacter votre médecin qui vous suit, qui vous a proposé d'entrer dans l'étude , et qui est le seul à connaître votre identité. Il servira d'intermédiaire avec l'équipe OPTIMON.

Après avoir lu cette note d’information, n’hésitez pas à poser à votre médecin toutes vos questions.



Annexe 8. Liste des Centres de Méthodologie et de G estion volontaires pour

participer à l'essai OPTIMON.

Liste mise à jour en février 2008.

Chef de service / Correspondant

(par ordre alphabétique après celui du coordonnateur)

Affiliation Ville

Roger Salamon / Geneviève Chêne CHU & INSERM U593 Bordeaux

Corinne Alberti CIE R. Debré Paris

Eric Bellissant CIC INSERM U0203 Rennes

Jacques Bénichou CHU & INSERM U657 Rouen

Fabrice Carrat INSERM U707 Paris

Gilles Chatellier URC & CIE HEGP Paris

Catherine Cornu CIC Lyon

Dominique Costagliola INSERM U720 Paris

Hervé Decousus CIC-EC Saint Etienne

Bruno Giraudeau CIC Tours

Francis Guillemin CIE Nancy

Alain Leizorovicz EA 3736, Faculté Laennec Lyon

Patrick Maison URC Henri Mondor Paris

Simone Mathoulon Institut Bergonié Bordeaux

Jean-Pierre Pignon CLCC Gustave Roussy Villejuif

Valérie Plättner DRC Lyon

Pierre Petit CIC Montpellier

Pierre-Marie Preux CHU Limoges

Olivier Rascol CIC Toulouse

Philippe Ravaud URC Bichat Paris

Jean-Marc Tréluyer URC Cochin Paris


© Bordeaux CHU OPTIMON, Comparison of monitoring strategies, finalised version for submission to the national PHRC, 29 April 2005 of 50

Annexe 9. Procédure d'anonymisation complète de l'i dentité du patient

• Effectuer successivement plusieurs tirages au sort sans remise de l'alphabet complet. Par exemple, 10 tirages successifs peuvent conduire à cette grille.

• Retenir les 2 premières lettres du nom et les 2 premières lettres du prénom du patient.

• Choisir un numéro de ligne. Il est conseillé, même si ce n'est pas indispensable, de varier le choix de la ligne à chaque patient.

• Repérer les 4 lettres retenues sur la première ligne. Faire correspondre les lettres dans l’ordre du nom et du prénom.

��

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

1 C J Q Y Z H V S U W M O D T X N F I A E G B K L R P

2 Q N B G V K P U X H W R Z M E I C L Y S O D J A T F

3 V G S M L C Z D B Q R U H A X P E K W I T Y F O N J

�� 4 K H L R I Y W Q S J U F A Z G O C B D P M V N E X T

5 E D C Z K B U M W R Q L V A T J G F H X P O S I N Y

6 C L M A D T G I Y O V X K B S Z J E W N Q F P U H R

7 A K J F S Z T C E D U Y O P G R M L I V X B H N W Q

8 D M P I Z L B V K S Q O T U J H R W G E C A Y F N X

9 W B T A S D Y U F E R Q I Z H V J N K G X P O L C M

10 A P Z N W C S G O B I Y D L M Q U R T J F K V X H E

DURAND PIERRE →→→→ D U P I →→→→ R M O S

evaluation of the efficacy and cost of two monitoring ... - protoc… · for public clinical...

Documents